Produced by Henry Flower, Adrian Mastronardi, Ramon Pajares
Box and the Online Distributed Proofreading Team at
http://www.pgdp.net (This file was produced from images
generously made available by The Internet Archive/American
Libraries.)






TRANSCRIBER'S NOTE

  * Italics are denoted by underscores as in _italics_.
  * Small caps are represented in upper case as in SMALL CAPS.
  * Letter spaced Greek text is enclosed in tildes as in ~καὶ τὰ
    λοιπά~.
  * Footnotes have been renumbered and placed at the end of the
    paragraph that includes their anchor.
  * Obvious printer errors have been silently corrected, after
    comparison with a later edition of this work. Greek text has
    also been corrected after checking with this later edition and
    with Perseus, when the reference was found.
  * Original spelling, hyphenation and punctuation have been kept,
    but variant spellings were made consistent when a predominant
    usage was found.
  * Nevertheless, inconsistent rendition of proper nouns (Acrisios
    and Akrisios, Mycênæ and Mykênæ, Athos and Athôs, Dædalus and
    Dædalos) have been kept, noting that this edition does not follow
    completely the rules for transcription of Greek names announced
    in the Preface, pp. xv-xvi.
  * In most cross-references, target pages do not correspond with
    pages in this edition.




[Illustration: GEORGE GROTE.]




  HISTORY OF GREECE.

  I. Legendary Greece.

  II. Grecian History to the Reign of
  Peisistratus at Athens.

  BY
  GEORGE GROTE, ESQ.

  VOL. I.

  REPRINTED FROM THE SECOND LONDON EDITION

  NEW YORK:
  HARPER & BROTHERS, PUBLISHERS,
  329 AND 331 PEARL STREET.
  1880.




PART I.—LEGENDARY GREECE

    Ἀνδρῶν ἡρώων θεῖον γένος, οἳ καλέονται
    Ἡμίθεοι προτἐρῃ γενέῃ.—HESIOD

PART II.—HISTORICAL GREECE.

    ... Πόλιες μερόπων ἀνθρώπων.—HOMER




PREFACE.


The first idea of this History was conceived many years ago, at a time
when ancient Hellas was known to the English public chiefly through
the pages of Mitford; and my purpose in writing it was to rectify
the erroneous statements as to matter of fact which that History
contained, as well as to present the general phenomena of the Grecian
world under what I thought a juster and more comprehensive point of
view. My leisure, however, was not at that time equal to the execution
of any large literary undertaking; nor is it until within the last
three or four years that I have been able to devote to the work that
continuous and exclusive labor, without which, though much may be done
to illustrate detached points, no entire or complicated subject can
ever be set forth in a manner worthy to meet the public eye.

Meanwhile the state of the English literary world, in reference to
ancient Hellas, has been materially changed in more ways than one. If
my early friend Dr. Thirlwall’s History of Greece had appeared a few
years sooner, I should probably never have conceived the design of the
present work at all; I should certainly not have been prompted to the
task by any deficiencies, such as those which I felt and regretted in
Mitford. The comparison of the two authors affords, indeed, a striking
proof of the progress of sound and enlarged views respecting the
ancient world during the present generation. Having studied of course
the same evidences as Dr. Thirwall, I am better enabled than others
to bear testimony to the learning, the sagacity, and the candor which
pervade his excellent work: and it is the more incumbent on me to give
expression to this sentiment, since the particular points on which I
shall have occasion to advert to it will, unavoidably, be points of
dissent oftener than of coincidence.

The liberal spirit of criticism, in which Dr. Thirwall stands so much
distinguished from Mitford, is his own: there are other features of
superiority which belong to him conjointly with his age. For during
the generation since Mitford’s work, philological studies have been
prosecuted in Germany with remarkable success: the stock of facts and
documents, comparatively scanty, handed down from the ancient world,
has been combined and illustrated in a thousand different ways: and
if our witnesses cannot be multiplied, we at least have numerous
interpreters to catch, repeat, amplify, and explain their broken
and half-inaudible depositions. Some of the best writers in this
department—Boeckh, Niebuhr, O. Müller—have been translated into our
language; so that the English public has been enabled to form some
idea of the new lights thrown upon many subjects of antiquity by the
inestimable aid of German erudition. The poets, historians, orators,
and philosophers of Greece, have thus been all rendered both more
intelligible and more instructive than they were to a student in the
last century; and the general picture of the Grecian world may now be
conceived with a degree of fidelity, which, considering our imperfect
materials, it is curious to contemplate.

It is that general picture which an historian of Greece is required
first to embody in his own mind, and next to lay out before his
readers;—a picture not merely such as to delight the imagination by
brilliancy of coloring and depth of sentiment, but also suggestive
and improving to the reason. Not omitting the points of resemblance
as well as of contrast with the better-known forms of modern society,
he will especially study to exhibit the spontaneous movement of
Grecian intellect, sometimes aided but never borrowed from without,
and lighting up a small portion of a world otherwise clouded and
stationary. He will develop the action of that social system, which,
while insuring to the mass of freemen a degree of protection elsewhere
unknown, acted as a stimulus to the creative impulses of genius, and
left the superior minds sufficiently unshackled to soar above religious
and political routine, to overshoot their own age, and to become the
teachers of posterity.

To set forth the history of a people by whom the first spark was set to
the dormant intellectual capacities of our nature,—Hellenic phenomena,
as illustrative of the Hellenic mind and character,—is the task
which I propose to myself in the present work; not without a painful
consciousness how much the deed falls short of the will, and a yet
more painful conviction, that full success is rendered impossible by
an obstacle which no human ability can now remedy,—the insufficiency
of original evidence. For, in spite of the valuable expositions of
so many able commentators, our stock of information respecting the
ancient world still remains lamentably inadequate to the demands of an
enlightened curiosity. We possess only what has drifted ashore from the
wreck of a stranded vessel; and though this includes some of the most
precious articles amongst its once abundant cargo, yet if any man will
cast his eyes over the citations in Diogenes Laërtius, Athenæus, or
Plutarch, or the list of names in Vossius de Historicis Græcis, he will
see with grief and surprise how much larger is the proportion which,
through the enslavement of the Greeks themselves, the decline of the
Roman Empire, the change of religion, and the irruption of barbarian
conquerors, has been irrecoverably submerged. We are thus reduced to
judge of he whole Hellenic world, eminently multiform as it was, from
a few compositions; excellent, indeed, in themselves, but bearing too
exclusively the stamp of Athens. Of Thucydides and Aristotle, indeed,
both as inquirers into matter of fact, and as free from narrow local
feeling, it is impossible to speak too highly; but, unfortunately,
that work of the latter which would have given us the most copious
information regarding Grecian political life—his collection and
comparison of one hundred and fifty distinct town constitutions—has
not been preserved: and the brevity of Thucydides often gives us but
a single word where a sentence would not have been too much, and
sentences which we should be glad to see expanded into paragraphs.

Such insufficiency of original and trustworthy materials, as
compared with those resources which are thought hardly sufficient
for the historian of any modern kingdom, is neither to be concealed
nor extenuated, however much we may lament it. I advert to the
point here on more grounds than one. For it not only limits the
amount of information which an historian of Greece can give to his
readers,—compelling him to leave much of his picture an absolute
blank,—but it also greatly spoils the execution of the remainder. The
question of credibility is perpetually obtruding itself, and requiring
a decision, which, whether favorable or unfavorable, always introduces
more or less of controversy; and gives to those outlines, which the
interest of the picture requires to be straight and vigorous, a faint
and faltering character. Expressions of qualified and hesitating
affirmation are repeated until the reader is sickened; while the writer
himself, to whom this restraint is more painful still, is frequently
tempted to break loose from the unseen spell by which a conscientious
criticism binds him down,—to screw up the possible and probable
into certainty, to suppress counterbalancing considerations, and to
substitute a pleasing romance in place of half-known and perplexing
realities. Desiring, in the present work, to set forth all which can be
ascertained, together with such conjectures and inferences as can be
reasonably deduced from it, but nothing more,—I notice, at the outset,
that faulty state of the original evidence which renders discussions of
credibility, and hesitation in the language of the judge, unavoidable.
Such discussions, though the reader may be assured that they will
become less frequent as we advance into times better known, are
tiresome enough, even with the comparatively late period which I adopt
as the historical beginning; much more intolerable would they have
proved, had I thought it my duty to start from the primitive terminus
of Deukaliôn or Inachus, or from the unburied Pelasgi and Leleges,
and to subject the heroic ages to a similar scrutiny. I really know
nothing so disheartening or unrequited as the elaborate balancing of
what is called evidence,—the comparison of infinitesimal probabilities
and conjectures all uncertified,—in regard to these shadowy times and
persons.

The law respecting sufficiency of evidence ought to be the same for
ancient times as for modern; and the reader will find in this History
an application, to the former, of criteria analogous to those which
have been long recognized in the latter. Approaching, though with a
certain measure of indulgence, to this standard, I begin the real
history of Greece with the first recorded Olympiad, or 776 B. C. To
such as are accustomed to the habits once universal, and still not
uncommon, in investigating the ancient world, I may appear to be
striking off one thousand years from the scroll of history; but to
those whose canon of evidence is derived from Mr. Hallam, M. Sismondi,
or any other eminent historian of modern events, I am well assured that
I shall appear lax and credulous rather than exigent or sceptical. For
the truth is, that historical records, properly so called, do not begin
until long after this date: nor will any man, who candidly considers
the extreme paucity of attested facts for two centuries after 776 B.
C., be astonished to learn that the state of Greece in 900, 1000, 1100,
1200, 1300, 1400 B. C., etc.,—or any earlier century which it may
please chronologists to include in their computed genealogies,—cannot
be described to him upon anything like decent evidence. I shall hope,
when I come to the lives of Socrates and Plato, to illustrate one of
the most valuable of their principles,—that conscious and confessed
ignorance is a better state of mind, than the fancy, without the
reality, of knowledge. Meanwhile, I begin by making that confession,
in reference to the real world of Greece anterior to the Olympiads;
meaning the disclaimer to apply to anything like a general history,—not
to exclude rigorously every individual event.

The times which I thus set apart from the region of history are
discernible only through a different atmosphere,—that of epic poetry
and legend. To confound together these disparate matters is, in my
judgment, essentially unphilosophical. I describe the earlier times by
themselves, as conceived by the faith and feeling of the first Greeks,
and known only through their legends,—without presuming to measure how
much or how little of historical matter these legends may contain. If
the reader blame me for not assisting him to determine this,—if he ask
me why I do not undraw the curtain and disclose the picture,—I reply in
the words of the painter Zeuxis, when the same question was addressed
to him on exhibiting his master-piece of imitative art: “The curtain is
the picture.” What we now read as poetry and legend was once accredited
history, and the only genuine history which the first Greeks could
conceive or relish of their past time: the curtain conceals nothing
behind, and cannot, by any ingenuity, be withdrawn. I undertake only to
show it as it stands,—not to efface, still less to repaint it.

Three-fourths of the two volumes now presented to the public are
destined to elucidate this age of historical faith, as distinguished
from the later age of historical reason: to exhibit its basis in the
human mind,—an omnipresent religious and personal interpretation of
nature; to illustrate it by comparison with the like mental habit
in early modern Europe; to show its immense abundance and variety
of narrative matter, with little care for consistency between one
story and another; lastly, to set forth the causes which overgrew and
partially supplanted the old epical sentiment, and introduced, in the
room of literal faith, a variety of compromises and interpretations.

The legendary age of the Greeks receives its principal charm and
dignity from the Homeric poems: to these, therefore, and to the other
poems included in the ancient epic, an entire chapter is devoted,
the length of which must be justified by the names of the Iliad and
Odyssey. I have thought it my duty to take some notice of the Wolfian
controversy as it now stands in Germany, and have even hazarded some
speculations respecting the structure of the Iliad. The society and
manners of the heroic age, considered as known in a general way from
Homer’s descriptions and allusions, are also described and criticized.

I next pass to the historical age, beginning at 776 B. C.; prefixing
some remarks upon the geographical features of Greece. I try to make
out, amidst obscure and scanty indications, what the state of Greece
was at this period; and I indulge some cautious conjectures, founded
upon the earliest verifiable facts, respecting the steps immediately
antecedent by which that condition was brought about. In the present
volumes, I have only been able to include the history of Sparta and the
Peloponnesian Dorians, down to the age of Peisistratus and Crœsus. I
had hoped to have comprised in them the entire history of Greece down
to this last-mentioned period, but I find the space insufficient.

The history of Greece falls most naturally into six compartments, of
which the first may be looked at as a period of preparation for the
five following, which exhaust the free life of collective Hellas.

I. Period from 776 B. C. to 560 B. C., the accession of Peisistratus at
Athens and of Crœsus in Lydia.

II. From the accession of Peisistratus and Crœsus to the repulse of
Xerxes from Greece.

III. From the repulse of Xerxes to the close of the Peloponnesian war
and overthrow of Athens.

IV. From the close of the Peloponnesian war to the battle of Leuktra.

V. From the battle of Leuktra to that of Chæroneia.

VI. From the battle of Chæroneia to the end of the generation of
Alexander.

The five periods, from Peisistratus down to the death of Alexander and
of his generation, present the acts of an historical drama capable
of being recounted in perspicuous succession, and connected by a
sensible thread of unity. I shall interweave in their proper places
the important but outlying adventures of the Sicilian and Italian
Greeks,—introducing such occasional notices of Grecian political
constitutions, philosophy, poetry, and oratory, as are requisite to
exhibit the many-sided activity of this people during their short but
brilliant career.

After the generation of Alexander, the political action of Greece
becomes cramped and degraded,—no longer interesting to the reader, or
operative on the destinies of the future world. We may, indeed, name
one or two incidents, especially the revolutions of Agis and Kleomenês
at Sparta, which are both instructive and affecting; but as a whole,
the period, between 300 B. C. and the absorption of Greece by the
Romans, is of no interest in itself, and is only so far of value as
it helps us to understand the preceding centuries. The dignity and
value of the Greeks from that time forward belong to them only as
individual philosophers, preceptors, astronomers, and mathematicians,
literary men and critics, medical practitioners, etc. In all these
respective capacities, especially in the great schools of philosophical
speculation they still constitute the light of the Roman world; though,
as communities, they have lost their own orbit, and have become
satellites of more powerful neighbors.

I propose to bring down the history of the Grecian communities to
the year 300 B. C., or the close of the generation which takes its
name from Alexander the Great, and I hope to accomplish this in eight
volumes altogether. For the next two or three volumes I have already
large preparations made, and I shall publish my third (perhaps my
fourth) in the course of the ensuing winter.

There are great disadvantages in the publication of one portion of
a history apart from the remainder; for neither the earlier nor the
later phenomena can be fully comprehended without the light which each
mutually casts upon the other. But the practice has become habitual,
and is indeed more than justified by the well-known inadmissibility
of “long hopes” into the short span of human life. Yet I cannot but
fear that my first two volumes will suffer in the estimation of many
readers by coming out alone,—and that men who value the Greeks for
their philosophy, their politics, and their oratory, may treat the
early legends as not worth attention. And it must be confessed that the
sentimental attributes of the Greek mind—its religious and poetical
vein—here appear in disproportionate relief, as compared with its
more vigorous and masculine capacities,—with those powers of acting,
organizing, judging, and speculating, which will be revealed in the
forthcoming volumes. I venture, however, to forewarn the reader, that
there will occur numerous circumstances in the after political life of
the Greeks, which he will not comprehend unless he be initiated into
the course of their legendary associations. He will not understand
the frantic terror of the Athenian public during the Peloponnesian
war, on the occasion of the mutilation of the statues called Hermæ,
unless he enters into the way in which they connected their stability
and security with the domiciliation of the gods in the soil: nor will
he adequately appreciate the habit of the Spartan king on military
expeditions,—when he offered his daily public sacrifices on behalf of
his army and his country,—“always to perform this morning service
immediately before sunrise, in order that he might be beforehand in
obtaining the favor of the gods,”[1] if he be not familiar with the
Homeric conception of Zeus going to rest at night and awaking to rise
at early dawn from the side of the “white-armed Hêrê.” The occasion
will, indeed, often occur for remarking how these legends illustrate
and vivify the political phenomena of the succeeding times, and I have
only now to urge the necessity of considering them as the beginning of
a series,—not as an entire work.

  [1] Xenophon, Repub. Lacedæmon. cap. xiii. 3. Ἀεὶ δὲ, ὅταν
  θύηται, ἄρχεται μὲν τούτου τοῦ ἔργου ἔτι κνεφαῖος, προλαμβάνειν
  βουλόμενος τὴν τοῦ θεοῦ εὔνοιαν.


  LONDON, March 5, 1846.




PREFACE TO THE SECOND EDITION OF VOLUMES I. AND II.


In preparing a Second Edition of the first two volumes of my History,
I have profited by the remarks and corrections of various critics,
contained in Reviews, both English and foreign. I have suppressed, or
rectified, some positions which had been pointed out as erroneous,
or as advanced upon inadequate evidence. I have strengthened my
argument in some cases where it appeared to have been imperfectly
understood,—adding some new notes, partly for the purpose of enlarged
illustration, partly to defend certain opinions which had been called
in question. The greater number of these alterations have been made in
Chapters XVI. and XXI. of Part I., and in Chapter VI. of Part II.

I trust that these three Chapters, more full of speculation, and
therefore more open to criticism than any of the others, will thus
appear in a more complete and satisfactory form. But I must at the same
time add that they remain for the most part unchanged in substance, and
that I have seen no sufficient reason to modify my main conclusions
even respecting the structure of the Iliad, controverted though they
have been by some of my most esteemed critics.

In regard to the character and peculiarity of Grecian legend, as
broadly distinguished throughout these volumes from Grecian history,
I desire to notice two valuable publications with which I have
only become acquainted since the date of my first edition. One of
these is, A Short Essay on Primæval History, by John Kenrick, M. A.
(London, 1846, published just at the same time as these volumes,)
which illustrates with much acute reflection the general features
of legend, not only in Greece but throughout the ancient world,—see
especially pages 65, 84, 92, _et seq._ The other work is, Rambles and
Recollections of an Indian Official, by Colonel Sleeman,—first made
known to me through an excellent notice of my History in the Edinburgh
Review for October 1846. The description given by Colonel Sleeman, of
the state of mind now actually prevalent among the native population of
Hindostan, presents a vivid comparison, helping the modern reader to
understand and appreciate the legendary era of Greece. I have embodied
in the notes of this Second Edition two or three passages from Colonel
Sleeman’s instructive work: but the whole of it richly deserves perusal.

Having now finished six volumes of this History, without attaining
a lower point than the peace of Nikias, in the tenth year of the
Peloponnesian war,—I find myself compelled to retract the expectation
held out in the preface to my First Edition, that the entire work might
be completed in eight volumes. Experience proves to me how impossible
it is to measure beforehand the space which historical subjects will
require. All I can now promise is, that the remainder of the work shall
be executed with as much regard to brevity as is consistent with the
paramount duty of rendering it fit for public acceptance.


  London, April 3, 1849.




NAMES OF GODS, GODDESSES, AND HEROES.


Following the example of Dr. Thirlwall and other excellent scholars, I
call the Greek deities by their real Greek names, and not by the Latin
equivalents used among the Romans. For the assistance of those readers
to whom the Greek names may be less familiar, I here annex a table of
the one and the other.

  _Greek._               _Latin._

  Zeus,                   Jupiter.
  Poseidôn,               Neptune.
  Arês,                   Mars.
  Dionysus,               Bacchus.
  Hermês,                 Mercury.
  Hêlios,                 Sol.
  Hêphæstus,              Vulcan.
  Hadês,                  Pluto.

  Hêrê,                   Juno.
  Athênê,                 Minerva.
  Artemis,                Diana.
  Aphroditê,              Venus.
  Eôs,                    Aurora.
  Hestia,                 Vesta.
  Lêtô,                   Latona.
  Dêmêtêr,                Ceres.

  Hêraklês,               Hercules.
  Asklêpius,              Æsculapius.

A few words are here necessary respecting the orthography of Greek
names adopted in the above table and generally throughout this history.
I have approximated as nearly as I dared to the Greek letters in
preference to the Latin; and on this point I venture upon an innovation
which I should have little doubt of vindicating before the reason of
any candid English student. For the ordinary practice of substituting,
in a Greek name, the English C in place of the Greek K, is, indeed, so
obviously incorrect, that it admits of no rational justification. Our
own K, precisely and in every point, coincides with the Greek K: we
have thus the means of reproducing the Greek name to the eye as well
as to the ear, yet we gratuitously take the wrong letter in preference
to the right. And the precedent of the Latins is here against us
rather than in our favor, for their C really coincided in sound with
the Greek K, whereas our C entirely departs from it, and becomes an S,
before _e_, _i_, _æ_, _œ_, and _y_. Though our C has so far deviated in
sound from the Latin C, yet there is some warrant for our continuing
to use it in writing Latin names,—because we thus reproduce the name
to the eye, though not to the ear. But this is not the case when we
employ our C to designate the Greek K, for we depart here not less
from the visible than from the audible original; while we mar the
unrivalled euphony of the Greek language by that multiplied sibilation
which constitutes the least inviting feature in our own. Among German
philologists, the K is now universally employed in writing Greek names,
and I have adopted it pretty largely in this work, making exception for
such names as the English reader has been so accustomed to hear with
the C, that they may be considered as being almost Anglicised. I have,
farther, marked the long _e_ and the long _o_ (η, ω,) by a circumflex
(Hêrê) when they occur in the last syllable or in the penultimate of a
name.




CONTENTS

VOL. I.


PART I.

LEGENDARY GREECE.




CHAPTER I.

LEGENDS RESPECTING THE GODS.

  Opening of the mythical world.—How the mythes are to be
  told.—Allegory rarely admissible.—Zeus—foremost in Grecian
  conception.—The gods—how conceived: human type enlarged.—Past
  history of the gods fitted on to present conceptions.—Chaos.—Gæa
  and Uranos.—Uranos disabled.—Kronos and the Titans.—Kronos
  overreached.—Birth and safety of Zeus and his brethren.—Other
  deities.—Ambitious schemes of Zeus.—Victory of Zeus and
  his brethren over Kronos and the Titans.—Typhôeus.—Dynasty
  of Zeus.—His offspring.—General distribution of the divine
  race.—Hesiodic theogony—its authority.—Points of difference
  between Homer and Hesiod.—Homeric Zeus.—Amplified theogony of
  Zens.—Hesiodic mythes traceable to Krête and Delphi.—Orphic
  theogony.—Zeus and Phanês.—Zagreus.—Comparison of Hesiod and
  Orpheus.—Influence of foreign religions upon Greece—Especially
  in regard to the worship of Dêmêtêr and Dionysos.—Purification
  for homicide unknown to Homer.—New and peculiar religious
  rites.—Circulated by voluntary teachers and promising special
  blessings.—Epimenidês, Sibylla, Bakis.—Principal mysteries
  of Greece.—Ecstatic rites introduced from Asia 700-500 B.
  C.—Connected with the worship of Dionysos.—Thracian and Egyptian
  influence upon Greece.—Encouragement to mystic legends.—Melampus
  the earliest name as teacher of the Dionysiac rites.—Orphic
  sect, a variety of the Dionysiac mystics.—Contrast of the
  mysteries with the Homeric Hymns.—Hymn to Dionysos.—Alteration
  of the primitive Grecian idea of Dionysos.—Asiatic frenzy
  grafted on the joviality of the Grecian Dionysia.—Eleusinian
  mysteries.—Homeric Hymn to Dêmêtêr.—Temple of Eleusis, built
  by order of Dêmêtêr for her residence.—Dêmêtêr prescribes the
  mystic ritual of Eleusis.—Homeric Hymn a sacred Eleusinian
  record, explanatory of the details of divine service.—Importance
  of the mysteries to the town of Eleusis.—Strong hold of the
  legend upon Eleusinian feelings.—Different legends respecting
  Dêmêtêr elsewhere.—Expansion of the legends.—Hellenic importance
  of Dêmêtêr.—Legends of Apollo.—Delian Apollo.—Pythian
  Apollo.—Foundation legends of the Delphian oracle.—They served
  the purpose of historical explanation.—Extended worship of
  Apollo.—Multifarious local legends respecting Apollo.—Festivals
  and Agônes.—State of mind and circumstances out of which
  Grecian mythes arose.—Discrepancies in the legends little
  noticed.—Aphroditê.—Athênê.—Artemis.—Poseidôn.—Stories
  of temporary servitude imposed on
  gods.—Hêrê.—Hêphæstos.—Hestia.—Hermês.—Hermês inventor of the
  lyre.—Bargain between Hermês and Apollo.—Expository value of the
  Hymn.—Zeus.—Mythes arising out of the religious ceremonies.—Small
  part of the animal sacrificed.—Promêtheus had outwitted
  Zeus.—Gods, heroes, and men, appear together in the mythes.
                                                            _pages_ 1-64


CHAPTER II.

LEGENDS RELATING TO HEROES AND MEN.

  Races of men as they appear in the Hesiodic “Works and Days.”—The
  Golden.—The Silver.—The Brazen.—The Heroic.—The Iron.—Different
  both from the Theogony and from Homer.—Explanation of this
  difference.—Ethical vein of sentiment.—Intersected by the
  mythical.—The “Works and Days,” earliest didactic poem.—First
  Introduction of dæmons.—Changes in the idea of dæmons.—Employed
  in attacks on the pagan faith.—Functions of the Hesiodic
  dæmons.—Personal feeling which pervades the “Works and
  Days.”—Probable age of the poem.                                 64-73


CHAPTER III.

LEGEND OF THE IAPETIDS.

  Iapetids in Hesiod.—Promêtheus and Epimêtheus.—Counter-manœuvring
  of Promêtheus and Zeus.—Pandôra.—Pandôra in the
  Theogony.—General feeling of the poet.—Man wretched, but
  Zeus not to blame.—Mischiefs arising from women.—Punishment
  of Promêtheus.—The Promêtheus of Æschylus.—Locality in which
  Promêtheus was confined.                                         73-80


CHAPTER IV.

HEROIC LEGENDS.—GENEALOGY OF ARGUS.

  Structure and purposes of Grecian genealogies.—To connect the
  Grecian community with their common god.—Lower members of
  the genealogy historical—higher members non-historical.—The
  non-historical portion equally believed, and most valued by the
  Greeks.—Number of such genealogies—pervading every fraction
  of Greeks.—Argeian genealogy.—Inachus.—Phorôneus.—Argos
  Panoptês.—Iô.—Romance of Iô historicized by Persians and
  Phœnicians.—Legendary abductions of heroines adapted to
  the feelings prevalent during the Persian war.—Danaos and
  the Danaïdes.—Acrisios and Prœtos.—The Prœtides cured of
  frenzy by Melampus.—Acrisios, Danaê, and Zeus.—Perseus and
  the Gorgons.—Foundation of Mycênæ—commencement of Perseid
  dynasty.—Amphitryôn, Alkmênê, Sthenelos.—Zeus and Alkmênê.—Birth
  of Hêraklês.—Homeric legend of his birth: its expository
  value.—The Hêrakleids expelled.—Their recovery of Peloponnêsus
  and establishment in Argos, Sparta, and Messênia.                80-95


CHAPTER V.

DEUKALION, HELLEN, AND SONS OF HELLEN.

  Deukaliôn, son of Promêtheus.—Phthiôtis: his permanent
  seat.—General deluge.—Salvation of Deukaliôn and Pyrrha.—Belief
  in this deluge throughout Greece.—Hellên and Amphiktyôn.—Sons
  of Hellên: Dôrus, Xuthus, Æolus.—Amphiktyonic assembly.—Common
  solemnities and games.—Division of Hellas: Æolians, Dôrians,
  Iônians.—Large extent of Dôris implied in this genealogy.—This
  form of the legend harmonizes with the great establishments of
  the historical Dôrians.—Achæus—purpose which his name serves in
  the legend.—Genealogical diversities.                           96-105


CHAPTER VI.

THE ÆOLIDS, OR SONS AND DAUGHTERS OF ÆOLUS.

  Legends of Greece, originally isolated, afterwards
  thrown into series.—_Æolus._—His seven sons and five
  daughters.—1. _First Æolid line_—Salmôneus, Tyrô.—Pelias
  and Nêleus.—Pêrô, Bias, and Melampus.—Periklymenos.—Nestor
  and his exploits.—Nêleids down to Kodrus.—_Second Æolid
  line_—Krêtheus.—Admêtus and Alcêstis.—Pêleus and the wife
  of Acastus.—Pelias and Jasôn.—Jasôn and Mêdea.—Mêdea at
  Corinth.—_Third Æolid line_—Sisyphus.—Corinthian genealogy
  of Eumêlus.—Coalescence of different legends about Mêdea and
  Sisyphus.—Bellerophôn.—_Fourth Æolid line_—Athamas.—Phryxus
  and Hellê.—Inô and Palæmôn—Isthmian games.—Local root of the
  legend of Athamas.—Traces of ancient human sacrifices.—Athamas
  in the district near Orchomenos.—Eteoklês—festival of the
  Charitêsia.—Foundation and greatness of Orchomenos.—Overthrow by
  Hêraklês and the Thebans.—Trophônius and Agamêdês.—Ascalaphos and
  Ialmenos.—Discrepancies in the Orchomenian genealogy.—Probable
  inferences as to the ante-historical Orchomenos.—Its early wealth
  and industry.—Emissaries of the lake Kôpaïs.—Old Amphiktyony at
  Kalauria.—Orchomenos and Thebês.—Alcyonê and Kêyx.—Canacê.—The
  Alôids.—Calycê.—Elis and Ætôlia.—Eleian genealogy.—Augeas.—The
  Molionid brothers.—Variations in the Eleian genealogy.—Ætôlian
  genealogy.—Œneus, Meleager, Tydeus.—Legend of Meleager in
  Homer.—How altered by the poets after Homer.—Althæa and the
  burning brand.—Grand Kalydônian boar-hunt.—Atalanta.—Relics of
  the boar long preserved at Tegea.—Atalanta vanquished in the race
  by stratagem.—Deianeira.—Death of Hêraklês.—Tydeus.—Old age of
  Œneus.—Discrepant genealogies.                                 105-153


CHAPTER VII.

THE PELOPIDS.

  Misfortunes and celebrity of the Pelopids.—Pelops—eponym of
  Peloponnêsus.—Deduction of the sceptre of Pelops.—Kingly
  attributes of the family.—Homeric Pelops.—Lydia, Pisa, etc.,
  post-Homeric additions.—Tantalus.—Niobê.—Pelops and Œnomaus,
  king of Pisa.—Chariot victory of Pelops—his principality at
  Pisa.—Atreus, Thyestês, Chrysippus.—Family horrors among the
  Pelopids.—Agamemnôn and Menelaus.—Orestês.—The goddess Hêrê and
  Mykênæ.—Legendary importance of Mykênæ.—Its decline coincident
  with the rise of Argos and Sparta.—Agamemnôn and Orestês
  transferred to Sparta.                                         153-167


CHAPTER VIII.

LACONIAN AND MESSENIAN GENEALOGIES.

  Lelex—autochthonous in Lacônia.—Tyndareus and Lêda.—Offspring
  of Lêda.—1. Castôr, Timandra, Klytæmnêstra, 2. Pollux,
  Helen.—Castôr and Pollux.—Legend of the Attic Dekeleia.—Idas and
  Lynkeus.—Great functions and power of the Dioskuri.—Messênian
  genealogy.—Periêrês—Idas and Marpêssa.                         168-173


CHAPTER IX.

ARCADIAN GENEALOGY.

  Pelasgus.—Lykaôn and his fifty sons.—Legend of Lykaôn—ferocity
  punished by the gods.—Deep religious faith of Pausanias.—His view
  of past and present world.—Kallistô and Arkas.—Azan, Apheidas,
  Elatus.—Aleus, Augê, Telephus.—Ancæus.—Echemus.—Echemus
  kills Hyllus.—Hêrakleids repelled from Peloponnêsus.—Korônis and
  Asklêpius.—Extended worship of Asklêpius—numerous legends.—Machaôn
  and Podaleirius.—Numerous Asklêpiads, or descendants from
  Asklêpius.—Temples of Asklêpius—sick persons healed there.     173-183


CHAPTER X

ÆAKUS AND HIS DESCENDANTS.—ÆGINA, SALAMIS, AND PHTHIA.

  Æakus—son of Zeus and Ægina.—Offspring of Æakus—Pêleus,
  Telamôn, Phôkus.—Prayers of Æakus—procure relief for
  Greece—Phôkus killed by Pêleus and Telamôn.—Telamôn, banished,
  goes to Salamis.—Pêleus—goes to Phthia—his marriage with
  Thetis.—Neoptolemus.—Ajax, his son Philæus the eponymous hero of
  a dême in Attica.—Teukrus banished, settles in Cyprus.—Diffusion
  of the Æakid genealogy.                                        184-190


CHAPTER XI.

ATTIC LEGENDS AND GENEALOGIES.

  Erechtheus—autochthonous.—Attic legends—originally from
  different roots—each dême had its own.—Little noticed by
  the old epic poets.—Kekrops.—Kranaus—Pandiôn.—Daughters of
  Pandiôn—Proknê, Philomêla.—Legend of Têreus.—Daughters
  of Erechtheus—Prokris.—Kreüsa.—Oreithyia, the wife of
  Boreas.—Prayers of the Athenians to Boreas—his gracious
  help in their danger.—Erechtheus and Eumolpus.—Voluntary
  self-sacrifice of the three daughters of Erechtheus.—Kreüsa
  and Iôn.—Sons of Pandiôn—Ægeus, etc.—Thêseus.—His legendary
  character refined.—Plutarch—his way of handling the matter
  of legend.—Legend of the Amazons.—Its antiquity and
  prevalence.—Glorious achievements of the Amazons.—Their
  ubiquity.—Universally received as a portion of the Greek
  past.—Amazons produced as present by the historians of
  Alexander.—Conflict of faith and reason in the historical critics.
                                                                 191-217

CHAPTER XII.

KRETAN LEGENDS.—MINOS AND HIS FAMILY.

  Minôs and Rhadamanthus, sons of Zeus.—Europê.—Pasiphaê
  and the Minôtaur.—Scylla and Nisus.—Death of Androgeos,
  and anger of Minôs against Athens.—Athenian victims for
  the Minôtaur.—Self-devotion of Thêseus—he kills the
  Minôtaur.—Athenian commemorative ceremonies.—Family of
  Minôs.—Minôs and Dædalus—flight of the latter to Sicily.—Minôs
  goes to retake him, but is killed.—Semi-Krêtan settlements
  elsewhere—connected with this voyage of Minôs.—Sufferings of the
  Krêtans afterwards from the wrath of Minôs.—Portrait of Minôs—how
  varied.—Affinity between Krête and Asia Minor.                 218-230


CHAPTER XIII.

ARGONAUTIC EXPEDITION.

  Ship Argô in the Odyssey.—In Hesiod and Eumêlus.—Jasôn and
  his heroic companions.—Lêmnos.—Adventures at Kyzikus, in
  Bithynia, etc.—Hêraklês and Hylas.—Phineus.—Dangers of the
  Symplêgades.—Arrival at Kolchis.—Conditions imposed by Æêtês
  as the price of the golden fleece.—Perfidy of Æêtês—flight of
  the Argonauts and Mêdea with the fleece.—Pursuit of Æêtês—the
  Argonauts saved by Mêdea.—Return of the Argonauts—circuitous and
  perilous.—Numerous and wide-spread monuments referring to the
  voyage.—Argonautic legend generally.—Fabulous geography—gradually
  modified as real geographical knowledge increased.—Transposition
  of epical localities.—How and when the Argonautic voyage
  became attached to Kolchis.—Æêtês and Circê.—Return of the
  Argonauts—different versions.—Continued faith in the voyage—basis
  of truth determined by Strabo.                                 231-256


CHAPTER XIV.

LEGENDS OF THEBES.

  Abundant legends of Thêbes.—Amphiôn and Zethus, Homeric founders
  of Kadmus and Bœôtus—both distinct legends.—Thêbes.—How Thêbes
  was founded by Kadmus.—Five primitive families at Thêbes called
  Sparti.—The four daughters of Kadmus: 1. Inô; 2. Semelê; 3.
  Autonoê and her son Actæôn; 4. Agavê and her son Pentheus.—He
  resists the god Dionysus—his miserable end.—Labdakus, Antiopê,
  Amphiôn, and Zêthus.—Laius—Œdipus—Legendary celebrity of Œdipus
  and his family.—The Sphinx.—Eteoklês and Polynikês.—Old epic
  poems on the sieges of Thêbes.                                 256-269


SIEGES OF THEBES.

  Curse pronounced by the devoted Oedipus upon his sons.—Novelties
  introduced by Sophoklês.—Death of Oedipus—quarrel of Eteoklês and
  Polynikês for the sceptre.—Polynikês retires to Argos—aid given
  to him by Adrastus.—Amphiaräus and Eriphylê.—Seven chiefs of the
  army against Thêbes.—Defeat of the Thêbans in the field—heroic
  devotion of Menœkus.—Single combat of Eteoklês and Polynikês,
  in which both perish.—Repulse and destruction of the Argeian
  chiefs—all except Adrastus—Amphiaräus is swallowed up in the
  earth.—Kreôn, king of Thêbes, forbids the burial of Polynikês
  and the other fallen Argeian chiefs.—Devotion and death of
  Antigonê.—The Athenians interfere to procure the interment of
  the fallen chiefs.—Second siege of Thêbes by Adrastus with
  the Epigoni, or sons of those slain in the first.—Victory
  of the Epigoni—capture of Thêbes.—Worship of Adrastus at
  Sikyôn—how abrogated by Kleisthenês.—Alkmæôn—his matricide and
  punishment.—Fatal necklace of Eriphylê.                        269-284


CHAPTER XV.

LEGEND OF TROY.

  Great extent and variety of the tale of Troy.—Dardanus,
  son of Zeus.—Ilus, founder of Ilium.—Walls of Ilium
  built by Poseidôn.—Capture of Ilium by Hêraklês.—Priam
  and his offspring.—Paris—his judgment on the three
  goddesses.—Carries off Helen from Sparta.—Expedition of
  the Greeks to recover her.—Heroes from all parts of Greece
  combined under Agamemnôn.—Achilles and Odysseus.—The Grecian
  host mistakes Teuthrania for Troy—Telephus.—Detention
  of the Greeks at Aulis—Agamemnon and Iphigeneia.—First
  success of the Greeks on landing near Troy.—Brisêis
  awarded to Achilles.—Palamêdês—his genius, and treacherous
  death.—Epic chronology—historicized.—Period of the
  Homeric Iliad.—Hectôr killed by Achilles.—New allies of
  Troy—Penthesileia.—Memnôn—killed by Achilles.—Death of
  Achilles.—Funeral games celebrated in honor of him.—Quarrel
  about his panoply.—Odysseus prevails and Ajax kills
  himself.—Philoktêtês and Neoptolemus.—Capture of the
  Palladium.—The wooden horse.—Destruction of Troy.—Distribution of
  the captives among the victors.—Helen restored to Menelaus—lives
  in dignity at Sparta—passes to a happy immortality.—Blindness
  and cure of the poet Stesichorus—alteration of the legend about
  Helen.—Egyptian tale about Helen—tendency to historicize.—Return
  of the Greeks from Troy.—Their sufferings—anger of the
  gods.—Wanderings of the heroes in all directions.—Memorials of
  them throughout the Grecian world.—Odysseus—his final adventures
  and death.—Æneas and his descendants.—Different stories about
  Æneas.—Æneadæ at Skêpsis.—Ubiquity of Æneas.—Antenôr.—Tale of
  Troy—its magnitude and discrepancies.—Trojan war—essentially
  legendary—its importance as an item in Grecian national
  faith.—Basis of history for it—possible, and nothing
  more.—Historicizing innovations—Dio Chrysostom.—Historical
  Ilium.—Generally received and visited as the town of
  Priam.—Respect shown to it by Alexander.—Successors of
  Alexander—foundation of Alexandreia Trôas.—The Romans—treat Ilium
  with marked respect.—Mythical legitimacy of Ilium—first called
  in question by Dêmêtrius of Skêpsis and Hestiæa.—Supposed Old
  Ilium, or real Troy, distinguished from New Ilium.—Strabo alone
  believes in Old Ilium as the real Troy—other authors continue in
  the old faith—the moderns follow Strabo.—The mythical faith not
  shaken by topographical impossibilities.—Historical Trôas and
  the Teukrians.—Æolic Greeks in the Trôad—the whole territory
  gradually Æolized.—Old date, and long prevalence of the worship
  of Apollo Sminthius.—Asiatic customs and religion—blended with
  Hellenic.—Sibylline prophecies.—Settlements from Milêtus,
  Mitylênê, and Athens.                                          284-340


CHAPTER XVI.

GRECIAN MYTHES, AS UNDERSTOOD, FELT, AND INTERPRETED BY THE GREEKS
THEMSELVES.

  The mythes formed the entire mental stock of the early
  Greeks.—State of mind out of which they arose.—Tendency to
  universal personification.—Absence of positive knowledge—supplied
  by personifying faith.—Multitude and variety of quasi-human
  personages.—What we read as poetical fancies, were to the Greeks
  serious realities.—The gods and heroes—their chief agency cast
  back into the past, and embodied in the mythes.—Marked and
  manifold types of the Homeric gods.—Stimulus which they afforded
  to the mythopœic faculty.—Easy faith in popular and plausible
  stories.—Poets—receive their matter from the divine inspiration
  of the Muse.—Meaning of the word _mythe_—original—altered.—Matter
  of actual history—uninteresting to early Greeks.—Mythical
  faith and religious point of view—paramount in the Homeric
  age.—Gradual development of the scientific point of view—its
  opposition to the religious.—Mythopœic age—anterior to this
  dissent.—Expansive force of Grecian intellect.—Transition
  towards positive and present fact.—The poet becomes the organ
  of present time instead of past.—Iambic, elegiac, and lyric
  poets.—Influence of the opening of Egypt to Grecian commerce,
  B. C. 660.—Progress—historical, geographical, social—from that
  period to B. C. 500.—Altered standard of judgment, ethical
  and intellectual.—Commencement of physical science—Thalês,
  Xenophanês, Pythagoras.—Impersonal nature conceived as an
  object of study.—Opposition between scientific method and the
  religious feeling of the multitude.—How dealt with by different
  philosophers.—Socratês.—Hippocratês.—Anaxagoras.—Contrasted
  with Grecian religious belief.—Treatment of Socratês by
  the Athenians.—Scission between the superior men and the
  multitude—important in reference to the mythes.—The mythes
  accommodated to a new tone of feeling and judgment.—The
  poets and logographers.—Pindar.—Tragic poets.—Æschylus and
  Sophoklês.—Tendencies of Æschylus in regard to the old
  legends.—He maintains undiminished the grandeur of the
  mythical world.—Euripidês—accused of vulgarizing the mythical
  heroes, and of introducing exaggerated pathos, refinement,
  and rhetoric.—The logographers—Pherekydês, etc.—Hekatæus—the
  mythes rationalized.—The historians—Herodotus.—Earnest piety
  of Herodotus—his mystic reserve.—His views of the mythical
  world.—His deference for Egypt and Egyptian statements.—His
  general faith in the mythical heroes and eponyms—yet combined
  with scepticism as to matters of fact—His remarks upon the
  miraculous foundation of the oracle at Dôdôna.—His remarks
  upon Melampus and his prophetic powers.—His remarks upon
  the Thessalian legend of Tempê.—Allegorical interpretation
  of the mythes—more and more esteemed and applied.—Divine
  legends allegorized.—Heroic legends historicized.—Limits
  to this interpreting process.—Distinction between gods and
  dæmons—altered and widened by Empedoclês.—Admission of dæmons as
  partially evil beings—effect of such admission.—Semi-historical
  interpretation—utmost which it can accomplish.—Some positive
  certificate indispensable as a constituent of historical
  proof—mere popular faith insufficient.—Mistake of ascribing to an
  unrecording age the historical sense of modern times.—Matter of
  tradition uncertified from the beginning.—Fictitious matter of
  tradition does not imply fraud or imposture.—Plausible fiction
  often generated and accredited by the mere force of strong and
  common sentiment, even in times of instruction.—Allegorical
  theory of the mythes—traced by some up to an ancient priestly
  caste.—Real import of the mythes supposed to be preserved in
  the religious mysteries.—Supposed ancient meaning is really
  a modern interpretation.—Triple theology of the pagan world.
  Treatment and use of the mythes according to Plato.—His views
  as to the necessity and use of fiction.—He deals with the
  mythes as expressions of feeling and imagination—sustained
  by religious faith, and not by any positive basis.—Grecian
  antiquity essentially a religious conception.—Application of
  chronological calculation divests it of this character.—Mythical
  genealogies all of one class, and all on a level in respect
  to evidence.—Grecian and Egyptian genealogies.—Value of
  each is purely subjective, having especial reference to
  the faith of the people.—Gods and men undistinguishable in
  Grecian antiquity.—General recapitulation.—General public
  of Greece—familiar with their local mythes, careless of
  recent history.—Religious festivals—their commemorative
  influence.—Variety and universality of mythical relics.—The
  mythes in their bearing on Grecian art.—Tendency of works of art
  to intensify the mythical faith.                               340-461


CHAPTER XVII.

THE GRECIAN MYTHICAL VEIN COMPARED WITH THAT OF MODERN EUROPE.

  Μῦθος—_Sage_—an universal manifestation of the human
  mind.—Analogy of the Germans and Celts with the
  Greeks.—Differences between them.—Grecian poetry
  matchless.—Grecian progress self-operated.—German progress
  brought about by violent influences from without.—Operation of
  the Roman civilization and of Christianity upon the primitive
  German mythes.—Alteration in the mythical genealogies—Odin and
  the other gods degraded into men.—Grecian Paganism—what would
  have been the case, if it had been supplanted by Christianity
  in 500 B. C.—Saxo Grammaticus and Snorro Sturleson contrasted
  with Pherekydês and Hellanikus.—Mythopœic tendencies in modern
  Europe still subsisting, but forced into a new channel: 1.
  Saintly ideal; 2. Chivalrous ideal.—Legends of the Saints—their
  analogy with the Homeric theology.—Chivalrous ideal—Romances
  of Charlemagne and Arthur.—Accepted as realities of the
  fore-time.—Teutonic and Scandinavian epic—its analogy with the
  Grecian.—Heroic character and self-expanding subject common to
  both.—Points of distinction between the two—epic of the Middle
  Ages neither stood so completely alone, nor was so closely
  interwoven with religion, as the Grecian.—History of England—how
  conceived down to the seventeenth century—began with Brute the
  Trojan.—Earnest and tenacious faith manifested in the defence of
  this early history.—Judgment of Milton.—Standard of historical
  evidence—raised in regard to England—not raised in regard to
  Greece.—Milton’s way of dealing with the British fabulous history
  objectionable.—Two ways open of dealing with the Grecian mythes:
  1, to omit them; or, 2, to recount them as mythes.—Reasons for
  preferring the latter.—Triple partition of past time by Varro.
                                                                 461-489




HISTORY OF GREECE.




PART I.

LEGENDARY GREECE.




CHAPTER I.

LEGENDS RESPECTING THE GODS.


The mythical world of the Greeks opens with the gods, anterior as well
as superior to man: it gradually descends, first to heroes, and next
to the human race. Along with the gods are found various monstrous
natures, ultra-human and extra-human, who cannot with propriety be
called gods, but who partake with gods and men in the attributes
of free-will, conscious agency, and susceptibility of pleasure and
pain,—such as the Harpies, the Gorgons, the Grææ, the Sirens, Scylla
and Charybdis, Echidna, Sphinx, Chimæra, Chrysaor, Pegasus, the
Cyclôpes, the Centaurs, etc. The first acts of what may be termed
the great mythical cycle describe the proceedings of these gigantic
agents—the crash and collision of certain terrific and overboiling
forces, which are ultimately reduced to obedience, or chained up, or
extinguished, under the more orderly government of Zeus, who supplants
his less capable predecessors, and acquires precedence and supremacy
over gods and men—subject, however to certain social restraints from
the chief gods and goddesses around him, as well as to the custom of
occasionally convoking and consulting the divine agora.

I recount these events briefly, but literally, treating them simply
as mythes springing from the same creative imagination, addressing
themselves to analogous tastes and feelings, and depending upon the
same authority, as the legends of Thebes and Troy. It is the inspired
voice of the Muse which reveals and authenticates both, and from which
Homer and Hesiod alike derive their knowledge—the one, of the heroic,
the other, of the divine, foretime. I maintain, moreover, fully, the
character of these great divine agents as Persons, which is the light
in which they presented themselves to the Homeric or Hesiodic audience.
Uranos, Nyx, Hypnos and Oneiros (Heaven, Night, Sleep and Dream), are
Persons, just as much as Zeus and Apollo. To resolve them into mere
allegories, is unsafe and unprofitable: we then depart from the point
of view of the original hearers, without acquiring any consistent or
philosophical point of view of our own.[2] For although some of the
attributes and actions ascribed to these persons are often explicable
by allegory the whole series and system of them never are so: the
theorist who adopts this course of explanation finds that, after one
or two simple and obvious steps, the path is no longer open, and he
is forced to clear a way for himself by gratuitous refinements and
conjectures. The allegorical persons and attributes are always found
mingled with other persons and attributes not allegorical; but the
two classes cannot be severed without breaking up the whole march of
the mythical events, nor can any explanation which drives us to such
a necessity be considered as admissible. To suppose indeed that these
legends could be all traced by means of allegory into a coherent
body of physical doctrine, would be inconsistent with all reasonable
presumptions respecting the age or society in which they arose. Where
the allegorical mark is clearly set upon any particular character,
or attribute, or event, to that extent we may recognize it; but we
can rarely venture to divine further, still less to alter the legends
themselves on the faith of any such surmises. The theogony of the
Greeks contains some cosmogonic ideas; but it cannot be considered
as a system of cosmogony, or translated into a string of elementary,
planetary, or physical changes.

  [2] It is sufficient, here, to state this position briefly: more
  will be said respecting the allegorizing interpretation in a
  future chapter.

In the order of legendary chronology, Zeus comes after Kronos and
Uranos; but in the order of Grecian conception, Zeus is the prominent
person, and Kronos and Uranos are inferior and introductory precursors,
set up in order to be overthrown and to serve as mementos of the
prowess of their conqueror. To Homer and Hesiod, as well as to the
Greeks universally, Zeus is the great and predominant god, “the father
of gods and men,” whose power none of the other gods can hope to
resist, or even deliberately think of questioning. All the other gods
have their specific potency and peculiar sphere of action and duty,
with which Zeus does not usually interfere; but it is he who maintains
the lineaments of a providential superintendence, as well over the
phænomena of Olympus as over those of earth. Zeus and his brothers
Poseidôn and Hadês have made a division of power: he has reserved the
æther and the atmosphere to himself—Poseidôn has obtained the sea—and
Hadês the under-world or infernal regions; while earth, and the events
which pass upon earth, are common to all of them, together with free
access to Olympus.[3]

  [3] See Iliad, viii. 405, 463; xv. 20, 130, 185. Hesiod, Theog.
  885.

  This unquestioned supremacy is the general representation of
  Zeus: at the same time the conspiracy of Hêrê, Poseidôn, and
  Athênê against him, suppressed by the unexpected apparition of
  Briareus as his ally, is among the exceptions. (Iliad, i. 400.)
  Zeus is at one time vanquished by Titan, but rescued by Hermês.
  (Apollodôr. i. 6, 3).

Zeus, then, with his brethren and colleagues, constitute the present
gods, whom Homer and Hesiod recognize as in full dignity and
efficiency. The inmates of this divine world are conceived upon the
model, but not upon the scale, of the human. They are actuated by the
full play and variety of those appetites, sympathies, passions and
affections, which divide the soul of man; invested with a far larger
and indeterminate measure of power, and an exemption as well from
death as (with some rare exceptions) from suffering and infirmity.
The rich and diverse types thus conceived, full of energetic movement
and contrast, each in his own province, and soaring confessedly above
the limits of experience, were of all themes the most suitable for
adventure and narrative, and operated with irresistible force upon the
Grecian fancy. All nature was then conceived as moving and working
through a number of personal agents, amongst whom the gods of Olympus
were the most conspicuous; the reverential belief in Zeus and Apollo
being only one branch of this omnipresent personifying faith. The
attributes of all these agents had a tendency to expand themselves into
illustrative legends—especially those of the gods, who were constantly
invoked in the public worship. Out of this same mental source sprang
both the divine and heroic mythes—the former being often the more
extravagant and abnormous in their incidents, in proportion as the
general type of the gods was more vast and awful than that of the
heroes.

As the gods have houses and wives like men, so the present dynasty
of gods must have a past to repose upon;[4] and the curious and
imaginative Greek, whenever he does not find a recorded past ready
to his hand, is uneasy until he has created one. Thus the Hesiodic
theogony explains, with a certain degree of system and coherence, first
the antecedent circumstances under which Zeus acquired the divine
empire, next the number of his colleagues and descendants.

  [4] Arist. Polit. i. 1. ὥσπερ δὲ καὶ τὰ εἴδη ἑαυτοῖς ἀφομοιοῦσιν
  ἄνθρωποι, οὕτως καὶ τοὺς βίους, τῶν θεῶν.

First in order of time (we are told by Hesiod) came Chaos; next Gæa,
the broad, firm, and flat Earth, with deep and dark Tartarus at her
base. Erôs (Love), the subduer of gods as well as men, came immediately
afterwards.[5]

  [5] Hesiod, Theog. 116. Apollodôrus begins with Uranos and Gæa
  (i. 1.); he does not recognize Erôs, Nyx, or Erebos.

From Chaos sprung Erebos and Nyx; from these latter Æthêr and Hêmera.
Gæa also gave birth to Uranos, equal in breadth to herself, in order to
serve both as an overarching vault to her, and as a residence for the
immortal gods; she further produced the mountains, habitations of the
divine nymphs, and Pontus, the barren and billowy sea.

Then Gæa intermarried with Uranos, and from this union came a numerous
offspring—twelve Titans and Titanides, three Cyclôpes, and three
Hekatoncheires or beings with a hundred hands each. The Titans were
Oceanus, Kœos, Krios, Hyperiôn, Iapetos, and Kronos: the Titanides,
Theia, Rhea, Themis, Mnêmosynê, Phœbê, and Têthys. The Cyclôpes were
Brontês, Steropês, and Argês,—formidable persons, equally distinguished
for strength and for manual craft, so that they made the thunder
which afterwards formed the irresistible artillery of Zeus.[6] The
Hekatoncheires were Kottos, Briareus, and Gygês, of prodigious bodily
force.

  [6] Hesiod, Theog. 140, 156. Apollod. _ut sup._

Uranos contemplated this powerful brood with fear and horror; as fast
as any of them were born, he concealed them in cavities of the earth,
and would not permit them to come out. Gæa could find no room for them,
and groaned under the pressure: she produced iron, made a sickle,
and implored her sons to avenge both her and themselves against the
oppressive treatment of their father. But none of them, except Kronos,
had courage to undertake the deed: he, the youngest and the most
daring, was armed with the sickle and placed in suitable ambush by the
contrivance of Gæa. Presently night arrived, and Uranos descended to
the embraces of Gæa: Kronos then emerged from his concealment, cut off
the genitals of his father, and cast the bleeding member behind him
far away into the sea.[7] Much of the blood was spilt upon the earth,
and Gæa in consequence gave birth to the irresistible Erinnys, the
vast and muscular Gigantes, and the Melian nymphs. Out of the genitals
themselves, as they swam and foamed upon the sea, emerged the goddess
Aphroditê, deriving her name from the foam out of which she had sprung.
She first landed at Kythêra, and then went to Cyprus: the island felt
her benign influence, and the green herb started up under her soft and
delicate tread. Erôs immediately joined her, and partook with her the
function of suggesting and directing the amorous impulses both of gods
and men.[8]

  [7] Hesiod, Theog. 160, 182. Apollod. i. 1, 4.

  [8] Hesiod, Theog. 192. This legend respecting the birth of
  Aphroditê seems to have been derived partly from her name (ἀφρὸς,
  _foam_), partly from the surname Urania, Ἀφροδίτη Οὐρανία, under
  which she was so very extensively worshipped, especially both in
  Cyprus and Cythêra, seemingly originated in both islands by the
  Phœnicians. Herodot. i. 105. Compare the instructive section in
  Boeckh’s Metrologie, c. iv. § 4.

Uranos being thus dethroned and disabled, Kronos and the Titans
acquired their liberty and became predominant: the Cyclôpes and the
Hekatoncheires had been cast by Uranos into Tartarus, and were still
allowed to remain there.

Each of the Titans had a numerous offspring: Oceanus, especially,
marrying his sister Têthys, begat three thousand daughters, the Oceanic
nymphs, and as many sons: the rivers and springs passed for his
offspring. Hyperiôn and his sister Theia had for their children Hêlios,
Selênê, and Eôs; Kœos with Phœbê begat Lêtô and Asteria; the children
of Krios were Astræos, Pallas, and Persês,—from Astræos and Eôs sprang
the winds Zephyrus, Boreas, and Notus. Iapetos, marrying the Oceanic
nymph Clymenê, counted as his progeny the celebrated Promêtheus,
Epimêtheus, Menœtius, and Atlas. But the offspring of Kronos were the
most powerful and transcendent of all. He married his sister Rhea, and
had by her three daughters—Hestia, Dêmêtêr, and Hêrê—and three sons,
Hadês, Poseidôn, and Zeus, the latter at once the youngest and the
greatest.

But Kronos foreboded to himself destruction from one of his own
children, and accordingly, as soon as any of them were born, he
immediately swallowed them and retained them in his own belly. In this
manner had the first five been treated, and Rhea was on the point of
being delivered of Zeus. Grieved and indignant at the loss of her
children, she applied for counsel to her father and mother, Uranos and
Gæa, who aided her to conceal the birth of Zeus. They conveyed her by
night to Lyktus in Crête, hid the new-born child in a woody cavern
on Mount Ida, and gave to Kronos, in place of it, a stone wrapped in
swaddling clothes, which he greedily swallowed, believing it to be
his child. Thus was the safety of Zeus ensured.[9] As he grew up his
vast powers fully developed themselves: at the suggestion of Gæa, he
induced Kronos by stratagem to vomit up, first the stone which had
been given to him,—next, the five children whom he had previously
devoured. Hestia, Dêmêtêr, Hêrê, Poseidôn and Hadês, were thus allowed
to grow up along with Zeus; and the stone to which the latter owed
his preservation was placed near the temple of Delphi, where it ever
afterwards stood, as a conspicuous and venerable memorial to the
religious Greek.[10]

  [9] Hesiod, Theog. 452, 487. Apollod. i. 1, 6.

  [10] Hesiod, Theog. 498.—

    Τὸν μὲν Ζεὺς στήριξε κατὰ χθονὸς εὐρυοδείης
    Πυθοῖ ἐν ἠγαθέῃ, γυάλοις ὑπὸ Παρνησοῖο,
    Σῆμ᾽ ἔμεν ἐξοπίσω, θαῦμα θνητοῖσι βροτοῖσι.

We have not yet exhausted the catalogue of beings generated during this
early period, anterior to the birth of Zeus. Nyx, alone and without
any partner, gave birth to a numerous progeny: Thanatos, Hypnos and
Oneiros; Mômus and Oïzys (Grief); Klôthô, Lachesis and Atropos, the
three Fates; the retributive and equalizing Nemesis; Apatê and Philotês
(Deceit and amorous Propensity), Gêras (Old Age) and Eris (Contention).
From Eris proceeded an abundant offspring, all mischievous and
maleficent: Ponos (Suffering), Lêthê, Limos (Famine), Phonos and Machê
(Slaughter and Battle), Dysnomia and Atê (Lawlessness and reckless
Impulse), and Horkos, the ever watchful sanctioner of oaths, as well as
the inexorable punisher of voluntary perjury.[11]

  [11] Hesiod, Theog. 212-232.

Gæa, too, intermarrying with Pontus, gave birth to Nereus, the just
and righteous old man of the sea; to Thaumas, Phorkys and Kêtô. From
Nereus, and Doris daughter of Oceanus, proceeded the fifty Nereids
or Sea-nymphs. Thaumus also married Elektra daughter of Oceanus, and
had by her Iris and the two Harpies, Allô and Okypetê,—winged and
swift as the winds. From Phorkys and Kêtô sprung the Dragon of the
Hesperides, and the monstrous Grææ and Gorgons: the blood of Medusa,
one of the Gorgons, when killed by Perseus, produced Chrysaor and the
horse Pegasus: Chrysaor and Kallirrhoê gave birth to Geryôn as well
as to Echidna,—a creature half-nymph and half-serpent, unlike both to
gods and to men. Other monsters arose from the union of Echidna with
Typhaôn,—Orthros, the two-headed dog of Geryôn; Cerberus, the dog
of Hadês, with fifty heads, and the Lernæan Hydra. From the latter
proceeded the Chimæra, the Sphinx of Thêbes, and the Nemean lion.[12]

  [12] Hesiod, Theog. 240-320. Apollodôr. i. 2, 6, 7.

A powerful and important progeny, also, was that of Styx, daughter
of Oceanus, by Pallas; she had Zêlos and Nikê (Imperiousness and
Victory), and Kratos and Bia (Strength and Force). The hearty and early
coöperation of Styx and her four sons with Zeus was one of the main
causes which enabled him to achieve his victory over the Titans.

Zeus had grown up not less distinguished for mental capacity than for
bodily force. He and his brothers now determined to wrest the power
from the hands of Kronos and the Titans, and a long and desperate
struggle commenced, in which all the gods and all the goddesses took
part. Zeus convoked them to Olympus, and promised to all who would aid
him against Kronos, that their functions and privileges should remain
undisturbed. The first who responded to the call, came with her four
sons, and embraced his cause, was Styx. Zeus took them all four as his
constant attendants, and conferred upon Styx the majestic distinction
of being the Horkos, or oath-sanctioner of the Gods,—what Horkos was to
men, Styx was to the Gods.[13]

  [13] Hesiod, Theog. 385-403.

Still further to strengthen himself, Zeus released the other Uranids
who had been imprisoned in Tartarus by their father,—the Cyclôpes and
the Centimanes,—and prevailed upon them to take part with him against
the Titans. The former supplied him with thunder and lightning, and the
latter brought into the fight their boundless muscular strength.[14]
Ten full years did the combat continue; Zeus and the Kronids occupying
Olympus, and the Titans being established on the more southerly
mountain-chain of Othrys. All nature was convulsed, and the distant
Oceanus, though he took no part in the struggle, felt the boiling,
the noise, and the shock, not less than Gæa and Pontus. The thunder
of Zeus, combined with the crags and mountains torn up and hurled by
the Centimanes, at length prevailed, and the Titans were defeated and
thrust down into Tartarus. Iapetos, Kronos, and the remaining Titans
(Oceanus excepted) were imprisoned, perpetually and irrevocably, in
that subterranean dungeon, a wall of brass being built around them by
Poseidôn, and the three Centimanes being planted as guards. Of the two
sons of Iapetos, Menœtius was made to share this prison, while Atlas
was condemned to stand for ever at the extreme west, and to bear upon
his shoulders the solid vault of heaven.[15]

  [14] Hesiod, Theog. 140, 624, 657. Apollodôr. i. 2, 4.

  [15] The battle with the Titans, Hesiod, Theog. 627-735. Hesiod
  mentions nothing about the Gigantes and the Gigantomachia:
  Apollodôrus, on the other hand, gives this latter in some detail,
  but despatches the Titans in a few words (i. 2, 4; i. 6, 1). The
  Gigantes seem to be only a second edition of the Titans,—a sort
  of duplication to which the legendary poets were often inclined.

Thus were the Titans subdued, and the Kronids with Zeus at their head
placed in possession of power. They were not, however, yet quite
secure; for Gæa, intermarrying with Tartarus, gave birth to a new and
still more formidable monster called Typhôeus, of such tremendous
properties and promise, that, had he been allowed to grow into full
development, nothing could have prevented him from vanquishing all
rivals and becoming supreme. But Zeus foresaw the danger, smote him at
once with a thunderbolt from Olympus, and burnt him up: he was cast
along with the rest into Tartarus, and no further enemy remained to
question the sovereignty of the Kronids.[16]

  [16] Hesiod, Theog. 820-869. Apollod. i. 6, 3. He makes Typhôn
  very nearly victorious over Zeus. Typhôeus, according to Hesiod,
  is father of the irregular, violent, and mischievous winds:
  Notus, Boreas, Argestês and Zephyrus, are of divine origin (870).

With Zeus begins a new dynasty and a different order of beings. Zeus,
Poseidôn, and Hadês agree upon the distribution before noticed, of
functions and localities: Zeus retaining the Æthêr and the atmosphere,
together with the general presiding function; Poseidôn obtaining
the sea, and administering subterranean forces generally; and Hadês
ruling the under-world or region in which the half-animated shadows of
departed men reside.

It has been already stated, that in Zeus, his brothers and his sisters,
and his and their divine progeny, we find the _present_ Gods; that
is, those, for the most part, whom the Homeric and Hesiodic Greeks
recognized and worshipped. The wives of Zeus were numerous as well as
his offspring. First he married Mêtis, the wisest and most sagacious of
the goddesses; but Gæa and Uranos forewarned him that if he permitted
himself to have children by her, they would be stronger than himself
and dethrone him. Accordingly when Mêtis was on the point of being
delivered of Athênê, he swallowed her up, and her wisdom and sagacity
thus became permanently identified with his own being.[17] His head
was subsequently cut open, in order to make way for the exit and
birth of the goddess Athênê.[18] By Themis, Zeus begat the Hôræ, by
Eurynomê, the three Charities or Graces; by Mnêmosynê, the Muses; by
Lêtô (Latona), Apollo and Artemis; and by Dêmêtêr, Persephonê. Last of
all he took for his wife Hêrê, who maintained permanently the dignity
of queen of the Gods; by her he had Hêbê, Arês, and Eileithyia. Hermês
also was born to him by Maia, the daughter of Atlas: Hêphæstos was born
to Hêrê, according to some accounts, by Zeus; according to others, by
her own unaided generative force.[19] He was born lame, and Hêrê was
ashamed of him: she wished to secrete him away, but he made his escape
into the sea, and found shelter under the maternal care of the Nereids
Thetis and Eurynome.[20] Our enumeration of the divine race, under the
presidency of Zeus, will thus give us,[21]—

  [17] Hesiod, Theog. 885-900.

  [18] Apollod. i. 3, 6.

  [19] Hesiod, Theog. 900-944.

  [20] Homer, Iliad, xviii. 397.

  [21] See Burckhardt, Homer, und Hesiod. Mythologie, sect. 102.
  (Leipz. 1844).

1. The twelve great gods and goddesses of Olympus,—Zeus, Poseidôn,
Apollo, Arês, Hêphæstos, Hermês, Hêrê, Athênê, Artemis, Aphroditê,
Hestia, Dêmêtêr.

2. An indefinite number of other deities, not included among the
Olympic, seemingly because the number _twelve_ was complete without
them, but some of them not inferior in power and dignity to many of
the twelve:—Hadês, Hêlios, Hekatê, Dionysos, Lêtô, Diônê, Persephonê,
Selênê, Themis, Eôs, Harmonia, the Charities, the Muses, the
Eileithyiæ, the Mœræ, the Oceanids and the Nereids, Proteus, Eidothea,
the Nymphs, Leukothea, Phorkys, Æolus, Nemesis, etc.

3. Deities who perform special services to the greater gods:—Iris,
Hêbê, the Horæ, etc.

4. Deities whose personality is more faintly and unsteadily
conceived:—Atê, the Litæ, Eris, Thanatos, Hypnos, Kratos, Bia, Ossa,
etc.[22] The same name is here employed sometimes to designate
the person, sometimes the attribute or event not personified,—an
unconscious transition of ideas, which, when consciously performed, is
called Allegory.

  [22] Λιμὸς—_Hunger_—is a person, in Hesiod, Opp. Di. 299.

5. Monsters, offspring of the Gods:—the Harpies, the Gorgons, the Grææ,
Pegasus, Chrysaor, Echidna, Chimæra, the Dragon of the Hesperides,
Cerberus, Orthros, Geryôn, the Lernæan Hydra, the Nemean lion, Scylla
and Charybdis, the Centaurs, the Sphinx, Xanthos and Balios the
immortal horses, etc.

From the gods we slide down insensibly, first to heroes, and then to
men; but before we proceed to this new mixture, it is necessary to
say a few words on the theogony generally. I have given it briefly as
it stands in the Hesiodic Theogonia, because that poem—in spite of
great incoherence and confusion, arising seemingly from diversity of
authorship as well as diversity of age—presents an ancient and genuine
attempt to cast the divine foretime into a systematic sequence. Homer
and Hesiod were the grand authorities in the pagan world respecting
theogony; but in the Iliad and Odyssey nothing is found except passing
allusions and implications, and even in the Hymns (which were commonly
believed in antiquity to be the productions of the same author as the
Iliad and the Odyssey) there are only isolated, unconnected narratives.
Accordingly men habitually took their information respecting their
theogonic antiquities from the Hesiodic poem, where it was ready laid
out before them; and the legends consecrated in that work acquired
both an extent of circulation and a firm hold on the national faith,
such as independent legends could seldom or never rival. Moreover the
scrupulous and sceptical Pagans, as well as the open assailants of
Paganism in later times, derived their subjects of attack from the same
source; so that it has been absolutely necessary to recount in their
naked simplicity the Hesiodic stories, in order to know what it was
that Plato deprecated and Xenophanês denounced. The strange proceedings
ascribed to Uranos, Kronos and Zeus, have been more frequently alluded
to, in the way of ridicule or condemnation, than any other portion of
the mythical world.

But though the Hesiodic theogony passed as orthodox among the later
Pagans,[23] because it stood before them as the only system anciently
set forth and easily accessible, it was evidently not the only system
received at the date of the poem itself. Homer knows nothing of Uranos,
in the sense of an arch-God anterior to Kronos. Uranos and Gæa, like
Oceanus, Têthys and Nyx, are with him great and venerable Gods, but
neither the one nor the other present the character of predecessors of
Kronos and Zeus.[24] The Cyclôpes, whom Hesiod ranks as sons of Uranos
and fabricators of thunder, are in Homer neither one nor the other;
they are not noticed in the Iliad at all, and in the Odyssey they are
gross gigantic shepherds and cannibals, having nothing in common with
the Hesiodic Cyclops except the one round central eye.[25] Of the
three Centimanes enumerated by Hesiod, Briareus only is mentioned in
Homer, and to all appearance, not as the son of Uranos, but as the son
of Poseidôn; not as aiding Zeus in his combat against the Titans, but
as rescuing him at a critical moment from a conspiracy formed against
him by Hêrê, Poseidôn and Athênê.[26] Not only is the Hesiodic Uranos
(with the Uranids) omitted in Homer, but the relations between Zeus
and Kronos are also presented in a very different light. No mention is
made of Kronos swallowing his young children: on the contrary, Zeus
is the eldest of the three brothers instead of the youngest, and the
children of Kronos live with him and Rhea: there the stolen intercourse
between Zeus and Hêrê first takes place without the knowledge of their
parents.[27] When Zeus puts Kronos down into Tartarus, Rhea consigns
her daughter Hêrê to the care of Oceanus: no notice do we find of any
terrific battle with the Titans as accompanying that event. Kronos,
Iapetos, and the remaining Titans are down in Tartarus, in the lowest
depths under the earth, far removed from the genial rays of Hêlios; but
they are still powerful and venerable, and Hypnos makes Hêrê swear an
oath in their name, as the most inviolable that he can think of.[28]

  [23] See Göttling, Præfat. ad Hesiod. p. 23.

  [24] Iliad, xiv. 249; xix. 259. Odyss. v. 184. Oceanus and Têthys
  seem to be presented in the Iliad as the primitive Father and
  Mother of the Gods:—

    Ὠκεανόν τε θεῶν γένεσιν, καὶ μητέρα Τηθύν. (xiv. 201).

  [25] Odyss. ix. 87.

  [26] Iliad, i. 401.

  [27] Iliad, xiv. 203-295; xv. 204.

  [28] Iliad, viii. 482; xiv. 274-279. In the Hesiodic Opp. et Di.,
  Kronos is represented as ruling in the Islands of the Blest in
  the neighborhood of Oceanus (v. 168).

In Homer, then, we find nothing beyond the simple fact that Zeus thrust
his father Kronos together with the remaining Titans into Tartarus;
an event to which he affords us a tolerable parallel in certain
occurrences even under the presidency of Zeus himself. For the other
gods make more than one rebellious attempt against Zeus, and are only
put down, partly by his unparalleled strength, partly by the presence
of his ally the Centimane Briareus. Kronos, like Laërtes or Pêleus,
has become old, and has been supplanted by a force vastly superior to
his own. The Homeric epic treats Zeus as present, and, like all the
interesting heroic characters, a father must be assigned to him: that
father has once been the chief of the Titans, but has been superseded
and put down into Tartarus along with the latter, so soon as Zeus and
the superior breed of the Olympic gods acquired their full development.

That antithesis between Zeus and Kronos—between the Olympic gods and
the Titans—which Homer has thus briefly brought to view, Hesiod has
amplified into a theogony, with many things new, and some things
contradictory to his predecessor; while Eumêlus or Arktinus in the
poem called Titanomachia (now lost) also adopted it as their special
subject.[29] As Stasinus, Arktinus, Leschês, and others, enlarged the
Legend of Troy by composing poems relating to a supposed time anterior
to the commencement, or subsequent to the termination of the Iliad,—as
other poets recounted adventures of Odysseus subsequent to his landing
in Ithaka,—so Hesiod enlarged and systematized, at the same time that
he corrupted, the skeleton theogony which we find briefly indicated
in Homer. There is violence and rudeness in the Homeric gods, but the
great genius of Grecian epic is no way accountable for the stories
of Uranos and Kronos,—the standing reproach against Pagan legendary
narrative.

  [29] See the few fragments of the Titanomachia, in Düntzer, Epic.
  Græc. Fragm. p. 2; and Hyne, ad Apollodôr. I. 2. Perhaps there
  was more than one poem on the subject, though it seems that
  Athenæus had only read one (viii. p. 277).

  In the Titanomachia, the generations anterior to Zeus were still
  further lengthened by making Uranos the son of Æthêr (Fr. 4.
  Düntzer). Ægæon was also represented as son of Pontus and Gæa,
  and as having fought in the ranks of the Titans; in the Iliad he
  (the same who is called Briareus) is the fast ally of Zeus.

  A _Titanographia_ was ascribed to Musæus (Schol. Apollôn. Rhod.
  iii. 1178; compare Lactant. de Fals. Rel. i. 21).

How far these stories are the invention of Hesiod himself is impossible
to determine.[30] They bring us down to a cast of fancy more coarse
and indelicate than the Homeric, and more nearly resembling some of
the Holy Chapters (ἱεροὶ λόγοι) of the more recent mysteries, such
(for example) as the tale of Dionysos Zagreus. There is evidence
in the Theogony itself that the author was acquainted with local
legends current both at Krête and at Delphi; for he mentions both the
mountain-cave in Krête wherein the new-born Zeus was hidden, and the
stone near the Delphian temple—the identical stone which Kronos had
swallowed—“placed by Zeus himself as a sign and wonder to mortal men.”
Both these two monuments, which the poet expressly refers to, and
had probably seen, imply a whole train of accessory and explanatory
local legends—current probably among the priests of Krête and Delphi,
between which places, in ancient times, there was an intimate religious
connection. And we may trace further in the poem,—that which would
be the natural feeling of Krêtan worshippers of Zeus,—an effort to
make out that Zeus was justified in his aggression on Kronos, by the
conduct of Kronos himself both towards his father and towards his
children: the treatment of Kronos by Zeus appears in Hesiod as the
retribution foretold and threatened by the mutilated Uranos against the
son who had outraged him. In fact the relations of Uranos and Gæa are
in almost all their particulars a mere copy and duplication of those
between Kronos and Rhea, differing only in the mode whereby the final
catastrophe is brought about. Now castration was a practice thoroughly
abhorrent both to the feelings and to the customs of Greece;[31] but
it was seen with melancholy frequency in the domestic life as well as
in the religious worship of Phrygia and other parts of Asia, and it
even became the special qualification of a priest of the Great Mother
Cybelê,[32] as well as of the Ephesian Artemis. The employment of the
sickle ascribed to Kronos seems to be the product of an imagination
familiar with the Asiatic worship and legends, which were connected
with and partially resembled the Krêtan.[33] And this deduction becomes
the more probable when we connect it with the first genesis of iron,
which Hesiod mentions to have been produced for the express purpose of
fabricating the fatal sickle; for metallurgy finds a place in the early
legends both of the Trojan and of the Krêtan Ida, and the three Idæan
Dactyls, the legendary inventors of it, are assigned sometimes to one
and sometimes to the other.[34]

  [30] That the Hesiodic Theogony is referable to an age
  considerably later than the Homeric poems, appears now to be the
  generally admitted opinion; and the reasons for believing so are,
  in my opinion, satisfactory. Whether the Theogony is composed by
  the same author as the Works and Days is a disputed point. The
  Bœotian literati in the days of Pausanias decidedly denied the
  identity, and ascribed to their Hesiod only the Works and Days:
  Pausanias himself concurs with them (ix. 31. 4; ix. 35. 1), and
  Völcker (Mithologie des Japetisch. Geschlechts, p. 14) maintains
  the same opinion, as well as Göttling (Præf. ad Hesiod. xxi.): K.
  O. Müller (History of Grecian Literature, ch. 8. § 4) thinks that
  there is not sufficient evidence to form a decisive opinion.

  Under the name of Hesiod (in that vague language which is usual
  in antiquity respecting authorship, but which modern critics
  have not much mended by speaking of the Hesiodic school, sect,
  or family) passed many different poems, belonging to three
  classes quite distinct from each other, but all disparate from
  the Homeric epic:—1. The poems of legend cast into historical and
  genealogical series, such as the Eoiai, the Catalogue of Women,
  etc. 2. The poems of a didactic or ethical tendency, such as
  the Works and Days, the Precepts of Cheirôn, the Art of Augural
  Prophecy, etc. 3. Separate and short mythical compositions,
  such as the Shield of Hêraklês, the Marriage of Keyx (which,
  however, was of disputed authenticity, Athenæ. ii. p. 49), the
  Epithalamium of Pêleus and Thetis, etc. (See Marktscheffel,
  Præfat. ad Fragment. Hesiod. p. 89).

  The Theogony belongs chiefly to the first of these classes, but
  it has also a dash of the second in the legend of Promêtheus,
  etc.: moreover in the portion which respects Hekatê, it has both
  a mystic character and a distinct bearing upon present life and
  customs, which we may also trace in the allusions to Krête and
  Delphi. There seems reason to place it in the same age with the
  Works and Days, perhaps in the half century preceding 700 B.
  C., and little, if at all, anterior to Archilochus. The poem
  is evidently conceived upon one scheme, yet the parts are so
  disorderly and incoherent, that it is difficult to say how much
  is interpolation. Hermann has well dissected the exordium; see
  the preface to Gaisford’s Hesiod (Poetæ Minor. p. 63).

  K. O. Müller tells us (_ut sup._ p. 90), “The Titans, according
  to the notions of Hesiod, represent a system of things in which
  elementary beings, natural powers, and notions of order and
  regularity are united to form a whole. The Cyclôpes denote the
  transient disturbances of this order of nature by storms, and the
  Hekatoncheires, or hundred-handed Giants, signify the fearful
  power of the greater revolutions of nature.” The poem affords
  little presumption that any such ideas were present to the mind
  of its author, as, I think, will be seen if we read 140-155,
  630-745.

  The Titans, the Cyclôpes, and the Hekatoncheires, can no more
  be construed into physical phænomena than Chrysaor, Pegasus,
  Echidna, the Grææ, or the Gorgons. Zeus, like Hêraklês, or Jasôn,
  or Perseus, if his adventures are to be described, must have
  enemies, worthy of himself and his vast type, and whom it is
  some credit for him to overthrow. Those who contend with him or
  assist him must be conceived on a scale fit to be drawn on the
  same imposing canvas: the dwarfish proportions of man will not
  satisfy the sentiment of the poet or his audience respecting the
  grandeur and glory of the gods. To obtain creations of adequate
  sublimity for such an object, the poet may occasionally borrow
  analogies from the striking accidents of physical nature, and
  when such an allusion manifests itself clearly, the critic does
  well to point it out. But it seems to me a mistake to treat
  these approximations to physical phænomena as forming the _main
  scheme_ of the poet,—to look for them everywhere, and to presume
  them where there is little or no indication.

  [31] The strongest evidences of this feeling are exhibited in
  Herodotus, iii. 48; viii. 105. See an example of this mutilation
  inflicted upon a youth named Adamas by the Thracian king Kotys,
  in Aristot. Polit. v. 8, 12, and the tale about the Corinthian
  Periander, Herod. iii. 48.

  It is an instance of the habit, so frequent among the Attic
  tragedians, of ascribing Asiatic or Phrygian manners to the
  Trojans, when Sophoclês in his lost play Troilus (ap. Jul.
  Poll. x. 165) introduced one of the characters of his drama as
  having been castrated by order of Hecuba, Σκαλμῇ γὰρ ὄρχεις
  βασιλὶς ἐκτέμνουσ᾽ ἐμούς,—probably the Παιδαγωγὸς, or guardian
  and companion of the youthful Troilus. See Welcker, Griechisch.
  Tragöd. vol. i. p. 125.

  [32] Herodot. viii. 105, εὐνοῦχοι. Lucian, De Deâ Syriâ, c. 50.
  Strabo, xiv. pp. 640-641.

  [33] Diodôr. v. 64. Strabo, x. p. 460. Hoeckh, in his learned
  work Krêta (vol. i. books 1 and 2), has collected all the
  information attainable respecting the early influences of Phrygia
  and Asia Minor upon Krête: nothing seems ascertainable except the
  general fact; all the particular evidences are lamentably vague.

  The worship of the Diktæan Zeus seemed to have originally
  belonged to the Eteokrêtes, who were not Hellens, and were more
  akin to the Asiatic population than to the Hellenic. Strabo, x.
  p. 478. Hoeckh, Krêta, vol. i. p. 139.

  [34] Hesiod, Theogon. 161,

    Αἶψα δὲ ποιήσασα γένος πολιοῦ ἀδάμαντος,
    Τεῦξε μέγα δρέπανον, etc.

  See the extract from the old poem _Phorônis_ ap. Schol. Apoll.
  Rhod. 1129; and Strabo, x. p. 472.

As Hesiod had extended the Homeric series of gods by prefixing the
dynasty of Uranos to that of Kronos, so the Orphic theogony lengthened
it still further.[35] First came Chronos, or Time, as a person, after
him Æthêr and Chaos, out of whom Chronos produced the vast mundane egg.
Hence emerged in process of time the first-born god Phanês, or Mêtis,
or Hêrikapæos, a person of double sex, who first generated the Kosmos,
or mundane system, and who carried within him the seed of the gods.
He gave birth to Nyx, by whom he begat Uranos and Gæa; as well as to
Hêlios and Selêne.[36]

  [35] See the scanty fragments of the Orphic theogony in Hermann’s
  edition of the Orphica, pp. 448, 504, which it is difficult to
  understand and piece together, even with the aid of Lobeck’s
  elaborate examination (Aglaophamus, p. 470, etc.). The passages
  are chiefly preserved by Proclus and the later Platonists,
  who seem to entangle them almost inextricably with their own
  philosophical ideas.

  The first few lines of the Orphic Argonautica contain a brief
  summary of the chief points of the theogony.

  [36] See Lobeck, Aglaoph. p. 472-476, 490-500, Μῆτιν σπέρμα
  φέροντα θεῶν κλυτὸν Ἠρικεπαῖον; again, Θῆλυς καὶ γενέτωρ κρατερὸς
  θεὸς Ἠρικέπαιος. Compare Lactant. iv. 8, 4: Suidas, v. Φάνης:
  Athenagoras, xx. 296: Diodôr. i. 27.

  This egg figures, as might be expected, in the cosmogony set
  forth by the Birds, Aristophan. Av. 695. Nyx gives birth to an
  egg, out of which steps the golden Erôs, from Erôs and Chaos
  spring the race of birds.

From Uranos and Gæa sprang the three Mœræ, or Fates, the three
Centimanes and the three Cyclôpes: these latter were cast by Uranos
into Tartarus, under the foreboding that they would rob him of his
dominion. In revenge for this maltreatment of her sons, Gæa produced of
herself the fourteen Titans, seven male and seven female: the former
were Kœos, Krios, Phorkys, Kronos, Oceanus, Hyperiôn and Iapetos;
the latter were Themis, Têthys, Mnêmosynê, Theia, Diônê, Phœbê and
Rhea.[37] They received the name of Titans because they avenged upon
Uranos the expulsion of their elder brothers. Six of the Titans, headed
by Kronos the most powerful of them all, conspiring against Uranos,
castrated and dethroned him: Oceanus alone stood aloof and took no part
in the aggression. Kronos assumed the government and fixed his seat
on Olympos; while Oceanus remained apart, master of his own divine
stream.[38] The reign of Kronos was a period of tranquillity and
happiness, as well as of extraordinary longevity and vigor.

  [37] Lobeck, Ag. p. 504. Athenagor. xv. p. 64.

  [38] Lobeck, Ag. p. 507. Plato, Timæus, p. 41. In the Διονύσου
  τρόφοι of Æschylus, the old attendants of the god Dionysos were
  said to have been cut up and boiled in a caldron, and rendered
  again young, by Medeia. Pherecydês and Simonidês said that Jasôn
  himself had been so dealt with. Schol. Aristoph. Equit. 1321.

Kronos and Rhea gave birth to Zeus and his brothers and sisters. The
concealment and escape of the infant Zeus, and the swallowing of the
stone by Kronos, are given in the Orphic Theogony substantially in
the same manner as by Hesiod, only in a style less simple and more
mysticized. Zeus is concealed in the cave of Nyx, the seat of Phanês
himself, along with Eidê and Adrasteia, who nurse and preserve him,
while the armed dance and sonorous instruments of the Kurêtes prevent
his infant cries from reaching the ears of Kronos. When grown up, he
lays a snare for his father, intoxicates him with honey, and having
surprised him in the depth of sleep, enchains and castrates him.[39]
Thus exalted to the supreme mastery, he swallowed and absorbed into
himself Mêtis, or Phanês, with all the preëxisting elements of things,
and then generated all things anew out of his own being and conformably
to his own divine ideas.[40] So scanty are the remains of this system,
that we find it difficult to trace individually the gods and goddesses
sprung from Zeus beyond Apollo, Dionysos, and Persephonê,—the latter
being confounded with Artemis and Hekatê.

  [39] Lobeck, p. 514. Porphyry, de Antro Nympharum, c. 16. φησὶ
  γὰρ παρ᾽ Ὀρφεῖ ἡ Νὺξ, τῷ Διῒ ὑποτιθεμένη τὸν διὰ τοῦ μέλιτος
  δόλον,

    Εὖτ᾽ ἂν δή μιν ἴδηαι ὑπὸ δρυσὶν ὑψικόμοισι
    Ἔργοισιν μεθύοντα μελισσάων ἐριβόμβων,
    Αὔτικά μιν δῆσον.

    Ὃ καὶ πάσχει ὁ Κρόνος καὶ δεθεὶς ἐκτέμνεται, ὡς Οὐρανός.

  Compare Timæus ap. Schol. Apoll. Rhod. iv. 983.

  [40] The Cataposis of Phanês by Zeus one of the most memorable
  points of the Orphic Theogony. Lobeck, p. 519.; also Fragm. vi.
  p. 456 of Hermann’s Orphica.

  From this absorption and subsequent reproduction of all things by
  Zeus, flowed the magnificent string of Orphic predicates about
  him,—

    Ζεὺς ἀρχὴ, Ζεὺς μέσσα, Διὸς δ᾽ ἐκ πάντα τέτυκται,—

  an allusion to which is traceable even in Plato, de Legg. iv. p.
  715. Plutarch, de Defectu Oracul. T. ix. p. 379. c. 48. Diodôrus
  (i. 11) is the most ancient writer remaining to us who mentions
  the name of Phanês, in a line cited as proceeding from Orpheus;
  wherein, however, Phanês is identified with Dionysos. Compare
  Macrobius, Saturnal. i. 18.

But there is one new personage, begotten by Zeus, who stands
preëminently marked in the Orphic Theogony, and whose adventures
constitute one of its peculiar features. Zagreus, “the horned child,”
is the son of Zeus by his own daughter Persephonê: he is the favorite
of his father, a child of magnificent promise, and predestined, if he
grow up, to succeed to supreme dominion as well as to the handling
of the thunderbolt. He is seated, whilst an infant, on the throne
beside Zeus, guarded by Apollo and the Kurêtes. But the jealous Hêrê
intercepts his career and incites the Titans against him, who, having
first smeared their faces with plaster, approach him on the throne,
tempt his childish fancy with playthings, and kill him with a sword
while he is contemplating his face in a mirror. They then cut up his
body and boil it in a caldron, leaving only the heart, which is picked
up by Athênê and carried to Zeus, who in his wrath strikes down the
Titans with thunder into Tartarus; whilst Apollo is directed to collect
the remains of Zagreus and bury them at the foot of Mount Parnassus.
The heart is given to Semelê, and Zagreus is born again from her under
the form of Dionysos.[41]

  [41] About the tale of Zagreus, see Lobeck, p. 552, _sqq._ Nonnus
  in his Dionysiaca has given many details about it:—

    Ζαγρέα γειναμένη κέροεν βρέφος, etc. (vi. 264).

  Clemens Alexandrin. Admonit. ad Gent. p. 11, 12, Sylb. The story
  was treated both by Callimachus and by Euphoriôn, Etymolog. Magn.
  v. Ζαγρεὺς, Schol. Lycophr. 208. In the old epic poem Alkmæônis
  or Epigoni, Zagreus is a surname of Hadês. See Fragm. 4, p. 7,
  ed. Düntzer. Respecting the Orphic Theogony generally, Brandis
  (Handbuch der Geschichte der Griechisch-Römisch. Philosophie,
  c. xvii., xviii.), K. O. Müller (Prolegg. Mythol. pp. 379-396),
  and Zoega (Abhandlungen, v. pp. 211-263) may be consulted with
  much advantage. Brandis regards this Theogony as _considerably
  older_ than the first Ionic philosophy, which is a higher
  antiquity than appears probable: some of the ideas which it
  contains, such, for example, as that of the Orphic egg, indicate
  a departure from the string of purely personal generations
  which both Homer and Hesiod exclusively recount, and a resort
  to something like physical analogies. On the whole, we cannot
  reasonably claim for it more than half a century above the age of
  Onomakritus. The Theogony of Pherekydês of Syros seems to have
  borne some analogy to the Orphic. See Diogen. Laërt. i. 119,
  Sturz. Fragment. Pherekyd. § 5-6, Brandis, Handbuch, _ut sup._ c.
  xxii. Pherekydês partially deviated from the mythical track or
  personal successions set forth by Hesiod. ἐπεὶ οἵ γε ~μεμιγμένοι~
  αὐτῶν καὶ ~τῷ μὴ μυθικῶς~ ἅπαντα λέγειν, οἷον Φερεκύδης καὶ
  ἑτεροί τινες, etc. (Aristot. Metaphys. N. p. 301, ed. Brandis).
  Porphyrias, de Antro Nymphar. c. 31, καὶ τοῦ Συρίου Φερεκύδου
  μυχοὺς καὶ βόθρους καὶ ἄντρα καὶ θύρας καὶ πύλας λέγοντος, καὶ
  διὰ τούτων αἰνιττομένου τὰς τῶν ψυχῶν γενέσεις καὶ ἀπογενέσεις,
  etc. Eudêmus the Peripatetic, pupil of Aristotle, had drawn up
  an account of the Orphic Theogony as well as of the doctrines of
  Pherekydês, Akusilaus and others, which was still in the hands of
  the Platonists of the fourth century, though it is now lost. The
  extracts which we find seem all to countenance the belief that
  the Hesiodic Theogony formed the basis upon which they worked.
  See about Akusilaus, Plato, Sympos. p. 178. Clem. Alex. Strom. p.
  629.

Such is the tissue of violent fancies comprehended under the title of
the Orphic Theogony, and read as such, it appears, by Plato, Isokratês
and Aristotle. It will be seen that it is based upon the Hesiodic
Theogony, but according to the general expansive tendency of Grecian
legend, much new matter is added: Zeus has in Homer one predecessor, in
Hesiod two, and in Orpheus four.

The Hesiodic Theogony, though later in date than the Iliad and Odyssey,
was coeval with the earliest period of what may be called Grecian
history, and certainly of an age earlier than 700 B. C. It appears to
have been widely circulated in Greece, and being at once ancient and
short, the general public consulted it as their principal source of
information respecting divine antiquity. The Orphic Theogony belongs to
a later date, and contains the Hesiodic ideas and persons, enlarged and
mystically disguised: its vein of invention was less popular, adapted
more to the contemplation of a sect specially prepared than to the
taste of a casual audience, and it appears accordingly to have obtained
currency chiefly among purely speculative men.[42] Among the majority
of these latter, however, it acquired greater veneration, and above all
was supposed to be of greater antiquity, than the Hesiodic. The belief
in its superior antiquity (disallowed by Herodotus, and seemingly also
by Aristotle[43]), as well as the respect for its contents, increased
during the Alexandrine age and through the declining centuries of
Paganism, reaching its maximum among the New-Platonists of the third
and fourth century after Christ: both the Christian assailants, as
well as the defenders, of paganism, treated it as the most ancient and
venerable summary of the Grecian faith. Orpheus is celebrated by Pindar
as the harper and companion of the Argonautic maritime heroes: Orpheus
and Musæus, as well as Pamphôs and Olên, the great supposed authors
of theogonic, mystical, oracular, and prophetic verses and hymns,
were generally considered by literary Greeks as older than either
Hesiod or Homer:[44] and such was also the common opinion of modern
scholars until a period comparatively recent. It has now been shown,
on sufficient ground, that the compositions which passed under these
names emanate for the most part from poets of the Alexandrine age, and
subsequent to the Christian æra; and that even the earliest among them,
which served as the stock on which the later additions were engrafted,
belong to a period far more recent than Hesiod; probably to the century
preceding Onomakritus (B. C. 610-510). It seems, however, certain,
that both Orpheus and Musæus were names of established reputation at
the time when Onomakritus flourished; and it is distinctly stated by
Pausanias that the latter was himself the author of the most remarkable
and characteristic mythe of the Orphic Theogony—the discerption of
Zagreus by the Titans, and his resurrection as Dionysos.[45]

  [42] The Orphic Theogony is never cited in the ample Scholia on
  Homer, though Hesiod is often alluded to. (See Lobeck, Aglaoph.
  p. 540). Nor can it have been present to the minds of Xenophanês
  and Herakleitus, as representing any widely diffused Grecian
  belief: the former, who so severely condemned Homer and Hesiod,
  would have found Orpheus much more deserving of his censure: and
  the latter could hardly have omitted Orpheus from his memorable
  denunciation:—Πολυμαθίη νόον οὐ διδάσκει· Ἡσίοδον γὰρ ἂν ἐδίδαξε
  καὶ Πυθαγόρην, αὖτις δὲ Ξενοφάνεά τε καὶ Ἑκαταῖον. Diog. Laër.
  ix. 1. Isokratês treats Orpheus as the most censurable of all
  the poets. See Busiris, p. 229; ii. p. 309, Bekk. The Theogony
  of Orpheus, as conceived by Apollônius Rhodius (i. 504) in the
  third century B. C., and by Nigidius in the first century B.
  C. (Servius ad Virgil. Eclog. iv. 10), seems to have been on a
  more contracted scale than that which is given in the text. But
  neither of them notice the tale of Zagreus, which we know to be
  as old as Onomakritus.

  [43] This opinion of Herodotus is implied in the remarkable
  passage about Homer and Hesiod, ii. 53, though he never once
  names Orpheus—only alluding once to “Orphic ceremonies,” ii. 81.
  He speaks more than once of the prophecies of Musæus. Aristotle
  denied the past existence and reality of Orpheus. See Cicero de
  Nat. Deor. i. 38.

  [44] Pindar Pyth. iv. 177. Plato seems to consider Orpheus as
  more ancient than Homer. Compare Theætêt. p. 179; Cratylus, p.
  402; De Republ. ii. p. 364. The order in which Aristophanês (and
  Hippias of Elis, ap. Clem. Alex. Str. vi. p. 624) mentions them
  indicates the same view, Ranæ, 1030. It is unnecessary to cite
  the later chronologers, among whom the belief in the antiquity
  of Orpheus was universal; he was commonly described as son of
  the Muse Calliopê. Androtiôn seems to have denied that he was
  a Thracian, regarding the Thracians as incurably stupid and
  illiterate. Androtiôn, Fragm. 36, ed. Didot. Ephorus treated
  him as having been a pupil of the Idæan Dactyls of Phrygia (see
  Diodôr. v. 64), and as having learnt from them his τελετὰς and
  μυστήρια, which he was the first to introduce into Greece. The
  earliest mention which we find of Orpheus, is that of the poet
  Ibycus (about B. C. 530), ὀνομάκλυτον Ὀρφῆν. Ibyci Fragm. 9, p.
  341, ed. Schneidewin.

  [45] Pausan. viii. 37, 3. Τιτᾶνας δὲ πρῶτον ἐς ποίησιν ἐσήγαγεν
  Ὅμηρος, θεοὺς εἶναι σφᾶς ὑπὸ τῷ καλουμένῳ Ταρτάρῳ· καὶ ἔστιν
  ἐν Ἡρᾶς ὅρκῳ τὰ ἔπη· παρὰ δὲ Ὁμήρου Ὀνομάκριτος, παραλαβὼν τῶν
  Τιτάνων τὸ ὄνομα, Διονύσῳ τε συνέθηκεν ὄργια, καὶ εἶναι τοὺς
  Τιτᾶνας τῷ Διονύσῳ τῶν παθημάτων ἐποίησεν αὐτουργούς. Both
  the date, the character and the function of Onomakritus are
  distinctly marked by Herodotus, vii. 6.

The names of Orpheus and Musæus (as well as that of Pythagoras,[46]
looking at one side of his character) represent facts of importance
in the history of the Grecian mind—the gradual influx of Thracian,
Phrygian, and Egyptian, religious ceremonies and feelings, and the
increasing diffusion of special mysteries,[47] schemes for religious
purification, and orgies (I venture to anglicize the Greek word,
which contains in its original meaning no implication of the ideas
of excess to which it was afterwards diverted) in honor of some
particular god—distinct both from the public solemnities and from the
gentile solemnities of primitive Greece,—celebrated apart from the
citizens generally, and approachable only through a certain course of
preparation and initiation—sometimes even forbidden to be talked of in
the presence of the uninitiated, under the severest threats of divine
judgment. Occasionally such voluntary combinations assumed the form
of permanent brotherhoods, bound together by periodical solemnities
as well as by vows of an ascetic character: thus the Orphic life (as
it was called) or regulation of the Orphic brotherhood, among other
injunctions partly arbitrary and partly abstinent, forbade animal food
universally, and on certain occasions, the use of woollen clothing.[48]
The great religious and political fraternity of the Pythagoreans,
which acted so powerfully on the condition of the Italian cities, was
one of the many manifestations of this general tendency, which stands
in striking contrast with the simple, open-hearted, and demonstrative
worship of the Homeric Greeks.

  [46] Herodotus believed in the derivation both of the Orphic and
  Pythagorean regulations from Egypt—ὁμολογέουσι δὲ ταῦτα τοῖσι
  Ὀρφικοῖσι καλεομένοισι καὶ Βακχικοῖσι, ἐοῦσι δὲ Αἰγυπτίοισι
  (ii. 81). He knows the names of those Greeks who have borrowed
  from Egypt the doctrine of the metempsychosis, but he will not
  mention them (ii. 123): he can hardly allude to any one but
  the Pythagoreans, many of whom he probably knew in Italy. See
  the curious extract from Xenophanês respecting the doctrine of
  Pythagoras, Diogen. Laërt. viii. 37; and the quotation from the
  Silli of Timôn, Πυθαγόραν δὲ γοήτος ἀποκλίναντ᾽ ἐπὶ δόξαν, etc.
  Compare Porphyr. in Vit. Pythag. c. 41.

  [47] Aristophan. Ran. 1030.—

    Ὀρφεὺς μὲν γὰρ τελετάς θ᾽ ἡμῖν κατέδειξε, φόνων τ᾽ ἀπέχεσθαι·
    Μουσαῖος τ᾽, ἐξακέσεις τε νόσων καὶ χρησμούς· Ἡσίοδος δὲ,
    Γῆς ἐργασίας, καρπῶν ὥρας, ἀρότους· ὁ δὲ θεῖος Ὅμηρος
    Ἀπὸ τοῦ τιμὴν καὶ κλέος ἔσχεν, πλὴν τοῦθ᾽, ὅτι χρήστ᾽ ἐδίδασκεν,
    Ἀρετὰς, τάξεις, ὁπλίσεις ἀνδρῶν; etc.

  The same general contrast is to be found in Plato, Protagoras,
  p. 316; the opinion of Pausanias, ix. 30, 4. The poems of Musæus
  seem to have borne considerable analogy to the Melampodia
  ascribed to Hesiod (see Clemen. Alex. Str. vi. p. 628); and
  healing charms are ascribed to Orpheus as well as to Musæus. See
  Eurip. Alcestis, 986.

  [48] Herod. ii. 81; Euripid. Hippol. 957, and the curious
  fragment of the lost Κρῆτες of Euripidês. Ὀρφικοὶ βίοι, Plato,
  Legg. vii. 782.

Festivals at seed-time and harvest—at the vintage and at the opening
of the new wine—were doubtless coeval with the earliest habits of the
Greeks; the latter being a period of unusual joviality. Yet in the
Homeric poems, Dionysos and Dêmêtêr, the patrons of the vineyard and
the cornfield, are seldom mentioned, and decidedly occupy little place
in the imagination of the poet as compared with the other gods: nor are
they of any conspicuous importance even in the Hesiodic Theogony. But
during the interval between Hesiod and Onomakritus, the revolution in
the religious mind of Greece was such as to place both these deities
in the front rank. According to the Orphic doctrine, Zagreus, son of
Persephonê, is destined to be the successor of Zeus, and although the
violence of the Titans intercepts this lot, yet even when he rises
again from his discerption under the name of Dionysos, he is the
colleague and coëqual of his divine father.

This remarkable change, occurring as it did during the sixth and a
part of the seventh century before the Christian æra, may be traced
to the influence of communication with Egypt (which only became
fully open to the Greeks about B. C. 660), as well as with Thrace,
Phrygia, and Lydia. From hence new religious ideas and feelings were
introduced, which chiefly attached themselves to the characters of
Dionysos and Dêmêtêr. The Greeks identified these two deities with the
great Egyptian Osiris and Isis, so that what was borrowed from the
Egyptian worship of the two latter naturally fell to their equivalents
in the Grecian system.[49] Moreover the worship of Dionysos (under
what name cannot be certainly made out) was indigenous in Thrace,[50]
as that of the Great Mother was in Phyrgia, and in Lydia—together
with those violent ecstasies and manifestations of temporary frenzy,
and that clashing of noisy instruments, which we find afterwards
characterizing it in Greece. The great masters of the pipe—as well as
the dythyramb,[51] and indeed the whole musical system appropriated to
the worship of Dionysos, which contrasted so pointedly with the quiet
solemnity of the Pæan addressed to Apollo—were all originally Phrygian.

  [49] Herodot. ii. 42, 59, 144.

  [50] Herodot. v. 7, vii. 111; Euripid. Hecub. 1249, and Rhêsus,
  969, and the Prologue to the Bacchæ; Strabo, x. p. 470; Schol.
  ad Aristophan. Aves, 874; Eustath. ad Dionys. Perieg. 1069;
  Harpocrat. v. Σάβοι; Photius, Εὐοῖ Σαβοῖ. The “Lydiaca” of Th.
  Menke (Berlin, 1843) traces the early connection between the
  religion of Dionysos and that of Cybelê, c. 6, 7. Hoeckh’s Krêta
  (vol. i. p. 128-134) is instructive respecting the Phrygian
  religion.

  [51] Aristotle, Polit. viii. 7, 9. Πᾶσα γὰρ Βάκχεια καὶ πᾶσα ἡ
  τοιαύτη κίνησις μάλιστα τῶν ὀργάνων ἐστὶν ἐν τοῖς αὐλοῖς· τῶν δ᾽
  ἁρμονίων ἐν τοῖς Φρυγιστὶ μέλεσι λαμβάνει ταῦτα τὸ πρέπον, οἷον ὁ
  διθύραμβος δοκεῖ ὁμολογουμένως εἶναι Φρύγιον. Eurip. Bacch. 58.—

    Αἴρεσθε τἀπιχώρι᾽ ἐν πόλει Φρυγῶν
    Τύμπανα, Ῥέας τε μητρὸς ἐμὰ θ᾽ εὑρήματα, etc.

  Plutarch, Εἰ. in Delph. c. 9; Philochor. Fr. 21, ed. Didot,
  p. 389. The complete and intimate manner in which Euripidês
  identifies the Bacchic rites of Dionysos with the Phrygian
  ceremonies in honor of the Great Mother, is very remarkable. The
  fine description given by Lucretius (ii. 600-640) of the Phrygian
  worship is much enfeebled by his unsatisfactory allegorizing.

From all these various countries, novelties, unknown to the Homeric
men, found their way into the Grecian worship: and there is one amongst
them which deserves to be specially noticed, because it marks the
generation of the new class of ideas in their theology. Homer mentions
many persons guilty of private or involuntary homicide, and compelled
either to go into exile or to make pecuniary satisfaction; but he
never once describes any of them to have either received or required
purification for the crime.[52] Now in the time subsequent to Homer,
purification for homicide comes to be considered as indispensable:
the guilty person is regarded as unfit for the society of man or the
worship of the gods until he has received it, and special ceremonies
are prescribed whereby it is to be administered. Herodotus tells us
that the ceremony of purification was the same among the Lydians and
among the Greeks:[53] we know that it formed no part of the early
religion of the latter, and we may perhaps reasonably suspect that
they borrowed it from the former. The oldest instance known to us of
expiation for homicide was contained in the epic poem of the Milesian
Arktinus,[54] wherein Achillês is purified by Odysseus for the
murder of Thersitês: several others occurred in the later or Hesiodic
epic—Hêraklês, Pêleus, Bellerophôn, Alkmæôn, Amphiktyôn, Pœmander,
Triopas,—from whence they probably passed through the hands of the
logographers to Apollodôrus, Diodôrus, and others.[55] The purification
of the murderer was originally operated, not by the hands of any priest
or specially sanctified man, but by those of a chief or king, who goes
through the appropriate ceremonies in the manner recounted by Herodotus
in his pathetic narrative respecting Crœsus and Adrastus.

  [52] Schol. ad Iliad, xi. 690—οὐ διὰ τὰ καθάρσια Ἰφίτου πορθεῖται
  ἡ Πύλος, ἐπεί τοι Ὀδυσσεὺς μείζων Νέστορος, καὶ παρ᾽ Ὁμήρῳ οὐκ
  οἴδαμεν φονέα καθαιρόμενον, ἀλλ᾽ ἀντιτίνοντα ἢ φυγαδευόμενον.
  The examples are numerous, and are found both in the Iliad and
  the Odyssey. Iliad, ii. 665 (_Tlêpolemos_); xiii. 697 (_Medôn_);
  xiii. 574 (_Epeigeus_); xxiii. 89 (_Patroclos_); Odyss. xv.
  224 (_Theoclymenos_); xiv. 380 (an _Ætôlian_). Nor does the
  interesting mythe respecting the functions of Atê and the Litæ
  harmonize with the subsequent doctrine about the necessity of
  purification. (Iliad, ix. 498).

  [53] Herodot. i. 35—ἔστι δὲ παραπλησίη ἡ κάθαρσις τοῖσι Λυδοῖσι
  καὶ τοῖσι Ἕλλησι. One remarkable proof, amongst many, of the deep
  hold which this idea took of the greatest minds in Greece, that
  serious mischief would fall upon the community if family quarrels
  or homicide remained without religious expiation, is to be found
  in the objections which Aristotle urges against the community of
  women proposed in the Platonic Republic. It could not be known
  what individuals stood in the relation of father, son or brother:
  if, therefore, wrong or murder of kindred should take place, the
  appropriate religious atonements (αἱ νομιζόμεναι λύσεις) could
  not be applied, and the crime would go unexpiated. (Aristot.
  Polit. ii. 1, 14. Compare Thucyd. i. 125-128).

  [54] See the Fragm. of the Æthiopis of Arktinus, in Düntzer’s
  Collection, p. 16.

  [55] The references for this are collected in Lobeck’s
  Aglaophamos. Epimetr. ii. ad Orphica, p. 968.

The idea of a special taint of crime, and of the necessity as well
as the sufficiency of prescribed religious ceremonies as a means of
removing it, appears thus to have got footing in Grecian practice
subsequent to the time of Homer. The peculiar rites or orgies,
composed or put together by Onomakritus, Methapus,[56] and other men
of more than the ordinary piety, were founded upon a similar mode
of thinking, and adapted to the same mental exigencies. They were
voluntary religious manifestations, superinduced upon the old public
sacrifices of the king or chiefs on behalf of the whole society, and
of the father on his own family hearth—they marked out the details of
divine service proper to appease or gratify the god to whom they were
addressed, and to procure for the believers who went through them his
blessings and protection here or hereafter—the exact performance of
the divine service in all its specialty was held necessary, and thus
the priests or Hierophants, who alone were familiar with the ritual,
acquired a commanding position.[57] Generally speaking, these peculiar
orgies obtained their admission and their influence at periods of
distress, disease, public calamity and danger, or religious terror
and despondency, which appear to have been but too frequent in their
occurrence.

  [56] Pausanias (iv. 1, 5)—μετεκόσμησε γὰρ καὶ Μέθαπος τῆς
  τελετῆς (the Eleusinian Orgies, carried by Kaukon from Eleusis
  into Messênia), ἔστιν ἅ. Ὁ δὲ Μέθαπος γένος μὲν ἦν Ἀθηναῖος,
  τελετῆς τε καὶ ~ὀργίων παντοίων συνθέτης~. Again, viii. 37,
  3, Onomakritus Διονύσῳ ~συνέθηκεν~ ὄργια, etc. This is another
  expression designating the same idea as the Rhêsus of Euripidês,
  944.—

    Μυστηρίων τε τῶν ἀποῤῥήτων φάνας
    Ἔδειξεν Ὀρφεύς.

  [57] Têlinês, the ancestor of the Syracusan despot Gelô,
  acquired great political power as possessing τὰ ἱρὰ τῶν χθονίων
  θεῶν (Herodot. vii. 153); he and his family became hereditary
  Hierophants of these ceremonies. How Têlinês acquired the ἱρὰ
  Herodotus cannot say—ὅθεν δὲ ἀυτὰ ἔλαβε, ἢ αὐτὸς ἐκτήσατο, τοῦτο
  οὐκ ἔχω εἶπαι. Probably there was a traditional legend, not
  inferior in sanctity to that of Eleusis, tracing them to the gift
  of Dêmêtêr herself.

The minds of men were prone to the belief that what they were suffering
arose from the displeasure of some of the gods, and as they found
that the ordinary sacrifices and worship were insufficient for their
protection, so they grasped at new suggestions proposed to them with
the view of regaining the divine favor.[58] Such suggestions were more
usually copied, either in whole or in part, from the religious rites
of some foreign locality, or from some other portion of the Hellenic
world; and in this manner many new sects or voluntary religious
fraternities, promising to relieve the troubled conscience and to
reconcile the sick or suffering with the offended gods, acquired
permanent establishment as well as considerable influence. They were
generally under the superintendence of hereditary families of priests,
who imparted the rites of confirmation and purification to communicants
generally; no one who went through the prescribed ceremonies being
excluded. In many cases, such ceremonies fell into the hands of
jugglers, who volunteered their services to wealthy men, and degraded
their profession as well by obtrusive venality as by extravagant
promises:[59] sometimes the price was lowered to bring them within
reach of the poor and even of slaves. But the wide diffusion, and the
number of voluntary communicants of these solemnities, proves how much
they fell in with the feeling of the time and how much respect they
enjoyed—a respect, which the more conspicuous establishments, such
as Eleusis and Samothrace, maintained for several centuries. And the
visit of the Kretan Epimenidês to Athens—in the time of Solôn, and at a
season of the most serious disquietude and dread of having offended the
gods—illustrates the tranquillizing effect of new orgies[60] and rites
of absolution, when enjoined by a man standing high in the favor of
the gods and reputed to be the son of a nymph. The supposed Erythræan
Sibyl, and the earliest collection of Sibylline prophecies,[61]
afterwards so much multiplied and interpolated, and referred (according
to Grecian custom) to an age even earlier than Homer, appear to belong
to a date not long posterior to Epimenidês. Other oracular verses, such
as those of Bakis, were treasured up in Athens and other cities: the
sixth century before the Christian æra was fertile in these kinds of
religious manifestations.

  [58] See Josephus cont. Apiôn. ii. c. 35; Hesych. Θεοὶ ξένιοι;
  Strabo, x. p. 471; Plutarch, Περὶ Δεισιδαιμον. c. iii. p. 166; c.
  vii. p. 167.

  [59] Plato, Republ. ii. p. 364; Demosthen. de Coronâ, c. 79,
  p. 313. The δεισιδαίμων of Theophrastus cannot be comfortable
  without receiving the Orphic communion monthly from the
  Orpheotelestæ (Theophr. Char. xvi.). Compare Plutarch, Περὶ τοῦ
  μὴ χρᾶν ἔμμετρα, etc., c. 25, p. 400. The comic writer Phrynichus
  indicates the existence of these rites of religious excitement,
  at Athens, during the Peloponnesian war. See the short fragment
  of his Κρόνος, ap. Schol. Aristoph. Aves, 989—

    Ἀνὴρ χορεύει, καὶ τὰ τοῦ καλῶς·
    Βούλει Διοπείθη μεταδράμω καὶ τύμπανα;

  Diopeithês was a χρησμόλογος, or collector and deliverer of
  prophecies, which he sung (or rather, perhaps, recited) with
  solemnity and emphasis, in public. ὥστε ποιοῦντες χρησμοὺς αὐτοὶ
  Διδόασ᾽ ᾄδειν Διοπείθει τῷ παραμαινομένῷ. (Ameipsias ap. Schol.
  Aristophan. _ut sup._, which illustrates Thucyd. ii. 21).

  [60] Plutarch, Solôn, c. 12; Diogen. Laërt. i. 110.

  [61] See Klausen, “Æneas und die Penaten:” his chapter on the
  connection between the Grecian and Roman Sibylline collections
  is among the most ingenious of his learned book. Book ii. pp.
  210-240; see Steph. Byz. v. Γέργις.

  To the same age belong the χρησμοὶ and καθαρμοὶ of Abaris and his
  marvellous journey through the air upon an arrow (Herodot. iv.
  36).

  Epimenidês also composed καθαρμοὶ in epic verse; his Κουρήτων
  and Κορυβάντων γένεσις, and his four thousand verses respecting
  Minôs and Rhadamanthys, if they had been preserved, would let us
  fully into the ideas of a religious mystic of that age respecting
  the antiquities of Greece. (Strabo, x. p. 474; Diogen. Laërt. i.
  10). Among the poems ascribed to Hesiod were comprised not only
  the Melampodia, but also ἔπη μαντικὰ and ἐξηγήσεις ἐπὶ τέρασιν.
  Pausan. ix. 31, 4.

Amongst the special rites and orgies of the character just described,
those which enjoyed the greatest Pan-Hellenic reputation were attached
to the Idæan Zeus in Krête, to Dêmêtêr at Eleusis, to the Kabeiri in
Samothrace, and to Dionysos at Delphi and Thebes.[62] That they were
all to a great degree analogous, is shown by the way in which they
unconsciously run together and become confused in the minds of various
authors: the ancient inquirers themselves were unable to distinguish
one from the other, and we must be content to submit to the like
ignorance. But we see enough to satisfy us of the general fact, that
during the century and a half which elapsed between the opening of
Egypt to the Greeks and the commencement of their struggle with the
Persian kings, the old religion was largely adulterated by importations
from Egypt, Asia Minor,[63] and Thrace. The rites grew to be more
furious and ecstatic, exhibiting the utmost excitement, bodily as well
as mental: the legends became at once more coarse, more tragical, and
less pathetic. The manifestations of this frenzy were strongest among
the women, whose religious susceptibilities were often found extremely
unmanageable,[64] and who had everywhere congregative occasional
ceremonies of their own, apart from the men—indeed, in the case of the
colonists, especially of the Asiatic colonists, the women had been
originally women of the country, and as such retained to a great degree
their non-Hellenic manners and feelings.[65] The god Dionysos,[66]
whom the legends described as clothed in feminine attire, and leading
a troop of frenzied women, inspired a temporary ecstasy, and those who
resisted the inspiration, being supposed to disobey his will, were
punished either by particular judgments or by mental terrors; while
those who gave full loose to the feeling, in the appropriate season and
with the received solemnities, satisfied his exigencies, and believed
themselves to have procured immunity from such disquietudes for the
future.[67] Crowds of women, clothed with fawn-skins and bearing the
sanctified thyrsus, flocked to the solitudes of Parnassus, or Kithærôn,
or Taygetus, during the consecrated triennial period, passed the
night there with torches, and abandoned themselves to demonstrations
of frantic excitement, with dancing and clamorous invocation of the
god: they were said to tear animals limb from limb, to devour the raw
flesh, and to cut themselves without feeling the wound.[68] The men
yielded to a similar impulse by noisy revels in the streets, sounding
the cymbals and tambourine, and carrying the image of the god in
procession.[69] It deserves to be remarked, that the Athenian women
never practised these periodical mountain excursions, so common among
the rest of the Greeks: they had their feminine solemnities of the
Thesmophoria,[70] mournful in their character and accompanied with
fasting, and their separate congregations at the temples of Aphroditê,
but without any extreme or unseemly demonstrations. The state festival
of the Dionysia, in the city of Athens, was celebrated with dramatic
entertainments, and the once rich harvest of Athenian tragedy and
comedy was thrown up under its auspices. The ceremonies of the Kurêtes
in Krête, originally armed dances in honor of the Idæan Zeus, seem also
to have borrowed from Asia so much of fury, of self-infliction, and
of mysticism, that they became at last inextricably confounded with
the Phrygian Korybantes or worshippers of the Great Mother; though it
appears that Grecian reserve always stopped short of the irreparable
self-mutilation of Atys.

  [62] Among other illustrations of this general resemblance, may
  be counted an epitaph of Kallimachus upon an aged priestess,
  who passed from the service of Dêmêtêr to that of the Kabeiri,
  then to that of Cybelê, having the superintendence of many young
  women. Kallimachus, Epigram. 42. p. 308, ed. Ernest.

  [63] Plutarch, (Defect. Oracul. c. 10, p. 415) treats these
  countries as the original seat of the worship of Dæmons (wholly
  or partially bad, and intermediate between gods and men), and
  their religious ceremonies as of a corresponding character: the
  Greeks were borrowers from them, according to him, both of the
  doctrine and of the ceremonies.

  [64] Strabo, vii. p. 297. Ἅπαντες γὰρ τῆς δεισιδαιμονίας ἀρχηγοὺς
  οἴονται τὰς γυναῖκας· αὐταὶ δὲ καὶ τοὺς ἄνδρας προκαλοῦνται ἐς
  τὰς ἐπὶ πλέον θεραπείας τῶν θεῶν, καὶ ἑορτὰς, καὶ ποτνιασμούς.
  Plato (De Legg. x. pp. 909, 910) takes great pains to restrain
  this tendency on the part of sick or suffering persons,
  especially women, to introduce new sacred rites into his city.

  [65] Herodot. i. 146. The wives of the Ionic original settlers at
  Miletos were Karian women, whose husbands they slew.

  The violences of the Karian worship are attested by what
  Herodotus says of the Karian residents in Egypt, at the festival
  of Isis at Busiris. The Egyptians at this festival manifested
  their feeling by beating themselves, the Karians by cutting their
  faces with knives (ii. 61). The Καρικὴ μοῦσα became proverbial
  for funeral wailings (Plato, Legg. vii. p. 800): the unmeasured
  effusions and demonstrations of sorrow for the departed, some
  times accompanied by cutting and mutilation self-inflicted by the
  mourner was a distinguishing feature in Asiatics and Egyptians
  as compared with Greeks. Plutarch, Consolat. ad Apollôn. c. 22,
  p. 123. Mournful feeling was, in fact, a sort of desecration of
  the genuine and primitive Grecian festival, which was a season
  of cheerful harmony and social enjoyment, wherein the god was
  believed to sympathize (εὐφροσύνη). See Xenophanes ap. Aristot.
  Rhetor. ii. 25; Xenophan. Fragm. 1. ed. Schneidewin; Theognis,
  776; Plutarch, De Superstit. p. 169. The unfavorable comments
  of Dionysius of Halicarnassus, in so far as they refer to the
  festivals of Greece, apply to the foreign corruptions, not to the
  native character, of Grecian worship.

  [66] The Lydian Hêraklês was conceived and worshipped as a man in
  female attire: this idea occurs often in the Asiatic religions.
  Mencke, Lydiaca, c. 8, p. 22. Διόνυσος ἄῤῥην καὶ θῆλυς. Aristid.
  Or. iv. p. 28; Æschyl. Fragm. Edoni, ap. Aristoph. Thesmoph. 135.
  Ποδαπὸς ὁ γύννις; τίς πάτρα; τίς ἡ στολή;

  [67] Melampos cures the women (whom Dionysos has struck mad for
  their resistance to his rites), παραλαβὼν τοὺς δυνατωτάτους τῶν
  νεανίων μετ᾽ ἀλαλαγμοῦ καί τινος ἐνθέου χορείας. Apollodôr. ii.
  2, 7. Compare Eurip. Bacch. 861.

  Plato (Legg. vii. p. 790) gives a similar theory of the
  healing effect of the Korybantic rites, which cured vague and
  inexplicable terrors of the mind by means of dancing and music
  conjoined with religious ceremonies—αἱ τὰ τῶν Κορυβάντων ἰάματα
  τελοῦσαι (the practitioners were women), αἱ τῶν ἐκφρόνων Βακχείων
  ἰάσεις—ἡ τῶν ἔξωθεν κρατεῖ κίνησις προσφερομένη τὴν ἐντὸς φοβερὰν
  οὖσαν καὶ μανικὴν κίνησιν—ὀρχουμένους δὲ καὶ αὐλουμένους μετὰ
  θεῶν, οἷς ἂν καλλιερήσαντες ἕκαστοι θύωσιν, κατειργάσατο ἀντὶ
  μανικῶν ἡμῖν διαθέσεων ἕξεις ἔμφρονας ἔχειν.

  [68] Described in the Bacchæ of Euripidês (140, 735, 1135,
  etc.). Ovid, Trist. iv. i. 41. “Utque suum Bacchis non sentit
  saucia vulnus, Cum furit Edonis exululata jugis.” In a fragment
  of the poet Alkman, a Lydian by birth, the Bacchanal nymphs are
  represented as milking the lioness, and making cheese of the
  milk, during their mountain excursions and festivals. (Alkman.
  Fragm. 14. Schn. Compare Aristid. Orat. iv. p. 29). Clemens
  Alexand. Admonit. ad Gent. p. 9, Sylb.; Lucian, Dionysos, c. 3,
  T. iii. p. 77, Hemsterh.

  [69] See the tale of Skylês in Herod. iv. 79, and Athenæus, x. p.
  445. Herodotus mentions that the Scythians abhorred the Bacchic
  ceremonies, accounting the frenzy which belonged to them to be
  disgraceful and monstrous.

  [70] Plutarch, De Isid. et Osir. c. 69, p. 378; Schol. ad
  Aristoph. Thesmoph. There were however Bacchic ceremonies
  practised to a certain extent by the Athenian women. (Aristoph.
  Lysist. 388).

The influence of the Thracian religion upon that of the Greeks cannot
be traced in detail, but the ceremonies contained in it were of a
violent and fierce character, like the Phrygian, and acted upon Hellas
in the same general direction as the latter. And the like may be said
of the Egyptian religion, which was in this case the more operative,
inasmuch as all the intellectual Greeks were naturally attracted to go
and visit the wonders on the banks of the Nile; the powerful effect
produced upon them is attested by many evidences, but especially by
the interesting narrative of Herodotus. Now the Egyptian ceremonies
were at once more licentious, and more profuse in the outpouring both
of joy and sorrow, than the Greek;[71] but a still greater difference
sprang from the extraordinary power, separate mode of life, minute
observances, and elaborate organization, of the priesthood. The
ceremonies of Egypt were multitudinous, but the legends concerning them
were framed by the priests, and as a general rule, seemingly, known
to the priests alone: at least they were not intended to be publicly
talked of, even by pious men. They were “holy stories,” which it was
sacrilege publicly to mention, and which from this very prohibition
only took firmer hold of the minds of the Greek visitors who heard
them. And thus the element of secrecy and mystic silence—foreign to
Homer, and only faintly glanced at in Hesiod—if it was not originally
derived from Egypt, at least received from thence its greatest
stimulus and diffusion. The character of the legends themselves was
naturally affected by this change from publicity to secrecy: the
secrets when revealed would be such as to justify by their own tenor
the interdict on public divulgation: instead of being adapted, like
the Homeric mythe, to the universal sympathies and hearty interest of
a crowd of hearers, they would derive their impressiveness from the
tragical, mournful, extravagant, or terror-striking character of the
incidents.[72] Such a tendency, which appears explicable and probable
even on general grounds, was in this particular case rendered still
more certain by the coarse taste of the Egyptian priests. That any
recondite doctrine, religious or philosophical, was attached to the
mysteries or contained in the holy stories, has never been shown,
and is to the last degree improbable though the affirmative has been
asserted by many learned men.

  [71] “Ægyptiaca numina fere plangoribus gaudent, Græca plerumque
  choreis, barbara autem strepitu cymbalistarum et tympanistarum
  et choraularum.” (Apuleius, De Genio Socratis, v. ii. p. 149,
  Oudend).

  [72] The legend of Dionysos and Prosymnos, as it stands in
  Clemens, could never have found place in an epic poem (Admonit.
  ad Gent. p. 22, Sylb.). Compare page 11 of the same work,
  where however he so confounds together Phrygian, Bacchic, and
  Eleusinian mysteries, that one cannot distinguish them apart.

  Demêtrius Phalêreus says about the legends belonging to
  these ceremonies—Διὸ καὶ τὰ μυστήρια λέγεται ἐν ἀλληγορίαις
  ~πρὸς ἔκπληξιν καὶ φρίκην~, ὥσπερ ἐν σκότῳ καὶ νυκτί. (De
  Interpretatione, c. 101).

Herodotus seems to have believed that the worship and ceremonies of
Dionysos generally were derived by the Greeks from Egypt, brought
over by Kadmus and taught by him to Melampus: and the latter appears
in the Hesiodic Catalogue as having cured the daughters of Prœtus of
the mental distemper with which they had been smitten by Dionysos for
rejecting his ritual. He cured them by introducing the Bacchic dance
and fanatical excitement: this mythical incident is the most ancient
mention of the Dionysiac solemnities presented in the same character
as they bear in Euripidês. It is the general tendency of Herodotus
to apply the theory of derivation from Egypt far too extensively to
Grecian institutions: the orgies of Dionysos were not originally
borrowed from thence, though they may have been much modified by
connection with Egypt as well as with Asia. The remarkable mythe
composed by Onomakritus respecting the dismemberment of Zagreus was
founded upon an Egyptian tale very similar respecting the body of
Osiris, who was supposed to be identical with Dionysos:[73] nor was it
unsuitable to the reckless fury of the Bacchanals during their state of
temporary excitement, which found a still more awful expression in the
mythe of Pentheus,—torn in pieces by his own mother Agavê at the head
of her companions in the ceremony, as an intruder upon the feminine
rites as well as a scoffer at the god.[74] A passage in the Iliad (the
authenticity of which has been contested, but even as an interpolation
it must be old)[75] also recounts how Lykurgus was struck blind by Zeus
for having chased away with a whip “the nurses of the mad Dionysos,”
and frightened the god himself into the sea to take refuge in the arms
of Thetis: and the fact, that Dionysos is so frequently represented in
his mythes as encountering opposition and punishing the refractory,
seems to indicate that his worship under its ecstatic form was a late
phænomenon and introduced not without difficulty. The mythical Thracian
Orpheus was attached as Eponymos to a new sect, who seem to have
celebrated the ceremonies of Dionysos with peculiar care, minuteness
and fervor, besides observing various rules in respect to food and
clothing. It was the opinion of Herodotus, that these rules, as well as
the Pythagorean, were borrowed from Egypt. But whether this be the fact
or not, the Orphic brotherhood is itself both an evidence, and a cause,
of the increased importance of the worship of Dionysos, which indeed is
attested by the great dramatic poets of Athens.

  [73] See the curious treatise of Plutarch, De Isid. et Osirid.
  c. 11-14. p. 356, and his elaborate attempt to allegorize the
  legend. He seems to have conceived that the Thracian Orpheus had
  first introduced into Greece the mysteries both of Dêmêtêr and
  Dionysos, copying them from those of Isis and Osiris in Egypt.
  See Fragm. 84, from one of his lost works, tom, v. p. 891, ed.
  Wyttenb.

  [74] Æschylus had dramatized the story of Pentheus as well as
  that of Lykurgus: one of his tetralogies was the Lykurgeia
  (Dindorf, Æsch. Fragm. 115). A short allusion to the story of
  Pentheus appears in Eumenid. 25. Compare Sophocl. Antigon. 985,
  and the Scholia.

  [75] Iliad, vi. 130. See the remarks of Mr. Payne Knight _ad loc._

The Homeric Hymns present to us, however, the religious ideas and
legends of the Greeks at an earlier period, when the enthusiastic and
mystic tendencies had not yet acquired their full development. Though
not referable to the same age or to the same author as either the
Iliad or the Odyssey, they do to a certain extent continue the same
stream of feeling, and the same mythical tone and coloring, as these
poems—manifesting but little evidence of Egyptian, Asiatic, or Thracian
adulterations. The difference is striking between the god Dionysos
as he appears in the Homeric hymn and in the Bacchæ of Euripidês.
The hymnographer describes him as standing on the sea-shore, in the
guise of a beautiful and richly-clothed youth, when Tyrrhenian pirates
suddenly approach: they seize and bind him and drag him on board
their vessel. But the bonds which they employ burst spontaneously,
and leave the god free. The steersman, perceiving this with affright,
points out to his companions that they have unwittingly laid hands
on a god,—perhaps Zeus himself, or Apollo, or Poseidôn. He conjures
them to desist, and to replace Dionysos respectfully on the shore,
lest in his wrath he should visit the ship with wind and hurricane:
but the crew deride his scruples, and Dionysos is carried prisoner
out to sea with the ship under full sail. Miraculous circumstances
soon attest both his presence and his power. Sweet-scented wine is
seen to flow spontaneously about the ship, the sail and mast appear
adorned with vine and ivy-leaves, and the oar-pegs with garlands. The
terrified crew now too late entreat the helmsman to steer his course
for the shore, and crowd round him for protection on the poop. But
their destruction is at hand: Dionysos assumes the form of a lion—a
bear is seen standing near him—this bear rushes with a loud roar upon
the captain, while the crew leap overboard in their agony of fright,
and are changed into dolphins. There remains none but the discreet
and pious steersman, to whom Dionysos addresses words of affectionate
encouragement, revealing his name, parentage and dignity.[76]

  [76] See Homer, Hymn 5, Διόνυσος ἢ Λῆσται.—The satirical drama
  of Euripidês, the Cyclôps, extends and alters this old legend.
  Dionysos is carried away by the Tyrrhenian pirates, and Silênus
  at the head of the Bacchanals goes everywhere in search of him
  (Eur. Cyc. 112). The pirates are instigated against him by the
  hatred of Hêrê, which appears frequently as a cause of mischief
  to Dionysos (Bacchæ, 286). Hêrê in her anger had driven him mad
  when a child, and he had wandered in this state over Egypt and
  Syria; at length he came to Cybela in Phrygia, was purified
  (καθαρθεὶς) by Rhea, and received from her female attire
  (Apollodôr. iii. 5, 1, with Heyne’s note). This seems to have
  been the legend adopted to explain the old verse of the Iliad, as
  well as the maddening attributes of the god generally.

  There was a standing antipathy between the priestesses and the
  religious establishments of Hêrê and Dionysos (Plutarch, Περὶ
  τῶν ἐν Πλαταίαις Δαιδάλων, c. 2, tom. v. p. 755, ed. Wytt).
  Plutarch ridicules the legendary reason commonly assigned for
  this, and provides a symbolical explanation which he thinks very
  satisfactory.

This hymn, perhaps produced at the Naxian festival of Dionysos,
and earlier than the time when the dithyrambic chorus became the
established mode of singing the praise and glory of that god, is
conceived in a spirit totally different from that of the Bacchic
Telatæ, or special rites which the Bacchæ of Euripidês so abundantly
extol,—rites introduced from Asia by Dionysos himself at the head of a
thiasus or troop of enthusiastic women,—inflaming with temporary frenzy
the minds of the women of Thebes,—not communicable except to those
who approach as pious communicants,—and followed by the most tragical
results to all those who fight against the god.[77] The Bacchic Teletæ,
and the Bacchic feminine frenzy, were importations from abroad,
as Euripidês represents them, engrafted upon the joviality of the
primitive Greek Dionysia; they were borrowed, in all probability, from
more than one source and introduced through more than one channel, the
Orphic life or brotherhood being one of the varieties. Strabo ascribes
to this latter a Thracian original, considering Orpheus, Musæus, and
Eumolpus as having been all Thracians.[78] It is curious to observe
how, in the Bacchæ of Euripidês, the two distinct and even conflicting
ideas of Dionysos come alternately forward; sometimes the old Grecian
idea of the jolly and exhilarating god of wine—but more frequently
the recent and imported idea of the terrific and irresistible god
who unseats the reason, and whose _œstrus_ can only be appeased by a
willing, though temporary obedience. In the fanatical impulse which
inspired the votaries of the Asiatic Rhea or Cybelê, or of the Thracian
Kotys, there was nothing of spontaneous joy; it was a sacred madness,
during which the soul appeared to be surrendered to a stimulus from
without, and accompanied by preternatural strength and temporary sense
of power,[79]—altogether distinct from the unrestrained hilarity of the
original Dionysia, as we see them in the rural demes of Attica, or in
the gay city of Tarentum. There was indeed a side on which the two bore
some analogy, inasmuch as, according to the religious point of view of
the Greeks, even the spontaneous joy of the vintage feast was conferred
by the favor and enlivened by the companionship of Dionysos. It was
upon this analogy that the framers of the Bacchic orgies proceeded
but they did not the less disfigure the genuine character of the old
Grecian Dionysia.

  [77] Eurip. Bacch. 325, 464, etc.

  [78] Strabo, x. p. 471. Compare Aristid. Or. iv. p. 28.

  [79] In the lost _Xantriæ_ of Æschylus, in which seems to
  have been included the tale of Pentheus, the goddess Λύσσα
  was introduced, stimulating the Bacchæ, and creating in them
  spasmodic excitement from head to foot: ἐκ ποδῶν δ᾽ ἄνω Ὑπέρχεται
  σπαραγμὸς εἰς ἄκρον κάρα, etc. (Fragm. 155, Dindorf). His tragedy
  called _Edoni_ also gave a terrific representation of the
  Bacchanals and their fury, exaggerated by the maddening music:
  Πίμπλησι μέλος, Μανίας ἐπαγωγὸν ὁμοκλάν (Fr. 54).

  Such also is the reigning sentiment throughout the greater
  part of the Bacchæ of Euripidês; it is brought out still more
  impressively in the mournful Atys of Catullus:—

    “Dea magna, Dea Cybele, Dindymi Dea, Domina,
    Procul a meâ tuus sit furor omnis, hera, domo:
    Alios age incitatos: alios age rabidos!”

  We have only to compare this fearful influence with the
  description of Dikæopolis and his exuberant joviality in the
  festival of the rural Dionysia (Aristoph. Acharn. 1051 _seq._;
  see also Plato. Legg. i. p. 637), to see how completely the
  foreign innovations recolored the old Grecian Dionysos,—Διόνυσος
  πολυγηθὴς,—who appears also in the scene of Dionysos and Ariadnê
  in the Symposion of Xenophôn, c. 9. The simplicity of the ancient
  Dionysiac processions is dwelt upon by Plutarch, De Cupidine
  Divitiarum, p. 527; and the original dithyramb addressed by
  Archilochus to Dionysos is an effusion of drunken hilarity
  (Archiloch. Frag. 69, Schneid.).

Dionysos is in the conception of Pindar the Paredros or companion in
worship of Dêmêtêr:[80] the worship and religious estimate of the
latter has by that time undergone as great a change as that of the
former, if we take our comparison with the brief description of Homer
and Hesiod: she has acquired[81] much of the awful and soul-disturbing
attributes of the Phrygian Cybelê. In Homer, Dêmêtêr is the goddess
of the corn-field, who becomes attached to the mortal man Jasiôn;
an unhappy passion, since Zeus, jealous of the connection between
goddesses and men, puts him to death. In the Hesiodic Theogony, Dêmêtêr
is the mother of Persephonê by Zeus, who permits Hadês to carry off
the latter as his wife: moreover Dêmêtêr has, besides, by Jasiôn a son
called Plutos, born in Krête. Even from Homer to Hesiod, the legend
of Dêmêtêr, has been expanded and her dignity exalted; according
to the usual tendency of Greek legend, the expansion goes on still
further. Through Jasiôn, Dêmêtêr becomes connected with the mysteries
of Samothrace; through Persephonê, with those of Eleusis. The former
connection it is difficult to follow out in detail, but the latter is
explained and traced to its origin in the Homeric Hymn to Dêmêtêr.

  [80] Pindar, Isthm. vi. 3. χαλκοκρότου πάρεδρον Δημήτερος,—the
  epithet marks the approximation of Dêmêtêr to the Mother of the
  Gods. ᾗ κροτάλων τυπάνων τ᾽ ἰαχὴ, σύν τε βρόμος αὐλῶν Εὔαδεν
  (Homer. Hymn, xiii.),—the Mother of the Gods was worshipped by
  Pindar himself along with Pan; she had in his time her temple and
  ceremonies at Thêbes (Pyth. iii. 78; Fragm. Dithyr. 5, and the
  Scholia _ad l._) as well as, probably, at Athens (Pausan. i. 3,
  3).

  Dionysos and Dêmêtêr are also brought together in the chorus of
  Sophoklês, Antigonê, 1072. μέδεις δὲ παγκοίνοις Ἐλευσινίας Δηοῦς
  ἐν κόλποις; and in Kallimachus, Hymn. Cerer. 70. Bacchus or
  Dionysos are in the Attic tragedians constantly confounded with
  the Dêmêtrian Iacchos, originally so different,—a personification
  of the mystic word shouted by the Eleusinian communicants. See
  Strabo, x. p. 468.

  [81] Euripidês in his Chorus in the Helena (1320 _seq._) assigns
  to Dêmêtêr all the attributes of Rhea, and blends the two
  completely into one.

Though we find different statements respecting the date as well as
the origin of the Eleusinian mysteries, yet the popular belief of
the Athenians, and the story which found favor at Eleusis, ascribed
them to the presence and dictation of the goddess Dêmêtêr herself;
just as the Bacchic rites are, according to the Bacchæ of Euripidês,
first communicated and enforced on the Greeks by the personal visit of
Dionysos to Thêbes, the metropolis of the Bacchic ceremonies.[82] In
the Eleusinian legend, preserved by the author of the Homeric Hymn,
she comes voluntarily and identifies herself with Eleusis; her past
abode in Krête being briefly indicated.[83] Her visit to Eleusis is
connected with the deep sorrow caused by the loss of her daughter
Persephonê, who had been seized by Hadês, while gathering flowers in
a meadow along with the Oceanic Nymphs, and carried off to become his
wife in the under-world. In vain did the reluctant Persephonê shriek
and invoke the aid of her father Zeus: he had consented to give her to
Hadês, and her cries were heard only by Hekatê and Hêlios. Dêmêtêr was
inconsolable at the disappearance of her daughter, but knew not where
to look for her: she wandered for nine days and nights with torches in
search of the lost maiden without success. At length Hêlios, the “spy
of gods and men,” revealed to her, in reply to her urgent prayer, the
rape of Persephonê, and the permission given to Hadês by Zeus. Dêmêtêr
was smitten with anger and despair: she renounced Zeus and the society
of Olympus, abstained from nectar and ambrosia, and wandered on earth
in grief and fasting until her form could no longer be known. In this
condition she came to Eleusis, then governed by the prince Keleos.
Sitting down by a well at the wayside in the guise of an old woman, she
was found by the daughters of Keleos, who came hither with their pails
of brass for water. In reply to their questions, she told them that she
had been brought by pirates from Krête to Thorikos, and had made her
escape; she then solicited from them succor and employment as a servant
or as a nurse. The damsels prevailed upon their mother Metaneira
to receive her, and to entrust her with the nursing of the young
Dêmophoôn, their late-born brother, the only son of Keleos. Dêmêtêr was
received into the house of Metaneira, her dignified form still borne
down by grief: she sat long silent and could not be induced either to
smile or to taste food, until the maid-servant Iambê, by jests and
playfulness, succeeded in amusing and rendering her cheerful. She would
not taste wine, but requested a peculiar mixture of barley-meal with
water and the herb mint.[84]

  [82] Sophocl. Antigon. Βακχᾶν μητρόπολιν Θήβαν.

  [83] Homer, Hymn. Cerer. 123. The Hymn to Dêmêtêr has been
  translated, accompanied with valuable illustrative notes, by J.
  H. Voss (Heidelb. 1826).

  [84] Homer, Hymn. Cerer. 202-210.

The child Dêmophoôn, nursed by Dêmêtêr, throve and grew up like a
god, to the delight and astonishment of his parents: she gave him no
food, but anointed him daily with ambrosia, and plunged him at night
in the fire like a torch, where he remained unburnt. She would have
rendered him immortal, had she not been prevented by the indiscreet
curiosity and alarm of Metaneira, who secretly looked in at night, and
shrieked with horror at the sight of her child in the fire.[85] The
indignant goddess, setting the infant on the ground, now revealed her
true character to Metaneira: her wan and aged look disappeared, and she
stood confest in the genuine majesty of her divine shape, diffusing
a dazzling brightness which illuminated the whole house. “Foolish
mother,” she said, “thy want of faith has robbed thy son of immortal
life. I am the exalted Dêmêtêr, the charm and comfort both of gods and
men: I was preparing for thy son exemption from death and old age; now
it cannot be but he must taste of both. Yet shall he be ever honored,
since he has sat upon my knee and slept in my arms. Let the people
of Eleusis erect for me a temple and altar on yonder hill above the
fountain; I will myself prescribe to them the orgies which they must
religiously perform in order to propitiate my favor.”[86]

  [85] This story was also told with reference to the Egyptian
  goddess Isis in her wanderings. See Plutarch, De Isid. et Osirid.
  c. 16, p. 357.

  [86] Homer, Hymn. Cerer. 274.—

    Ὄργια δ᾽ αὐτὴ ἐγὼν ὑποθήσομαι, ὡς ἂν ἔπειτα
    Εὐαγέως ἕρδοντες ἐμὸν νόον ἱλάσκησθε.

  The same story is told in regard to the infant Achilles. His
  mother Thetis was taking similar measures to render him immortal,
  when his father Pêleus interfered and prevented the consummation.
  Thetis immediately left him in great wrath (Apollôn. Rhod. iv.
  866).

The terrified Metaneira was incapable even of lifting up her child
from the ground; her daughters entered at her cries, and began to
embrace and tend their infant brother, but he sorrowed and could not
be pacified for the loss of his divine nurse. All night they strove to
appease the goddess.[87]

  [87] Homer, Hymn. 290.—

                  τοῦ δ᾽ οὐ μειλίσσετο θυμὸς,
    Χειρότεραι γὰρ δή μιν ἔχον τρόφοι ἠδὲ τιθῆναι.

Strictly executing the injunctions of Dêmêtêr, Keleos convoked the
people of Eleusis and erected the temple on the spot which she had
pointed out. It was speedily completed, and Dêmêtêr took up her abode
in it,—apart from the remaining gods, still pining with grief for the
loss of her daughter, and withholding her beneficent aid from mortals.
And thus she remained a whole year,—a desperate and terrible year:[88]
in vain did the oxen draw the plough, and in vain was the barley-seed
cast into the furrow,—Dêmêtêr suffered it not to emerge from the earth.
The human race would have been starved, and the gods would have been
deprived of their honors and sacrifice, had not Zeus found means to
conciliate her. But this was a hard task; for Dêmêtêr resisted the
entreaties of Iris and of all the other goddesses and gods whom Zeus
successively sent to her. She would be satisfied with nothing less than
the recovery of her daughter. At length Zeus sent Hermês to Hadês, to
bring Persephonê away: Persephonê joyfully obeyed, but Hadês prevailed
upon her before she departed to swallow a grain of pomegranate, which
rendered it impossible for her to remain the whole year away from
him.[89]

  [88] Homer, H. Cer. 305.—

    Αἰνότατον δ᾽ ἐνιαυτὸν ἐπὶ χθόνα πουλυβότειραν
    Ποίησ᾽ ἀνθρώποις, ἰδὲ κύντατον.

  [89] Hymn, v. 375.

With transport did Dêmêtêr receive back her lost daughter, and the
faithful Hekatê sympathized in the delight felt by both at the
reunion.[90] It was now an easier undertaking to reconcile her with
the gods. Her mother Rhea, sent down expressly by Zeus, descended from
Olympus on the fertile Rharan plain, then smitten with barrenness like
the rest of the earth: she succeeded in appeasing the indignation of
Dêmêtêr, who consented again to put forth her relieving hand. The
buried seed came up in abundance, and the earth was covered with fruit
and flowers. She would have wished to retain Persephonê constantly
with her, but this was impossible; and she was obliged to consent that
her daughter should go down for one-third of each year to the house of
Hadês, departing from her every spring at the time when the seed is
sown. She then revisited Olympus, again to dwell with the gods; but
before her departure, she communicated to the daughters of Keleos, and
to Keleos himself, together with Triptolemus, Dioklês and Eumolpus, the
divine service and the solemnities which she required to be observed in
her honor.[91] And thus began the venerable mysteries of Eleusis, at
her special command: the lesser mysteries, celebrated in February, in
honor of Persephonê; the greater, in August, to the honor of Dêmêtêr
herself. Both are jointly patronesses of the holy city and temple.

  [90] Hymn, v. 443.

  [91] Hymn, v. 475.—

            Ἡ δὲ κιοῦσα θεμιστοπόλοις βασιλεῦσι
    Δεῖξεν, Τριπτολέμῳ τε, Διοκλέϊ τε πληξίππῳ,
    Εὐμόλπου τε βίῃ, Κελέῳ θ᾽ ἡγήτορι λαῶν,
    Δρησμοσύνην ἱερῶν· καὶ ἐπέφραδεν ὄργια παισὶν
    Πρεσβυτέρῃς Κελέοιο, etc.

Such is a brief sketch of the temple legend of Eleusis, set forth at
length in the Homeric Hymn to Dêmêtêr. It is interesting not less as a
picture of the Mater Dolorosa (in the mouth of an Athenian, Dêmêtêr and
Persephonê were always the Mother and Daughter, by excellence), first
an agonized sufferer, and then finally glorified,—the weal and woe of
man being dependent upon her kindly feeling,—than as an illustration
of the nature and growth of Grecian legend generally. Though we now
read this Hymn as pleasing poetry, to the Eleusinians, for whom it
was composed, it was genuine and sacred history. They believed in the
visit of Dêmêtêr to Eleusis, and in the mysteries as a revelation from
her, as implicitly as they believed in her existence and power as a
goddess. The Eleusinian psalmist shares this belief in common with his
countrymen, and embodies it in a continuous narrative, in which the
great goddesses of the place, as well as the great heroic families,
figure in inseparable conjunction. Keleos is the son of the Eponymous
hero Eleusis, and his daughters, with the old epic simplicity, carry
their basins to the well for water. Eumolpus, Triptolemus, Dioklês,
heroic ancestors of the privileged families who continued throughout
the historical times of Athens to fulfil their special hereditary
functions in the Eleusinian solemnities, are among the immediate
recipients of inspiration from the goddess; but chiefly does she favor
Metaneira and her infant son Dêmophoôn, for the latter of whom her
greatest boon is destined, and intercepted only by the weak faith of
the mother. Moreover, every incident in the Hymn has a local coloring
and a special reference. The well, overshadowed by an olive-tree near
which Dêmêtêr had rested, the stream Kallichorus and the temple-hill,
were familiar and interesting places in the eyes of every Eleusinian;
the peculiar posset prepared from barley-meal with mint was always
tasted by the Mysts (or communicants) after a prescribed fast, as an
article in the ceremony,—while it was also the custom, at a particular
spot in the processional march, to permit the free interchange of
personal jokes and taunts upon individuals for the general amusement.
And these two customs are connected in the Hymn with the incidents,
that Dêmêtêr herself had chosen the posset as the first interruption of
her long and melancholy fast, and that her sorrowful thoughts had been
partially diverted by the coarse playfulness of the servant-maid Iambê.
In the enlarged representation of the Eleusinian ceremonies, which
became established after the incorporation of Eleusis with Athens, the
part of Iambê herself was enacted by a woman, or man in woman’s attire,
of suitable wit and imagination, who was posted on the bridge over
the Kephissos, and addressed to the passers-by in the procession,[92]
especially the great men of Athens, saucy jeers, probably not less
piercing than those of Aristophanês on the stage. The torch-bearing
Hekatê received a portion of the worship in the nocturnal ceremonies
of the Eleusinia: this too is traced, in the Hymn, to her kind and
affectionate sympathy with the great goddesses.

  [92] Aristophanês, Vesp. 1363. Hesych. v. Γεφυρίς. Suidas,
  v. Γεφυρίζων. Compare about the details of the ceremony,
  Clemens Alexandr. Admon. ad Gent. p. 13. A similar license of
  unrestrained jocularity appears in the rites of Dêmêtêr in Sicily
  (Diodôr. v. 4; see also Pausan. vii. 27, 4), and in the worship
  of Damia and Auxesia at Ægina (Herodot. v. 83).

Though all these incidents were sincerely believed by the Eleusinians
as a true history of the past, and as having been the real initiatory
cause of their own solemnities, it is not the less certain that they
are simply mythes or legends, and not to be treated as history,
either actual or exaggerated. They do not take their start from
realities of the past, but from realities of the present, combined
with retrospective feeling and fancy, which fills up the blank of
the aforetime in a manner at once plausible and impressive. What
proportion of fact there may be in the legend, or whether there be
any at all, it is impossible to ascertain and useless to inquire;
for the story did not acquire belief from its approximation to real
fact, but from its perfect harmony with Eleusinian faith and feeling,
and from the absence of any standard of historical credibility. The
little town of Eleusis derived all its importance from the solemnity
of the Dêmêtria, and the Hymn which we have been considering (probably
at least as old as 600 B. C.) represents the town as it stood before
its absorption into the larger unity of Athens, which seems to have
produced an alteration of its legends and an increase of dignity in
its great festival. In the faith of an Eleusinian, the religious as
well as the patriotic antiquities of his native town were connected
with this capital solemnity. The divine legend of the sufferings of
Dêmêtêr and her visit to Eleusis was to him that which the heroic
legend of Adrastus and the Siege of Thêbes was to a Sikyonian, or
that of Erechtheus and Athênê to an Athenian grouping together in the
same scene and story the goddess and the heroic fathers of the town.
If our information were fuller, we should probably find abundance of
other legends respecting the Dêmêtria: the Gephyræi of Athens, to whom
belonged the celebrated Harmodios and Aristogeitôn, and who possessed
special Orgies of Dêmêtêr the Sorrowful, to which no man foreign to
their Gens was ever admitted,[93] would doubtless have told stories
not only different but contradictory; and even in other Eleusinian
mythes we discover Eumolpus as king of Eleusis, son of Poseidôn, and
a Thracian, completely different from the character which he bears in
the Hymn before us.[94] Neither discrepancies nor want of evidence, in
reference to alleged antiquities, shocked the faith of a non-historical
public. What they wanted was a picture of the past, impressive to their
feelings and plausible to their imagination; and it is important to
the reader to remember, while he reads either the divine legends which
we are now illustrating or the heroic legends to which we shall soon
approach, that he is dealing with a past which never was present,—a
region essentially mythical, neither approachable by the critic nor
mensurable by the chronologer.

  [93] Herodot. v. 61.

  [94] Pausan. i. 38, 3; Apollodôr. iii. 15, 4. Heyne in his
  Note admits several persons named Eumolpus. Compare Isokratês,
  Panegyr. p. 55. Philochorus the Attic antiquary could not have
  received the legend of the Eleusinian Hymn, from the different
  account which he gave respecting the rape of Persephonê (Philoch.
  Fragm. 46, ed. Didot), and also respecting Keleos (Fr. 28,
  _ibid._).

The tale respecting the visit of Dêmêtêr, which was told by the
ancient Gens, called the Phytalids,[95] in reference to another temple
of Dêmêtêr between Athens and Eleusis, and also by the Megarians in
reference to a Dêmêtrion near their city, acquired under the auspices
of Athens still further extension. The goddess was reported to have
first communicated to Triptolemus at Eleusis the art of sowing corn,
which by his intervention was disseminated all over the earth. And thus
the Athenians took credit to themselves for having been the medium of
communication from the gods to man of all the inestimable blessings
of agriculture, which they affirmed to have been first exhibited on
the fertile Rharian plain near Eleusis. Such pretensions are not to
be found in the old Homeric hymn. The festival of the Thesmophoria,
celebrated in honor of Dêmêtêr Thesmophoros at Athens, was altogether
different from the Eleusinia, in this material respect, as well as
others, that all males were excluded, and women only were allowed to
partake in it: the surname Thesmophoros gave occasion to new legends
in which the goddess was glorified as the first authoress of laws and
legal sanctions to mankind.[96] This festival, for women apart and
alone, was also celebrated at Paros, at Ephesus, and in many other
parts of Greece.[97]

  [95] Phytalus, the Eponym or godfather of this gens, had received
  Dêmêtêr as a guest in his house, when she first presented mankind
  with the fruit of the fig-tree. (Pausan. i. 37, 2.)

  [96] Kallimach. Hymn. Cerer. 19. Sophoklês, Triptolemos, Frag. 1.
  Cicero, Legg. ii. 14, and the note of Servius ad Virgil. Æn. iv.
  58.

  [97] Herodot. vi. 16, 134. ἕρκος Θεσμοφόρου Δήμητρος—τὸ ἐς ἔρσενα
  γόνον ἄῤῥητα ἱερά.

Altogether, Dêmêtêr and Dionysos, as the Grecian counterparts of the
Egyptian Isis and Osiris, seem to have been the great recipients of the
new sacred rites borrowed from Egypt, before the worship of Isis in
her own name was introduced into Greece: their solemnities became more
frequently recluse and mysterious than those of the other deities. The
importance of Dêmêtêr to the collective nationality of Greece may be
gathered from the fact that her temple was erected at Thermopylæ, the
spot where the Amphiktyonic assemblies were held, close by the temple
of the Eponymous hero Amphiktyôn himself, and under the surname of the
Amphiktyonic Dêmêtêr.[98]

  [98] Herodot. vii. 200.

We now pass to another and not less important celestial
personage—Apollo.

The legends of Dêlos and Delphi, embodied in the Homeric Hymn to
Apollo, indicate, if not a greater dignity, at least a more widely
diffused worship of that god than even of Dêmêtêr. The Hymn is, in
point of fact, an aggregate of two separate compositions, one emanating
from an Ionic bard at Dêlos, the other from Delphi. The first details
the birth, the second the mature divine efficiency, of Apollo; but
both alike present the unaffected charm as well as the characteristic
peculiarities of Grecian mythical narrative. The hymnographer sings,
and his hearers accept in perfect good faith, a history of the past;
but it is a past, imagined partly as an introductory explanation to
the present, partly as a means of glorifying the god. The island
of Dêlos was the accredited birth-place of Apollo, and is also the
place in which he chiefly delights, where the great and brilliant
Ionic festival is periodically convened in his honor. Yet it is a
rock narrow, barren, and uninviting: how came so glorious a privilege
to be awarded to it? This the poet takes upon himself to explain.
Lêtô, pregnant with Apollo, and persecuted by the jealous Hêrê, could
find no spot wherein to give birth to her offspring. In vain did she
address herself to numerous places in Greece, the Asiatic coast and the
intermediate islands; all were terrified at the wrath of Hêrê, and
refused to harbor her. As a last resort, she approached the rejected
and repulsive island of Dêlos, and promised that, if shelter were
granted to her in her forlorn condition, the island should become the
chosen resort of Apollo as well as the site of his temple with its
rich accompanying solemnities.[99] Dêlos joyfully consented, but not
without many apprehensions that the potent Apollo would despise her
unworthiness, and not without exacting a formal oath from Lêtô,—who
was then admitted to the desired protection, and duly accomplished her
long and painful labor. Though Diônê, Rhea, Themis and Amphitritê came
to soothe and succor her, yet Hêrê kept away the goddess presiding
over childbirth, Eileithyia, and thus cruelly prolonged her pangs. At
length Eileithyia came, and Apollo was born. Hardly had Apollo tasted,
from the hands of Themis, the immortal food, nectar and ambrosia,
when he burst at once his infant bands, and displayed himself in full
divine form and strength, claiming his characteristic attributes
of the bow and the harp, and his privileged function of announcing
beforehand to mankind the designs of Zeus. The promise made by Lêtô to
Dêlos was faithfully performed: amidst the numberless other temples
and groves which men provided for him, he ever preferred that island
as his permanent residence, and there the Ionians with their wives
and children, and all their “bravery,” congregated periodically from
their different cities to glorify him. Dance and song and athletic
contests adorned the solemnity, and the countless ships, wealth, and
grace of the multitudinous Ionians had the air of an assembly of gods.
The Delian maidens, servants of Apollo, sang hymns to the glory of
the god, as well as of Artemis and Lêtô, intermingled with adventures
of foregone men and women, to the delight of the listening crowd. The
blind itinerant bard of Chios (composer of this the Homeric hymn,
and confounded in antiquity with the author of the Iliad) had found
honor and acceptance at this festival, and commends himself, in a
touching farewell strain, to the remembrance and sympathy of the Delian
maidens.[100]

  [99] According to another legend, Lêtô was said to have been
  conveyed from the Hyperboreans to Dêlos in twelve days, in
  the form of a she-wolf, to escape the jealous eye of Hêrê. In
  connection with this legend, it was affirmed that the she-wolves
  always brought forth their young only during these twelve days in
  the year (Aristot. Hist. Animal. vii. 35).

  [100] Hom. Hymn. Apoll. i. 179.

But Dêlos was not an oracular spot: Apollo did not manifest himself
there as revealer of the futurities of Zeus. A place must be found
where this beneficent function, without which mankind would perish
under the innumerable doubts and perplexities of life, may be exercised
and rendered available. Apollo himself descends from Olympus to
make choice of a suitable site: the hymnographer knows a thousand
other adventures of the god which he might sing, but he prefers this
memorable incident, the charter and patent of consecration for the
Delphian temple. Many different places did Apollo inspect; he surveyed
the country of the Magnêtes and the Perrhæbians, came to Iôlkos,
and passed over from thence to Eubœa and the plain of Lelanton. But
even this fertile spot did not please him: he crossed the Euripus to
Bœotia, passed by Teumêssus and Mykalêssus, and the then inaccessible
and unoccupied forest on which the city of Thêbes afterwards stood.
He next proceeded to Onchêstos, but the grove of Poseidôn was already
established there; next across the Kêphissus to Okalea, Haliartus,
and the agreeable plain and much-frequented fountain of Delphusa, or
Tilphusa. Pleased with the place, Apollo prepared to establish his
oracle there, but Tilphusa was proud of the beauty of her own site,
and did not choose that her glory should be eclipsed by that of the
god.[101] She alarmed him with the apprehension that the chariots
which contended in her plain, and the horses and mules which watered
at her fountain would disturb the solemnity of his oracle; and she
thus induced him to proceed onward to the southern side of Parnassus,
overhanging the harbor of Krissa. Here he established his oracle, in
the mountainous site not frequented by chariots and horses, and near to
a fountain, which however was guarded by a vast and terrific serpent,
once the nurse of the monster Typhaôn. This serpent Apollo slew with an
arrow, and suffered its body to rot in the sun: hence the name of the
place, Pythô,[102] and the surname of the Pythian Apollo. The plan of
his temple being marked out, it was built by Trophônios and Agamêdês,
aided by a crowd of forward auxiliaries from the neighborhood. He now
discovered with indignation, however, that Tilphusa had cheated him,
and went back with swift step to resent it. “Thou shalt not thus,”
he said, “succeed in thy fraud and retain thy beautiful water; the
glory of the place shall be mine, and not thine alone.” Thus saying,
he tumbled down a crag upon the fountain, and obstructed her limpid
current: establishing an altar for himself in a grove hard by near
another spring, where men still worship him as Apollo Tilphusios,
because of his severe vengeance upon the once beautiful Tilphusa.[103]

  [101] Hom. Hymn. Apoll. 262.

  [102] Hom. Hymn. 363—πύθεσθαι, _to rot_.

  [103] Hom. Hymn. Apoll. 381.

Apollo next stood in need of chosen ministers to take care of his
temple and sacrifice, and to pronounce his responses at Pythô.
Descrying a ship, “containing many and good men,” bound on traffic from
the Minoian Knossus in Krête, to Pylus in Peloponnêsus, he resolved
to make use of the ship and her crew for his purpose. Assuming the
shape of a vast dolphin, he splashed about and shook the vessel so
as to strike the mariners with terror, while he sent a strong wind,
which impelled her along the coast of Peloponnêsus into the Corinthian
Gulf, and finally to the harbor of Krissa, where she ran aground.
The affrighted crew did not dare to disembark: but Apollo was seen
standing on the shore in the guise of a vigorous youth, and inquired
who they were, and what was their business. The leader of the Krêtans
recounted in reply their miraculous and compulsory voyage, when Apollo
revealed himself as the author and contriver of it, announcing to them
the honorable function and the dignified post to which he destined
them.[104] They followed him by his orders to the rocky Pythô on
Parnassus, singing the solemn Io-Paian such as it is sung in Krête,
while the god himself marched at their head, with his fine form and
lofty step, playing on the harp. He showed them the temple and site
of the oracle, and directed them to worship him as Apollo Delphinios,
because they had first seen him in the shape of a dolphin. “But how,”
they inquired, “are we to live in a spot where there is neither corn,
nor vine, nor pasturage?” “Ye silly mortals,” answered the god, “who
look only for toil and privation, know that an easier lot is yours. Ye
shall live by the cattle whom crowds of pious visitors will bring to
the temple: ye shall need only the knife to be constantly ready for
sacrifice.[105] Your duty will be to guard my temple, and to officiate
as ministers at my feasts: but if ye be guilty of wrong or insolence,
either by word or deed, ye shall become the slaves of other men, and
shall remain so forever. Take heed of the word and the warning.”

  [104] Hom. Hymn. Apoll. 475 _sqq._

  [105] Homer. Hymn. Apoll. 535.—

    Δεξιτέρῃ μάλ᾽ ἕκαστος ἔχων ἐν χειρὶ μάχαιραν
    Σφάζειν αἰεὶ μῆλα· τὰ δ᾽ ἄφθονα πάντα πάρεσται,
    Ὅσσα ἐμοίγ᾽ ἀγάγωσι περίκλυτα φῦλ᾽ ἀνθρώπων.

Such are the legends of Dêlos and Delphi, according to the Homeric Hymn
to Apollo. The specific functions of the god, and the chief localities
of his worship, together with the surnames attached to them, are
thus historically explained, being connected with his past acts and
adventures. Though these are to us only interesting poetry, yet to
those who heard them sung they possessed all the requisites of history,
and were fully believed as such, not because they were partially
founded in reality, but because they ran in complete harmony with the
feelings; and, so long as that condition was fulfilled, it was not
the fashion of the time to canvass truth or falsehood. The narrative
is purely personal, without any discernible symbolized doctrine or
allegory, to serve as a supposed ulterior purpose: the particular deeds
ascribed to Apollo grow out of the general preconceptions as to his
attributes, combined with the present realities of his worship. It is
neither history nor allegory, but simple mythe or legend.

The worship of Apollo is among the most ancient, capital, and strongly
marked facts of the Grecian world, and widely diffused over every
branch of the race. It is older than the Iliad or Odyssey, in the
latter of which both Pythô and Dêlos are noted, though Dêlos is not
named in the former. But the ancient Apollo is different in more
respects than one from the Apollo of later times. He is in an especial
manner the god of the Trojans, unfriendly to the Greeks, and especially
to Achilles; he has, moreover, only two primary attributes, his bow and
his prophetic powers, without any distinct connection either with the
harp, or with medicine, or with the sun, all which in later times he
came to comprehend. He is not only, as Apollo Karneius, the chief god
of the Doric race, but also (under the surname of Patrôus) the great
protecting divinity of the gentile tie among the Ionians:[106] he is
moreover the guide and stimulus to Grecian colonization, scarcely any
colony being ever sent out without encouragement and direction from the
oracle at Delphi: Apollo Archêgetês is one of his great surnames.[107]
His temple lends sanctity to the meetings of the Amphiktyonic assembly,
and he is always in filial subordination and harmony with his father
Zeus: Delphi and Olympia are never found in conflict. In the Iliad,
the warm and earnest patrons of the Greeks are Hêrê, Athênê, and
Poseidôn: here too Zeus and Apollo are seen in harmony, for Zeus is
decidedly well-inclined to the Trojans, and reluctantly sacrifices
them to the importunity of the two great goddesses.[108] The worship
of the Sminthian Apollo, in various parts of the Troad and the
neighboring territory, dates before the earliest periods of Æolic
colonization:[109] hence the zealous patronage of Troy ascribed to him
in the Iliad. Altogether, however, the distribution and partialities
of the gods in that poem are different from what they become in later
times,—a difference which our means of information do not enable us
satisfactorily to explain. Besides the Delphian temple, Apollo had
numerous temples throughout Greece, and oracles at Abæ in Phôkis, on
the Mount Ptôon, and at Tegyra in Bœotia, where he was said to have
been born,[110] at Branchidæ near Milêtus, at Klarus in Asia Minor, and
at Patara in Lykia. He was not the only oracular god: Zeus at Dodona
and at Olympia gave responses also: the gods or heroes Trophônius,
Amphiaraus, Amphilochus, Mopsus, etc., each at his own sanctuary and
in his own prescribed manner, rendered the same service.

  [106] Harpocration v. Ἀπόλλων πατρῶος and Ἑρκεῖος Ζεύς. Apollo
  Delphinios also belongs to the Ionic Greeks generally. Strabo,
  iv. 179.

  [107] Thucydid. vi. 3; Kallimach. Hymn. Apoll. 56.—

          Φοῖβος γὰρ ἀεὶ πολίεσσι φιληδεῖ
    Κτιζομέναις, αὐτὸς δὲ θεμείλια Φοῖβος ὑφαίνει.

  [108] Iliad, iv. 30-46.

  [109] Iliad, i. 38, 451; Stephan. Byz. Ἵλιον, Τένεδος. See also
  Klausen, Æneas und die Penaten, b. i. p. 69. The worship of
  Apollo Sminthios and the festival of the Sminthia at Alexandria
  Troas lasted down to the time of Menander the rhêtôr, at the
  close of the third century after Christ.

  [110] Plutarch. Defect. Oracul. c. 5, p. 412; c. 8, p. 414;
  Steph. Byz. v. Τεγύρα. The temple of the Ptôan Apollo had
  acquired celebrity before the days of the poet Asius. Pausan. ix.
  23, 3.

The two legends of Delphi and Dêlos, above noticed, form of course
a very insignificant fraction of the narratives which once existed
respecting the great and venerated Apollo. They serve only as
specimens, and as very early specimens,[111] to illustrate what
these divine mythes were, and what was the turn of Grecian faith and
imagination. The constantly recurring festivals of the gods caused
an incessant demand for new mythes respecting them, or at least for
varieties and reproductions of the old mythes. Even during the third
century of the Christian æra, in the time of the rhêtôr Menander,
when the old forms of Paganism were waning and when the stock of
mythes in existence was extremely abundant, we see this demand in
great force; but it was incomparably more operative in those earlier
times when the creative vein of the Grecian mind yet retained its
pristine and unfaded richness. Each god had many different surnames,
temples, groves, and solemnities; with each of which was connected
more or less of mythical narrative, originally hatched in the prolific
and spontaneous fancy of a believing neighborhood, to be afterwards
expanded, adorned and diffused by the song of the poet. The earliest
subject of competition[112] at the great Pythian festival was the
singing of a hymn in honor of Apollo: other _agones_ were subsequently
added, but the ode or hymn constituted the fundamental attribute of
the solemnity: the Pythia at Sikyôn and elsewhere were probably framed
on a similar footing. So too at the ancient and celebrated Charitêsia,
or festival of the Charites, at Orchomenos, the rivalry of the poets
in their various modes of composition both began and continued as the
predominant feature:[113] and the inestimable treasures yet remaining
to us of Attic tragedy and comedy, are gleanings from the once numerous
dramas exhibited at the solemnity of the Dionysia. The Ephesians gave
considerable rewards for the best hymns in honor of Artemis, to be sung
at her temple.[114] And the early lyric poets of Greece, though their
works have not descended to us, devoted their genius largely to similar
productions, as may be seen by the titles and fragments yet remaining.

  [111] The legend which Ephorus followed about the establishment
  of the Delphian temple was something radically different from
  the Homeric Hymn (Ephori Fragm. 70, ed. Didot): his narrative
  went far to politicize and rationalize the story. The progeny of
  Apollo was very numerous, and of the most diverse attributes; he
  was father of the Korybantes (Pherekydes, Fragm. 6, ed. Didot),
  as well as of Asklêpios and Aristæus (Schol. Apollôn. Rhod. ii.
  500; Apollodôr. iii. 10, 3).

  [112] Strabo, ix. p. 421. Menander the Rhetor (Ap. Walz. Coll.
  Rhett. t. ix. p. 136) gives an elaborate classification of hymns
  to the gods, distinguishing them into nine classes,—κλητικοὶ,
  ἀποπεμπτικοὶ, φυσικοὶ, μυθικοὶ, γενεαλογικοὶ, πεπλασμένοι,
  εὐκτικοὶ, ἀπευκτικοὶ, μικτοί:—the second class had reference to
  the temporary absences or departure of a god to some distant
  place, which were often admitted in the ancient religion. Sappho
  and Alkman in their _kletic_ hymns invoked the gods from many
  different places,—τὴν μὲν γὰρ Ἄρτεμιν ἐκ μυρίων μὲν ὄρεων,
  μυρίων δὲ πόλεων, ἔτι δὲ ποτάμων, ἀνακαλεῖ,—also Aphroditê and
  Apollo, etc. All these songs were full of adventures and details
  respecting the gods,—in other words of legendary matter.

  [113] Pindar, Olymp. xiv.; Boeckh, Staatshaushaltung der Athener,
  Appendix, § xx. p. 357.

  [114] Alexander Ætolus, apud Macrobium, Saturn. v. 22.

Both the Christian and the Mahomedan religions have begun during the
historical age, have been propagated from one common centre, and have
been erected upon the ruins of a different pre-existing faith. With
none of these particulars did Grecian Paganism correspond. It took rise
in an age of imagination and feeling simply, without the restraints,
as well as without the aid, of writing or records, of history or
philosophy: it was, as a general rule, the spontaneous product of many
separate tribes and localities, imitation and propagation operating
as subordinate causes; it was moreover a primordial faith, as far as
our means of information enable us to discover. These considerations
explain to us two facts in the history of the early Pagan mind: first,
the divine mythes, the matter of their religion, constituted also
the matter of their earliest history; next, these mythes harmonized
with each other only in their general types, but differed incurably
in respect of particular incidents. The poet who sung a new adventure
of Apollo, the trace of which he might have heard in some remote
locality, would take care that it should be agreeable to the general
conceptions which his hearers entertained respecting the god. He would
not ascribe the cestus or amorous influences to Athênê, nor armed
interference and the ægis to Aphroditê; but, provided he maintained
this general keeping, he might indulge his fancy without restraint in
the particular events of the story.[115] The feelings and faith of
his hearers went along with him, and there were no critical scruples
to hold them back: to scrutinize the alleged proceedings of the gods
was repulsive, and to disbelieve them impious. And thus these divine
mythes, though they had their root simply in religious feelings, and
though they presented great discrepancies of fact, served nevertheless
as primitive matter of history to an early Greek: they were the only
narratives, at once publicly accredited and interesting, which he
possessed. To them were aggregated the heroic mythes (to which we shall
proceed presently),—indeed the two are inseparably blended, gods,
heroes and men almost always appearing in the same picture,—analogous
both in their structure and their genesis, and differing chiefly in the
circumstance that they sprang from the type of a hero instead of from
that of a god.

  [115] The birth of Apollo and Artemis from Zeus and Lêtô is among
  the oldest and most generally admitted facts in the Grecian
  divine legends. Yet Æschylus did not scruple to describe Artemis
  publicly as daughter of Dêmêtêr (Herodot. ii. 156; Pausan. viii.
  37, 3). Herodotus thinks that he copied this innovation from the
  Egyptians, who affirmed that Apollo and Artemis were the sons of
  Dionysos and Isis.

  The number and discrepancies of the mythes respecting each god
  are attested by the fruitless attempts of learned Greeks to
  escape the necessity of rejecting any of them by multiplying
  homonymous personages,—three persons named Zeus; five named
  Athênê; six named Apollo, etc. (Cicero. de Natur. Deor. iii. 21:
  Clemen. Alexand. Admon. ad Gent. p. 17).

We are not to be astonished if we find Aphroditê, in the Iliad, born
from Zeus and Dionê,—and in the Theogony of Hesiod, generated from the
foam on the sea after the mutilation of Uranos; nor if in the Odyssey
she appears as the wife of Hêphæstos, while in the Theogony the latter
is married to Aglaia, and Aphroditê is described as mother of three
children by Arês.[116] The Homeric hymn to Aphroditê details the legend
of Aphroditê and Anchisês, which is presupposed in the Iliad as the
parentage of Æneas: but the author of the hymn, probably sung at one
of the festivals of Aphroditê in Cyprus, represents the goddess as
ashamed of her passion for a mortal, and as enjoining Anchisês under
severe menaces not to reveal who the mother of Æneas was;[117] while
in the Iliad she has no scruple in publicly owning him, and he passes
everywhere as her acknowledged son. Aphroditê is described in the hymn
as herself cold and unimpressible, but ever active and irresistible
in inspiring amorous feelings to gods, to men, and to animals. Three
goddesses are recorded as memorable exceptions to her universal
empire,—Athênê, Artemis, and Hestia or Vesta. Aphroditê was one of the
most important of all the goddesses in the mythical world; for the
number of interesting, pathetic and tragical adventures deducible from
misplaced or unhappy passion was of course very great; and in most
of these cases the intervention of Aphroditê was usually prefixed,
with some legend to explain why she manifested herself. Her range of
action grows wider in the later epic and lyric and tragic poets than in
Homer.[118]

  [116] Hesiod, Theogon. 188, 934, 945; Homer, Iliad, v. 371;
  Odyss. viii. 268.

  [117] Homer, Hymn. Vener. 248, 286; Homer, Iliad, v. 320, 386.

  [118] A large proportion of the Hesiodic epic related to the
  exploits and adventures of the heroic women,—the Catalogue of
  Women and the Eoiai embodied a string of such narratives. Hesiod
  and Stesichorus explained the conduct of Helen and Klytæmnêstra
  by the anger of Aphroditê, caused by the neglect of their
  father Tyndareus to sacrifice to her (Hesiod, Fragm. 59, ed.
  Düntzer; Stesichor. Fragm. 9, ed. Schneidewin): the irresistible
  ascendency of Aphroditê is set forth in the Hippolytus of
  Euripidês not less forcibly than that of Dionysos in the Bacchæ.
  The character of Daphnis the herdsman, well-known from the first
  Idyll of Theocritus, and illustrating the destroying force of
  Aphroditê, appears to have been first introduced into Greek
  poetry by Stesichorus (see Klausen, Æneas und die Penaten, vol.
  i. pp. 526-529). Compare a striking piece among the Fragmenta
  Incerta of Sophoklês (Fr. 63, Brunck) and Euripid. Troad. 946,
  995, 1048. Even in the Opp. et Di. of Hesiod, Aphroditê is
  conceived rather as a disturbing and injurious influence (v. 65).

  Adonis owes his renown to the Alexandrine poets and their
  contemporary sovereigns (see Bion’s Idyll and the Adoniazusæ of
  Theocritus). The favorites of Aphroditê, even as counted up by
  the diligence of Clemens Alexandrinus, are however very few in
  number. (Admonitio ad Gent. p. 12, Sylb.)

Athênê, the man-goddess,[119] born from the head of Zeus, without
a mother and without feminine sympathies, is the antithesis
partly of Aphroditê, partly of the effeminate or womanized god
Dionysos—the latter is an importation from Asia, but Athênê is a Greek
conception—the type of composed, majestic and unrelenting force. It
appears however as if this goddess had been conceived in a different
manner in different parts of Greece. For we find ascribed to her, in
some of the legends, attributes of industry and home-keeping; she is
represented as the companion of Hêphæstos, patronizing handicraft, and
expert at the loom and the spindle: the Athenian potters worshipped
her along with Promêtheus. Such traits of character do not square with
the formidable ægis and the massive and crushing spear which Homer and
most of the mythes assign to her. There probably were at first at least
two different types of Athênê, and their coalescence has partially
obliterated the less marked of the two.[120] Athênê is the constant
and watchful protectress of Hêraklês: she is also locally identified
with the soil and people of Athens, even in the Iliad: Erechtheus, the
Athenian, is born of the earth, but Athênê brings him up, nourishes
him, and lodges him in her own temple, where the Athenians annually
worship him with sacrifice and solemnities.[121] It was altogether
impossible to make Erechtheus son of Athênê,—the type of the goddess
forbade it; but the Athenian mythe-creators, though they found this
barrier impassable, strove to approach to it as near as they could,
and the description which they give of the birth of Erichthonios, at
once un-Homeric and unseemly, presents something like the phantom of
maternity.[122]

  [119] Ἀνδροθέᾳ δῶρον ... Ἀθάνᾳ Simmias Rhodius; Πέλεκυς, ap.
  Hephæstion. c. 9. p. 54, Gaisford.

  [120] Apollodôr. ap. Schol. ad Sophokl. Œdip. vol. 57; Pausan.
  i. 24, 3; ix. 26, 3; Diodôr. v. 73; Plato, Legg. xi. p. 920. In
  the Opp. et Di. of Hesiod, the carpenter is the servant of Athênê
  (429): see also Phereklos the τέκτων in the Iliad, v. 61: compare
  viii. 385; Odyss. viii. 493; and the Homeric Hymn, to Aphroditê,
  v. 12. The learned article of O. Müller (in the Encyclopædia of
  Ersch and Gruber, since republished among his Kleine Deutsche
  Schriften, p. 134 _seq._), _Pallas Athênê_, brings together all
  that can be known about this goddess.

  [121] Iliad, ii. 546; viii. 362.

  [122] Apollodôr. iii. 4, 6. Compare the vague language of Plato,
  Kritias, c. iv., and Ovid, Metamorph. ii. 757.

The huntress Artemis, in Arcadia and in Greece proper generally,
exhibits a well-defined type with which the legends respecting her
are tolerably consistent. But the Ephesian as well as the Tauric
Artemis partakes more of the Asiatic character, and has borrowed the
attributes of the Lydian Great Mother as well as of an indigenous
Tauric Virgin:[123] this Ephesian Artemis passed to the colonies of
Phokæa and Milêtus.[124] The Homeric Artemis shares with her brother
Apollo in the dexterous use of the far-striking bow, and sudden death
is described by the poet as inflicted by her gentle arrow. The jealousy
of the gods at the withholding of honors and sacrifices, or at the
presumption of mortals in contending with them,—a point of character so
frequently recurring in the types of the Grecian gods,—manifests itself
in the legends of Artemis: the memorable Kalydônian boar is sent by her
as a visitation upon Œneus, because he had omitted to sacrifice to her,
while he did honor to other gods.[125] The Arcadian heroine Atalanta is
however a reproduction of Artemis, with little or no difference, and
the goddess is sometimes confounded even with her attendant nymphs.

  [123] Herodot. iv. 103; Strabo, xii. p. 534; xiii. p. 650. About
  the Ephesian Artemis, see Guhl, Ephesiaca (Berlin, 1843), p.
  79 _sqq._; Aristoph. Nub. 590; Autokrates in Tympanistis apud
  Ælian. Hist. Animal. xii. 9; and Spanheim ad Kallimach. Hymn.
  Dian. 36. The dances in honor of Artemis sometimes appear to
  have approached to the frenzied style of Bacchanal movement. See
  the words of Timotheus ap. Plutarch. de Audiend. Poet. p. 22, c.
  4, and περὶ Δεισιδ. c. 10, p. 170, also Aristoph. Lysist. 1314.
  They seem to have been often celebrated in the solitudes of the
  mountains, which were the favorite resort of Artemis (Kallimach.
  Hymn. Dian. 19), and these ὀρειβάσιαι were always causes
  predisposing to fanatical excitement.

  [124] Strabo, iv. p. 179.

  [125] Iliad, ix. 529.

The mighty Poseidôn, the earth-shaker and the ruler of the sea, is
second only to Zeus in power, but has no share in those imperial and
superintending capacities which the Father of gods and men exhibits.
He numbers a numerous heroic progeny, usually men of great corporeal
strength, and many of them belonging to the Æolic race: the great
Neleid family of Pylus trace their origin up to him; and he is also
the father of Polyphêmus the Cyclôps, whose well-earned suffering
he cruelly revenges upon Odysseus. The island of Kalaureia is his
Dêlos,[126] and there was held in it an old local Amphiktyony, for the
purpose of rendering to him joint honor and sacrifice: the isthmus of
Corinth, Helikê in Achaia, and Onchêstos in Bœotia, are also residences
which he much affects, and where he is solemnly worshipped. But the
abode which he originally and specially selected for himself was the
Acropolis of Athens, where by a blow of his trident he produced a
well of water in the rock: Athênê came afterwards and claimed the
spot for herself, planting in token of possession the olive-tree
which stood in the sacred grove of Pandrosos: and the decision either
of the autochthonous Cecrops, or of Erechtheus, awarded to her the
preference, much to the displeasure of Poseidôn. Either on this
account, or on account of the death of his son Eumolpus, slain in
assisting the Eleusinians against Erechtheus, the Attic mythes ascribed
to Poseidôn great enmity against the Erechtheid family, which he is
asserted to have ultimately overthrown: Theseus, whose glorious reign
and deeds succeeded to that family, is said to have been really his
son.[127] In several other places,—in Ægina, Argos and Naxos,—Poseidôn
had disputed the privileges of patron-god with Zeus, Hêrê and Dionysos:
he was worsted in all, but bore his defeat patiently.[128] Poseidôn
endured a long slavery, in common with Apollo, gods as they were,[129]
under Laomedôn, king of Troy, at the command and condemnation of Zeus:
the two gods rebuilt the walls of the city, which had been destroyed
by Hêraklês. When their time was expired, the insolent Laomedôn
withheld from them the stipulated reward, and even accompanied its
refusal with appalling threats; and the subsequent animosity of the
god against Troy was greatly determined by the sentiment of this
injustice.[130] Such periods of servitude, inflicted upon individual
gods, are among the most remarkable of all the incidents in the divine
legends. We find Apollo on another occasion condemned to serve Admêtus,
king of Pheræ, as a punishment for having killed the Cyclôpes, and
Hêraklês also is sold as a slave to Omphalê. Even the fierce Arês,
overpowered and imprisoned for a long time by the two Alôids,[131] is
ultimately liberated only by extraneous aid. Such narratives attest
the discursive range of Grecian fancy in reference to the gods, as
well as the perfect commingling of things and persons, divine and
human, in their conceptions of the past. The god who serves is for
the time degraded: but the supreme god who commands the servitude is
in the like proportion exalted, whilst the idea of some sort of order
and government among these superhuman beings was never lost sight of.
Nevertheless the mythes respecting the servitude of the gods became
obnoxious afterwards, along with many others, to severe criticism on
the part of philosophers.

  [126] Strabo, viii. p. 374. According to the old poem called
  Eumolpia, ascribed to Musæus, the oracle of Delphi originally
  belonged to Poseidôn and Gæa, jointly: from Gæa it passed to
  Themis, and from her to Apollo, to whom Poseidôn also made over
  his share as a compensation for the surrender of Kalaureia to
  him. (Pausan. x. 5, 3).

  [127] Apollodôr. iii. 14, 1; iii. 15, 3, 5.

  [128] Plutarch, Sympos. viii. 6, p. 741.

  [129] Iliad, ii. 716, 766; Euripid. Alkêstis, 2. See Panyasis,
  Fragm. 12, p. 24, ed. Düntzer.

  [130] Iliad, vii. 452; xxi. 459.

  [131] Iliad, v. 386.

The proud, jealous, and bitter Hêrê,—the goddess of the once-wealthy
Mykênæ, the _fax et focus_ of the Trojan war, and the ever-present
protectress of Jasôn in the Argonautic expedition,[132]—occupies an
indispensable station in the mythical world. As the daughter of Kronos
and wife of Zeus, she fills a throne from whence he cannot dislodge
her, and which gives her a right perpetually to grumble and to thwart
him.[133] Her unmeasured jealousy of the female favorites of Zeus, and
her antipathy against his sons, especially against Hêraklês, has been
the suggesting cause of innumerable mythes: the general type of her
character stands here clearly marked, as furnishing both stimulus and
guide to the mythopœic fancy. The “Sacred Wedding,” or marriage of Zeus
and Hêrê, was familiar to epithalamic poets long before it became a
theme for the spiritualizing ingenuity of critics.

  [132] Iliad, iv. 51; Odyss. xii. 72.

  [133] Iliad, i. 544; iv. 29-38; viii. 408.

Hêphæstos is the son of Hêrê without a father, and stands to her in
the same relation as Athênê to Zeus: her pride and want of sympathy
are manifested by her casting him out at once in consequence of his
deformity.[134] He is the god of fire, and especially of fire in its
practical applications to handicraft, and is indispensable as the
right-hand and instrument of the gods. His skill and his deformity
appear alternately as the source of mythical stories: wherever
exquisite and effective fabrication is intended to be designated,
Hêphæstos is announced as the maker, although in this function the
type of his character is reproduced in Dædalos. In the Attic legends
he appears intimately united both with Promêtheus and with Athênê, in
conjunction with whom he was worshipped at Kolônus near Athens. Lemnos
was the favorite residence of Hêphæstos; and if we possessed more
knowledge of this island and its town Hêphæstias, we should doubtless
find abundant legends detailing his adventures and interventions.

  [134] Iliad, xviii. 306.

The chaste, still, and home-keeping Hestia, goddess of the family
hearth, is far less fruitful in mythical narratives, in spite of her
very superior dignity, than the knavish, smooth-tongued, keen, and
acquisitive Hermês. His function of messenger of the gods brings him
perpetually on the stage, and affords ample scope for portraying the
features of his character. The Homeric hymn to Hermês describes the
scene and circumstances of his birth, and the almost instantaneous
manifestation, even in infancy, of his peculiar attributes; it explains
the friendly footing on which he stood with Apollo,—the interchange of
gifts and functions between them,—and lastly, the inviolate security
of all the wealth and offerings in the Delphian temple, exposed as
they were to thieves without any visible protection. Such was the
innate cleverness and talent of Hermês, that on the day he was born
he invented the lyre, stringing the seven chords on the shell of a
tortoise:[135] and he also stole the cattle of Apollo in Pieria,
dragging them backwards to his cave in Arcadia, so that their track
could not be detected. To the remonstrances of his mother Maia, who
points out to him the danger of offending Apollo, Hermês replies, that
he aspires to rival the dignity and functions of Apollo among the
immortals, and that if his father Zeus refuses to grant them to him,
he will employ his powers of thieving in breaking open the sanctuary
at Delphi, and in carrying away the gold and the vestments, the
precious tripods and vessels.[136] Presently Apollo discovers the loss
of his cattle, and after some trouble finds his way to the Kyllênian
cavern, where he sees Hermês asleep in his cradle. The child denies
the theft with effrontery, and even treats the surmise as a ridiculous
impossibility: he persists in such denial even before Zeus, who however
detects him at once, and compels him to reveal the place where the
cattle are concealed. But the lyre was as yet unknown to Apollo, who
has heard nothing except the voice of the Muses and the sound of the
pipe. So powerfully is he fascinated by hearing the tones of the lyre
from Hermês, and so eager to become possessed of it, that he is willing
at once to pardon the past theft, and even to conciliate besides the
friendship of Hermês.[137] Accordingly a bargain is struck between the
two gods and sanctioned by Zeus. Hermês surrenders to Apollo the lyre,
inventing for his own use the syrinx or panspipe, and receiving from
Apollo in exchange the golden rod of wealth, with empire over flocks
and herds as well as over horses and oxen and the wild animals of the
woods. He presses to obtain the gift of prophecy, but Apollo is under
a special vow not to impart that privilege to any god whatever: he
instructs Hermês however how to draw information, to a certain extent,
from the Mœræ or Fates themselves; and assigns to him, over and above,
the function of messenger of the gods to Hadês.

  [135] Homer. Hymn. Mercur. 18.—

    Ἤῳος γεγονὼς, μέσῳ ἥματι ἐγκιθάριζεν,
    Ἑσπέριος βοῦς κλέψεν ἑκηβόλου Ἀπόλλωνος, etc.

  [136] Homer. Hymn. Merc. 177.—

    Εἰμὶ γὰρ ἐς Πύθωνα, μέγαν δόμον ἀντιτορήσων,
    Ἔνθεν ἅλις τρίποδας περικαλλέας, ἠδὲ λέβητας
    Πορθήσω καὶ χρυσὸν, etc.

  [137] Homer. Hymn. Merc. 442-454.

Although Apollo has acquired the lyre, the particular object of his
wishes, he is still under apprehension that Hermês will steal it away
from him again, together with his bow, and he exacts a formal oath by
Styx as security. Hermês promises solemnly that he will steal none of
the acquisitions, nor ever invade the sanctuary of Apollo; while the
latter on his part pledges himself to recognize Hermês as his chosen
friend and companion, amongst all the other sons of Zeus, human or
divine.[138]

  [138] Homer. Hymn. Merc. 504-520.—

                          Καὶ τὸ μὲν Ἑρμῆς
    Λητοΐδην ἐφίλησε διαμπερὲς, ὡς ἔτι καὶ νῦν, etc.
       ·   ·   ·   ·   ·
    Καὶ τότε Μαίαδος υἱὸς ὑποσχόμενος κατένευσε
    Μή ποτ᾽ ἀποκλέψειν, ὅσ᾽ Ἑκήβολος ἐκτεάτισται,
    Μηδέ ποτ᾽ ἐμπελάσειν πυκίνῳ δόμῳ· αὐτὰρ Ἀπόλλων
    Λητοΐδης κατένευσεν ἐπ᾽ ἀρθμῷ καὶ φιλότητι
    Μή τινα φίλτερον ἄλλον ἐν ἀθανάτοισιν ἔσεσθαι
    Μήτε θεὸν, μήτ᾽ ἄνδρα Διὸς γόνον, etc.

So came to pass, under the sanction of Zeus, the marked favor shown
by Apollo to Hermês. But Hermês (concludes the hymnographer, with
frankness unusual in speaking of a god) “does very little good: he
avails himself of the darkness of night to cheat without measure the
tribes of mortal men.”[139]

  [139] Homer. Hymn. Merc. 574.—

    Παῦρα μὲν οὖν ὀνίνησι, τὸ δ᾽ ἄκριτον ἠπεροπεύει
    Νύκτα δι᾽ ὀρφναίην φῦλα θνητῶν ἀνθρώπων.

Here the general types of Hermês and Apollo, coupled with the present
fact that no thief ever approached the rich and seemingly accessible
treasures of Delphi, engender a string of expository incidents cast
into a quasi-historical form and detailing how it happened that Hermês
had bound himself by especial convention to respect the Delphian
temple. The types of Apollo seem to have been different in different
times and parts of Greece: in some places he was worshipped as Apollo
Nomios,[140] or the patron of pasture and cattle; and this attribute,
which elsewhere passed over to his son Aristæus, is by our hymnographer
voluntarily surrendered to Hermês, combined with the golden rod of
fruitfulness. On the other hand, the lyre did not originally belong to
the Far-striking King, nor is he at all an inventor: the hymn explains
both its first invention and how it came into his possession. And the
value of the incidents is thus partly expository, partly illustrative,
as expanding in detail the general preconceived character of the
Kyllênian god.

  [140] Kallimach. Hymn. Apoll. 47.

To Zeus more amours are ascribed than to any of the other
gods,—probably because the Grecian kings and chieftains were especially
anxious to trace their lineage to the highest and most glorious
of all,—each of these amours having its representative progeny on
earth.[141] Such subjects were among the most promising and agreeable
for the interest of mythical narrative, and Zeus as a lover thus became
the father of a great many legends, branching out into innumerable
interferences, for which his sons, all of them distinguished
individuals, and many of them persecuted by Hêrê, furnished the
occasion. But besides this, the commanding functions of the supreme
god, judicial and administrative, extending both over gods and men, was
a potent stimulus to the mythopœic activity. Zeus has to watch over
his own dignity,—the first of all considerations with a god: moreover
as Horkios, Xenios, Ktêsios, Meilichios, (a small proportion of his
thousand surnames,) he guaranteed oaths and punished perjurers, he
enforced the observance of hospitality, he guarded the family hoard
and the crop realized for the year, and he granted expiation to the
repentant criminal.[142] All these different functions created a
demand for mythes, as the means of translating a dim, but serious,
presentiment into distinct form, both self-explaining and communicable
to others. In enforcing the sanctity of the oath or of the tie of
hospitality, the most powerful of all arguments would be a collection
of legends respecting the judgments of Zeus Horkios or Xenios; the more
impressive and terrific such legends were, the greater would be their
interest, and the less would any one dare to disbelieve them. They
constituted the natural outpourings of a strong and common sentiment,
probably without any deliberate ethical intention: the preconceptions
of the divine agency, expanded into legend, form a product analogous to
the idea of the divine features and symmetry embodied in the bronze or
the marble statue.

  [141] Kallimach. Hymn. Jov. 79. Ἐκ δὲ Διὸς βασιλῆες, etc.

  [142] See Herodot. i. 44. Xenoph. Anabas. vii. 8, 4. Plutarch,
  Thêseus, c. 12.

But it was not alone the general type and attributes of the gods which
contributed to put in action the mythopœic propensities. The rites and
solemnities forming the worship of each god, as well as the details
of his temple and its locality, were a fertile source of mythes,
respecting his exploits and sufferings, which to the people who heard
them served the purpose of past history. The exegetes, or local guide
and interpreter, belonging to each temple, preserved and recounted to
curious strangers these traditional narratives, which lent a certain
dignity even to the minutiæ of divine service. Out of a stock of
materials thus ample, the poets extracted individual collections, such
as the “Causes” (Αἴτια) of Kallimachus, now lost, and such as the Fasti
of Ovid are for the Roman religious antiquities.[143]

  [143] Ovid, Fasti, iv. 211, about the festivals of Apollo:—

                “Priscique imitamina facti
    Æra Deæ comites raucaque terga movent.”

  And Lactantius, v. 19, 15. “Ipsos ritus ex rebus gestis (deorum)
  vel ex casibus vel etiam ex mortibus, natos:” to the same purpose
  Augustin. De Civ. D. vii. 18; Diodôr. iii. 56. Plutarch’s
  Quæstiones Græcæ et Romaicæ are full of similar tales, professing
  to account for existing customs, many of them religious and
  liturgic. See Lobeck, Orphica, p. 675.

It was the practice to offer to the gods in sacrifice the bones of the
victim only, inclosed in fat: how did this practice arise? The author
of the Hesiodic Theogony has a story which explains it: Promêtheus
tricked Zeus into an imprudent choice, at the period when the gods and
mortal men first came to an arrangement about privileges and duties
(in Mekônê). Promêtheus, the tutelary representative of man, divided
a large steer into two portions: on the one side he placed the flesh
and guts, folded up in the omentum and covered over with the skin: on
the other, he put the bones enveloped in fat. He then invited Zeus to
determine which of the two portions the gods would prefer to receive
from mankind. Zeus “with both hands” decided for and took the white
fat, but was highly incensed on finding that he had got nothing at the
bottom except the bones.[144] Nevertheless the choice of the gods was
now irrevocably made: they were not entitled to any portion of the
sacrificed animal beyond the bones and the white fat; and the standing
practice is thus plausibly explained.[145] I select this as one amongst
a thousand instances to illustrate the genesis of legend out of
religious practices. In the belief of the people, the event narrated in
the legend was the real producing cause of the practice: but when we
come to apply a sound criticism, we are compelled to treat the event as
existing only in its narrative legend, and the legend itself as having
been, in the greater number of cases, engendered by the practice,—thus
reversing the supposed order of production.

  [144] Hesiod, Theog. 550.—

    Φῆ ῥα δολοφρονέων· Ζεὺς δ᾽ ἄφθιτα μήδεα εἰδὼς
    Γνῶ ῥ᾽ οὐδ᾽ ἠγνοίησε δόλον· κακὰ δ᾽ ὄσσετο θυμῷ
    Θνητοῖς ἀνθρώποισι, τὰ καὶ τελέεσθαι ἔμελλεν.
    Χερσὶ δ᾽ ὅγ᾽ ἀμφοτέρῃσιν ἀνείλετο λευκὸν ἄλειφαρ·
    Χώσατο δὲ φρένας, ἀμφὶ χόλος δέ μιν ἵκετο θυμὸν,
    Ὡς ἴδεν ὀστέα λευκὰ βοὸς δολίῃ ἐπὶ τέχνῃ.

  In the second line of this citation, the poet tells us that Zeus
  saw through the trick, and was imposed upon by his own consent,
  foreknowing that after all the mischievous consequences of the
  proceeding would be visited on man. But the last lines, and
  indeed the whole drift of the legend, imply the contrary of this:
  Zeus was really taken in, and was in consequence very angry. It
  is curious to observe how the religious feelings of the poet
  drive him to save in words the prescience of Zeus, though in
  doing so he contradicts and nullifies the whole point of the
  story.

  [145] Hesiod, Theog. 557.—

    Ἐκ τοῦ δ᾽ ἀθανάτοισιν ἐπὶ χθονὶ φῦλ᾽ ἀνθρώπων
    Καίουσ᾽ ὀστέα λευκὰ θυηέντων ἐπὶ βωμῶν.

In dealing with Grecian mythes generally, it is convenient to
distribute them into such as belong to the Gods and such as belong
to the Heroes, according as the one or the other are the prominent
personages. The former class manifests, more palpably than the latter,
their real origin, as growing out of the faith and the feelings,
without any necessary basis, either of matter of fact or allegory:
moreover, they elucidate more directly the religion of the Greeks, so
important an item in their character as a people. But in point of fact,
most of the mythes present to us Gods, Heroes and Men, in juxtaposition
one with the other and the richness of Grecian mythical literature
arises from the infinite diversity of combinations thus opened out;
first by the three class-types, God, Hero, and Man; next by the strict
keeping with which each separate class and character is handled. We
shall now follow downward the stream of mythical time, which begins
with the Gods, to the Heroic legends, or those which principally
concern the Heroes and Heroines; for the latter were to the full as
important in legend as the former.




CHAPTER II.

LEGENDS RELATING TO HEROES AND MEN.


The Hesiodic theogony gives no account of anything like a creation of
man, nor does it seem that such an idea was much entertained in the
legendary vein of Grecian imagination; which commonly carried back
the present men by successive generations to some primitive ancestor,
himself sprung from the soil, or from a neighboring river or mountain,
or from a god, a nymph, etc. But the poet of the Hesiodic “Works and
Days” has given us a narrative conceived in a very different spirit
respecting the origin of the human race, more in harmony with the sober
and melancholy ethical tone which reigns through that poem.[146]

  [146] Hesiod, as cited in the Etymologicon Magnum (probably
  the Hesiodic Catalogue of Women, as Marktscheffel considers
  it, placing it Fragm. 133), gives the parentage of a certain
  _Brotos_, who must probably be intended as the first of men:
  Βρότος, ὡς μὲν Εὐήμερος ὁ Μεσσήνιος, ἀπὸ Βρότου τινὸς αὐτόχθονος·
  ὁ δὲ Ἡσίοδος, ἀπὸ Βρότου τοῦ Αἴθερος καὶ Ἡμέρας.

First (he tells us) the Olympic gods made the golden race,—good,
perfect, and happy men, who lived from the spontaneous abundance of the
earth, in ease and tranquillity like the gods themselves: they suffered
neither disease nor old age, and their death was like a gentle sleep.
After death they became, by the award of Zeus, guardian terrestrial
dæmons, who watch unseen over the proceedings of mankind—with the regal
privilege of dispensing to them wealth, and taking account of good and
bad deeds.[147]

  [147] Opp. Di. 120.—

    Αὐτὰρ ἐπειδὴ τοῦτο γένος κατὰ γαῖα κάλυψεν
    Τοὶ μὲν δαίμονές εἰσι Διὸς μεγάλου διὰ βουλὰς
    Ἐσθλοὶ, ἐπιχθόνιοι, φύλακες θνητῶν ἀνθρώπων·
    Οἵ ῥα φυλάσσουσίν τε δίκας καὶ σχέτλια ἔργα,
    Ἠέρα ἑσσάμενοι, πάντη φοιτῶντες ἐπ᾽ αἶαν
    Πλουτόδοται· καὶ τοῦτο γέρας βασιληΐον ἔσχον.

Next, the gods made the silver race,—unlike and greatly inferior, both
in mind and body, to the golden. The men of this race were reckless and
mischievous towards each other, and disdainful of the immortal gods,
to whom they refused to offer either worship or sacrifice. Zeus in his
wrath buried them in the earth: but there they still enjoy a secondary
honor, as the Blest of the under-world.[148]

  [148] Opp. Di. 140.—

    Αὐτὰρ ἐπεὶ καὶ τοῦτο γένος κατὰ γαῖα κάλυψε,
    Τοὶ μὲν ὑποχθόνιοι μάκαρες θνητοὶ καλέονται
    Δεύτεροι, ἀλλ᾽ ἔμπης τιμὴ καὶ τοῖσιν ὀπηδεῖ.

Thirdly, Zeus made the brazen race, quite different from the silver.
They were made of hard ash-wood, pugnacious and terrible; they were
of immense strength and adamantine soul, nor did they raise or touch
bread. Their arms, their houses, and their implements were all of
brass: there was then no iron. This race, eternally fighting, perished
by each other’s hands, died out, and descended without name or
privilege to Hadês.[149]

  [149] The ash was the wood out of which spear-handles were made
  (Iliad, xvi. 142): the Νύμφαι Μέλιαι are born along with the
  Gigantes and the Erinnyes (Theogon. 187),—“gensque virûm truncis
  et duro robore nata” (Virgil, Æneid, viii. 315),—_hearts of oak_.

Next, Zeus made a fourth race, far juster and better than the last
preceding. These were the Heroes or demigods, who fought at the sieges
of Troy and Thêbes. But this splendid stock also became extinct: some
perished in war, others were removed by Zeus to a happier state in the
islands of the Blest. There they dwell in peace and comfort, under
the government of Kronos, reaping thrice in the year the spontaneous
produce of the earth.[150]

  [150] Opp. Di. 157.—

    Ἀνδρῶν Ἡρώων θεῖον γένος, οἳ καλέονται
    Ἡμίθεοι προτέρῃ γενέῃ κατ᾽ ἀπείρονα γαῖαν.

The fifth race, which succeeds to the Heroes, is of iron: it is the
race to which the poet himself belongs, and bitterly does he regret it.
He finds his contemporaries mischievous, dishonest, unjust, ungrateful,
given to perjury, careless both of the ties of consanguinity and of the
behests of the gods: Nemesis and Ædôs (Ethical Self-reproach) have left
earth and gone back to Olympus. How keenly does he wish that his lot
had been cast either earlier or later![151] This iron race is doomed to
continual guilt, care, and suffering, with a small infusion of good;
but the time will come when Zeus will put an end to it. The poet does
not venture to predict what sort of race will succeed.

  [151] Opp. Di. 173.—

    Μηκέτ᾽ ἔπειτ᾽ ὤφειλον ἐγὼ πέμπτοισι μετεῖναι
    Ἀνδράσιν, ἀλλ᾽ ἢ πρόσθε θανεῖν, ἢ ἔπειτα γενέσθαι.
    Νῦν γὰρ δὴ γένος ἐστὶ σιδήρεον....

Such is the series of distinct races of men, which Hesiod, or the
author of the “Works and Days,” enumerates as having existed down to
his own time. I give it as it stands, without placing much confidence
in the various explanations which critics have offered. It stands
out in more than one respect from the general tone and sentiment of
Grecian legend: moreover the sequence of races is neither natural nor
homogeneous,—the heroic race not having any metallic denomination,
and not occupying any legitimate place in immediate succession to the
brazen. Nor is the conception of the dæmons in harmony either with
Homer or with the Hesiodic theogony. In Homer, there is scarcely any
distinction between gods and dæmons, while the gods are stated to
go about and visit the cities of men in various disguises for the
purpose of inspecting good and evil proceedings.[152] But in the poem
now before us, the distinction between gods and dæmons is generic.
The latter are invisible tenants of earth, remnants of the once happy
golden race whom the Olympic gods first made: the remnants of the
second or silver race are not dæmons, nor are they tenants of earth,
but they still enjoy an honorable posthumous existence as the Blest of
the under-world. Nevertheless the Hesiodic dæmons are in no way authors
or abettors of evil: on the contrary, they form the unseen police of
the gods, for the purpose of repressing wicked behavior in the world.

  [152] Odyss. xvii. 486.

We may trace, I think, in this quintuple succession of earthly races,
set forth by the author of the “Works and Days,” the confluence of
two veins of sentiment, not consistent one with the other, yet both
coëxisting in the author’s mind. The drift of his poem is thoroughly
didactic and ethical: though deeply penetrated with the injustice and
suffering which darken the face of human life, he nevertheless strives
to maintain, both in himself and in others, a conviction that on the
whole the just and laborious man will come off well,[153] and he
enforces in considerable detail the lessons of practical prudence and
virtue. This ethical sentiment, which dictates his appreciation of the
present, also guides his imagination as to the past. It is pleasing
to him to bridge over the chasm between the gods and degenerate man,
by the supposition of previous races,—the first altogether pure, the
second worse than the first, and the third still worse than the second;
and to show further how the first race passed by gentle death-sleep
into glorious immortality; how the second race was sufficiently wicked
to drive Zeus to bury them in the under-world, yet still leaving them
a certain measure of honor; while the third was so desperately violent
as to perish by its own animosities, without either name or honor of
any kind. The conception of the golden race passing after death into
good guardian dæmons, which some suppose to have been derived from a
comparison with oriental angels, presents itself to the poet partly as
approximating this race to the gods, partly as a means of constituting
a triple gradation of post-obituary existence, proportioned to the
character of each race whilst alive. The denominations of gold and
silver, given to the first two races, justify themselves, like those
given by Simonidês of Amorgos and by Phokylidês to the different
characters of women, derived from the dog, the bee, the mare, the ass,
and other animals; and the epithet of brazen is specially explained
by reference to the material which the pugnacious third race so
plentifully employed for their arms and other implements.

  [153] There are some lines, in which he appears to believe that,
  under the present wicked and treacherous rulers, it is not the
  interest of any man to be just (Opp. Di. 270):—

    Νῦν δὴ ἐγὼ μήτ᾽ αὐτὸς ἐν ἀνθρώποισι δίκαιος
    Εἴην, μήτ᾽ ἐμὸς υἱός· ἐπεὶ κακόν ἐστι δίκαιον
    Ἔμμεναι, εἰ μείζω γε δίκην ἀδικώτερος ἕξει·
    Ἀλλὰ τόδ᾽ οὔπω ἔολπα τελεῖν Δία τερπικέραυνον.

  On the whole, however, his conviction is to the contrary.

  Plutarch rejects the above four lines, seemingly on no other
  ground than because he thought them immoral and unworthy of
  Hesiod (see Proclus _ad loc._). But they fall in perfectly with
  the temper of the poem: and the rule of Plutarch is inadmissible,
  in determining the critical question of what is genuine or
  spurious.

So far we trace intelligibly enough the moralizing vein: we find the
revolutions of the past so arranged as to serve partly as an ethical
lesson, partly as a suitable preface to the present.[154] But fourth
in the list comes “the divine race of Heroes:” and here a new vein
of thought is opened by the poet. The symmetry of his ethical past
is broken up, in order to make way for these cherished beings of the
national faith. For though the author of the “Works and Days” was
himself of a didactic cast of thought, like Phokylidês, or Solôn,
or Theognis, yet he had present to his feelings, in common with his
countrymen, the picture of Grecian foretime, as it was set forth in
the current mythes, and still more in Homer and those other epical
productions which were then the only existing literature and history.
It was impossible for him to exclude, from his sketch of the past,
either the great persons or the glorious exploits which these poems
ennobled; and even if he himself could have consented to such an
exclusion, the sketch would have become repulsive to his hearers.
But the chiefs who figured before Thêbes and Troy could not be well
identified either with the golden, the silver, or the brazen race:
moreover it was essential that they should be placed in immediate
contiguity with the present race, because their descendants, real or
supposed, were the most prominent and conspicuous of existing men.
Hence the poet is obliged to assign to them the fourth place in the
series, and to interrupt the descending ethical movement in order to
interpolate them between the brazen and the iron race, with neither
of which they present any analogy. The iron race, to which the poet
himself unhappily belongs, is the legitimate successor, not of the
heroic, but of the brazen. Instead of the fierce and self-annihilating
pugnacity which characterizes the latter, the iron race manifests
an aggregate of smaller and meaner vices and mischiefs. It will not
perish by suicidal extinction—but it is growing worse and worse, and is
gradually losing its vigor, so that Zeus will not vouchsafe to preserve
much longer such a race upon the earth.

  [154] Aratus (Phænomen. 107) gives only three successive
  races,—the golden, silver, and brazen; Ovid superadds to these
  the iron race (Metamorph. i. 89-144): neither of them notice the
  heroic race.

  The observations both of Buttmann (Mythos der ältesten
  Menschengeschlechter, t. ii. p. 12 of the Mythologus) and of
  Völcker (Mythologie des Japetischen Geschlechts, § 6, pp.
  250-279) on this series of distinct races, are ingenious, and
  may be read with profit. Both recognize the disparate character
  of the fourth link in the series, and each accounts for it in a
  different manner. My own view comes nearer to that of Völcker,
  with some considerable differences; amongst which one is, that
  he rejects the verses respecting the dæmons, which seem to me
  capital parts of the whole scheme.

We thus see that the series of races imagined by the poet of the
“Works and Days” is the product of two distinct and incongruous veins
of imagination,—the didactic or ethical blending with the primitive
mythical or epical. His poem is remarkable as the most ancient didactic
production of the Greeks, and as one of the first symptoms of a
new tone of sentiment finding its way into their literature, never
afterwards to become extinct. The tendency of the “Works and Days” is
anti-heroic: far from seeking to inspire admiration for adventurous
enterprise, the author inculcates the strictest justice, the most
unremitting labor and frugality, and a sober, not to say anxious,
estimate of all the minute specialties of the future. Prudence and
probity are his means,—practical comfort and happiness his end. But
he deeply feels, and keenly exposes, the manifold wickedness and
short-comings of his contemporaries, in reference to this capital
standard. He turns with displeasure from the present men, not because
they are too feeble to hurl either the spear of Achilles or some
vast boundary-stone, but because they are rapacious, knavish, and
unprincipled.

The dæmons first introduced into the religious atmosphere of the
Grecian world by the author of the “Works and Days,” as generically
different from the gods, but as essentially good, and as forming the
intermediate agents and police between gods and men,—are deserving
of attention as the seed of a doctrine which afterwards underwent
many changes, and became of great importance, first as one of the
constituent elements of pagan faith, then as one of the helps to its
subversion. It will be recollected that the buried remnants of the
half-wicked silver race, though they are not recognized as dæmons,
are still considered as having a substantive existence, a name, and
dignity, in the under-world. The step was easy, to treat them as
dæmons also, but as dæmons of a defective and malignant character:
this step was made by Empedoclês and Xenocratês, and to a certain
extent countenanced by Plato.[155] There came thus to be admitted among
the pagan philosophers dæmons both good and bad, in every degree:
and these dæmons were found available as a means of explaining many
phænomena for which it was not convenient to admit the agency of the
gods. They served to relieve the gods from the odium of physical and
moral evils, as well as from the necessity of constantly meddling in
small affairs; and the objectionable ceremonies of the pagan world were
defended upon the ground that in no other way could the exigencies of
such malignant beings be appeased. They were most frequently noticed
as causes of evil, and thus the name (_dæmon_) came insensibly to
convey with it a bad sense,—the idea of an evil being as contrasted
with the goodness of a god. So it was found by the Christian writers
when they commenced their controversy with paganism. One branch of
their argument led them to identify the pagan gods with dæmons in the
evil sense, and the insensible change in the received meaning of the
word lent them a specious assistance. For they could easily show that
not only in Homer, but in the general language of early pagans, all
the gods generally were spoken of as dæmons—and therefore, verbally
speaking, Clemens and Tatian seemed to affirm nothing more against Zeus
or Apollo than was employed in the language of paganism itself. Yet the
audience of Homer or Sophoklês would have strenuously repudiated the
proposition, if it had been put to them in the sense which the word
_dæmon_ bore in the age and among the circle of these Christian writers.

  [155] See this subject further mentioned—_infra_, chap. xvi. p.
  565.

In the imagination of the author of the “Works and Days,” the dæmons
occupy an important place, and are regarded as being of serious
practical efficiency. When he is remonstrating with the rulers around
him upon their gross injustice and corruption, he reminds them of the
vast number of these immortal servants of Zeus who are perpetually on
guard amidst mankind, and through whom the visitations of the gods will
descend even upon the most potent evil doers.[156] His supposition that
the dæmons were not gods, but departed men of the golden race, allowed
him to multiply their number indefinitely, without too much cheapening
the divine dignity.

  [156] Opp. Di. 252. Τρὶς γὰρ μύριοί εἰσιν ἐπὶ χθονὶ πουλυβοτείρῃ,
  etc.

As this poet has been so much enslaved by the current legends as
to introduce the Heroic race into a series to which it does not
legitimately belong, so he has under the same influence inserted in
another part of his poem the mythe of Pandora and Promêtheus,[157] as
a means of explaining the primary diffusion, and actual abundance,
of evil among mankind. Yet this mythe can in no way consist with his
quintuple scale of distinct races, and is in fact a totally distinct
theory to explain the same problem,—the transition of mankind from
a supposed state of antecedent happiness to one of present toil and
suffering. Such an inconsistency is not a sufficient reason for
questioning the genuineness of either passage; for the two stories,
though one contradicts the other, both harmonize with that central
purpose which governs the author’s mind,—a querulous and didactic
appreciation of the present. That such was his purpose appears not only
from the whole tenor of his poem, but also from the remarkable fact
that his own personality, his own adventures and kindred, and his own
sufferings, figure in it conspicuously. And this introduction of self
imparts to it a peculiar interest. The father of Hesiod came over from
the Æolic Kymê, with the view of bettering his condition, and settled
at Askra in Bœotia, at the foot of Mount Helicon. After his death his
two sons divided the family inheritance: but Hesiod bitterly complains
that his brother Persês cheated and went to law with him, and obtained
through corrupt judges an unjust decision. He farther reproaches his
brother with a preference for the suits and unprofitable bustle of the
agora, at a time when he ought to be laboring for his subsistence in
the field. Askra indeed was a miserable place, repulsive both in summer
and winter. Hesiod had never crossed the sea, except once from Aulis to
Eubœa, whither he went to attend the funeral games of Amphidamas, the
chief of Chalkis: he sung a hymn, and gained as prize a tripod, which
he consecrated to the muses in Helicon.[158]

  [157] Opp. Di. 50-105.

  [158] Opp. Di. 630-650, 27-45.

These particulars, scanty as they are, possess a peculiar value, as
the earliest authentic memorandum respecting the doing or suffering of
any actual Greek person. There is no external testimony at all worthy
of trust respecting the age of the “Works and Days:” Herodotus treats
Hesiod and Homer as belonging to the same age, four hundred years
before his own time; and there are other statements besides, some
placing Hesiod at an earlier date than Homer, some at a later. Looking
at the internal evidences, we may observe that the pervading sentiment,
tone and purpose of the poem is widely different from that of the Iliad
and Odyssey, and analogous to what we read respecting the compositions
of Archilochus and the Amorgian Simonidês. The author of the “Works
and Days” is indeed a preacher and not a satirist: but with this
distinction, we find in him the same predominance of the present and
the positive, the same disposition to turn the muse into an exponent of
his own personal wrongs, the same employment of Æsopic fable by way of
illustration, and the same unfavorable estimate of the female sex,[159]
all of which may be traced in the two poets above mentioned, placing
both of them in contrast with the Homeric epic. Such an internal
analogy, in the absence of good testimony, is the best guide which we
can follow in determining the date of the “Works and Days,” which we
should accordingly place shortly after the year 700 B. C. The style
of the poem might indeed afford a proof that the ancient and uniform
hexameter, though well adapted to continuous legendary narrative or to
solemn hymns, was somewhat monotonous when called upon either to serve
a polemical purpose or to impress a striking moral lesson. When poets,
then the only existing composers, first began to apply their thoughts
to the cut and thrust of actual life, aggressive or didactic, the verse
would be seen to require a new, livelier and smarter metre; and out
of this want grew the elegiac and the iambic verse, both seemingly
contemporaneous, and both intended to supplant the primitive hexameter
for the short effusions then coming into vogue.

  [159] Compare the fable (αἶνος) in the “Works and Days,” v. 200,
  with those in Archilochus, Fr. xxxviii. and xxxix., Gaisford,
  respecting the fox and the ape; and the legend of Pandôra (v. 95
  and v. 705) with the fragment of Simonidês of Amorgos respecting
  women (Fr. viii. ed. Welcker, v. 95-115); also Phokylidês ap.
  Stobæum Florileg. lxxi.

  Isokratês assimilates the character of the “Works and Days” to
  that of Theognis and Phokylidês (ad Nikokl. Or. ii. p. 23).




CHAPTER III.

LEGEND OF THE IAPETIDS.


The sons of the Titan god Iapetus, as described in the Hesiodic
theogony, are Atlas, Menœtius, Promêtheus and Epimêtheus.[160] Of
these, Atlas alone is mentioned by Homer in the Odyssey, and even he
not as the son of Iapetus: the latter himself is named in the Iliad as
existing in Tartarus along with Kronos. The Homeric Atlas “knows the
depths of the whole sea, and keeps by himself those tall pillars which
hold the heaven apart from the earth.”[161]

  [160] Hesiod, Theog. 510.

  [161] Hom. Odyss. i. 120.—

    Ἄτλαντος θυγατὴρ ὀλοόφρονος, ὅστε θαλάσσης
    Πάσης βένθεα οἶδε, ἔχει δέ τε κίονας αὐτὸς
    Μακρὰς, αἳ γαῖάν τε καὶ οὐρανὸν ἀμφὶς ἔχουσιν.

As the Homeric theogony generally appears much expanded in Hesiod, so
also does the family of Iapetus, with their varied adventures. Atlas is
here described, not as the keeper of the intermediate pillars between
heaven and earth, but as himself condemned by Zeus to support the
heaven on his head and hands;[162] while the fierce Menœtius is thrust
down to Erebus as a punishment for his ungovernable insolence. But the
remaining two brothers, Promêtheus and Epimêtheus, are among the most
interesting creations of Grecian legend, and distinguished in more than
one respect from all the remainder.

  [162] Hesiod, Theog. 516.—

    Ἄτλας δ᾽ οὐρανὸν εὐρὺν ἔχει κρατερῆς ὑπ᾽ ἀνάγκης
    Ἑστηὼς, κεφαλῇ τε καὶ ἀκαμάτοισι χέρεσσι.

  Hesiod stretches far beyond the simplicity of the Homeric
  conception.

First, the main battle between Zeus and the Titan gods is a contest of
force purely and simply—mountains are hurled and thunder is launched,
and the victory remains to the strongest. But the competition between
Zeus and Promêtheus is one of craft and stratagem: the victory does
indeed remain to the former, but the honors of the fight belong to the
latter. Secondly, Promêtheus and Epimêtheus (the fore-thinker and the
after-thinker[163]) are characters stamped at the same mint and by
the same effort, the express contrast and antithesis of each other.
Thirdly, mankind are here expressly brought forward, not indeed as
active partners in the struggle, but as the grand and capital subjects
interested,—as gainers or sufferers by the result. Promêtheus appears
in the exalted character of champion of the human race, even against
the formidable superiority of Zeus.

  [163] Pindar extends the family of Epimêtheus and gives him a
  daughter, Πρόφασις (Pyth. v. 25), _Excuse_, the offspring of
  After-thought.

In the primitive or Hesiodic legend, Promêtheus is not the creator
or moulder of man; it is only the later additions which invest him
with this character.[164] The race are supposed as existing, and
Promêtheus, a member of the dispossessed body of Titan gods, comes
forward as their representative and defender. The advantageous bargain
which he made with Zeus on their behalf, in respect to the partition of
the sacrificial animals, has been recounted in the preceding chapter.
Zeus felt that he had been outwitted, and was exceeding wroth. In his
displeasure he withheld from mankind the inestimable comfort of fire,
so that the race would have perished, had not Promêtheus stolen fire,
in defiance of the command of the Supreme Ruler, and brought it to men
in the hollow of a ferule.[165]

  [164] Apollodôr. i. 7. 1. Nor is he such either in Æschylus, or
  in the Platonic fable (Protag. c. 30), though this version became
  at last the most popular. Some hardened lumps of clay, remnants
  of that which had been employed by Promêtheus in moulding man,
  were shown to Pausanias at Panopeus in Phokis (Paus. x. 4, 3).

  The first Epigram of Erinna (Anthol. i. p. 58, ed. Branck) seems
  to allude to Promêtheus as moulder of man. The expression of
  Aristophanês (Aves, 689)—πλάσματα πηλοῦ—does not necessarily
  refer to Promêtheus.

  [165] Hesiod, Theog. 566; Opp. Di. 52.

Zeus was now doubly indignant, and determined to play off a still more
ruinous stratagem. Hêphæstos, by his direction, moulded the form of
a beautiful virgin; Athênê dressed her, Aphroditê and the Charities
bestowed upon her both ornament and fascination, while Hermês infused
into her the mind of a dog, a deceitful spirit, and treacherous
words.[166] The messenger of the gods conducted this “fascinating
mischief” to mankind, at a time when Promêtheus was not present.
Now Epimêtheus had received from his brother peremptory injunctions
not to accept from the hands of Zeus any present whatever; but the
beauty of Pandôra (so the newly-formed female was called) was not to
be resisted. She was received and admitted among men, and from that
moment their comfort and tranquillity was exchanged for suffering of
every kind.[167] The evils to which mankind are liable had been before
enclosed in a cask in their own keeping: Pandôra in her malice removed
the lid of the cask, and out flew these thousand evils and calamities,
to exercise forever their destroying force. Hope alone remained
imprisoned, and therefore without efficacy, as before—the inviolable
lid being replaced before she could escape. Before this incident
(says the legend) men had lived without disease or suffering; but now
both earth and sea are full of mischiefs, while maladies of every
description stalk abroad by day as well as by night,[168] without any
hope for man of relief to come.

  [166] Theog. 580; Opp. Di. 50-85.

  [167] Opp. Di. 81-90.

  [168] Opp. Di. 93. Pandôra does not _bring with her_ the cask,
  as the common version of this story would have us suppose: the
  cask exists fast closed in the custody of Epimêtheus, or of man
  himself, and Pandôra commits the fatal treachery of removing
  the lid. The case is analogous to that of the closed bag of
  unfavorable winds which Æolus gives into the hands of Odysseus,
  and which the guilty companions of the latter force open, to the
  entire ruin of his hopes (Odyss. x. 19-50). The idea of the two
  casks on the threshold of Zeus, lying ready for dispensation—one
  full of evils the other of benefits—is Homeric (Iliad, xxiv.
  527):—

    Δοιοὶ γάρ τε πίθοι κατακείαται ἐν Διὸς οὔδει, etc.

  Plutarch assimilates to this the πίθος opened by Pandôra,
  Consolat. ad Apollôn. c. 7. p. 105. The explanation here given
  of the Hesiodic passage relating to Hope, is drawn from an able
  article in the Wiener Jahrbücher, vol. 109 (1845), p. 220,
  Ritter; a review of Schömmann’s translation of the Promêtheus of
  Æschylus. The diseases and evils are inoperative so long as they
  remain shut up in the cask: the same mischief-making influence
  which lets them out to their calamitous work, takes care that
  Hope shall still continue a powerless prisoner in the inside.

The Theogony gives the legend here recounted, with some
variations—leaving out the part of Epimêtheus altogether, as well as
the cask of evils. Pandôra is the ruin of man, simply as the mother
and representative of the female sex.[169] And the variations are
thus useful, as they enable us to distinguish the essential from the
accessory circumstances of the story.

  [169] Theog. 590.—

    Ἐκ τῆς γὰρ γένος ἐστὶ γυναικῶν θηλυτεράων,
    Τῆς γὰρ ὀλώιόν ἐστι γένος· καὶ φῦλα γυναικῶν,
    Πῆμα μέγα θνητοῖσι μετ᾽ ἀνδράσι ναιετάουσι, etc.

“Thus (says the poet, at the conclusion of his narrative) it is
not possible to escape from the purposes of Zeus.”[170] His mythe,
connecting the calamitous condition of man with the malevolence of the
supreme god, shows, first, by what cause such an unfriendly feeling
was raised; next, by what instrumentality its deadly results were
brought about. The human race are not indeed the creation, but the
protected flock of Promêtheus, one of the elder or dispossessed Titan
gods: when Zeus acquires supremacy, mankind along with the rest become
subject to him, and are to make the best bargain they can respecting
worship and service to be yielded. By the stratagem of their advocate
Promêtheus, Zeus is cheated into such a partition of the victims as is
eminently unprofitable to him; whereby his wrath is so provoked, that
he tries to subtract from man the use of fire. Here however his scheme
is frustrated by the theft of Promêtheus: but his second attempt is
more successful, and he in his turn cheats the unthinking Epimêtheus
into the acceptance of a present (in spite of the peremptory interdict
of Promêtheus) by which the whole of man’s happiness is wrecked. This
legend grows out of two feelings; partly as to the relations of the
gods with man, partly as to the relation of the female sex with the
male. The present gods are unkind towards man, but the old gods, with
whom man’s lot was originally cast, were much kinder—and the ablest
among them stands forward as the indefatigable protector of the race.
Nevertheless, the mere excess of his craft proves the ultimate ruin of
the cause which he espouses. He cheats Zeus out of a fair share of the
sacrificial victim, so as both to provoke and justify a retaliation
which he cannot be always at hand to ward off: the retaliation is, in
his absence, consummated by a snare laid for Epimêtheus and voluntarily
accepted. And thus, though Hesiod ascribes the calamitous condition
of man to the malevolence of Zeus, his piety suggests two exculpatory
pleas for the latter: mankind have been the first to defraud Zeus of
his legitimate share of the sacrifice—and they have moreover been
consenting parties to their own ruin. Such are the feelings, as to
the relation between the gods and man, which have been one of the
generating elements of this legend. The other element, a conviction
of the vast mischief arising to man from women, whom yet they cannot
dispense with, is frequently and strongly set forth in several of the
Greek poets—by Simonidês of Amorgos and Phokylidês, not less than by
the notorious misogynist Euripidês.

  [170] Opp. Di. 105.—

      Οὕτως οὔτι πῆ ἐστὶ Διὸς νόον ἐξαλέασθαι.

But the miseries arising from woman, however great they might be,
did not reach Promêtheus himself. For him, the rash champion who
had ventured “to compete in sagacity”[171] with Zeus, a different
punishment was in store. Bound by heavy chains to a pillar, he remained
fast imprisoned for several generations: every day did an eagle prey
upon his liver, and every night did the liver grow afresh for the
next day’s suffering. At length Zeus, eager to enhance the glory of
his favorite son Hêraclês, permitted the latter to kill the eagle and
rescue the captive.[172]

  [171] Theog. 534. Οὕνεκ᾽ ἐρίζετο βουλὰς ὑπερμενέϊ Κρονίωνι.

  [172] Theog. 521-532.

Such is the Promêthean mythe as it stands in the Hesiodic poems; its
earliest form, as far as we can trace. Upon it was founded the sublime
tragedy of Æschylus, “The Enchained Promêtheus,” together with at
least one more tragedy, now lost, by the same author.[173] Æschylus
has made several important alterations; describing the human race, not
as having once enjoyed and subsequently lost a state of tranquillity
and enjoyment, but as originally feeble and wretched. He suppresses
both the first trick played off by Promêtheus upon Zeus respecting
the partition of the victim—and the final formation and sending of
Pandôra—which are the two most marked portions of the Hesiodic story;
while on the other hand he brings out prominently and enlarges upon
the theft of fire,[174] which in Hesiod is but slightly touched. If
he has thus relinquished the antique simplicity of the story, he has
rendered more than ample compensation by imparting to it a grandeur of
_idéal_, a large reach of thought combined with appeals to our earnest
and admiring sympathy, and a pregnancy of suggestion in regard to the
relations between the gods and man, which soar far above the Hesiodic
level—and which render his tragedy the most impressive, though not
the most artistically composed, of all Grecian dramatic productions.
Promêtheus there appears not only as the heroic champion and sufferer
in the cause and for the protection of the human race, but also as
the gifted teacher of all the arts, helps, and ornaments of life,
amongst which fire is only one:[175] all this against the will and in
defiance of the purpose of Zeus, who, on acquiring his empire, wished
to destroy the human race and to beget some new breed.[176] Moreover,
new relations between Promêtheus and Zeus are superadded by Æschylus.
At the commencement of the struggle between Zeus and the Titan gods,
Promêtheus had vainly attempted to prevail upon the latter to conduct
it with prudence; but when he found that they obstinately declined
all wise counsel, and that their ruin was inevitable, he abandoned
their cause and joined Zeus. To him and to his advice Zeus owed the
victory: yet the monstrous ingratitude and tyranny of the latter is
now manifested by nailing him to a rock, for no other crime than
because he frustrated the purpose of extinguishing the human race, and
furnished to them the means of living with tolerable comfort.[177] The
new ruler Zeus, insolent with his victory over the old gods, tramples
down all right, and sets at naught sympathy and obligation, as well
towards gods as towards man. Yet the prophetic Promêtheus, in the midst
of intense suffering, is consoled by the foreknowledge that the time
will come when Zeus must again send for him, release him, and invoke
his aid, as the sole means of averting from himself dangers otherwise
insurmountable. The security and means of continuance for mankind
have now been placed beyond the reach of Zeus—whom Promêtheus proudly
defies, glorying in his generous and successful championship,[178]
despite the terrible price which he is doomed to pay for it.

  [173] Of the tragedy called Προμηθεὺς Λυόμενος some few fragments
  yet remain: Προμηθεὺς Πύρφορος was a satyric drama, according to
  Dindorf. Welcker recognizes a third tragedy, Προμηθεὺς Πύρφορος,
  and a satyric drama, Προμηθεὺς Πυρκαεύς (Die Griechisch.
  Tragödien, vol. i. p. 30). The story of Promêtheus had also been
  handled by Sapphô in one of her lost songs (Servius ad Virgil.
  Eclog. vi. 42).

  [174] Apollodôrus too mentions only the theft of fire (i. 7. 1).

  [175] Æsch. Prom. 442-506.—

    Πᾶσαι τέχναι βροτοῖσιν ἐκ Προμηθέως.

  [176] Æsch. Prom. 231.—

          βροτῶν δὲ τῶν ταλαιπώρων λόγον
    Οὐκ ἔσχεν οὐδέν᾽, ἀλλ᾽ ἀϊστώσας γένος
    Τὸ πᾶν, ἔχρῃζεν ἄλλο φιτῦσαι νέον.

  [177] Æsch. Prom. 198-222. 123.—

    διὰ τὴν λίαν φιλότητα βροτῶν.

  [178] Æsch. Prom. 169-770.

As the Æschylean Promêtheus, though retaining the old lineaments, has
acquired a new coloring, soul and character, so he has also become
identified with a special locality. In Hesiod, there is no indication
of the place in which he is imprisoned; but Æschylus places it in
Scythia,[179] and the general belief of the Greeks supposed it to be on
Mount Caucasus. So long and so firmly did this belief continue, that
the Roman general Pompey, when in command of an army in Kolchis, made
with his companion, the literary Greek Theophanês, a special march to
view the spot in Caucasus where Promêtheus had been transfixed.[180]

  [179] Prometh. 2. See also the Fragments of the Promêtheus
  Solutus, 177-179, ed. Dindorf, where Caucasus is specially named;
  but v. 719 of the Promêtheus Vinctus seems to imply that Mount
  Caucasus is a place different from that to which the suffering
  prisoner is chained.

  [180] Appian, Bell. Mithridat. c. 103.




CHAPTER IV.

HEROIC LEGENDS.—GENEALOGY OF ARGOS.


Having briefly enumerated the gods of Greece, with their chief
attributes as described in legend, we come to those genealogies which
connected them with historical men.

In the retrospective faith of a Greek, the ideas of worship and
ancestry coalesced. Every association of men, large or small, in whom
there existed a feeling of present union, traced back that union to
some common initial progenitor; that progenitor being either the common
god whom they worshipped, or some semi-divine person closely allied
to him. What the feelings of the community require is, a continuous
pedigree to connect them with this respected source of existence,
beyond which they do not think of looking back. A series of names,
placed in filiation or fraternity, together with a certain number of
family or personal adventures ascribed to some of the individuals
among them, constitute the ante-historical past through which the
Greek looks back to his gods. The names of this genealogy are, to a
great degree, gentile or local names familiar to the people,—rivers,
mountains, springs, lakes, villages, demes, etc.,—embodied as persons,
and introduced as acting or suffering. They are moreover called kings
or chiefs, but the existence of a body of subjects surrounding them
is tacitly implied rather than distinctly set forth; for their own
personal exploits or family proceedings constitute for the most part
the whole matter of narrative. And thus the genealogy was made to
satisfy at once the appetite of the Greeks for romantic adventure, and
their demand for an unbroken line of filiation between themselves and
the gods. The eponymous personage, from whom the community derive their
name, is sometimes the begotten son of the local god, sometimes an
indigenous man sprung from the earth, which is indeed itself divinized.

It will be seen from the mere description of these genealogies that
they included elements human and historical, as well as elements divine
and extra-historical. And if we could determine the time at which any
genealogy was first framed, we should be able to assure ourselves
that the men then represented as present, together with their fathers
and grandfathers, were real persons of flesh and blood. But this is
a point which can seldom be ascertained; moreover, even if it could
be ascertained, we must at once set it aside, if we wish to look at
the genealogy in the point of view of the Greeks. For to them, not
only all the members were alike real, but the gods and heroes at the
commencement were in a certain sense the most real; at least, they were
the most esteemed and indispensable of all. The value of the genealogy
consisted, not in its length, but in its continuity; not (according
to the feeling of modern aristocracy) in the power of setting out a
prolonged series of human fathers and grandfathers, but in the sense
of ancestral union with the primitive god. And the length of the
series is traceable rather to humility, inasmuch as the same person
who was gratified with the belief that he was descended from a god in
the fifteenth generation, would have accounted it criminal insolence
to affirm that a god was his father or grandfather. In presenting to
the reader those genealogies which constitute the supposed primitive
history of Hellas, I make no pretence to distinguish names real and
historical from fictitious creations; partly because I have no evidence
upon which to draw the line, and partly because by attempting it I
should altogether depart from the genuine Grecian point of view.

Nor is it possible to do more than exhibit a certain selection of
such as were most current and interesting; for the total number of
them which found place in Grecian faith exceeds computation. As a
general rule, every deme, every gens, every aggregate of men accustomed
to combined action, religious or political, had its own. The small
and unimportant demes into which Attica was divided had each its
ancestral god and heroes, just as much as the great Athens herself.
Even among the villages of Phokis, which Pausanias will hardly permit
himself to call towns, deductions of legendary antiquity were not
wanting. And it is important to bear in mind, when we are reading the
legendary genealogies of Argos, or Sparta, or Thêbes, that these are
merely samples amidst an extensive class, all perfectly analogous,
and all exhibiting the religious and patriotic retrospect of some
fraction of the Hellenic world. They are no more matter of historical
tradition than any of the thousand other legendary genealogies which
men delighted to recall to memory at the periodical festivals of their
gens, their deme, or their village.

With these few prefatory remarks, I proceed to notice the most
conspicuous of the Grecian heroic pedigrees, and first, that of Argos.

The earliest name in Argeian antiquity is that of Inachus, the son
of Oceanus and Têthys, who gave his name to the river flowing under
the walls of the town. According to the chronological computations of
those who regarded the mythical genealogies as substantive history,
and who allotted a given number of years to each generation, the reign
of Inachus was placed 1986 B. C., or about 1100 years prior to the
commencement of the recorded Olympiads.[181]

  [181] Apollodôr. ii. 1. Mr. Fynes Clinton does not admit the
  historical reality of Inachus; but he places Phorôneus seventeen
  generations, or 570 years prior to the Trojan war, 978 years
  earlier than the first recorded Olympiad. See Fasti Hellenici,
  vol. iii. c. 1. p. 19.

The sons of Inachus were Phorôneus and Ægialeus; both of whom however
were sometimes represented as autochthonous men, the one in the
territory of Argos, the other in that of Sikyôn. Ægialeus gave his
name to the north-western region of the Peloponnêsus, on the southern
coast of the Corinthian Gulf.[182] The name of Phorôneus was of great
celebrity in the Argeian mythical genealogies, and furnished both the
title and the subject of the ancient poem called Phorônis, in which he
is styled “the father of mortal men.”[183] He is said to have imparted
to mankind, who had before him lived altogether isolated, the first
notion and habits of social existence, and even the first knowledge of
fire: his dominion extended over the whole Peloponnêsus. His tomb at
Argos, and seemingly also the place called the Phorônic city, in which
he formed the first settlement of mankind, were still shown in the days
of Pausanias.[184] The offspring of Phorôneus, by the nymph Teledikê,
were Apis and Niobê. Apis, a harsh ruler, was put to death by Thelxiôn
and Telchin, having given to Peloponnêsus the name of Apia:[185] he
was succeeded by Argos, the son of his sister Niobê by the god Zeus.
From this sovereign Peloponnêsus was denominated Argos. By his wife
Evadnê, daughter of Strymôn,[186] he had four sons, Ekbasus, Peiras,
Epidaurus, and Kriasus. Ekbasus was succeeded by his son Agênôr,
and he again by his son Argos Panoptês,—a very powerful prince who
is said to have had eyes distributed over all his body, and to have
liberated Peloponnêsus from several monsters and wild animals which
infested it:[187] Akusilaus and Æschylus make this Argos an earth-born
person, while Pherekydês reports him as son of Arestôr. Iasus was the
son of Argos Panoptês by Ismênê, daughter of Asôpus. According to the
authors whom Apollodôrus and Pausanias prefer, the celebrated Iô was
his daughter: but the Hesiodic epic (as well as Akusilaus) represented
her as daughter of Peiras, while Æschylus and Kastor the chronologist
affirmed the primitive king Inachus to have been her father.[188] A
favorite theme, as well for the ancient genealogical poets as for the
Attic tragedians, were the adventures of Iô, of whom, while priestess
of Hêrê, at the ancient and renowned Hêræon between Mykênæ and Argos,
Zeus became amorous. When Hêrê discovered the intrigue and taxed him
with it, he denied the charge, and metamorphosed Iô into a white cow.
Hêrê, requiring that the cow should be surrendered to her, placed her
under the keeping of Argos Panoptês; but this guardian was slain by
Hermês, at the command of Zeus: and Hêrê then drove the cow Iô away
from her native land by means of the incessant stinging of a gad-fly,
which compelled her to wander without repose or sustenance over an
immeasurable extent of foreign regions. The wandering Iô gave her name
to the Ionian Gulf, traversed Epirus and Illyria, passed the chain
of Mount Hæmus and the lofty summits of Caucasus, and swam across
the Thracian or Cimmerian Bosporus (which also from her derived its
appellation) into Asia. She then went through Scythia, Cimmeria, and
many Asiatic regions, until she arrived in Egypt, where Zeus at length
bestowed upon her rest, restored her to her original form, and enabled
her to give birth to his black son Epaphos.[189]

  [182] Pausan. ii. 5, 4.

  [183] See Düntzer, Fragm. Epic. Græc. p. 57. The Argeian author
  Akusilaus treated Phorôneus as the first of men, Fragm. 14.
  Didot ap. Clem. Alex. Stromat i. p. 321. Φορωνῆες, a synonym for
  Argeians; Theocrit. Idyll. xxv. 200.

  [184] Apollodôr. ii. 1, 1; Pausan. ii. 15, 5; 19, 5; 20, 3.

  [185] Apis in Æschylus is totally different: ἰατρόμαντις or
  medical charmer, son of Apollo, who comes across the gulf from
  Naupactus, purifies the territory of Argos from noxious monsters,
  and gives to it the name of _Apia_ (Æschyl. Suppl. 265). Compare
  Steph. Byz. v. Ἀπίη; Soph. Œdip. Colon. 1303. The name Ἀπία for
  Peloponnêsus remains still a mystery, even after the attempt of
  Buttmann (Lexilogus, s. 19) to throw light upon it.

  Eusebius asserts that Niobê was the wife of Inachus and mother of
  Phorôneus, and pointedly contradicts those who call her daughter
  of Phorôneus—φασὶ δέ τινες Νιόβην Φορωνέως εἶναι θυγατέρα, ὅπερ
  οὐκ ἀληθές (Chronic. p. 23, ed. Scalig.): his positive tone is
  curious, upon such a matter.

  Hellanikus in his Argolica stated that Phorôneus had three sons,
  Pelasgus, Iasus and Agênôr, who at the death of their father
  divided his possessions by lot. Pelasgus acquired the country
  near the river Erasinus, and built the citadel of Larissa: Iasus
  obtained the portion near to Elis. After their decease, the
  younger brother Agênôr invaded and conquered the country, at the
  head of a large body of horse. It was from these three persons
  that Argos derived three epithets which are attached to it in the
  Homeric poems—Ἄργος Πελασγικὸν, Ἴασον, Ἱππόβοτον (Hellanik. Fr.
  38, ed. Didot; Phavorin. v. Ἄργος). This is a specimen of the way
  in which legendary persons as well as legendary events were got
  up to furnish an explanation of Homeric epithets: we may remark
  as singular, that Hellanikus seems to apply Πελασγικὸν Ἄργος to a
  portion of Peloponnêsus, while the Homeric Catalogue applies it
  to Thessaly.

  [186] Apollod. _l. c._ The mention of Strymôn seems connected
  with Æschylus Suppl. 255.

  [187] Akusil. Fragm. 17, ed. Didot; Æsch. Prometh. 568; Pherekyd.
  Fragm. 22, ed. Didot; Hesiod. Ægimius. Fr. 2, p. 56, ed. Düntzer:
  among the varieties of the story, one was that Argos was changed
  into a peacock (Schol. Aristoph. Aves, 102). Macrobius (i. 19)
  considers Argos as an allegorical expression of the starry
  heaven; an idea which Panofska also upholds in one of the recent
  Abhandlungen of the Berlin Academy, 1837, p. 121 _seq._

  [188] Apollod. ii. 1, 1; Pausan. ii. 16, 1; Æsch. Prom. v.
  590-663.

  [189] Æschyl. Prom. v. 790-850; Apollod. ii. 1. Æschylus in the
  Supplices gives a different version of the wanderings of Iô from
  that which appears in the Promêtheus: in the former drama he
  carries her through Phrygia, Mysia, Lydia, Pamphylia and Cilicia
  into Egypt (Supplic. 544-566): nothing is there said about
  Promêtheus, or Caucasus or Scythia, etc.

  The track set forth in the Supplices is thus geographically
  intelligible, that in the Promêtheus (though the most noticed of
  the two) defies all comprehension, even as a consistent fiction;
  nor has the erudition of the commentators been successful in
  clearing it up. See Schutz, Excurs. iv. ad Prometh. Vinct.
  pp. 144-149; Welcker, Æschylische Trilogie, pp. 127-146, and
  especially Völcker, Mythische Geographie der Griech. und Römer,
  part i. pp. 3-13.

  The Greek inhabitants at Tarsus in Cilicia traced their origin
  to Argos: their story was, that Triptolemus had been sent forth
  from that town in quest of the wandering Iô, that he had followed
  her to Tyre, and then renounced the search in despair. He and
  his companions then settled partly at Tarsus, partly at Antioch
  (Strabo, xiv. 673; xv. 750). This is the story of Kadmos and
  Eurôpê inverted, as happens so often with the Grecian mythes.

  Homer calls Hermês Ἀργειφόντης; but this epithet hardly affords
  sufficient proof that he was acquainted with the mythe of Iô,
  as Völcker supposes: it cannot be traced higher than Hesiod.
  According to some authors, whom Cicero copies, it was on account
  of the murder of Argos that Hermês was obliged to leave Greece
  and go into Egypt: then it was that he taught the Egyptians laws
  and letters (De Natur. Deor. iii. 22).

Such is a general sketch of the adventures which the ancient poets,
epic, lyric, and tragic, and the logographers after them, connect
with the name of the Argeian Iô,—one of the numerous tales which the
fancy of the Greeks deduced from the amorous dispositions of Zeus and
the jealousy of Hêrê. That the scene should be laid in the Argeian
territory appears natural, when we recollect that both Argos and Mykênæ
were under the special guardianship of Hêrê, and that the Hêræon
between the two was one of the oldest and most celebrated temples
in which she was worshipped. It is useful to compare this amusing
fiction with the representation reported to us by Herodotus, and
derived by him as well from Phœnician as from Persian antiquarians,
of the circumstances which occasioned the transit of Iô from Argos
to Egypt,—an event recognized by all of them as historical matter of
fact. According to the Persians, a Phœnician vessel had arrived at
the port near Argos, freighted with goods intended for sale to the
inhabitants of the country. After the vessel had remained a few days,
and disposed of most of her cargo, several Argeian women, and among
them Iô the king’s daughter, coming on board to purchase, were seized
and carried off by the crew, who sold Iô in Egypt.[190] The Phœnician
antiquarians, however, while they admitted the circumstance that Iô
had left her own country in one of their vessels, gave a different
color to the whole by affirming that she emigrated voluntarily, having
been engaged in an amour with the captain of the vessel, and fearing
that her parents might come to the knowledge of her pregnancy. Both
Persians and Phœnicians described the abduction of Iô as the first of a
series of similar acts between Greeks and Asiatics, committed each in
revenge for the preceding. First came the rape of Eurôpê from Phœnicia
by Grecian adventurers,—perhaps, as Herodotus supposed, by Krêtans:
next, the abduction of Mêdeia from Kolchis by Jasôn, which occasioned
the retaliatory act of Paris, when he stole away Helena from Menelaos.
Up to this point the seizures of women by Greeks from Asiatics, and by
Asiatics from Greeks, had been equivalents both in number and in wrong.
But the Greeks now thought fit to equip a vast conjoint expedition to
recover Helen, in the course of which they took and sacked Troy. The
invasions of Greece by Darius and Xerxes were intended, according to
the Persian antiquarians, as a long-delayed retribution for the injury
inflicted on the Asiatics by Agamemnôn and his followers.[191]

  [190] The story in Parthênius (Narrat. 1) is built upon this
  version of Iô’s adventures.

  [191] Herodot. i. 1-6. Pausanias (ii. 15, 1) will not undertake
  to determine whether the account given by Herodotus, or that of
  the old legend, respecting the cause which carried Iô from Argos
  to Egypt, is the true one: Ephorus (ap. Schol. Apoll. Rhod. ii.
  168) repeats the abduction of Iô to Egypt, by the Phœnicians,
  subjoining a strange account of the Etymology of the name
  Bosporus. The remarks of Plutarch on the narrative of Herodotus
  are curious: he adduces as one proof of the κακοήθεια (bad
  feeling) of Herodotus, that the latter inserts so discreditable
  a narrative respecting Iô, daughter of Inachus, “whom all Greeks
  believe to have been divinized by foreigners, to have given name
  to seas and straits, and to be the source of the most illustrious
  regal families.” He also blames Herodotus for rejecting Epaphus,
  Iô, Iasus and Argos, as highest members of the Perseid genealogy.
  He calls Herodotus φιλοβάρβαρος (Plutarch, De Malign. Herodoti,
  c. xi. xii. xiv. pp. 856, 857).

The account thus given of the adventures of Iô, when contrasted with
the genuine legend, is interesting, as it tends to illustrate the
phænomenon which early Grecian history is constantly presenting to
us,—the way in which the epical furniture of an unknown past is recast
and newly colored so as to meet those changes which take place in the
retrospective feelings of the present. The religious and poetical
character of the old legend disappears: nothing remains except the
names of persons and places, and the voyage from Argos to Egypt: we
have in exchange a sober, quasi-historical narrative, the value of
which consists in its bearing on the grand contemporary conflicts
between Persia and Greece, which filled the imagination of Herodotus
and his readers.

To proceed with the genealogy of the kings of Argos, Iasus was
succeeded by Krotôpus, son of his brother Agênôr; Krotôpus by
Sthenelas, and he again by Gelanôr.[192] In the reign of the latter,
Danaos came with his fifty daughters from Egypt to Argos; and here we
find another of those romantic adventures which so agreeably decorate
the barrenness of the mythical genealogies. Danaos and Ægyptos were
two brothers descending from Epaphos, son of Iô: Ægyptos had fifty
sons, who were eager to marry the fifty daughters of Danaos, in spite
of the strongest repugnance of the latter. To escape such a necessity,
Danaos placed his fifty daughters on board of a penteconter (or vessel
with fifty oars) and sought refuge at Argos; touching in his voyage at
the island of Rhodes, where he erected a statue of Athênê at Lindos,
which was long exhibited as a memorial of his passage. Ægyptos and
his sons followed them to Argos and still pressed their suit, to which
Danaos found himself compelled to assent; but on the wedding night he
furnished each of his daughters with a dagger, and enjoined them to
murder their husbands during the hour of sleep. His orders were obeyed
by all, with the single exception of Hypermnêstra, who preserved her
husband Lynkeus, incurring displeasure and punishment from her father.
He afterwards, however, pardoned her; and when, by the voluntary
abdication of Gelanôr, he became king of Argos, Lynkeus was recognized
as his son-in-law and ultimately succeeded him. The remaining
daughters, having been purified by Athênê and Hermês, were given in
marriage to the victors in a gymnic contest publicly proclaimed. From
Danaos was derived the name of Danai, applied to the inhabitants of the
Argeian territory,[193] and to the Homeric Greeks generally.

  [192] It would be an unprofitable fatigue to enumerate the
  multiplied and irreconcilable discrepancies in regard to every
  step of this old Argeian genealogy. Whoever desires to see
  them brought together, may consult Schubart, Quæstiones in
  Antiquitatem Heroicam, Marburg, 1832, capp. 1 and 2.

  The remarks which Schubart makes (p. 35) upon Petit-Radel’s
  Chronological Tables will be assented to by those who follow
  the unceasing string of contradictions, without any sufficient
  reason to believe that any one of them is more worthy of trust
  than the remainder, which he has cited:—“Videant alii, quomodo
  genealogias heroicas, et chronologiæ rationes, in concordiam
  redigant. Ipse abstineo, probe persuasus, stemmata vera,
  historiæ fide comprobata, in systema chronologiæ redigi posse:
  at ore per sæcula tradita, a poetis reficta, sæpe mutata, prout
  fabula postulare videbatur, ab historiarum deinde conditoribus
  restituta, scilicet, brevi, qualia prostant stemmata—chronologiæ
  secundum annos distributæ vincula semper recusatura esse.”

  [193] Apollod. ii. 1. The Supplices of Æschylus is the commencing
  drama of a trilogy on this subject of the Danaïdes,—Ἱκετίδες,
  Αἰγύπτιοι, Δαναΐδες. Welcker, Griechisch. Tragödien, vol. i. p.
  48: the two latter are lost. The old epic poem called Danaïs or
  Danaïdes, which is mentioned in the Tabula Iliaca as containing
  5000 verses, has perished, and is unfortunately very little
  alluded to: see Düntzer, Epic. Græc. Fragm. p. 3; Welcker, Der
  Episch. Kyklus, p. 35.

From the legend of the Danaïdes we pass to two barren names of kings,
Lynkeus and his son Abas. The two sons of Abas were Akrisios and
Prœtos, who, after much dissension, divided between them the Argeian
territory; Akrisios ruling at Argos, and Prœtos at Tiryns. The families
of both formed the theme of romantic stories. To pass over for the
present the legend of Bellerophôn, and the unrequited passion which
the wife of Prœtos conceived for him, we are told that the daughters
of Prœtos, beautiful, and solicited in marriage by suitors from all
Greece, were smitten with leprosy and driven mad, wandering in unseemly
guise throughout Peloponnêsus. The visitation had overtaken them,
according to Hesiod, because they refused to take part in the Bacchic
rites; according to Pherekydês and the Argeian Akusilaus,[194] because
they had treated scornfully the wooden statue and simple equipments of
Hêrê: the religious character of the old legend here displays itself in
a remarkable manner. Unable to cure his daughters, Prœtos invoked the
aid of the renowned Pylian prophet and leech, Melampus son of Amythaôn,
who undertook to remove the malady on condition of being rewarded
with the third part of the kingdom. Prœtos indignantly refused these
conditions: but the state of his daughters becoming aggravated and
intolerable, he was compelled again to apply to Melampus; who, on the
second request, raised his demands still higher, and required another
third of the kingdom for his brother Bias. These terms being acceded
to, he performed his part of the covenant. He appeased the wrath of
Hêrê by prayer and sacrifice; or, according to another account, he
approached the deranged women at the head of a troop of young men,
with shouting and ecstatic dance,—the ceremonies appropriate to the
Bacchic worship of Dionysos,—and in this manner effected their cure.
Melampus, a name celebrated in many different Grecian mythes, is the
legendary founder and progenitor of a great and long-continued family
of prophets. He and his brother Bias became kings of separate portions
of the Argeian territory: he is recognized as ruler there even in the
Odyssey, and the prophet Theoklymenos, his grandson, is protected and
carried to Ithaca by Telemachus.[195] Herodotus also alludes to the
cure of the women, and to the double kingdom of Melampus and Bias
in the Argeian land: he recognizes Melampus as the first person who
introduced to the knowledge of the Greeks the name and worship of
Dionysos, with its appropriate sacrifices and phallic processions. Here
again he historicizes various features of the old legend in a manner
not unworthy of notice.[196]

  [194] Apollod. 1. c.; Pherekyd. ap. Schol. Hom. Odyss. xv. 225;
  Hesiod, Fragm. Marktsch. Fr. 36, 37, 38. These Fragments belong
  to the Hesiodic Catalogue of Women: Apollodôrus seems to refer
  to some other of the numerous Hesiodic poems. Diodôrus (iv. 68)
  assigns the anger of Dionysos as the cause.

  [195] Odyss. xv. 240-256.

  [196] Herod. ix. 34; ii. 49: compare Pausan. ii. 18, 4. Instead
  of the Prœtides, or daughters of Prœtos, it is the Argeian
  women generally whom he represents Melampus as having cured,
  and the Argeians generally who send to Pylus to invoke his aid:
  the heroic personality which pervades the primitive story has
  disappeared.

  Kallimachus notices the Prœtid virgins as the parties suffering
  from madness, but he treats Artemis as the healing influence
  (Hymn. ad Dianam 235).

But Danaê, the daughter of Akrisios, with her son Perseus acquired
still greater celebrity than her cousins the Prœtides. An oracle had
apprized Akrisios that his daughter would give birth to a son by whose
hand he would himself be slain. To guard against this danger, he
imprisoned Danaê in a chamber of brass under ground. But the god Zeus
had become amorous of her, and found means to descend through the roof
in the form of a shower of gold: the consequence of his visits was the
birth of Perseus. When Akrisios discovered that his daughter had given
existence to a son, he enclosed both the mother and the child in a
coffer, which he cast into the sea.[197] The coffer was carried to the
isle of Seriphos, where Diktys, brother of the king Polydektês, fished
it up, and rescued both Danaê and Perseus. The exploits of Perseus,
when he grew up, against the three Phorkides or daughters of Phorkys,
and the three Gorgons, are among the most marvellous and imaginative
in all Grecian legend: they bear a stamp almost Oriental. I shall
not here repeat the details of those unparalleled hazards which the
special favor of Athênê enabled him to overcome, and which ended in his
bringing back from Libya the terrific head of the Gorgon Medusa, endued
with the property of turning every one who looked upon it into stone.
In his return, he rescued Andromeda, daughter of Kêpheus, who had been
exposed to be devoured by a sea-monster, and brought her back as his
wife. Akrisios trembled to see him after this victorious expedition,
and retired into Thessaly to avoid him; but Perseus followed him
thither, and having succeeded in calming his apprehensions, became
competitor in a gymnic contest where his grandfather was among the
spectators. By an incautious swing of his quoit, he unintentionally
struck Akrisios, and caused his death: the predictions of the oracle
were thus at last fulfilled. Stung with remorse at the catastrophe,
and unwilling to return to Argos, which had been the principality of
Akrisios, Perseus made an exchange with Megapenthês, son of Prœtos king
of Tiryns. Megapenthês became king of Argos, and Perseus of Tiryns:
moreover, the latter founded, within ten miles of Argos, the far-famed
city of Mykênæ. The massive walls of this city, like those of Tiryns,
of which remains are yet to be seen, were built for him by the Lykian
Cyclôpes.[198]

  [197] The beautiful fragment of Simonidês (Fragm. vii. ed.
  Gaisford. Poet. Min.), describing Danaê and the child thus
  exposed, is familiar to every classical reader.

  [198] Paus. ii. 15, 4; ii. 16, 5. Apollod. ii. 2. Pherekyd.
  Fragm. 26, Dind.

We here reach the commencement of the Perseid dynasty of Mykênæ. It
should be noticed, however, that there were among the ancient legends
contradictory accounts of the foundation of this city. Both the Odyssey
and the Great Eoiai enumerated, among the heroines, Mykênê, the Eponyma
of the city; the former poem classifying her with Tyrô and Alkmênê, the
latter describing her as the daughter of Inachus and wife of Arestôr.
And Akusilaus mentioned an Eponymus Mykêneus, the son of Spartôn and
grandson of Phorôneus.[199]

  [199] Odyss. ii. 120. Hesiod. Fragment. 154. Marktscheff.—Akusil.
  Fragm. 16. Pausan. ii. 16, 4. Hekatæus derived the name of the
  town from the μύκης of the sword of Perseus (Fragm. 360, Dind.).
  The Schol. ad Eurip. Orest. 1247, mentions Mykêneus as son of
  Spartôn, but grandson of Phêgeus the brother of Phorôneus.

The prophetic family of Melampus maintained itself in one of the
three parts of the divided Argeian kingdom for five generations, down
to Amphiaraos and his sons Alkmæôn and Amphilochos. The dynasty of
his brother Bias, and that of Megapenthês, son of Prœtos, continued
each for four generations: a list of barren names fills up the
interval.[200] The Perseids of Mykênæ boasted a descent long and
glorious, heroic as well as historical, continuing down to the last
sovereigns of Sparta.[201] The issue of Perseus was numerous: his son
Alkæos was father of Amphitryôn; another of his sons, Elektryôn, was
father of Alkmênê;[202] a third, Sthenelos, father of Eurystheus.

  [200] Pausan. ii. 18, 4.

  [201] Herodot. vi. 53.

  [202] In the Hesiodic Shield of Hêraklês, Alkmênê is distinctly
  mentioned as daughter of Elektryôn; the genealogical poet, Asios,
  called her the daughter of Amphiaraos and Eriphyle (Asii Fragm.
  4, ed. Markt. p. 412). The date of Asios cannot be precisely
  fixed; but he may be probably assigned to an epoch between the
  30th and 40th Olympiad.

  Asios must have adopted a totally different legend respecting the
  birth of Hêraklês and the circumstances preceding it, among which
  the deaths of her father and brothers are highly influential. Nor
  could he have accepted the received chronology of the sieges of
  Thêbes and Troy.

After the death of Perseus, Alkæos and Amphitryôn dwelt at Tiryns.
The latter became engaged in a quarrel with Elektryôn respecting
cattle, and in a fit of passion killed him:[203] moreover the piratical
Taphians from the west coast of Akarnania invaded the country, and
slew the sons of Elektryôn, so that Alkmênê alone was left of that
family. She was engaged to wed Amphitryôn; but she bound him by oath
not to consummate the marriage until he had avenged upon the Têleboæ
the death of her brothers. Amphitryôn, compelled to flee the country
as the murderer of his uncle, took refuge in Thêbes, whither Alkmênê
accompanied him: Sthenelos was left in possession of Tiryns. The
Kadmeians of Thêbes, together with the Locrians and Phocians, supplied
Amphitryôn with troops, which he conducted against the Têleboæ and
the Taphians:[204] yet he could not have subdued them without the aid
of Komæthô, daughter of the Taphian king Pterelaus, who conceived a
passion for him, and cut off from her father’s head the golden lock
to which Poseidôn had attached the gift of immortality.[205] Having
conquered and expelled his enemies, Amphitryôn returned to Thêbes,
impatient to consummate his marriage: but Zeus on the wedding-night
assumed his form and visited Alkmênê before him: he had determined to
produce from her a son superior to all his prior offspring,—“a specimen
of invincible force both to gods and men.”[206] At the proper time,
Alkmênê was delivered of twin sons: Hêraklês the offspring of Zeus,—the
inferior and unhonored Iphiklês, offspring of Amphitryôn.[207]

  [203] So runs the old legend in the Hesiodic Shield of Hêraklês
  (12-82). Apollodôrus (or Pherekydês, whom he follows) softens
  it down, and represents the death of Elektryôn as accidentally
  caused by Amphitryôn. (Apollod. ii. 4, 6. Pherekydês, Fragm. 27,
  Dind.)

  [204] Hesiod, Scut. Herc. 24. Theocrit. Idyll. xxiv. 4. Teleboas,
  the Eponym of these marauding people, was son of Poseidôn
  (Anaximander ap. Athenæ. xi. p. 498).

  [205] Apollod. ii. 4, 7. Compare the fable of Nisus at Megara,
  _infra_, chap. xii. p. 302.

  [206] Hesiod, Scut. Herc. 29. ὄφρα θεοῖσιν Ἀνδράσι τ᾽ ἀλφηστῇσιν
  ἀρῆς ἀλκτῆρα φυτεύσῃ.

  [207] Hesiod. Sc. H. 50-56.

When Alkmênê was on the point of being delivered at Thêbes, Zeus
publicly boasted among the assembled gods, at the instigation of the
mischief-making Atê, that there was on that day about to be born on
earth, from his breed, a son who should rule over all his neighbors.
Hêrê treated this as an empty boast, calling upon him to bind himself
by an irremissible oath that the prediction should be realized. Zeus
incautiously pledged his solemn word; upon which Hêrê darted swiftly
down from Olympus to the Achaic Argos, where the wife of Sthenelos
(son of Perseus, and therefore grandson of Zeus) was already seven
months gone with child. By the aid of the Eileithyiæ, the special
goddesses of parturition, she caused Eurystheus, the son of Sthenelos,
to be born before his time on that very day, while she retarded the
delivery of Alkmênê. Then returning to Olympus, she announced the
fact to Zeus: “The good man Eurystheus, son of the Perseid Sthenelos,
is this day born of thy loins: the sceptre of the Argeians worthily
belongs to him.” Zeus was thunderstruck at the consummation which he
had improvidently bound himself to accomplish. He seized Atê his evil
counsellor by the hair, and hurled her forever away from Olympus: but
he had no power to avert the ascendency of Eurystheus and the servitude
of Hêraklês. “Many a pang did he suffer, when he saw his favorite son
going through his degrading toil in the tasks imposed upon him by
Eurystheus.”[208]

  [208] Homer, Iliad, xix. 90-133; also viii. 361.—

    Τὴν αἰεὶ στενάχεσχ᾽, ὅθ᾽ ἑὸν φίλον υἱὸν ὁρῷτο
    Ἔργον ἀεικὲς ἔχοντα, ὑπ᾽ Εὐρυσθῆος ἀέθλων.

The legend, of unquestionable antiquity, here transcribed from the
Iliad, is one of the most pregnant and characteristic in the Grecian
mythology. It explains, according to the religious ideas familiar to
the old epic poets, both the distinguishing attributes and the endless
toil and endurances of Hêraklês,—the most renowned and most ubiquitous
of all the semi-divine personages worshipped by the Hellênes,—a being
of irresistible force, and especially beloved by Zeus, yet condemned
constantly to labor for others and to obey the commands of a worthless
and cowardly persecutor. His recompense is reserved to the close of
his career, when his afflicting trials are brought to a close: he is
then admitted to the godhead and receives in marriage Hêbê.[209] The
twelve labors, as they are called, too notorious to be here detailed,
form a very small fraction of the exploits of this mighty being, which
filled the Hêrakleian epics of the ancient poets. He is found not
only in most parts of Hellas, but throughout all the other regions
then known to the Greeks, from Gadês to the river Thermôdôn in the
Euxine and to Scythia, overcoming all difficulties and vanquishing
all opponents. Distinguished families are everywhere to be traced
who bear his patronymic, and glory in the belief that they are his
descendants. Among Achæans, Kadmeians, and Dôrians, Hêraklês is
venerated: the latter especially treat him as their principal hero,—the
Patron Hero-God of the race: the Hêrakleids form among all Dôrians a
privileged gens, in which at Sparta the special lineage of the two
kings was included.

His character lends itself to mythes countless in number as well as
disparate in their character. The irresistible force remains constant,
but it is sometimes applied with reckless violence against friends as
well as enemies, sometimes devoted to the relief of the oppressed. The
comic writers often brought him out as a coarse and stupid glutton,
while the Athênian philosopher Prodikos, without at all distorting
the type, extracted from it the simple, impressive, and imperishable
apologue still known as the Choice of Hercules.

After the death and apotheosis of Hêraklês, his son Hyllos and his
other children were expelled and persecuted by Eurystheus: the fear of
his vengeance deterred both the Trachinian king Kêyx and the Thêbans
from harboring them, and the Athênians alone were generous enough to
brave the risk of offering them shelter. Eurystheus invaded Attica,
but perished in the attempt by the hand of Hyllos, or by that of
Iolaos, the old companion and nephew of Hêraklês.[210] The chivalrous
courage which the Athênians had on this occasion displayed in behalf
of oppressed innocence, was a favorite theme for subsequent eulogy by
Attic poets and orators.

  [209] Hesiod, Theogon. 951, τελέσας στονόεντας ἀέθλους. Hom.
  Odyss. xi. 620; Hesiod, Eœæ, Fragm. 24, Düntzer, p. 36,
  πονηρότατον καὶ ἄριστον.

  [210] Apollod. ii. 8, 1; Hecatæ. ap. Longin. c. 27; Diodôr. iv.
  57.

All the sons of Eurystheus lost their lives in the battle along with
him, so that the Perseid family was now represented only by the
Hêrakleids, who collected an army and endeavored to recover the
possessions from which they had been expelled. The united forces of
Iônians, Achæans, and Arcadians, then inhabiting Peloponnêsus, met
the invaders at the isthmus, when Hyllos, the eldest of the sons of
Hêraklês, proposed that the contest should be determined by a single
combat between himself and any champion of the opposing army. It
was agreed, that if Hyllos were victorious, the Hêrakleids should
be restored to their possessions—if he were vanquished, that they
should forego all claim for the space of a hundred years, or fifty
years, or three generations,—for in the specification of the time,
accounts differ. Echemos, the hero of Tegea in Arcadia, accepted the
challenge, and Hyllos was slain in the encounter; in consequence of
which the Hêrakleids retired, and resided along with the Dôrians
under the protection of Ægimios, son of Dôrus.[211] As soon as the
stipulated period of truce had expired, they renewed their attempt upon
Peloponnêsus conjointly with the Dôrians, and with complete success:
the great Dôrian establishments of Argos, Sparta, and Messênia were
the result. The details of this victorious invasion will be hereafter
recounted.

  [211] Herodot. ix. 26; Diodôr. iv. 58.

Sikyôn, Phlios, Epidauros, and Trœzen[212] all boasted of respected
eponyms and a genealogy of dignified length, not exempt from the usual
discrepancies—but all just as much entitled to a place on the tablet
of history as the more renowned Æolids or Hêrakleids. I omit them here
because I wish to impress upon the reader’s mind the salient features
and character of the legendary world,—not to load his memory with a
full list of legendary names.

  [212] Pausan. ii. 5, 5; 12, 5; 26, 3. His statements indicate how
  much the predominance of a powerful neighbor like Argos tended to
  alter the genealogies of these inferior towns.




CHAPTER V.

DEUKALION, HELLEN, AND SONS OF HELLEN.


In the Hesiodic Theogony, as well as in the “Works and Days,” the
legend of Promêtheus and Epimêtheus presents an import religious,
ethical, and social, and in this sense it is carried forward by
Æschylus; but to neither of the characters is any genealogical function
assigned. The Hesiodic Catalogue of Women brought both of them into the
stream of Grecian legendary lineage, representing Deukaliôn as the son
of Promêtheus and Pandôra, and seemingly his wife Pyrrha as daughter of
Epimêtheus.[213]

  [213] Schol. ad Apollôn. Rhod. iii. 1085. Other accounts of the
  genealogy of Deukaliôn are given in the Schol. ad Homer. Odyss.
  x. 2, on the authority both of Hesiod and Akusilaus.

Deukaliôn is important in Grecian mythical narrative under two points
of view. First, he is the person specially saved at the time of the
general deluge: next, he is the father of Hellên, the great eponym of
the Hellenic race; at least this was the more current story, though
there were other statements which made Hellên the son of Zeus.

The name of Deukaliôn is originally connected with the Lokrian towns
of Kynos and Opus, and with the race of the Leleges, but he appears
finally as settled in Thessaly, and ruling in the portion of that
country called Phthiôtis.[214] According to what seems to have been the
old legendary account, it is the deluge which transferred him from the
one to the other; but according to another statement, framed in more
historicizing times, he conducted a body of Kurêtes and Leleges into
Thessaly, and expelled the prior Pelasgian occupants.[215]

  [214] Hesiodic Catalog. Fragm. xi.; Gaisf. lxx. Düntzer—

    Ἤτοι γὰρ Λοκρὸς Λελέγων ἡγήσατο λαῶν,
    Τούς ῥά ποτε Κρονίδης Ζεὺς ἄφθιτα μήδεα εἰδὼς,
    Λεκτοὺς ἐκ γαίης λάας πόρε Δευκαλίωνι.

  The reputed lineage of Deukaliôn continued in Phthia down to
  the time of Dikæarchus, if we may judge from the old Phthiot
  Pherekratês, whom he introduced in one of his dialogues as a
  disputant, and whom he expressly announced as a descendant of
  Deukaliôn (Cicero, Tuscul. Disp. i. 10).

  [215] The latter account is given by Dionys. Halic. i. 17; the
  former seems to have been given by Hellanikus, who affirmed that
  the ark after the deluge stopped upon Mount Othrys, and not
  upon Mount Parnassus (Schol. Pind. _ut. sup._) the former being
  suitable for a settlement in Thessaly.

  Pyrrha is the eponymous heroine of Pyrrhæa or Pyrrha, the ancient
  name of a portion of Thessaly (Rhianus, Fragm. 18, p. 71, ed.
  Düntzer).

  Hellanikus had written a work, now lost, entitled Δευκαλιώνεια:
  all the fragments of it which are cited have reference to places
  in Thessaly, Lokris and Phokis. See Preller, ad Hellanitum, p.
  12 (Dörpt. 1840). Probably Hellanikus is the main source of the
  important position occupied by Deukaliôn in Grecian legend.
  Thrasybulus and Akestodôrus represented Deukaliôn as having
  founded the oracle of Dôdôna, immediately after the deluge (Etm.
  Mag. v. Δωδωναῖος).

The enormous iniquity with which earth was contaminated—as Apollodôrus
says, by the then existing brazen race, or as others say, by the fifty
monstrous sons of Lykaôn—provoked Zeus to send a general deluge.[216]
An unremitting and terrible rain laid the whole of Greece under water,
except the highest mountain-tops, whereon a few stragglers found
refuge. Deukaliôn was saved in a chest or ark, which he had been
forewarned by his father Promêtheus to construct. After floating for
nine days on the water, he at length landed on the summit of Mount
Parnassus. Zeus having sent Hermês to him, promising to grant whatever
he asked, he prayed that men and companions might be sent to him in
his solitude: accordingly Zeus directed both him and Pyrrha to cast
stones over their heads: those cast by Pyrrha became women, those by
Deukaliôn men. And thus the “stony race of men” (if we may be allowed
to translate an etymology which the Greek language presents exactly,
and which has not been disdained by Hesiod, by Pindar, by Epicharmus,
and by Virgil) came to tenant the soil of Greece.[217] Deukaliôn on
landing from the ark sacrificed a grateful offering to Zeus Phyxios,
or the God of escape; he also erected altars in Thessaly to the twelve
great gods of Olympus.[218]

  [216] Apollodôrus connects this deluge with the wickedness of the
  brazen race in Hesiod, according to the practice general with
  the logographers of stringing together a sequence out of legends
  totally unconnected with each other (i. 7, 2).

  [217] Hesiod, Fragm. 135. ed. Markts. ap. Strabo. vii. p. 322,
  where the word λάας, proposed by Heyne as the reading of the
  unintelligible text, appears to me preferable to any of the
  other suggestions. Pindar, Olymp. ix. 47. Ἄτερ δ᾽ Εὐνᾶς ὁμόδαμον
  Κτησάσθαν λίθινον γόνον· Λαοὶ δ᾽ ὠνόμασθεν. Virgil, Georgic i.
  63. “Unde homines nati, durum genus.” Epicharmus ap. Schol.
  Pindar. Olymp. ix. 56. Hygin. f. 153. Philochorus retained the
  etymology, though he gave a totally different fable, nowise
  connected with Deukaliôn, to account for it; a curious proof how
  pleasing it was to the fancy of the Greek (see Schol. ad Pind. 1.
  c. 68).

  [218] Apollod. i. 7, 2. Hellanic. Fragm. 15. Didot. Hellanikus
  affirmed that the ark rested on Mount Othrys, not on Mount
  Parnassus (Fragm. 16. Didot). Servius (ad Virgil. Eclog. vi. 41)
  placed it on Mount Athôs—Hyginus (f. 153) on Mount Ætna.

The reality of this deluge was firmly believed throughout the
historical ages of Greece: the chronologers, reckoning up by
genealogies, assigned the exact date of it, and placed it at the
same time as the conflagration of the world by the rashness of
Phaëtôn, during the reign of Krotôpas king of Argus, the seventh from
Inachus.[219] The meteorological work of Aristotle admits and reasons
upon this deluge as an unquestionable fact, though he alters the
locality by placing it west of Mount Pindus, near Dôdôna and the river
Achelôus.[220] He at the same time treats it as a physical phenomenon,
the result of periodical cycles in the atmosphere, thus departing from
the religious character of the old legend, which described it as a
judgment inflicted by Zeus upon a wicked race. Statements founded upon
this event were in circulation throughout Greece even to a very late
date. The Megarians affirmed that Megaros, their hero, son of Zeus by
a local nymph, had found safety from the waters on the lofty summit
of their mountain Geraneia, which had not been completely submerged.
And in the magnificent temple of the Olympian Zeus at Athens, a cavity
in the earth was shown, through which it was affirmed that the waters
of the deluge had retired. Even in the time of Pausanias, the priests
poured into this cavity holy offerings of meal and honey.[221] In this,
as in other parts of Greece, the idea of the Deukalionian deluge was
blended with the religious impressions of the people and commemorated
by their sacred ceremonies.

  [219] Tatian adv. Græc. c. 60, adopted both by Clemens and
  Eusebius. The Parian marble placed this deluge in the reign of
  Kranaos at Athens, 752 years before the first recorded Olympiad,
  and 1528 years before the Christian æra; Apollodôrus also places
  it in the reign of Kranaos, and in that of Nyctimus in Arcadia
  (iii. 8, 2; 14, 5).

  The deluge and the _ekpyrosis_ or conflagration are connected
  together also in Servius ad Virgil. Bucol. vi. 41: he refines
  both of them into a “mutationem temporum.”

  [220] Aristot. Meteorol. i. 14. Justin rationalizes the fable
  by telling us that Deukaliôn was king of Thessaly, who provided
  shelter and protection to the fugitives from the deluge (ii. 6,
  11).

  [221] Pausan. i. 18, 7; 40, 1. According to the Parian marble (s.
  5), Deukaliôn had come to Athens after the deluge, and had there
  himself founded the temple of the Olympian Zeus. The etymology
  and allegorization of the names of Deukaliôn and Pyrrha, given by
  Völcker in his ingenious Mythologie des Iapetischen Geschlechts
  (Giessen, 1824), p. 343, appears to me not at all convincing.

The offspring of Deukaliôn and Pyrrha were two sons, Hellên and
Amphiktyôn, and a daughter, Prôtogeneia, whose son by Zeus was
Aëthlius: it was however maintained by many, that Hellên was the
son of Zeus and not of Deukaliôn. Hellên had by a nymph three sons,
Dôrus, Xuthus, and Æolus. He gave to those who had been before called
Greeks,[222] the name of Hellênes, and partitioned his territory
among his three children. Æolus reigned in Thessaly; Xuthus received
Peloponnêsus, and had by Creüsa as his sons, Achæus and Iôn; while
Dôrus occupied the country lying opposite to the Peloponnêsus, on
the northern side of the Corinthian Gulf. These three gave to the
inhabitants of their respective countries the names of Æolians, Achæans
and Iônians, and Dôrians.[223]

  [222] Such is the statement of Apollodôrus (i. 7, 3); but I
  cannot bring myself to believe that the name (Γραϊκοὶ) Greeks is
  at all old in the legend, or that the passage of Hesiod, in which
  Græcus and Latinus purport to be mentioned, is genuine.

  See Hesiod, Theogon. 1013, and Catalog. Fragm. xxix. ed.
  Göttling, with the note of Göttling; also Wachsmuth, Hellen.
  Alterth. i. 1. p. 311, and Bernhardy, Griech. Literat. vol. i. p.
  167.

  [223] Apollod. i. 7, 4.

Such is the genealogy as we find it in Apollodôrus. In so far as
the names and filiation are concerned, many points in it are given
differently, or implicitly contradicted, by Euripidês and other
writers. Though as literal and personal history it deserves no notice,
its import is both intelligible and comprehensive. It expounds and
symbolizes the first fraternal aggregation of Hellênic men, together
with their territorial distribution and the institutions which they
collectively venerated.

There were two great holding-points in common for every section of
Greeks. One was the Amphiktyonic assembly, which met half-yearly,
alternately at Delphi and at Thermopylæ; originally and chiefly for
common religious purposes, but indirectly and occasionally embracing
political and social objects along with them. The other was, the public
festivals or games, of which the Olympic came first in importance;
next, the Pythian, Nemean and Isthmian,—institutions which combined
religious solemnities with recreative effusion and hearty sympathies,
in a manner so imposing and so unparalleled. Amphiktyôn represents
the first of these institutions, and Aëthlius the second. As the
Amphiktyonic assembly was always especially connected with Thermopylæ
and Thessaly, Amphiktyôn is made the son of the Thessalian Deukaliôn;
but as the Olympic festival was nowise locally connected with
Deukaliôn, Aëthlius is represented as having Zeus for his father, and
as touching Deukaliôn only through the maternal line. It will be seen
presently, that the only matter predicted respecting Aëthlius is, that
he settled in the territory of Elis, and begat Endymiôn: this brings
him into local contact with the Olympic games, and his function is then
ended.

Having thus got Hellas as an aggregate with its main cementing forces,
we march on to its subdivision into parts, through Æolus, Dôrus and
Xuthus, the three sons of Hellen;[224] a distribution which is far
from being exhaustive: nevertheless, the genealogists whom Apollodôrus
follows recognize no more than three sons.

  [224] How literally and implicitly even the ablest Greeks
  believed in eponymous persons, such as Hellên and Iôn, as the
  real progenitors of the races called after him, may be seen by
  this, that Aristotle gives this common descent as the definition
  of γένος (Metaphysic. iv. p. 118, Brandis):—

  Γένος λέγεται, τὸ μὲν ... τὸ δὲ, ἀφ᾽ οὗ ἂν ὦσι πρώτου κινήσαντος
  εἰς τὸ εἶναι. Οὕτω γὰρ λέγονται οἱ μὲν, Ἕλληνες τὸ γένος, οἱ δὲ,
  Ἴωνες· τῷ, οἱ μὲν ἀπὸ Ἕλληνος, οἱ δὲ ἀπὸ Ἴωνος, εἶναι πρώτου
  γεννήσαντος.

The genealogy is essentially post-Homeric; for Homer knows Hellas and
the Hellênes only in connection with a portion of Achaia Phthiôtis.
But as it is recognized in the Hesiodic Catalogue[225]—composed
probably within the first century after the commencement of recorded
Olympiads, or before 676 B. C.—the peculiarities of it, dating from
so early a period, deserve much attention. We may remark, first,
that it seems to exhibit to us Dôrus and Æolus as the only pure and
genuine offspring of Hellên. For their brother Xuthus is not enrolled
as an eponymus; he neither founds nor names any people; it is only
his sons Achæus and Iôn, after his blood has been mingled with that
of the Erechtheid Kreüsa, who become eponyms and founders, each of
his own separate people. Next, as to the territorial distribution,
Xuthus receives Peloponnêsus from his father, and unites himself with
Attica (which the author of this genealogy seems to have conceived as
originally unconnected with Hellên) by his marriage with the daughter
of the indigenous hero, Erechtheus. The issue of this marriage, Achæus
and Iôn, present to us the population of Peloponnêsus and Attica
conjointly as related among themselves by the tie of brotherhood, but
as one degree more distant both from Dôrians and Æolians. Æolus reigns
over the regions about Thessaly, and called the people in those parts
Æolians; while Dôrus occupies “the country over against Peloponnêsus on
the opposite side of the Corinthian Gulf,” and calls the inhabitants
after himself, Dôrians.[226] It is at once evident that this
designation is in no way applicable to the confined district between
Parnassus and Œta, which alone is known by the name of Dôris, and its
inhabitants by that of Dôrians, in the historical ages. In the view of
the author of this genealogy, the Dôrians are the original occupants of
the large range of territory north of the Corinthian Gulf, comprising
Ætôlia, Phôkis, and the territory of the Ozolian Lokrians. And this
farther harmonizes with the other legend noticed by Apollodôrus,
when he states that Ætolus, son of Endymiôn, having been forced to
expatriate from Peloponnêsus, crossed into the Kurêtid territory,[227]
and was there hospitably received by Dôrus, Laodokus and Polypœtês,
sons of Apollo and Phthia. He slew his hosts, acquired the territory,
and gave to it the name of Ætôlia: his son Pleurôn married Xanthippê,
daughter of Dôrus; while his other son, Kalydôn, marries Æolia,
daughter of Amythaôn. Here again we have the name of Dôrus, or the
Dôrians, connected with the tract subsequently termed Ætôlia. That
Dôrus should in one place be called the son of Apollo and Phthia,
and in another place the son of Hellên by a nymph, will surprise no
one accustomed to the fluctuating personal nomenclature of these old
legends: moreover the name of Phthia is easy to reconcile with that of
Hellên, as both are identified with the same portion of Thessaly, even
from the days of the Iliad.

  [225] Hesiod, Fragm. 8. p. 278, ed. Marktsch.—

    Ἕλληνος δ᾽ ἐγένοντο θεμιστόπολοι βασιλῆες
    Δῶρός τε, Ξοῦθός τε, καὶ Αἴολος ἱππιοχάρμης.
    Αἰολίδαι δ᾽ ἐγένοντο θεμιστόπολοι βασιλῆες
    Κρηθεὺς ἠδ᾽ Ἀθάμας καὶ Σίσυφος αἰολομήτης
    Σαλμωνεύς τ᾽ ἄδικος καὶ ὑπέρθυμος Περιήρης.

  [226] Apollod. i. 7, 3. Ἕλληνος δὲ καὶ Νύμφης Ὀρσήϊδος (?),
  Δῶρος, Ξοῦθος, Αἴολος. Αὐτὸς μὲν οὖν ἀφ᾽ αὑτοῦ τοὺς καλουμένους
  Γραϊκοὺς προσηγόρευσεν Ἕλληνας, τοῖς δὲ παισὶν ἐμέρισε τὴν χώραν.
  Καὶ Ξοῦθος μὲν λαβὼν τὴν Πελοπόννησον, ἐκ Κρεούσης τῆς Ἐρεχθέως
  Ἀχαιὸν ἐγέννησε καὶ Ἴωνα, ἀφ᾽ ὧν Ἀχαιοὶ καὶ Ἴωνες καλοῦνται.
  Δῶρος δὲ, ~τὴν πέραν χώραν Πελοποννήσου λαβὼν, τοὺς κατοίκους
  ἀφ᾽ ἑαυτοῦ Δωριεῖς ἐκάλεσεν~. Αἴολος δὲ, βασιλεύων τῶν περὶ τὴν
  Θετταλίαν τόπων, τοὺς ἐνοικοῦντας Αἰολεῖς προσηγόρευσε.

  Strabo (viii. p. 383) and Conôn (Narr. 27), who evidently copy
  from the same source, represent Dôrus as going to settle in the
  territory property known as Dôris.

  [227] Apollod. i. 7, 6. Αἰτωλὸς ... φυγὼν εἰς τὴν Κουρήτιδα
  χώραν, κτείνας τοὺς ὑποδεξαμένους Φθίας καὶ Ἀπόλλωνος υἱοὺς,
  Δῶρον καὶ Λαόδοκον καὶ Πολυποίτην, ἀφ᾽ ἑαυτοῦ τὴν χώραν Αἰτωλίαν
  ἐκάλεσε. Again, i. 8, 1. Πλευρὼν (son of Ætôlus) γήμας Ξανθίππην
  τὴν Δώρου, παῖδα ἐγέννησεν Ἀγήνορα.

This story, that the Dôrians were at one time the occupants, or
the chief occupants, of the range of territory between the river
Achelôus and the northern shore of the Corinthian Gulf, is at least
more suitable to the facts attested by historical evidence than the
legends given in Herodotus, who represents the Dôrians as originally
in the Phthiôtid; then as passing under Dôrus, the son of Hellên,
into the Histiæôtid, under the mountains of Ossa and Olympus; next,
as driven by the Kadmeians into the regions of Pindus; from thence
passing into the Dryopid territory, on Mount Œta; lastly, from thence
into Peloponnêsus.[228] The received story was, that the great
Dôrian establishments in Peloponnêsus were formed by invasion from
the north, and that the invaders crossed the gulf from Naupaktus,—a
statement which, however disputable with respect to Argos, seems highly
probable in regard both to Sparta and Messênia. That the name of
Dôrians comprehended far more than the inhabitants of the insignificant
tetrapolis of Dôris Proper, must be assumed, if we believe that they
conquered Sparta and Messênia: both the magnitude of the conquest
itself, and the passage of a large portion of them from Naupaktus,
harmonize with the legend as given by Apollodôrus, in which the Dôrians
are represented as the principal inhabitants of the northern shore of
the gulf. The statements which we find in Herodotus, respecting the
early migrations of the Dôrians, have been considered as possessing
greater historical value than those of the fabulist Apollodôrus. But
both are equally matter of legend, while the brief indications of the
latter seem to be most in harmony with the facts which we afterwards
find attested by history.

  [228] Herod. i. 56.

It has already been mentioned that the genealogy which makes Æolus,
Xuthus and Dôrus sons of Hellên, is as old as the Hesiodic Catalogue;
probably also that which makes Hellên son of Deukaliôn. Aëthlius also
is an Hesiodic personage: whether Amphiktyôn be so or not, we have no
proof.[229] They could not have been introduced into the legendary
genealogy until after the Olympic games and the Amphiktyonic council
had acquired an established ascendancy and universal reverence
throughout Greece.

  [229] Schol. Apollôn. Rhod. iv. 57. Τὸν δὲ Ἐνδυμίωνα Ἠσίοδος μὲν
  Ἀεθλίου τοῦ Διὸς καὶ Καλύκης παῖδα λέγει.... Καὶ Πείσανδρος δὲ
  τὰ αὐτά φησι, καὶ Ἀκουσίλαος, καὶ Φερεκύδης, καὶ Νίκανδρος ἐν
  δευτέρῳ Αἰτωλικῶν, καὶ Θεόπομπος ἐν Ἐποποιΐαις.

  Respecting the parentage of Hellên, the references to Hesiod are
  very confused. Compare Schol. Homer. Odyss. x. 2, and Schol.
  Apollôn. Rhod. iii. 1086. See also Hellanic. Frag. 10. Didot.

  Apollodôrus, and Pherekydês before him (Frag. 51. Didot),
  called Protôgeneia daughter of Deukaliôn; Pindar (Olymp. ix.
  64) designated her as daughter of Opus. One of the stratagems
  mentioned by the Scholiast to get rid of this genealogical
  discrepancy was, the supposition that Deukaliôn had two names
  (διώνυμος); that he was also named Opus. (Schol. Pind. Olymp. ix.
  85).

  That the Deukalidæ or posterity of Deukaliôn reigned in Thessaly,
  was mentioned both by Hesiod and Hekatæus, ap. Schol. Apollôn.
  Rhod. iv. 265.

Respecting Dôrus the son of Hellên, we find neither legends nor
legendary genealogy; respecting Xuthus, very little beyond the tale
of Kreüsa and Iôn, which has its place more naturally among the
Attic fables. Achæus however, who is here represented as the son of
Xuthus, appears in other stories with very different parentage and
accompaniments. According to the statement which we find in Dionysius
of Halicarnassus, Achæus, Phthius and Pelasgus are sons of Poseidôn and
Larissa. They migrate from Peloponnêsus into Thessaly, and distribute
the Thessalian territory between them, giving their names to its
principal divisions: their descendants in the sixth generation were
driven out of that country by the invasion of Deukaliôn at the head
of the Kurêtes and the Leleges.[230] This was the story of those who
wanted to provide an eponymus for the Achæans in the southern districts
of Thessaly: Pausanias accomplishes the same object by different
means, representing Achæus, the son of Xuthus as having gone back
to Thessaly and occupied the portion of it to which his father was
entitled. Then, by way of explaining how it was that there were Achæans
at Sparta and at Argos, he tells us that Archander and Architelês, the
sons of Archæus, came back from Thessaly to Peloponnêsus, and married
two daughters of Danaus: they acquired great influence at Argos and
Sparta, and gave to the people the name of Achæans after their father
Achæus.[231]

  [230] Dionys. H. A. R. i. 17.

  [231] Pausan. vii. 1, 1-3. Herodotus also mentions (ii. 97)
  Archander, son of Phthius and grandson of Achæus, who married the
  daughter of Danaus. Larcher (Essai sur la Chronologie d’Hérodote,
  ch. x. p. 321) tells us that this cannot be the Danaus who came
  from Egypt, the father of the fifty daughters, who must have
  lived two centuries earlier, as may be proved by chronological
  arguments: this must be another Danaus, according to him.

  Strabo seems to give a different story respecting the Achæans in
  Peloponnêsus: he says that they were the original population of
  the peninsula, that they came in from Phthia with Pelops, and
  inhabited Laconia, which was from them called Argos Achaicum,
  and that on the conquest of the Dôrians, they moved into Achaia
  properly so called, expelling the Iônians therefrom (Strabo, viii
  p. 365). This narrative is, I presume, borrowed from Ephorus.

Euripidês also deviates very materially from the Hesiodic genealogy
in respect to these eponymous persons. In the drama called Iôn, he
describes Iôn as son of Kreüsa by Apollo, but adopted by Xuthus:
according to him, the real sons of Xuthus and Kreüsa are Dôrus and
Achæus,[232]—eponyms of the Dôrians and Achæans in the interior of
Peloponnêsus. And it is a still more capital point of difference, that
he omits Hellên altogether—making Xuthus an Achæan by race, the son of
Æolus, who is the son of Zeus.[233] This is the more remarkable, as
in the fragments of two other dramas of Euripidês, the Melanippê and
the Æolus, we find Hellên mentioned both as father of Æolus and son of
Zeus.[234] To the general public even of the most instructed city of
Greece, fluctuations and discrepancies in these mythical genealogies
seem to have been neither surprising nor offensive.

  [232] Eurip. Ion, 1590.

  [233] Eurip. Ion, 64.

  [234] See the Fragments of these two plays in Matthiae’s edition;
  compare Welcker, Griechisch. Tragöd. v. ii. p. 842. If we may
  judge from the Fragments of the Latin Melanippê of Ennius (see
  Fragm. 2, ed. Bothe), Hellên was introduced as one of the
  characters of the piece.




CHAPTER VI.

THE ÆOLIDS, OR SONS AND DAUGHTERS OF ÆOLUS.


If two of the sons of Hellên, Dôrus and Xuthus, present to us families
comparatively unnoticed in mythical narrative, the third son, Æolus,
richly makes up for the deficiency. From him we pass to his seven sons
and five daughters, amidst a great abundance of heroic and poetical
incident.

In dealing however with these extensive mythical families, it is
necessary to observe, that the legendary world of Greece, in the manner
in which it is presented to us, appears invested with a degree of
symmetry and coherence which did not originally belong to it. For the
old ballads and stories which were sung or recounted at the multiplied
festivals of Greece, each on its own special theme, have been lost: the
religious narratives, which the Exegêtês of every temple had present
to his memory, explanatory of the peculiar religious ceremonies and
local customs in his own town or Dême, have passed away: all these
primitive elements, originally distinct and unconnected, are removed
out of our sight, and we possess only an aggregate result, formed by
many confluent streams of fable, and connected together by the agency
of subsequent poets and logographers. Even the earliest agents in this
work of connecting and systematizing—the Hesiodic poets—have been
hardly at all preserved. Our information respecting Grecian mythology
is derived chiefly from the prose logographers who followed them, and
in whose works, since a continuous narrative was above all things
essential to them, the fabulous personages are woven into still more
comprehensive pedigrees, and the original isolation of the legends
still better disguised. Hekatæus, Pherekydês, Hellanikus, and Akusilaus
lived at a time when the idea of Hellas as one great whole, composed
of fraternal sections, was deeply rooted in the mind of every Greek;
and when the fancy of one or a few great families, branching out widely
from one common stem, was more popular and acceptable than that of a
distinct indigenous origin in each of the separate districts. These
logographers, indeed, have themselves been lost; but Apollodôrus and
the various scholiasts, our great immediate sources of information
respecting Grecian mythology, chiefly borrowed from them: so that the
legendary world of Greece is in fact known to us through them, combined
with the dramatic and Alexandrine poets, their Latin imitators, and the
still later class of scholiasts—except indeed such occasional glimpses
as we obtain from the Iliad and the Odyssey, and the remaining Hesiodic
fragments, which exhibit but too frequently a hopeless diversity when
confronted with the narratives of the logographers.

Though Æolus (as has been already stated) is himself called the son
of Hellên along with Dôrus and Xuthus, yet the legends concerning the
Æolids, far from being dependent upon this genealogy, are not all even
coherent with it: moreover the name of Æolus in the legend is older
than that of Hellên, inasmuch as it occurs both in the Iliad and
Odyssey.[235] Odysseus sees in the under-world the beautiful Tyrô,
daughter of Salmôneus, and wife of Krêtheus, son of Æolus.

  [235] Iliad, vi. 154. Σίσυφος Αἰολίδης, etc. Again Odyss. xi.
  234.—

    Ἔνθ᾽ ἤτοι πρώτην Τυρὼ ἴδον ~εὐπατέρειαν~,
    Ἣ φάτο Σαλμωνῆος ἀμύμονος ἔκγονος εἶναι,
    Φῆ δὲ Κρηθῆος γυνὴ ἔμμεναι ~Αἰολίδαο~.

Æolus is represented as having reigned in Thessaly: his seven sons were
Krêtheus, Sisyphus, Athamas, Salmôneus, Deiôn, Magnês and Periêrês: his
five daughters, Canacê, Alcyonê, Peisidikê, Calycê and Perimêdê. The
fables of this race seem to be distinguished by a constant introduction
of the god Poseidôn, as well as by an unusual prevalence of haughty and
presumptuous attributes among the Æolid heroes, leading them to affront
the gods by pretences of equality, and sometimes even by defiance. The
worship of Poseidôn must probably have been diffused and preëminent
among a people with whom these legends originated.


SECTION I.—SONS OF ÆOLUS.

Salmôneus is not described in the Odyssey as son of Æolus, but he is
so denominated both in the Hesiodic Catalogue, and by the subsequent
logographers. His daughter Tyrô became enamoured of the river
Enipeus, the most beautiful of all streams that traverse the earth:
she frequented the banks assiduously, and there the god Poseidôn
found means to indulge his passion for her, assuming the character
of the river god himself. The fruit of this alliance were the twin
brothers, Pelias and Nêleus: Tyrô afterwards was given in marriage
to her uncle Krêtheus, another son of Æolus, by whom she had Æsôn,
Pherês and Amythaôn—all names of celebrity in the heroic legends.[236]
The adventures of Tyrô formed the subject of an affecting drama of
Sophoklês, now lost. Her father had married a second wife, named
Sidêrô, whose cruel counsels induced him to punish and torture his
daughter on account of her intercourse with Poseidôn. She was shorn
of her magnificent hair, beaten and ill-used in various ways, and
confined in a loathsome dungeon. Unable to take care of her two
children, she had been compelled to expose them immediately on their
birth in a little boat on the river Enipeus; they were preserved by the
kindness of a herdsman, and when grown up to manhood, rescued their
mother, and revenged her wrongs by putting to death the iron-hearted
Sidêrô.[237] This pathetic tale respecting the long imprisonment of
Tyrô is substituted by Sophoklês in place of the Homeric legend, which
represented her to have become the wife of Krêtheus and mother of a
numerous offspring.[238]

  [236] Homer, Odyss. xi. 234-257; xv. 226.

  [237] Diodôrus, iv. 68. Sophoklês, Fragm. 1. Τυρώ. Σαφῶς Σιδηρὼ
  καὶ φέρουσα τοὔνομα. The genius of Sophoklês is occasionally
  seduced by this play upon the etymology of a name, even in the
  most impressive scenes of his tragedies. See Ajax, 425. Compare
  Hellanik. Fragm. p. 9, ed. Preller. There was a first and second
  edition of the Tyrô—τῆς δευτέρας Τυροῦς. Schol. ad Aristoph.
  Av. 276. See the few fragments of the lost drama in Dindorf’s
  Collection, p. 53. The plot was in many respects analogous to the
  Antiopê of Euripidês.

  [238] A third story, different both from Homer and from
  Sophoklês, respecting Tyrô, is found in Hyginus (Fab. lx.): it
  is of a tragical cast, and borrowed, like so many other tales in
  that collection, from one of the lost Greek dramas.

Her father, the unjust Salmôneus, exhibited in his conduct the most
insolent impiety towards the gods. He assumed the name and title even
of Zeus, and caused to be offered to himself the sacrifices destined
for that god: he also imitated the thunder and lightning, by driving
about with brazen caldrons attached to his chariot and casting lighted
torches towards heaven. Such wickedness finally drew upon him the wrath
of Zeus, who smote him with a thunderbolt, and effaced from the earth
the city which he had founded, with all its inhabitants.[239]

  [239] Apollod. i. 9, 7. Σαλμωνεύς τ᾽ ἄδικος καὶ ὑπέρθυμος
  Περιήρης. Hesiod, Fragm. Catal. 8. Marktscheffel.

  Where the city of Salmôneus was situated, the ancient
  investigators were not agreed; whether in the Pisatid, or in
  Elis, or in Thessaly (see Strabo, viii. p. 356). Euripidês in
  his Æolus placed him on the banks of the Alpheius (Eurip. Fragm.
  Æol. 1). A village and fountain in the Pisatid bore the name
  of Salmônê; but the mention of the river Enipeus seems to mark
  Thessaly as the original seat of the legend. But the _naïveté_ of
  the tale preserved by Apollodôrus (Virgil in the Æneid, vi. 586,
  has retouched it) marks its ancient date: the final circumstance
  of that tale was, that the city and its inhabitants were
  annihilated.

  Ephorus makes Salmôneus king of the Epeians and of the Pisatæ
  (Fragm. 15, ed. Didot).

  The lost drama of Sophoklês, called Σαλμωνεὺς, was a δρᾶμα
  σατυρικόν See Dindorf’s Fragm. 483.

Pelias and Nêleus, “both stout vassals of the great Zeus,” became
engaged in dissension respecting the kingdom of Iôlkos in Thessaly.
Pelias got possession of it, and dwelt there in plenty and prosperity;
but he had offended the goddess Hêrê by killing Sidêrô upon her altar,
and the effects of her wrath were manifested in his relations with his
nephew Jasôn.[240]

  [240] Hom. Od. xi. 280. Apollod. i. 9, 9. κρατέρω θεραπόντε Διὸς,
  etc.

Nêleus quitted Thessaly, went into Peloponnêsus, and there founded
the kingdom of Pylos. He purchased by immense marriage presents, the
privilege of wedding the beautiful Chlôris, daughter of Amphiôn, king
of Orchomenos, by whom he had twelve sons and but one daughter[241]—the
fair and captivating Pêrô, whom suitors from all the neighborhood
courted in marriage. But Nêleus, “the haughtiest of living men,”[242]
refused to entertain the pretensions of any of them: he would grant his
daughter only to that man who should bring to him the oxen of Iphiklos,
from Phylakê in Thessaly. These precious animals were carefully
guarded, as well by herdsmen as by a dog whom neither man nor animal
could approach. Nevertheless, Bias, the son of Amythaôn, nephew of
Nêleus, being desperately enamored of Pêrô, prevailed upon his brother
Melampus to undertake for his sake the perilous adventure, in spite of
the prophetic knowledge of the latter, which forewarned him that though
he would ultimately succeed, the prize must be purchased by severe
captivity and suffering. Melampus, in attempting to steal the oxen,
was seized and put in prison; from whence nothing but his prophetic
powers rescued him. Being acquainted with the language of worms, he
heard these animals communicating to each other, in the roof over his
head, that the beams were nearly eaten through and about to fall in.
He communicated this intelligence to his guards, and demanded to be
conveyed to another place of confinement, announcing that the roof
would presently fall in and bury them. The prediction was fulfilled,
and Phylakos, father of Iphiklos, full of wonder at this specimen of
prophetic power, immediately caused him to be released. He further
consulted him respecting the condition of his son Iphiklos, who was
childless; and promised him the possession of the oxen on condition of
his suggesting the means whereby offspring might be ensured. A vulture
having communicated to Melampus the requisite information, Podarkês,
the son of Iphiklos, was born shortly afterwards. In this manner
Melampus obtained possession of the oxen, and conveyed them to Pylos,
obtaining for his brother Bias the hand of Pêrô.[243] How this great
legendary character, by miraculously healing the deranged daughters of
Prœtos, procured both for himself and for Bias dominion in Argos, has
been recounted in a preceding chapter.

  [241] Diodôr. iv. 68.

  [242] Νηλέα τε μεγάθυμον, ἀγαυότατον ζωόντων (Hom. Odyss. xv.
  228).

  [243] Hom. Od. xi. 278; xv. 234. Apollod. i. 9, 12. The basis of
  this curious romance is in the Odyssey, amplified by subsequent
  poets. There are points however in the old Homeric legend, as
  it is briefly sketched in the fifteenth book of the Odyssey,
  which seem to have been subsequently left out or varied. Nêleus
  seizes the property of Melampus during his absence; the latter,
  returning with the oxen from Phylakê, revenges himself upon
  Nêleus for the injury. Odyss. xv. 233.

Of the twelve sons of Nêleus, one at least, Periklymenos,—besides the
ever-memorable Nestôr,—was distinguished for his exploits as well as
for his miraculous gifts. Poseidôn, the divine father of the race, had
bestowed upon him the privilege of changing his form at pleasure into
that of any bird, beast, reptile, or insect.[244] He had occasion for
all these resources, and he employed them for a time with success in
defending his family against the terrible indignation of Hêraklês,
who, provoked by the refusal of Nêleus to perform for him the ceremony
of purification after his murder of Iphitus, attacked the Nêleids
at Pylos. Periklymenos by his extraordinary powers prolonged the
resistance, but the hour of his fate was at length brought upon him
by the intervention of Athênê, who pointed him out to Hêraklês while
he was perched as a bee upon the hero’s chariot. He was killed, and
Hêraklês became completely victorious, overpowering Poseidôn, Hêrê,
Arês, and Hadês, and even wounding the three latter, who assisted in
the defence. Eleven of the sons of Nêleus perished by his hand, while
Nestôr, then a youth, was preserved only by his accidental absence at
Gerêna, away from his father’s residence.[245]

  [244] Hesiod, Catalog. ap. Schol. Apollôn. Rhod. i. 156; Ovid,
  Metam. xii. p. 556; Eustath. ad Odyss. xi. p. 284. Poseidôn
  carefully protects Antilochus son of Nestôr, in the Iliad, xiii.
  554-563.

  [245] Hesiod, Catalog. ap. Schol. Ven. ad Iliad. ii. 336; and
  Steph. Byz. v. Γερηνία; Homer, Il. v. 392; xi. 693; Apollodôr.
  ii. 7, 3; Hesiod, Scut. Herc. 360; Pindar, Ol. ix. 32.

  According to the Homeric legend, Nêleus himself was not killed
  by Hêraklês: subsequent poets or logographers, whom Apollodôrus
  follows, seem to have thought it an injustice, that the offence
  given by Nêleus himself should have been avenged upon his sons
  and not upon himself; they therefore altered the legend upon this
  point, and rejected the passage in the Iliad as spurious (see
  Schol. Ven. ad Iliad. xi. 682).

  The refusal of purification by Nêleus to Hêraklês is a genuine
  legendary cause: the commentators, who were disposed to spread a
  coating of history over these transactions, introduced another
  cause,—Nêleus, as king of Pylos, had aided the Orchomenians in
  their war against Hêraklês and the Thêbans (see Sch. Ven. ad
  Iliad. xi. 689).

  The neighborhood of Pylos was distinguished for its ancient
  worship both of Poseidôn and of Hadês: there were abundant local
  legends respecting them (see Strabo, viii. pp. 344, 345).

The proud house of the Nêleids was now reduced to Nestôr; but Nestôr
singly sufficed to sustain its eminence. He appears not only as the
defender and avenger of Pylos against the insolence and rapacity
of his Epeian neighbors in Elis, but also as aiding the Lapithæ in
their terrible combat against the Centaurs, and as companion of
Thêseus, Peirithöus, and the other great legendary heroes who preceded
the Trojan war. In extreme old age his once marvellous power of
handling his weapons has indeed passed away, but his activity remains
unimpaired, and his sagacity as well as his influence in counsel is
greater than ever. He not only assembles the various Grecian chiefs
for the armament against Troy, perambulating the districts of Hellas
along with Odysseus, but takes a vigorous part in the siege itself,
and is of preëminent service to Agamemnôn. And after the conclusion
of the siege, he is one of the few Grecian princes who returns to his
original dominions, and is found, in a strenuous and honored old age,
in the midst of his children and subjects,—sitting with the sceptre
of authority on the stone bench before his house at Pylos,—offering
sacrifice to Poseidôn, as his father Nêleus had done before him,—and
mourning only over the death of his favorite son Antilochus, who had
fallen, along with so many brave companions in arms, in the Trojan
war.[246]

  [246] About Nestôr, Iliad, i. 260-275; ii. 370; xi. 670-770;
  Odyss. iii. 5, 110, 409.

After Nestôr the line of the Nêleids numbers undistinguished
names,—Bôrus, Penthilus, and Andropompus,—three successive generations
down to Melanthus, who on the invasion of Peloponnêsus by the
Herakleids, quitted Pylos and retired to Athens, where he became king,
in a manner which I shall hereafter recount. His son Kodrus was the
last Athênian king; and Nêleus, one of the sons of Kodrus, is mentioned
as the principal conductor of what is called the Ionic emigration from
Athens to Asia Minor.[247] It is certain that during the historical
age, not merely the princely family of the Kodrids in Milêtus, Ephesus,
and other Ionic cities, but some of the greatest families even in
Athens itself, traced their heroic lineage through the Nêleids up to
Poseidôn: and the legends respecting Nestôr and Periklymenos would
find especial favor amidst Greeks with such feelings and belief. The
Kodrids at Ephesus, and probably some other Ionic towns, long retained
the title and honorary precedence of kings, even after they had lost
the substantial power belonging to the office. They stood in the same
relation, embodying both religious worship and supposed ancestry,
to the Nêleids and Poseidôn, as the chiefs of the Æolic colonies to
Agamemnôn and Orestês. The Athenian despot Peisistratus was named after
the son of Nestôr in the Odyssey; and we may safely presume that the
heroic worship of the Nêleids was as carefully cherished at the Ionic
Milêtus as at the Italian Metapontum.[248]

  [247] Hellanik. Fragm. 10, ed. Didot; Pausan. vii. 2, 3; Herodot.
  v. 65; Strabo, xiv. p. 633. Hellanikus, in giving the genealogy
  from Nêleus to Melanthus, traces it through Periklymenos and not
  through Nestôr: the words of Herodotus imply that _he_ must have
  included Nestôr.

  [248] Herodot. v. 67; Strabo, vi. p. 264; Mimnermus, Fragm. 9,
  Schneidewin.

Having pursued the line of Salmôneus and Nêleus to the end of its
lengendary career, we may now turn back to that of another son of
Æolus, Krêtheus,—a line hardly less celebrated in respect of the heroic
names which it presents. Alkêstis, the most beautiful of the daughters
of Pelias,[249] was promised by her father in marriage to the man
that could bring him a lion and a boar tamed to the yoke and drawing
together. Admêtus, son of Pherês, the eponymus of Pheræ in Thessaly,
and thus grandson of Krêtheus, was enabled by the aid of Apollo to
fulfil this condition, and to win her;[250] for Apollo happened at that
time to be in his service as a slave (condemned to this penalty by Zeus
for having put to death the Cyclôpes), in which capacity he tended the
herds and horses with such success, as to equip Eumêlus (the son of
Admêtus) to the Trojan war with the finest horses in the Grecian army.
Though menial duties were imposed upon him, even to the drudgery of
grinding in the mill,[251] he yet carried away with him a grateful and
friendly sentiment towards his mortal master, whom he interfered to
rescue from the wrath of the goddess Artemis, when she was indignant at
the omission of her name in his wedding sacrifices. Admêtus was about
to perish by a premature death, when Apollo, by earnest solicitation
to the Fates, obtained for him the privilege that his life should be
prolonged, if he could find any person to die a voluntary death in his
place. His father and his mother both refused to make this sacrifice
for him, but the devoted attachment of his wife Alkêstis disposed her
to embrace with cheerfulness the condition of dying to preserve her
husband. She had already perished, when Hêraklês, the ancient guest and
friend of Admêtus, arrived during the first hour of lamentation; his
strength and daring enabled him to rescue the deceased Alkêstis even
from the grasp of Thanatos (Death), and to restore her alive to her
disconsolate husband.[252]

  [249] Iliad, ii. 715.

  [250] Apollodôr. i. 9, 15; Eustath. ad Iliad. ii. 711.

  [251] Euripid. Alkêst. init. Welcker; Griechisch. Tragœd. (p.
  344) on the lost play of Sophoklês called Admêtus or Alkêstis;
  Hom. Iliad., ii. 766; Hygin. Fab. 50-51 (Sophoklês, Fr. Inc.
  730; Dind. ap. Plutarch. Defect. Orac. p. 417). This tale of the
  temporary servitude of particular gods, by order of Zeus as a
  punishment for misbehavior, recurs not unfrequently among the
  incidents of the mythical world. The poet Panyasis (ap. Clem.
  Alexand. Adm. ad Gent. p. 23)—

    Τλῆ μὲν Δημήτηρ, τλῆ δὲ κλυτὸς Ἀμφιγυήεις,
    Τλῆ δὲ Ποσειδάων, τλῆ δ᾽ ἀργυρότοξος Ἀπολλὼν
    Ἀνδρὶ παρὰ θνητῷ θητεύσεμεν εἰς ἐνιαυτόν·
    Τλῆ δὲ καὶ ὀβριμόθυμος Ἄρης ὑπὸ πατρὸς ἀνάγκης.

  The old legend followed out the fundamental idea with remarkable
  consistency: Laômedôn, as the temporary master of Poseidôn and
  Apollo, threatens to bind them hand and foot, to sell them in the
  distant islands, and to cut off the ears of both, when they come
  to ask for their stipulated wages (Iliad, xxi. 455). It was a
  new turn given to the story by the Alexandrine poets, when they
  introduced the motive of love, and made the servitude voluntary
  on the part of Apollo (Kallimachus, Hymn. Apoll. 49; Tibullus,
  Elegii. 3, 11-30).

  [252] Eurip. Alkêstis, Arg.; Apollod. i. 9, 15. To bring this
  beautiful legend more into the color of history, a new version
  of it was subsequently framed: Hêraklês was eminently skilled in
  medicine, and saved the life of Alkêstis when she was about to
  perish from a desperate malady (Plutarch. Amator c. 17. vol. iv.
  p. 53, Wytt.).

The son of Pelias, Akastus, had received and sheltered Pêleus when
obliged to fly his country in consequence of the involuntary murder of
Eurytiôn. Krêthêis, the wife of Akastus, becoming enamored of Pêleus,
made to him advances which he repudiated. Exasperated at his refusal,
and determined to procure his destruction, she persuaded her husband
that Pêleus had attempted her chastity: upon which Akastus conducted
Pêleus out upon a hunting excursion among the woody regions of Mount
Pêlion, contrived to steal from him the sword fabricated and given
by Hêphæstos, and then left him, alone and unarmed, to perish by the
hands of the Centaurs or by the wild beasts. By the friendly aid of the
Centaur Cheirôn, however, Pêleus was preserved, and his sword restored
to him: returning to the city, he avenged himself by putting to death
both Akastus and his perfidious wife.[253]

  [253] The legend of Akastus and Pêleus was given in great detail
  in the Catalogue of Hesiod (Catalog. Fragm. 20-21, Marktscheff.);
  Schol. Pindar Nem. iv. 95. Schol. Apoll. Rhod. i. 224; Apollod.
  iii. 13, 2.

But amongst all the legends with which the name of Pelias is connected,
by far the most memorable is that of Jasôn and the Argonautic
expedition. Jasôn was son of Æsôn, grandson of Krêtheus, and thus
great-grandson of Æolus. Pelias, having consulted the oracle respecting
the security of his dominion at Iôlkos, had received in answer a
warning to beware of the man who should appear before him with only
one sandal. He was celebrating a festival in honor of Poseidôn, when
it so happened that Jasôn appeared before him with one of his feet
unsandaled: he had lost one sandal in wading through the swollen
current of the river Anauros. Pelias immediately understood that this
was the enemy against whom the oracle had forewarned him. As a means
of averting the danger, he imposed upon Jasôn the desperate task of
bringing back to Iôlkos the Golden Fleece,—the fleece of that ram
which had carried Phryxos from Achaia to Kolchis, and which Phryxos
had dedicated in the latter country as an offering to the god Arês.
The result of this injunction was the memorable expedition—of the ship
Argô and her crew called the Argonauts, composed of the bravest and
noblest youths of Greece—which cannot be conveniently included among
the legends of the Æolids, and is reserved for a separate chapter.

The voyage of the Argô was long protracted, and Pelias, persuaded
that neither the ship nor her crew would ever return, put to death
both the father and mother of Jasôn, together with their infant son.
Æsôn, the father, being permitted to choose the manner of his own
death, drank bull’s blood while performing a sacrifice to the gods.
At length, however, Jasôn did return, bringing with him not only the
golden fleece, but also Mêdea, daughter of Æêtês, king of Kolchis,
as his wife,—a woman distinguished for magical skill and cunning, by
whose assistance alone the Argonauts had succeeded in their project.
Though determined to avenge himself upon Pelias, Jasôn knew he could
only succeed by stratagem: he remained with his companions at a short
distance from Iôlkos, while Mêdea, feigning herself a fugitive from
his ill-usage, entered the town alone, and procured access to the
daughters of Pelias. By exhibitions of her magical powers she soon
obtained unqualified ascendency over their minds. For example, she
selected from the flocks of Pelias a ram in the extremity of old age,
cut him up and boiled him in a caldron with herbs, and brought him
out in the shape of a young and vigorous lamb:[254] the daughters of
Pelias were made to believe that their old father could in like manner
be restored to youth. In this persuasion they cut him up with their
own hands and cast his limbs into the caldron, trusting that Mêdea
would produce upon him the same magical effect. Mêdea pretended that an
invocation to the moon was a necessary part of the ceremony: she went
up to the top of the house as if to pronounce it, and there lighting
the fire-signal concerted with the Argonauts, Jasôn and his companions
burst in and possessed themselves of the town. Satisfied with having
thus revenged himself, Jasôn yielded the principality of Iôlkos to
Akastus, son of Pelias, and retired with Mêdea to Corinth. Thus did
the goddess Hêrê gratify her ancient wrath against Pelias: she had
constantly watched over Jasôn, and had carried the “all-notorious” Argô
through its innumerable perils, in order that Jasôn might bring home
Mêdea to accomplish the ruin of his uncle.[255] The misguided daughters
of Pelias departed as voluntary exiles to Arcadia: Akastus his son
celebrated splendid funeral games in honor of his deceased father.[256]

  [254] This incident was contained in one of the earliest
  dramas of Euripidês, the Πελίαδες, now lost. Moses of Chorênê
  (Progymnasm. ap. Maii ad Euseb. p. 43), who gives an extract
  from the argument, says that the poet “extremos mentiendi fines
  attingit.”

  The Ῥιζότομοι of Sophoklês seems also to have turned upon the
  same catastrophe (see Fragm. 479, Dindorf.).

  [255] The kindness of Hêrê towards Jasôn seems to be older in
  the legend than her displeasure against Pelias; at least it is
  specially noticed in the Odyssey, as the great cause of the
  escape of the ship Argô: Ἀλλ᾽ Ἥρη παρέπεμψεν, ἐπεὶ φίλος ἦεν
  Ἰήσων. (xii. 70). In the Hesiodic Theogony Pelias stands to
  Jasôn in the same relation as Eurystheus to Hêraklês,—a severe
  taskmaster as well as a wicked and insolent man,—ὑβριστὴς Πελίης
  καὶ ἀτάσθαλος, ὀβριμόεργος. (Theog. 995). Apollônius Rhodius
  keeps the wrath of Hêrê against Pelias in the foreground, i. 14;
  iii. 1134; iv. 242; see also Hygin, f. 13.

  There is great diversity in the stories given of the proximate
  circumstances connected with the death of Pelias: Eurip. Mêd.
  491; Apollodôr. i. 9, 27; Diodôr. iv. 50-52; Ovid, Metam. vii.
  162, 203, 297, 347; Pausan. viii. 11, 2; Schol. ad Lycoph. 175.

  In the legend of Akastus and Pêleus as recounted above, Akastus
  was made to perish by the hand of Pêleus. I do not take upon me
  to reconcile these contradictions.

  Pausanias mentions that he could not find in any of the poets,
  so far as he had read, the names of the daughters of Pelias, and
  that the painter Mikôn had given to them names (ὀνόματα δ᾽ αὐταῖς
  ποιητὴς μὲν ἔθετο οὐδεὶς, ὅσα γ᾽ ἐπελεξάμεθα ἡμεῖς, etc., Pausan.
  viii. 11, 1). Yet their names are given in the authors whom
  Diodôrus copied; and Alkêstis, at any rate, was most memorable.
  Mikôn gave the names Asteropeia and Antinoê, altogether different
  from those in Diodôrus. Both Diodôrus and Hyginus exonerate
  Alkêstis from all share in the death of her father (Hygin. f. 24).

  The old poem called the Νόστοι (see Argum. ad Eurip. Mêd.,
  and Schol. Aristophan. Equit. 1321) recounted, that Mêdea had
  boiled in a caldron the old Æsôn, father of Jasôn, with herbs
  and incantations, and that she had brought him out young and
  strong. Ovid copies this (Metam. vii. 162-203). It is singular
  that Pherêkydês and Simonidês said that she had performed this
  process upon Jasôn himself (Schol. Aristoph. _l. c._). Diogenes
  (ap. Stobæ. Florileg. t. xxix. 92) rationalizes the story,
  and converts Mêdea from an enchantress into an improving and
  regenerating preceptress. The death of Æsôn, as described in the
  text, is given from Diodôrus and Apollodôrus. Mêdea seems to have
  been worshipped as a goddess in other places besides Corinth
  (see Athenagor. Legat. pro Christ. 12; Macrobius, i. 12, p. 247,
  Gronov.).

  [256] These funeral games in honor of Pelias were among the most
  renowned of the mythical incidents: they were celebrated in a
  special poem by Stesichorus, and represented on the chest of
  Kypselus at Olympia. Kastôr, Meleager, Amphiaraos, Jasôn, Pêleus,
  Mopsos, etc. contended in them (Pausan. v. 17. 4; Stesichori
  Fragm. 1. p. 54, ed. Klewe; Athên. iv. 172). How familiar
  the details of them were to the mind of a literary Greek is
  indirectly attested by Plutarch, Sympos. v. 2, vol. iii. p. 762,
  Wytt.

Jasôn and Mêdea retired from Iôlkos to Corinth, where they resided ten
years: their children were—Medeius, whom the Centaur Cheirôn educated
in the regions of Mount Pêlion,[257]—and Mermerus and Pherôs, born at
Corinth. After they had resided there ten years in prosperity, Jasôn
set his affections on Glaukê, daughter of Kreôn[258] king of Corinth;
and as her father was willing to give her to him in marriage, he
determined to repudiate Mêdea, who received orders forthwith to leave
Corinth. Stung with this insult and bent upon revenge, Mêdea prepared
a poisoned robe, and sent it as a marriage present to Glaukê: it was
unthinkingly accepted and put on, and the body of the unfortunate bride
was burnt up and consumed. Kreôn, her father, who tried to tear from
her the burning garment, shared her fate and perished. The exulting
Mêdea escaped by means of a chariot with winged serpents furnished to
her by her grandfather Hêlios: she placed herself under the protection
of Ægêus at Athens, by whom she had a son named Mêdus. She left her
young children in the sacred enclosure of the Akræan Hêrê, relying on
the protection of the altar to ensure their safety; but the Corinthians
were so exasperated against her for the murder of Kreôn and Glaukê,
that they dragged the children away from the altar and put them to
death. The miserable Jasôn perished by a fragment of his own ship Argô,
which fell upon him while he was asleep under it,[259] being hauled on
shore, according to the habitual practice of the ancients.

  [257] Hesiod, Theogon. 998.

  [258] According to the Schol. ad Eurip. Mêd. 20, Jasôn marries
  the daughter of Hippotês the son of Kreôn, who is the son
  of Lykæthos. Lykæthos, after the departure of Bellerophôn
  from Corinth, reigned twenty-seven years; then Kreôn reigned
  thirty-five years; then came Hippotês.

  [259] Apollodôr. i. 9, 27; Diodôr. iv. 54. The Mêdea of
  Euripidês, which has fortunately been preserved to us, is
  too well known to need express reference. He makes Mêdea the
  destroyer of her own children, and borrows from this circumstance
  the most pathetic touches of his exquisite drama. Parmeniskôs
  accused him of having been bribed by the Corinthians to give
  this turn to the legend; and we may regard the accusation as a
  proof that the older and more current tale imputed the murder of
  the children to the Corinthians (Schol. Eurip. Mêd. 275, where
  Didymos gives the story out of the old poem of Kreophylos). See
  also Ælian, V. H. v. 21; Pausan. ii. 3, 6.

  The most significant fact in respect to the fable is, that the
  Corinthians celebrated periodically a propitiatory sacrifice to
  Hêrê Akræa and to Mermerus and Pherês, as an atonement for the
  sin of having violated the sanctuary of the altar. The legend
  grew out of this religious ceremony, and was so arranged as to
  explain and account for it (see Eurip. Mêd. 1376, with the Schol.
  Diodôr. iv. 55).

  Mermerus and Pherês were the names given to the children of Mêdea
  and Jasôn in the old Naupaktian Verses; in which, however, the
  legend must have been recounted quite differently, since they
  said that Jasôn and Mêdea had gone from Iôlkos, not to Corinth,
  but to Corcyra; and that Mermerus had perished in hunting on the
  opposite continent of Epirus. Kinæthôn again, another ancient
  genealogical poet, called the children of Mêdea and Jasôn Eriôpis
  and Mêdos (Pausan. ii. 3, 7). Diodôrus gives them different names
  (iv. 34). Hesiod, in the Theogony, speaks only of Medeius as the
  son of Jasôn.

  Mêdea does not appear either in the Iliad or Odyssey: in the
  former, we find Agamêdê, daughter of Augeas, “who knows all the
  poisons (or medicines) which the earth nourishes” (Iliad, xi.
  740); in the latter, we have Circê, sister of Æêtês, father of
  Mêdea, and living in the Ææan island (Odyss. x. 70). Circê is
  daughter of the god Hêlios, as Mêdea is his grand-daughter,—she
  is herself a goddess. She is in many points the parallel of
  Mêdea; she forewarns and preserves Odysseus throughout his
  dangers, as Mêdea aids Jasôn: according to the Hesiodic story,
  she has two children by Odysseus, Agrius and Latinus (Theogon.
  1001).

  Odysseus goes to Ephyrê to Ilos the son of Mermerus, to procure
  poison for his arrows: Eustathius treats this Mermerus as the
  son of Mêdea (see Odyss. i. 270, and Eust.). As Ephyrê is the
  legendary name of Corinth, we may presume this to be a thread of
  the same mythical tissue.

The first establishment at Ephyrê, or Corinth, had been founded by
Sisyphus, another of the sons of Æolus, brother of Salmôneus and
Krêtheus.[260] The Æolid Sisyphus was distinguished as an unexampled
master of cunning and deceit. He blocked up the road along the isthmus,
and killed the strangers who came along it by rolling down upon them
great stones from the mountains above. He was more than a match even
for the arch thief Autolycus, the son of Hermês, who derived from his
father the gift of changing the color and shape of stolen goods, so
that they could no longer be recognized: Sisyphus, by marking his sheep
under the foot, detected Autolycus when he stole them, and obliged
him to restore the plunder. His penetration discovered the amour of
Zeus with the nymph Ægina, daughter of the river-god Asôpus. Zeus had
carried her off to the island of Œnônê (which subsequently bore the
name of Ægina); upon which Asôpus, eager to recover her, inquired of
Sisyphus whither she was gone: the latter told him what had happened,
on condition that he should provide a spring of water on the summit of
the Acro-Corinthus. Zeus, indignant with Sisyphus for this revelation,
inflicted upon him in Hadês the punishment of perpetually heaving up a
hill a great and heavy stone, which, so soon as it attained the summit,
rolled back again in spite of all his efforts, with irresistible force
into the plain.[261]

  [260] See Euripid. Æol.—Fragm. 1, Dindorf; Dikæarch. Vit. Græc.
  p. 22.

  [261] Respecting Sisyphus, see Apollodôr. i. 9, 3; iii. 12, 6.
  Pausan. ii. 5, 1. Schol. ad Iliad. i. 180. Another legend about
  the amour of Sisyphus with Tyrô, is in Hygin. fab. 60, and about
  the manner in which he overreached even Hadês (Pherekydês ap.
  Schol. Iliad. vi. 153). The stone rolled by Sisyphus in the
  under-world appears in Odyss. xi. 592. The name of Sisyphus was
  given during the historical age to men of craft and stratagem,
  such as Derkyllidês (Xenoph. Hellenic. iii. 1, 8). He passed
  for the real father of Odysseus, though Heyne (ad Apollodôr.
  i. 9, 3) treats this as another Sisyphus, whereby he destroys
  the suitableness of the predicate as regards Odysseus. The
  duplication and triplication of synonymous personages is an
  ordinary resource for the purpose of reducing the legends into a
  seeming chronological sequence.

  Even in the days of Eumêlus a religious mystery was observed
  respecting the tombs of Sisyphus and Nêleus,—the latter had also
  died at Corinth,—no one could say where they were buried (Pausan.
  ii. 2, 2).

  Sisyphus even overreached Persephonê, and made his escape from
  the under-world (Theognis, 702).

In the application of the Æolid genealogy to Corinth, Sisyphus, the
son of Æolus, appears as the first name: but the old Corinthian poet
Eumêlus either found or framed an heroic genealogy for his native city
independent both of Æolus and Sisyphus. According to this genealogy,
Ephyrê, daughter of Oceanus and Têthys, was the primitive tenant of
the Corinthian territory, Asôpus of the Sikyônian: both were assigned
to the god Hêlios, in adjusting a dispute between him and Poseidôn,
by Briareus. Hêlios divided the territory between his two sons Æêtês
and Alôeus: to the former he assigned Corinth, to the latter Sikyôn.
Æêtês, obeying the admonition of an oracle, emigrated to Kolchis,
leaving his territory under the rule of Bunos, the son of Hermês,
with the stipulation that it should be restored whenever either he
or any of his descendants returned. After the death of Bunos, both
Corinth and Sikyôn were possessed by Epôpeus, son of Alôeus, a wicked
man. His son Marathôn left him in disgust and retired into Attica,
but returned after his death and succeeded to his territory, which he
in turn divided between his two sons Corinthos and Sikyôn, from whom
the names of the two districts were first derived. Corinthos died
without issue, and the Corinthians then invited Mêdea from Iôlkos as
the representative of Æêtês: she with her husband Jasôn thus obtained
the sovereignty of Corinth.[262] This legend of Eumêlus, one of the
earliest of the genealogical poets, so different from the story adopted
by Neophrôn or Euripidês, was followed certainly by Simonidês and
seemingly by Theopompus.[263] The incidents in it are imagined and
arranged with a view to the supremacy of Mêdea; the emigration of Æêtês
and the conditions under which he transferred his sceptre, being so
laid out as to confer upon Mêdea an hereditary title to the throne.
The Corinthians paid to Mêdea and to her children solemn worship,
either divine or heroic, in conjunction with Hêrê Akræa,[264] and this
was sufficient to give to Mêdea a prominent place in the genealogy
composed by a Corinthian poet, accustomed to blend together gods,
heroes and men in the antiquities of his native city. According to
the legend of Eumêlus, Jasôn became (through Mêdea) king of Corinth;
but she concealed the children of their marriage in the temple of
Hêrê, trusting that the goddess would render them immortal. Jasôn,
discovering her proceedings, left her and retired in disgust to Iôlkos;
Mêdea also, being disappointed in her scheme, quitted the place,
leaving the throne in the hands of Sisyphus, to whom, according to
the story of Theopompus, she had become attached.[265] Other legends
recounted, that Zeus had contracted a passion for Mêdea, but that she
had rejected his suit from fear of the displeasure of Hêrê; who, as
a recompense for such fidelity, rendered her children immortal:[266]
moreover Mêdea had erected, by special command of Hêrê, the celebrated
temple of Aphroditê at Corinth. The tenor of these fables manifests
their connection with the temple of Hêrê: and we may consider the
legend of Mêdea as having been originally quite independent of that of
Sisyphus, but fitted on to it, in seeming chronological sequence, so as
to satisfy the feelings of those Æolids of Corinth who passed for his
descendants.

  [262] Pausan. ii. 1, 1; 3, 10. Schol. ad Pindar, Olymp. xiii. 74.
  Schol. Lycoph. 174-1024. Schol. Apoll. Rhod. iv. 1212.

  [263] Simonid. ap. Schol. ad Eurip. Mêd. 10-20; Theopompus,
  Fragm. 340, Didot; though Welcker (Der Episch. Cycl. p. 29)
  thinks that this does not belong to the historian Theopompus.
  Epimenidês also followed the story of Eumêlus in making Æêtês a
  Corinthian (Schol. ad Apoll. Rhod. iii. 242).

  [264] Περὶ δὲ τῆς εἰς Κόρινθον μετοικήσεως, Ἵππυς ἐκτίθεται καὶ
  Ἑλλάνικος· ὅτι δὲ βεβασίλευκε τῆς Κορίνθου ἡ Μήδεια, Εὔμηλος
  ἱστορεῖ καὶ Σιμωνίδης· Ὅτι δὲ καὶ ἀθάνατος ἦν ἡ Μήδεια, Μουσαῖος
  ἐν τῷ περὶ Ἰσθμίων ἱστορεῖ, ἅμα καὶ περὶ τῶν τῆς Ἀκραίας Ἥρας
  ἑορτῶν ἐκτιθείς. (Schol. Eurip. Mêd. 10). Compare also v. 1376
  of the play itself, with the Scholia and Pausan. ii. 3, 6. Both
  Alkman and Hesiod represented Mêdea as a goddess (Athenagoras,
  Legatia pro Christianis, p. 54, ed. Oxon.).

  [265] Pausan. ii. 3, 10; Schol. Pindar. Olymp. xiii. 74.

  [266] Schol. Pindar. Olymp. xiii. 32-74; Plutarch, De Herodot.
  Malign. p. 871.

Sisyphus had for his sons Glaukos and Ornytiôn. From Glaukos sprang
Bellerophôn, whose romantic adventures commence with the Iliad, and
are further expanded by subsequent poets: according to some accounts
he was really the son of Poseidôn, the prominent deity of the Æolid
family.[267] The youth and beauty of Bellerophôn rendered him the
object of a strong passion on the part of the Anteia, wife of Prœtos
king of Argos. Finding her advances rejected, she contracted a violent
hatred towards him, and endeavored by false accusations to prevail upon
her husband to kill him. Prœtos refused to commit the deed under his
own roof, but despatched him to his son-in-law the king of Lykia in
Asia Minor, putting into his hands a folded tablet full of destructive
symbols. Conformably to these suggestions, the most perilous
undertakings were imposed upon Bellerophôn. He was directed to attack
the monster Chimæra and to conquer the warlike Solymi as well as the
Amazons: as he returned victorious from these enterprises, an ambuscade
was laid for him by the bravest Lykian warriors, all of whom he slew.
At length the Lykian king recognized him “as the genuine son of a
god,” and gave him his daughter in marriage together with half of his
kingdom. The grand-children of Bellerophôn, Glaukos and Sarpêdôn,—the
latter a son of his daughter Laodameia by Zeus,—combat as allies of
Troy against the host of Agamemnon.[268] Respecting the winged Pegasus,
Homer says nothing; but later poets assigned to Bellerophôn this
miraculous steed, whose parentage is given in the Hesiodic Theogony,
as the instrument both of his voyage and of his success.[269] Heroic
worship was paid at Corinth to Bellerophôn, and he seems to have
been a favorite theme of recollection not only among the Corinthians
themselves, but also among the numerous colonists whom they sent out to
other regions.[270]

  [267] Pindar, Olymp. xiii. 98. and Schol. ad 1; Schol. ad Iliad,
  vi. 155; this seems to be the sense of Iliad, vi. 191.

  The lost drama called _Iobatês_ of Sophoklês, and the two by
  Euripidês called _Sthenebœa_ and _Bellerophôn_, handled the
  adventures of this hero. See the collection of the few fragments
  remaining in Dindorf, Fragm. Sophok. 280; Fragm. Eurip. p.
  87-108; and Hygin. fab. 67.

  Welcker (Griechische Tragöd. ii. p. 777-800) has ingeniously put
  together all that can be divined respecting the two plays of
  Euripidês.

  Völcker seeks to make out that Bellerophôn is identical with
  Poseidôn Hippios,—a separate personification of one of the
  attributes of the god Poseidôn. For this conjecture he gives some
  plausible grounds (Mythologie des Japetisch. Geschlechts, p. 129
  _seq._).

  [268] Iliad, vi. 155-210.

  [269] Hesiod, Theogon. 283.

  [270] Pausan. ii. 2, 4. See Pindar, Olymp. xiii. 90, addressed
  to Xenophôn the Corinthian, and the Adoniazusæ of the Syracusan
  Theocritus, a poem in which common Syracusan life and feeling are
  so graphically depicted, Idyll xv. 91.—

                            Συρακοσίαις ἐπιτάσσεις;
    Ὡς δ᾽ εἰδῇς καὶ τοῦτο, Κορίνθιαι εἶμες ἄνωθεν
    Ὡς καὶ ὁ Βελλερόφων· Πελοποννασιστὶ λαλεῦμες.

From Ornytiôn, the son of Sisyphus, we are conducted through a series
of three undistinguished family names,—Thoas, Damophôn, and the
brothers Propodas and Hyanthidas,—to the time of the Dôrian occupation
of Corinth[271], which will be hereafter recounted.

  [271] Pausan. ii. 4, 3.

We now pass from Sisyphus and the Corinthian fables to another son of
Æolus, Athamas, whose family history is not less replete with mournful
and tragical incidents, abundantly diversified by the poets. Athamas,
we are told, was king of Orchomenos; his wife Nephelê was a goddess,
and he had by her two children, Phryxus and Hellê. After a certain
time he neglected Nephelê, and took to himself as a new wife Inô, the
daughter of Kadmus, by whom he had two sons, Learchus and Melikertês.
Inô, looking upon Phryxus with the hatred of a step-mother, laid a
snare for his life. She persuaded the women to roast the seed-wheat,
which, when sown in this condition, yielded no crop, so that famine
overspread the land. Athamas sent to Delphi to implore counsel and a
remedy: he received for answer, through the machinations of Inô with
the oracle, that the barrenness of the fields could not be alleviated
except by offering Phryxus as a sacrifice to Zeus. The distress of the
people compelled him to execute this injunction, and Phryxus was led as
a victim to the altar. But the power of his mother Nephelê snatched him
from destruction, and procured for him from Hermês a ram with a fleece
of gold, upon which he and his sister Hellê mounted and were carried
across the sea. The ram took the direction of the Euxine sea and
Kolchis: when they were crossing the Hellespont, Hellê fell off into
the narrow strait, which took its name from that incident. Upon this,
the ram, who was endued with speech, consoled the terrified Phryxus,
and ultimately carried him safe to Kolchis: Æêtês, king of Kolchis son
of the god Hêlios and brother of Circê, received Phryxus kindly, and
gave him his daughter Chalciopê in marriage. Phryxus sacrificed the ram
to Zeus Phyxios, and suspended the golden fleece in the sacred grove of
Arês.

Athamas—according to some both Athamas and Inô—were afterwards driven
mad by the anger of the goddess Hêrê; insomuch that the father shot
his own son Learchus, and would also have put to death his other son
Melikertês, if Inô had not snatched him away. She fled with the boy,
across the Megarian territory and Mount Geraneia, to the rock Moluris,
overhanging the Sarônic Gulf: Athamas pursued her, and in order to
escape him she leaped into the sea. She became a sea-goddess under the
title of Leukothea; while the body of Melikertês was cast ashore on the
neighboring territory of Schœnus, and buried by his uncle Sisyphus, who
was directed by the Nereïds to pay to him heroic honors under the name
of Palæmôn. The Isthmian games, one of the great periodical festivals
of Greece, were celebrated in honor of the god Poseidôn, in conjunction
with Palæmôn as a hero. Athamas abandoned his territory, and became the
first settler of a neighboring region called from him Athmantia, or the
Athamantian plain. [272]

  [272] Eurip. Mêd. 1250, with the Scholia, according to which
  story Inô killed both her children:—

    Ἴνω μανεῖσαν ἐκ θεῶν, ὅθ᾽ ἡ Διὸς
    Δάμαρ νιν ἐξέπεμψε δώματων ἄλῃ.

  Compare Valckenaer, Diatribe in Eurip.; Apollodôr. i. 9, 1-2;
  Schol. ad Pindar. Argum. ad Isthm. p. 180. The many varieties of
  the fable of Athamas and his family may be seen in Hygin. fab.
  1-5; Philostephanus ap. Schol. Iliad, vii. 86: it was a favorite
  subject with the tragedians, and was handled by Æschylus,
  Sophoklês and Euripidês in more than one drama (see Welcker,
  Griechische Tragöd. vol. i. p. 312-332; vol. ii. p. 612). Heyne
  says that the proper reading of the name is _Phrixus_, not
  _Phryxus_,—incorrectly, I think: Φρύξος connects the name both
  with the story of roasting the wheat (φρύγειν), and also with
  the country Φρυγία, of which it was pretended that Phryxus was
  the Eponymus. Inô, or Leukothea, was worshipped as a heroine at
  Megara as well as at Corinth (Pausan. i. 42, 3): the celebrity
  of the Isthmian games carried her worship, as well as that of
  Palæmôn, throughout most parts of Greece (Cicero, De Nat. Deor.
  iii. 16). She is the only personage of this family noticed
  either in the Iliad or Odyssey: in the latter poem she is a
  sea-goddess, who has once been a mortal, daughter of Kadmus; she
  saves Odysseus from imminent danger at sea by presenting to him
  her κρήδεμνον (Odyss. v. 433; see the refinements of Aristidês,
  Orat. iii. p. 27). The voyage of Phryxus and Hellê to Kolchis was
  related in the Hesiodic Eoiai: we find the names of the children
  of Phryxus by the daughter of Æêtês quoted from that poem (Schol.
  ad Apollôn. Rhod. ii. 1123) both Hesiod and Pherekydês mentioned
  the golden fleece of the ram (Eratosthen. Catasterism. 19;
  Pherekyd. Fragm. 53, Didot).

  Hekatæus preserved the romance of the speaking ram (Schol. Apoll.
  Rhod. i. 256) but Hellanikus dropped the story of Hellê having
  fallen into the sea: according to him she died at Pactyê in the
  Chersonesus (Schol. Apoll. Rhod. ii. 1144).

  The poet Asius seems to have given the genealogy of Athamas by
  Themistô much in the same manner as we find it in Apollodôrus
  (Pausan. ix. 23, 3).

  According to the ingenious refinements of Dionysius and
  Palæphatus (Schol. ad Apoll. Rhod. ii. 1144; Palæphat. de Incred.
  c. 31) the ram of Phryxus was after all a man named Krios, a
  faithful attendant who aided in his escape; others imagined a
  ship with a ram’s head at the bow.

The legend of Athamas connects itself with some sanguinary religious
rites and very peculiar family customs, which prevailed at Alos,
in Achaia Phthiôtis, down to a time[273] later than the historian
Herodotus, and of which some remnant existed at Orchomenos even in the
days of Plutarch. Athamas was worshipped at Alos as a hero, having
both a chapel and a consecrated grove, attached to the temple of Zeus
Laphystios. On the family of which he was the heroic progenitor, a
special curse and disability stood affixed. The eldest of the race was
forbidden to enter the prytaneion or government-house; and if he was
found within the doors of the building, the other citizens laid hold
of him on his going out, surrounded him with garlands, and led him in
solemn procession to be sacrificed as a victim at the altar of Zeus
Laphystios. The prohibition carried with it an exclusion from all the
public meetings and ceremonies, political as well as religious, and
from the sacred fire of the state: many of the individuals marked out
had therefore been bold enough to transgress it. Some had been seized
on quitting the building and actually sacrificed; others had fled the
country for a long time to avoid a similar fate.

  [273] Plutarch, Quæst. Græc. c. 38. p. 299. Schol. Apoll. Rhod.
  ii. 655.

The guides who conducted Xerxês and his army through southern Thessaly
detailed to him this existing practice, coupled with the local legend,
that Athamas, together with Inô, had sought to compass the death of
Phryxus, who however had escaped to Kolchis; that the Achæans had been
enjoined by an oracle to offer up Athamas himself as an expiatory
sacrifice to release the country from the anger of the gods; but that
Kytissoros, son of Phryxus, coming back from Kolchis, had intercepted
the sacrifice of Athamas,[274] whereby the anger of the gods remained
still unappeased, and an undying curse rested upon the family.[275]

  [274] Of the Athamas of Sophoklês, turning upon this intended,
  but not consummated sacrifice, little is known, except from a
  passage of Aristophanês and the Scholia upon it (Nubes, 258).—

            ἐπὶ τί στέφανον; οἴμοι, Σώκρατες,
    ὥσπερ με τὸν Ἀθάμανθ᾽ ὅπως μὴ θύσετε.

  Athamas was introduced in this drama with a garland on his head,
  on the point of being sacrificed as an expiation for the death of
  his son Phryxus, when Hêraklês interposes and rescues him.

  [275] Herodot. vii. 197. Plato, Minôs, p. 315.

That such human sacrifices continued to a greater or less extent,
even down to a period later than Herodotus, among the family who
worshipped Athamas as their heroic ancestor, appears certain: mention
is also made of similar customs in parts of Arcadia, and of Thessaly,
in honor of Pêleus and Cheirôn.[276] But we may reasonably presume,
that in the period of greater humanity which Herodotus witnessed,
actual sacrifice had become very rare. The curse and the legend still
remained, but were not called into practical working, except during
periods of intense national suffering or apprehension, during which
the religious sensibilities were always greatly aggravated. We cannot
at all doubt, that during the alarm created by the presence of the
Persian king with his immense and ill-disciplined host, the minds of
the Thessalians must have been keenly alive to all that was terrific in
their national stories, and all that was expiatory in their religious
solemnities. Moreover, the mind of Xerxês himself was so awe-struck by
the tale, that he reverenced the dwelling-place consecrated to Athamas.
The guides who recounted to him the romantic legend, gave it as the
historical and generating cause of the existing rule and practice: a
critical inquirer is forced (as has been remarked before) to reverse
the order of precedence, and to treat the practice as having been the
suggesting cause of its own explanatory legend.

  [276] Plato, Minôs, c. 5. Καὶ οἱ τοῦ Ἀθάμαντος ἔκγονοι, οἵας
  θυσίας θύουσιν, Ἕλληνες ὄντες. As a testimony to the fact still
  existing or believed to exist, this dialogue is quite sufficient,
  though not the work of Plato.

  Μόνιμος δ᾽ ἱστορεῖ, ἐν τῇ τῶν θαυμασίων συναγωγῇ, ἐν Πέλλῃ τῆς
  Θετταλίας Ἀχαιὸν ἄνθρωπον Πηλεῖ καὶ Χείρωνι καταθύεσθαι. (Clemens
  Alexand. Admon. ad Gent. p. 27, Sylb.) Respecting the sacrifices
  at the temple of Zeus Lykæus in Arcadia, see Plato, Republ.
  viii. p. 565. Pausanias (viii. p. 38, 5) seems to have shrunk,
  when he was upon the spot, even from inquiring what they were—a
  striking proof of the fearful idea which he had conceived of
  them. Plutarch (De Defectu Oracul. c. 14) speaks of τὰς πάλαι
  ποιουμένας ἀνθρωποθυσίας. The Schol. ad Lycophron. 229, gives a
  story of children being sacrificed to Melikertês at Tenedos; and
  Apollodôrus (ad Porphyr. de Abstinentiâ, ii. 55, see Apollod.
  Fragm. 20, ed. Didot) said that the Lacedæmonians had sacrificed
  a man to Arês—καὶ Λακεδαιμονίους φησὶν ὁ Ἀπολλόδωρος τῷ Ἄρει
  θύειν ἄνθρωπον. About Salamis in Cyprus, see Lactantius, De Falsâ
  Religione, i. c. 21. “Apud Cypri Salaminem, humanam hostiam Jovi
  Teucrus immolavit, idque sacrificium posteris tradidit: quod est
  nuper Hadriano imperante sublatum.”

  Respecting human sacrifices in historical Greece, consult a
  good section in K. F. Hermann’s Gottesdienstliche Alterthümer
  der Griechen (sect. 27). Such sacrifices had been a portion of
  primitive Grecian religion, but had gradually become obsolete
  everywhere—except in one or two solitary cases, which were spoken
  of with horror. Even in these cases, too, the reality of the
  fact, in later times, is not beyond suspicion.

The family history of Athamas, and the worship of Zeus Laphystios, are
expressly connected by Herodotus with Alos in Achæa Phthiôtis—one of
the towns enumerated in the Iliad as under the command of Achilles. But
there was also a mountain called Laphystion, and a temple and worship
of Zeus Laphystios between Orchomenos and Korôneia, in the northern
portion of the territory known in the historical ages as Bœotia. Here
also the family story of Athamas is localized, and Athamas is presented
to us as king of the districts of Korôneia, Haliartus and Mount
Laphystion: he is thus interwoven with the Orchomenian genealogy.[277]
Andreus (we are told), son of the river Pêneios, was the first person
who settled in the region: from him it received the name Andrêis.
Athamas, coming subsequently to Andreus, received from him the
territory of Korôneia and Haliartus with Mount Laphystion: he gave
in marriage to Andreus, Euippê, daughter of his son Leucôn, and the
issue of this marriage was Eteoklês, said to be the son of the river
Kêphisos. Korônos and Haliartus, grandsons of the Corinthian Sisyphus,
were adopted by Athamas, as he had lost all his children: but when his
grandson Presbôn, son of Phryxus, returned to him from Kolchis, he
divided his territory in such manner that Korônos and Haliartus became
the founders of the towns which bore their names. Almôn, the son of
Sisyphus, also received from Eteoklês a portion of territory, where he
established the village Almônes.[278]

  [277] Pausan. ix. 34, 4.

  [278] Pausan. ix. 34, 5.

With Eteoklês began, according to a statement in one of the Hesiodic
poems, the worship of the Charites or Graces, so long and so
solemnly continued at Orchomenos in the periodical festival of the
Charitêsia, to which many neighboring towns and districts seem to
have contributed.[279] He also distributed the inhabitants into two
tribes—Eteokleia and Kêphisias. He died childless, and was succeeded
by Almos, who had only two daughters, Chrysê and Chrysogeneia. The son
of Chrysê by the god Arês was Phlegyas, the father and founder of the
warlike and predatory Phlegyæ, who despoiled every one within their
reach, and assaulted not only the pilgrims on their road to Delphi,
but even the treasures of the temple itself. The offended god punished
them by continued thunder, by earthquakes, and by pestilence, which
extinguished all this impious race, except a scanty remnant who fled
into Phokis.

  [279] Ephorus, Fragm. 68, Marx.

Chrysogeneia, the other daughter of Almos, had for issue, by the god
Poseidôn, Minyas: the son of Minyas was Orchomenos. From these two was
derived the name both of Minyæ for the people, and of Orchomenos for
the town.[280] During the reign of Orchomenos, Hyêttus came to him from
Argos, having become an exile in consequence of the death of Molyros:
Orchomenos assigned to him a portion of land, where he founded the
village called Hyêttus.[281] Orchomenos, having no issue, was succeeded
by Klymenos, son of Presbôn, of the house of Athamas: Klymenos was
slain by some Thêbans during the festival of Poseidôn at Onchêstos; and
his eldest son, Erginus, to avenge his death, attacked the Thêbans with
his utmost force;—an attack, in which he was so successful, that the
latter were forced to submit, and to pay him an annual tribute.

  [280] Pausan. ix. 36, 1-3. See also a legend, about the three
  daughters of Minyas, which was treated by the Tanagræan poetess
  Korinna, the contemporary of Pindar (Antonin. Liberalis, Narr.
  x.).

  [281] This exile of Hyêttus was recounted in the Eoiai. Hesiod,
  Fragm. 148, Markt.

The Orchomenian power was now at its height: both Minyas and Orchomenos
had been princes of surpassing wealth, and the former had built
a spacious and durable edifice which he had filled with gold and
silver. But the success of Erginus against Thêbes was soon terminated
and reversed by the hand of the irresistible Hêraklês, who rejected
with disdain the claim of tribute, and even mutilated the envoys
sent to demand it: he not only emancipated Thêbes, but broke down
and impoverished Orchomenos.[282] Erginus in his old age married a
young wife, from which match sprang the illustrious heroes, or gods,
Trophônius and Agamêdês; though many (amongst whom is Pausanius
himself) believed Trophônius to be the son of Apollo.[283] Trophônius,
one of the most memorable persons in Grecian mythology, was worshipped
as a god in various places, but with especial sanctity as Zeus
Trophônius at Lebadeia: in his temple at this town, the prophetic
manifestations outlasted those of Delphi itself.[284] Trophônius and
Agamêdês, enjoying matchless renown as architects, built[285] the
temple of Delphi, the thalamus of Amphitryôn at Thêbes, as well as the
inaccessible vault of Hyrieus at Hyria, in which they are said to have
left one stone removable at pleasure, so as to reserve for themselves
a secret entrance. They entered so frequently, and stole so much gold
and silver, that Hyrieus, astonished at his losses, at length spread a
fine net, in which Agamêdês was inextricably caught: Trophônius cut off
his brother’s head and carried it away, so that the body, which alone
remained, was insufficient to identify the thief. Like Amphiaraos, whom
he resembles in more than one respect, Trophônius was swallowed up by
the earth near Lebadeia.[286]

  [282] Pausan. ix. 37, 2. Apollod. ii. 4, 11. Diodôr. iv. 10. The
  two latter tell us that Erginus was slain. Klymenê is among the
  wives and daughters of the heroes seen by Odysseus in Hadês: she
  is termed by the Schol. daughter of Minyas (Odyss. xi. 325).

  [283] Pausan. ix. 37, 1-3. Λέγεται δὲ ὁ Τροφώνιος Ἀπόλλωνος
  εἶναι, καὶ οὐκ Ἐργίνου· καὶ ἐγώ τε πείθομαι, καὶ ὅστις παρὰ
  Τροφώνιον ἦλθε δὴ μαντευσόμενος.

  [284] Plutarch, De Defectu Oracul. c. 5, p. 411. Strabo, ix.
  p. 414. The mention of the honeyed cakes, both in Aristophanês
  (Nub. 508) and Pausanias (ix. 39, 5), indicates that the curious
  preliminary ceremonies, for those who consulted the oracle
  of Trophônius, remained the same after a lapse of 550 years.
  Pausanias consulted it himself. There had been at one time an
  oracle of Teiresias at Orchomenos: but it had become silent at an
  early period (Plutarch. Defect. Oracul. c. 44, p. 434).

  [285] Homer. Hymn. Apoll. 296. Pausan. ix. 11, 1.

  [286] Pausan. ix. 37, 3. A similar story, but far more romantic
  and amplified, is told by Herodotus (ii. 121), respecting the
  treasury vault of Rhampsinitus, king of Egypt. Charax (ap.
  Schol. Aristoph. Nub. 508) gives the same tale, but places the
  scene in the treasury-vault of Augeas, king of Elis, which he
  says was built by Trophônius, to whom he assigns a totally
  different genealogy. The romantic adventures of the tale rendered
  it eminently fit to be interwoven at some point or another of
  legendary history, in any country.

From Trophônius and Agamêdês the Orchomenian genealogy passes to
Ascalaphos and Ialmenos, the sons of Arês by Astyochê, who are named
in the Catalogue of the Iliad as leaders of the thirty ships from
Orchomenos against Troy. Azeus, the grandfather of Astyochê in the
Iliad, is introduced as the brother of Erginus[287] by Pausanias, who
does not carry the pedigree lower.

  [287] Pausan. ix. 38, 6; 29, 1.

The genealogy here given out of Pausanias is deserving of the more
attention, because it seems to have been copied from the special
history of Orchomenos by the Corinthian Kallippus, who again borrowed
from the native Orchomenian poet, Chersias: the works of the latter
had never come into the hands of Pausanias. It illustrates forcibly
the principle upon which these mythical genealogies were framed, for
almost every personage in the series is an Eponymus. Andreus gave
his name to the country, Athamas to the Athamantian plain; Minyas,
Orchomenos, Korônus, Haliartus, Almos and Hyêttos, are each in like
manner connected with some name of people, tribe, town or village;
while Chrysê and Chrysogeneia have their origin in the reputed ancient
wealth of Orchomenos. Abundant discrepancies are found, however, in
respect to this old genealogy, if we look to other accounts. According
to one statement, Orchomenos was the son of Zeus by Isionê, daughter
of Danaus; Minyas was the son of Orchomenos (or rather of Poseidôn)
by Hermippê, daughter of Bœôtos; the sons of Minyas were Presbôn,
Orchomenos, Athamas and Diochthôndas.[288] Others represented Minyas as
son of Poseidôn by Kallirrhoê, an Oceanic nymph,[289] while Dionysius
called him son of Arês, and Aristodêmus, son of Aleas: lastly, there
were not wanting authors who termed both Minyas and Orchomenos sons
of Eteoklês.[290] Nor do we find in any one of these genealogies the
name of Amphiôn, the son of Iasus, who figures so prominently in the
Odyssey as king of Orchomenos, and whose beautiful daughter Chlôris is
married to Nêleus. Pausanias mentions him, but not as king, which is
the denomination given to him in Homer.[291]

  [288] Schol. Apollôn. Rhod. i. 230. Compare Schol. ad Lycophron.
  873.

  [289] Schol. Pindar, Olymp. xiv. 5.

  [290] Schol. Pindar, Isthm. i. 79. Other discrepancies in Schol.
  Vett. ad Iliad. ii. Catalog. 18.

  [291] Odyss. xi. 283. Pausan. ix. 36, 3.

The discrepancies here cited are hardly necessary in order to prove
that these Orchomenian genealogies possess no historical value. Yet
some probable inferences appear deducible from the general tenor of the
legends, whether the facts and persons of which they are composed be
real or fictitious.

Throughout all the historical age, Orchomenos is a member of the
Bœôtian confederation. But the Bœôtians are said to have been
immigrants into the territory which bore their name from Thessaly; and
prior to the time of their immigration, Orchomenos and the surrounding
territory appear as possessed by the Minyæ, who are recognized in that
locality both in the Iliad and in the Odyssey,[292] and from whom
the constantly recurring Eponymus, King Minyas, is borrowed by the
genealogists. Poetical legend connects the Orchomenian Minyæ on the one
side, with Pylos and Tryphylia in Peloponnêsus; on the other side, with
Phthiôtis and the town of Iôlkos in Thessaly; also with Corinth,[293]
through Sisyphus and his sons. Pherekydês represented Nêleus, king
of Pylos, as having also been king of Orchomenos.[294] In the region
of Triphylia, near to or coincident with Pylos, a Minyeian river is
mentioned by Homer; and we find traces of residents called Minyæ even
in the historical times, though the account given by Herodotus of the
way in which they came thither is strange and unsatisfactory.[295]

  [292] Iliad, ii. 5, 11. Odyss. xi. 283. Hesiod, Fragm. Eoiai,
  27, Düntz. Ἴξεν δ᾽ Ὀρχόμενον Μινυήϊον. Pindar, Olymp. xiv. 4.
  Παλαιγόνων Μινυᾶν ἐπίσκοποι. Herodot. i. 146. Pausanias calls
  them Minyæ even in their dealings with Sylla (ix. 30, 1).
  Buttmann, in his Dissertation (Über die Minyæ der Ältesten Zeit,
  in the Mythologus, Diss. xxi. p. 218), doubts whether the name
  Minyæ was ever a real name; but all the passages make against his
  opinion.

  [293] Schol. Apoll. Rhod. ii. 1186. i. 230. Σκήψιος δὲ Δημήτρός
  φησι τοὺς περὶ τὴν Ἰωλκὸν οἰκοῦντας Μινύας καλεῖσθαι; and i. 763.
  Τὴν γὰρ Ἰωλκὸν οἱ Μίνυαι ᾤκουν, ὥς φησι Σιμωνίδης ἐν Συμμικτοῖς:
  also Eustath. ad Iliad. ii. 512. Steph. Byz. v. Μινύα. Orchomenos
  and Pylos run together in the mind of the poet of the Odyssey,
  xi. 458.

  [294] Pherekyd. Fragm. 56, Didot. We see by the 55th Fragment of
  the same author, that he extended the genealogy of Phryxos to
  Pheræ in Thessaly.

  [295] Herodot. iv. 145. Strabo, viii. 337-347. Hom. Iliad, xi.
  721. Pausan. v. 1, 7. ποταμὸν Μινυήϊον, near Elis.

Before the great changes which took place in the inhabitants of
Greece from the immigration of the Thesprôtians into Thessaly, of
the Bœôtians into Bœôtia, and of the Dôrians and Ætôlians into
Peloponnêsus, at a date which we have no means of determining, the
Minyæ and tribes fraternally connected with them seem to have occupied
a large portion of the surface of Greece, from Iôlkos in Thessaly to
Pylos in the Peloponnêsus. The wealth of Orchomenos is renowned even
in the Iliad;[296] and when we study its topography in detail, we are
furnished with a probable explanation both of its prosperity and its
decay. Orchomenos was situated on the northern bank of the lake Kôpaïs,
which receives not only the river Kêphisos from the valleys of Phôkis,
but also other rivers from Parnassus and Helicôn. The waters of the
lake find more than one subterranean egress—partly through natural
rifts and cavities in the limestone mountains, partly through a tunnel
pierced artificially more than a mile in length—into the plain on the
north-eastern side, from whence they flow into the Eubœan sea near
Larymna:[297] and it appears that, so long as these channels were
diligently watched and kept clear, a large portion of the lake was in
the condition of alluvial land, preëminently rich and fertile. But when
the channels came to be either neglected, or designedly choked up by an
enemy, the water accumulated to such a degree, as to occupy the soil
of more than one ancient town, to endanger the position of Kôpæ, and
to occasion the change of the site of Orchomenos itself from the plain
to the declivity of Mount Hyphanteion. An engineer, Kratês, began the
clearance of the obstructed water-courses in the reign of Alexander the
Great, and by his commission—the destroyer of Thêbes being anxious to
reëstablish the extinct prosperity of Orchomenos. He succeeded so far
as partially to drain and diminish the lake, whereby the site of more
than one ancient city was rendered visible: but the revival of Thêbes
by Kassander, after the decease of Alexander, arrested the progress of
the undertaking, and the lake soon regained its former dimensions, to
contract which no farther attempt was made.[298]

  [296] Iliad, ix. 381.

  [297] See the description of these channels or Katabothra in
  Colonel Leake’s Travels in Northern Greece, vol. ii. c. 15, p.
  281-293, and still more elaborately in Fiedler, Reise durch alle
  Theile des Königreichs Griechenlands, Leipzig, 1840. He traced
  fifteen perpendicular shafts sunk for the purpose of admitting
  air into the tunnel, the first separated from the last by about
  5900 feet: they are now of course overgrown and stopped up (vol.
  i. p. 115).

  Forchhammer states the length of this tunnel as considerably
  greater than what is here stated. He also gives a plan of the
  Lake Kôpaïs with the surrounding region, which I have placed at
  the end of the second volume of this History. See also _infra_,
  vol. ii. ch. iii. p. 391.

  [298] We owe this interesting fact to Strabo, who is however both
  concise and unsatisfactory, viii. p. 406-407. It was affirmed
  that there had been two ancient towns, named Eleusis and Athênæ,
  originally founded by Cecrôps, situated on the lake, and thus
  overflowed (Steph. Byz. v. Ἀθῆναι Diogen. Laërt. iv. 23. Pausan.
  ix. 24, 2). For the plain or marsh near Orchomenos, see Plutarch,
  Sylla, c. 20-22.

According to the Thêban legend,[299] Hêraklês, after his defeat of
Erginus had blocked up the exit of the waters, and converted the
Orchomenian plain into a lake. The spreading of these waters is thus
connected with the humiliation of the Minyæ; and there can be little
hesitation in ascribing to these ancient tenants of Orchomenos,
before it became bœotized, the enlargement and preservation of these
protective channels. Nor could such an object have been accomplished,
without combined action and acknowledged ascendency on the part of
that city over its neighbors, extending even to the sea at Larymna,
where the river Kôphisos discharges itself. Of its extended influence,
as well as of its maritime activity, we find a remarkable evidence in
the ancient and venerated Amphiktyony at Kalauria. The little island
so named, near the harbor of Trœzên, in Peloponnêsus, was sacred
to Poseidôn, and an asylum of inviolable sanctity. At the temple
of Poseidôn, in Kalauria, there had existed, from unknown date, a
periodical sacrifice, celebrated by seven cities in common—Hermionê,
Epidaurus, Ægina, Athens, Prasiæ, Nauplia, and the Minyeian Orchomenos.
This ancient religious combination dates from the time when Nauplia
was independent of Argos, and Prasiæ of Sparta: Argos and Sparta,
according to the usual practice in Greece, continued to fulfil the
obligation each on the part of its respective dependent.[300] Six
out of the seven states are at once sea-towns, and near enough to
Kalauria to account for their participation in this Amphiktyony. But
the junction of Orchomenos, from its comparative remoteness, becomes
inexplicable, except on the supposition that its territory reached the
sea, and that it enjoyed a considerable maritime traffic—a fact which
helps to elucidate both its legendary connection with Iôlkos, and its
partnership in what is called the Iônic emigration.[301] The mythical
genealogy, whereby Ptôos, Schœneus and Erythrios are enumerated among
the sons of Athamas, goes farther to confirm the idea that the towns
and localities on the south-east of the lake recognized a fraternal
origin with the Orchomenian Minyæ, not less than Korôneia and Haliartus
on the south-west.[302]

  [299] Diodôr. iv. 18. Pausan. ix. 38, 5.

  [300] Strabo, viii. p. 374. Ἦν δὲ καὶ Ἀμφικτυονία τις περὶ τὸ
  ἱερὸν τοῦτο, ἕπτα πόλεων αἳ μετεῖχον τῆς θυσίας· ἦσαν δὲ Ἑρμιὼν,
  Ἐπίδαυρος, Αἴγινα, Ἀθῆναι, Πρασιεῖς, Ναυπλιεῖς, Ὀρχόμενος ὁ
  Μινύειος. Ὑπὲρ μὲν οὖν τῶν Ναυπλιέων Ἀργεῖοι, ὑπὲρ Πρασιέων δὲ
  Λακεδαιμόνιοι, ξυνετέλουν.

  [301] Pausan. ix. 17, 1; 26, 1.

  [302] See Müller, Orchomenos und die Minyer, p. 214. Pausan. ix.
  23, 3; 24, 3. The genealogy is as old as the poet Asios.

The great power of Orchomenos was broken down, and the city reduced to
a secondary and half-dependent position by the Bœôtians of Thêbes; at
what time, and under what circumstances, history has not preserved. The
story, that the Thêban hero, Hêraklês, rescued his native city from
servitude and tribute to Orchomenos, since it comes from a Kadmeian
and not from an Orchomenian legend, and since the details of it were
favorite subjects of commemoration in the Thêbian temples,[303]
affords a presumption that Thêbes was really once dependent on
Orchomenos. Moreover the savage mutilations inflicted by the hero on
the tribute-seeking envoys, so faithfully portrayed in his surname
Rhinokoloustês, infuse into the mythe a portion of that bitter feeling
which so long prevailed between Thêbes and Orchomenos, and which led
the Thêbans, as soon as the battle of Leuctra had placed supremacy in
their hands, to destroy and depopulate their rival.[304] The ensuing
generation saw the same fate retorted upon Thêbes, combined with
the restoration of Orchomenos. The legendary grandeur of this city
continued, long after it had ceased to be distinguished for wealth
and power, imperishably recorded both in the minds of the nobler
citizens and in the compositions of the poets; the emphatic language of
Pausanias shows how much he found concerning it in the old epic.[305]

  [303] Herod. i. 146. Pausan. vii. 2, 2.

  [304] Theocrit. xvi. 104.—

    Ὦ ᾿Ετεόκλειοι θύγατρες θεαὶ, αἱ Μινύειον
    ᾿Ορχόμενον φιλέοισαι, ἀπεχθόμενόν ποκα Θήβαις.

  The scholiast gives a sense to these words much narrower than
  they really bear. See Diodôr. xv. 79; Pausan. ix. 15. In the
  oration which Isokratês places in the mouth of a Platæan,
  complaining of the oppressions of Thêbes, the ancient servitude
  and tribute to Orchomenos is cast in the teeth of the Thêbans
  (Isokrat. Orat. Plataic. vol. iii. p. 32, Auger).

  [305] Pausan. ix. 34, 5. See also the fourteenth Olympic Ode
  of Pindar, addressed to the Orchomenian Asopikus. The learned
  and instructive work of K. O. Müller, Orchomenos und die
  Minyer, embodies everything which can be known respecting this
  once-memorable city; indeed the contents of the work extends much
  farther than its title promises.


SECTION II.—DAUGHTERS OF ÆOLUS.

With several of the daughters of Æolus memorable mythical pedigrees and
narratives are connected. Alcyonê married Kêyx, the son of Eôsphoros,
but both she and her husband displayed in a high degree the overweening
insolence common in the Æolic race. The wife called her husband Zeus,
while he addressed her as Hêrê, for which presumptuous act Zeus
punished them by changing both into birds.[306]

  [306] Apollodôr. i. 7, 4. A. Kêyx,—king of Trachin,—the friend
  of Hêraklês and protector of the Hêrakleids to the extent of
  his power (Hesiod, Scut. Hercul. 355-473: Apollodôr. ii. 7, 5;
  Hekatæ. Fragm. 353, Didot.).

Canacê had by the god Poseidôn several children, amongst whom were
Epôpeus and Alôeus.[307] Alôeus married Imphimêdea, who became
enamored of the god Poseidôn, and boasted of her intimacy with him.
She had by him two sons, Otos and Ephialtês, the huge and formidable
Alôids,—Titanic beings, nine fathoms in height and nine cubits in
breadth, even in their boyhood, before they had attained their full
strength. These Alôids defied and insulted the gods in Olympus; they
paid their court to Hêrê and Artemis, and they even seized and bound
Arês, confining him in a brazen chamber for thirteen months. No one
knew where he was, and the intolerable chain would have worn him to
death, had not Eribœa, the jealous stepmother of the Alôids, revealed
the place of his detention to Hermês, who carried him surreptitiously
away when at the last extremity; nor could Arês obtain any atonement
for such an indignity. Otos and Ephialtês even prepared to assault the
gods in heaven, piling up Ossa on Olympus and Pêlion on Ossa, in order
to reach them. And this they would have accomplished had they been
allowed to grow to their full maturity; but the arrows of Apollo put a
timely end to their short-lived career.[308]

  [307] Canacê, daughter of Æolus, is a subject of deep tragical
  interest both in Euripidês and Ovid. The eleventh Heroic Epistle
  of the latter, founded mainly on the lost tragedy of the former
  called Æolus, purports to be from Canacê to Macareus, and
  contains a pathetic description of the ill-fated passion between
  a brother and sister: see the fragments of the Æolus in Dindorf’s
  collection. In the tale of Kaunos and Byblis, both children of
  Milêtos, the results of an incestuous passion are different but
  hardly less melancholy (Parthenios, Narr. xi.).

  Makar, the son of Æolus, is the primitive settler of the island
  of Lesbos (Hom. Hymn. Apoll. 37): moreover in the Odyssey, Æolus
  son of Hippotês, the dispenser of the winds, has six sons and
  six daughters, and marries the former to the latter (Odyss. x.
  7). The two persons called Æolus are brought into connection
  genealogically (see Schol. ad Odyss. _l. c._, and Diodôr. iv.
  67), but it seems probable that Euripidês was the first to place
  the names of Macareus and Canacê in that relation which confers
  upon them their poetical celebrity. Sostratus (ap. Stobæum, t.
  614, p. 404) can hardly be considered to have borrowed from any
  older source than Euripidês. Welcker (Griech. Tragöd. vol. ii. p.
  860) puts together all that can be known respecting the structure
  of the lost drama of Euripidês.

  [308] Iliad, v. 386; Odyss. xi. 306; Apollodôr. i. 7, 4. So
  Typhôeus, in the Hesiodic Theogony, the last enemy of the gods,
  is killed before he comes to maturity (Theog. 837). For the
  different turns given to this ancient Homeric legend, see Heyne,
  ad Apollodôr. l. c, and Hyginus, f. 28. The Alôids were noticed
  in the Hesiodic poems (ap. Schol. Apoll. Rhod. i. 482). Odysseus
  does not see _them_ in Hadês, as Heyne by mistake says; he sees
  their mother Iphimêdea. Virgil (Æn. vi. 582) assigns to them a
  place among the sufferers of punishment in Tartarus.

  Eumêlus, the Corinthian poet, designated Alôeus as son of the god
  Hêlios and brother of Æêtês, the father of Mêdea (Eumêl. Fragm.
  2, Marktscheffel). The scene of their death was subsequently
  laid in Naxos (Pindar, Pyth. iv. 88): their tombs were seen
  at Anthêdôn in Bœôtia (Pausan. ix. 22, 4). The very curious
  legend alluded to by Pausanias from Hegesinoos, the author of
  an Atthis,—to the effect that Otos and Ephialtês were the first
  to establish the worship of the Muses in Helicôn, and that they
  founded Ascra along with Œoklos, the son of Poseidôn,—is one
  which we have no means of tracing farther (Pausan. ix. 29, I).

  The story of the Alôids, as Diodôrus gives it (v. 51, 52),
  diverges on almost every point: it is evidently borrowed from
  some Naxian archæologist, and the only information which we
  collect from it is, that Otos and Ephialtês received heroic
  honors at Naxos. The views of O. Müller (Orchomenos, p. 387)
  appear to me unusually vague and fanciful.

  Ephialtês takes part in the combat of the giants against the gods
  (Apollodôr. t. 6, 2), where Heyne remarks, as in so many other
  cases, “Ephialtês hic non confundendus cum altero Alôei filio;”
  an observation just indeed, if we are supposed to be dealing
  with personages and adventures historically real, but altogether
  misleading in regard to these legendary characters; for here the
  general conception of Ephialtês and his attributes is in both
  cases the same; but the particular adventures ascribed to him
  cannot be made to consist, as facts, one with the other.

The genealogy assigned to Calycê, another daughter of Æolus, conducts
us from Thessaly to Elis and Ætôlia. She married Aëthlius (the son
of Zeus by Prôtogeneia, daughter of Deukaliôn and sister of Hellên),
who conducted a colony out of Thessaly and settled in the territory
of Elis. He had for his son Endymiôn, respecting whom the Hesiodic
Catalogue and the Eoiai related several wonderful things. Zeus granted
him the privilege of determining the hour of his own death, and even
translated him into heaven, which he forfeited by daring to pay court
to Hêrê: his vision in this criminal attempt was cheated by a cloud,
and he was cast out into the under-world.[309] According to other
stories, his great beauty caused the goddess Sêlêne to become enamored
of him, and to visit him by night during his sleep:—the sleep of
Endymiôn became a proverbial expression for enviable, undisturbed, and
deathless repose.[310] Endymiôn had for issue (Pausanias gives us three
different accounts, and Apollodôrus a fourth, of the name of his wife)
Epeios, Ætôlus, Pæôn, and a daughter Eurykydê. He caused his three sons
to run a race on the stadium at Olympia, and Epeios, being victorious,
was rewarded by becoming his successor in the kingdom: it was after him
that the people were denominated Epeians.

  [309] Hesiod, Akusilaus and Pherekydês, ap. Schol. Apollôn. Rhod.
  iv, 57. Ἴν δ᾽ αὐτῷ θανάτου ταμίης. The Scholium is very full
  of matter, and exhibits many of the diversities in the tale of
  Endymiôn: see also Apollodôr i. 7, 5; Pausan. v. 1, 2; Conôn.
  Narr. 14.

  [310] Theocrit. iii. 49; xx. 35; where, however, Endymiôn is
  connected with Latmos in Caria (see Schol. _ad loc._).

Both the story here mentioned, and still more, the etymological
signification of the names Aëthlius and Endymiôn, seem plainly to
indicate (as has before been remarked) that this genealogy was not
devised until after the Olympic games had become celebrated and
notorious throughout Greece.

Epeios had no male issue, and was succeeded by his nephew Eleios, son
of Euykydê by the god Poseidôn: the name of the people was then changed
from Epeians to Eleians. Ætôlus, the brother of Epeios, having slain
Apis, son of Phorôneus, was compelled to flee from the country: he
crossed the Corinthian gulf and settled in the territory then called
Kurêtis, but to which he gave the name of Ætôlia.[311]

  [311] Pausan. v. 1. 3-6; Apollodôr. i. 7, 6.

The son of Eleios,—or, according to other accounts, of the god Hêlios,
of Poseidôn, or of Phorbas,[312]—is Augeas, whom we find mentioned
in the Iliad as king of the Epeians or Eleians. Nestôr gives a long
and circumstantial narrative of his own exploits at the head of his
Pylian countrymen against his neighbors the Epeians and their king
Augeas, whom he defeated with great loss, slaying Mulios, the king’s
son-in-law, and acquiring a vast booty.[313] Augeas was rich in all
sorts of rural wealth, and possessed herds of cattle so numerous, that
the dung of the animals accumulated in the stable or cattle enclosures
beyond all power of endurance. Eurystheus, as an insult to Hêraklês,
imposed upon him the obligation of cleansing this stable: the hero,
disdaining to carry off the dung upon his shoulders, turned the
course of the river Alpheios through the building, and thus swept the
encumbrance away.[314] But Augeas, in spite of so signal a service,
refused to Hêraklês the promised reward, though his son Phyleus
protested against such treachery, and when he found that he could not
induce his father to keep faith, retired in sorrow and wrath to the
island of Dulichiôn.[315] To avenge the deceit practised upon him,
Hêraklês invaded Elis; but Augeas had powerful auxiliaries, especially
his nephews, the two Molionids (sons of Poseidôn by Molionê, the wife
of Aktôr), Eurytos and Kteatos. These two miraculous brothers, of
transcendent force, grew together,—having one body, but two heads and
four arms.[316] Such was their irresistible might, that Hêraklês was
defeated and repelled from Elis: but presently the Eleians sent the two
Molionid brothers as _Theôri_ (sacred envoys) to the Isthmian games,
and Hêraklês, placing himself in ambush at Kleônæ, surprised and killed
them as they passed through. For this murderous act the Eleians in
vain endeavored to obtain redress both at Corinth and at Argos; which
is assigned as the reason for the self-ordained exclusion, prevalent
throughout all the historical age, that no Eleian athlête would ever
present himself as a competitor at the Isthmian games.[317] The
Molionids being thus removed, Hêraklês again invaded Elis, and killed
Augeas along with his children,—all except Phyleus, whom he brought
over from Dulichiôn, and put in possession of his father’s kingdom.
According to the more gentle narrative which Pausanias adopts, Augeas
was not killed, but pardoned at the request of Phyleus.[318] He was
worshipped as a hero[319] even down to the time of that author.

  [312] Apollodôr. ii. 5, 5; Schol. Apoll. Rhod. i. 172. In all
  probability, the old legend made Augeas the son of the god
  Hêlios: Hêlios, Augeas and Agamêdê are a triple series parallel
  to the Corinthian genealogy, Hêlios, Æêtês and Mêdia; not to
  mention that the etymology of Augeas connects him with Hêlios.
  Theocritus (xx. 55) designates him as the son of the god Hêlios,
  through whose favor his cattle are made to prosper and multiply
  with such astonishing success (xx. 117).

  [313] Iliad, xi. 670-760; Pherekyd. Fragm. 57, Didot.

  [314] Diodôr. iv. 13. Ὕβρεως ἕνεκεν Εὐρυσθεὺς προσέταξε καθᾶραι·
  ὁ δὲ Ἡρακλῆς τὸ μὲν τοῖς ὤμοις ἐξενεγκεῖν αὐτὴν ἀπεδοκίμασεν,
  ἐκκλίνων τὴν ἐκ τῆς ὕβρεως αἰσχύνην, etc. (Pausan. v. 1. 7;
  Apollodôr. ii. 5, 5).

  It may not be improper to remark that this fable indicates a
  purely pastoral condition, or at least a singularly rude state
  of agriculture; and the way in which Pausanias recounts it goes
  even beyond the genuine story: ὡς καὶ τὰ πολλὰ τῆς χώρας αὐτῷ ἤδη
  διατελεῖν ἀργὰ ὄντα ὑπὸ τῶν βοσκημάτων τῆς κόπρου. The slaves
  of Odysseus however know what use to make of the dung heaped
  before his outer fence (Odyss. xvii. 299); not so the purely
  carnivorous and pastoral Cyclôps (Odyss. ix. 329). The stabling
  into which the cattle go from their pasture, is called κόπρος in
  Homer,—Ἐλθούσας ἐς κόπρον, ἐπὴν βοτανῆς κορέσωνται (Odyss. x.
  411): compare Iliad, xviii. 575—Μυκηθμῷ δ᾽ ἀπὸ κόπρου ἐπεσσεύοντο
  πέδονδε.

  The Augeas of Theocritus has abundance of wheat-land and
  vineyard, as well as cattle: he ploughs his land three or four
  times, and digs his vineyard diligently (xx. 20-32).

  [315] The wrath and retirement of Phyleus is mentioned in the
  Iliad (ii. 633), but not the cause of it.

  [316] These singular properties were ascribed to them both in
  the Hesiodic poems and by Pherekydês (Schol. Ven. ad II. xi.
  715-750, et ad II. xxiii. 638), but not in the Iliad. The poet
  Ibykus (Fragm. 11, Schneid. ap. Athenæ. ii. 57) calls them ἅλικας
  ἰσοκεφάλους, ἐνιγυίους, Ἀμφοτέρους γεγαῶτας ἐν ὠέῳ ἀργυρέῳ.

  There were temples and divine honors to Zeus Moliôn (Lactantius.
  de Falsâ Religione, i. 22).

  [317] Pausan. v. 2, 4. The inscription cited by Pausanias
  proves that this was the reason assigned by the Eleian athlêtes
  themselves for the exclusion; but there were several different
  stories.

  [318] Apollodôr. ii. 7, 2. Diodôr. iv. 33. Pausan. v. 2, 2; 3,
  2. It seems evident from these accounts that the genuine legend
  represented Hêraklês as having been defeated by the Molionids:
  the unskilful evasions both of Apollodôrus and Diodôrus betray
  this. Pindar (Olymp. xi. 25-50) gives the story without any
  flattery to Hêraklês.

  [319] Pausan. v. 4, 1.

It was on occasion of this conquest of Elis, according to the old mythe
which Pindar has ennobled in a magnificent ode, that Hêraklês first
consecrated the ground of Olympia, and established the Olympic games.
Such at least was one of the many fables respecting the origin of that
memorable institution.[320]

  [320] The Armenian copy of Eusebius gives a different genealogy
  respecting Elis and Pisa: Aëthlius, Epeius, Endymiôn, Alexinus;
  next Œnomaus and Pêlops, then Hêraklês. Some counted ten
  generations, others _three_, between Hêraklês and Iphitus, who
  renewed the discontinued Olympic games (see Armen. Euseb. copy c.
  xxxii. p. 140).

Phyleus, after having restored order in Elis, retired again to
Dulichiôn, and left the kingdom to his brother Agasthenês, which again
brings us into the Homeric series. For Polyxenos, son of Agasthenês, is
one of the four commanders of the Epeian forty ships in the Iliad, in
conjunction with the two sons of Eurytos and Kteatos, and with Diôrês
son of Amarynceus. Megês, the son of Phyleus, commands the contingent
from Dulichiôn and the Echinades.[321] Polyxenos returns safe from
Troy, is succeeded by his son Amphimachos,—named after the Epeian chief
who had fallen before Troy,—and he again by another Eleios, in whose
time the Dôrians and the Hêrakleids invade Peloponnêsus.[322] These two
names, barren of actions or attributes, are probably introduced by the
genealogists whom Pausanias followed, to fill up the supposed interval
between the Trojan war and the Dôrian invasion.

  [321] Iliad, ii. 615-630.

  [322] Pausan. v. 3, 4.

We find the ordinary discrepancies in respect to the series and the
members of this genealogy. Thus some called Epeios son of Aëthlius,
others son of Endymiôn:[323] a third pedigree, which carries the
sanction of Aristotle and is followed by Conôn, designated Eleios,
the first settler of Elis, as son of Poseidôn and Eurypylê, daughter
of Endymiôn, and Epeios and Alexis as the two sons of Eleios.[324]
And Pindar himself, in his ode to Epharmostus the Locrian, introduces
with much emphasis another king of the Epeians named Opus, whose
daughter, pregnant by Zeus, was conveyed by that god to the old and
childless king Locrus: the child when born, adopted by Locrus and named
Opus, became the eponymous hero of the city so called in Locris.[325]
Moreover Hekatæus the Milesian not only affirmed (contrary both to the
Iliad and the Odyssey) that the Epeians and the Eleians were different
people, but also added that the Epeians had assisted Hêraklês in his
expedition against Augeas and Elis; a narrative very different from
that of Apollodôrus and Pausanias, and indicating besides that he must
have had before him a genealogy varying from theirs.[326]

  [323] Schol. Pindar, Olymp. ix. 86.

  [324] Schol. Ven. ad II. xi. 687; Conôn, Narrat. xv. ap. Scriptt.
  Mythogr. West p. 130.

  [325] Pindar, Olymp. ix. 62: Schol. _ibid._ 86. Ὀποῦντος ἠν
  θυγάτηρ Ἠλείων βασιλέως, ἣν Ἀριστοτέλης Καμβύσην καλεῖ.

  [326] Ἑκαταῖος δὲ ὁ Μιλήσιος ἑτέρους λέγει τῶν Ἠλείων τοὺς
  Ἐπείους· τῷ γοῦν Ἡρακλεῖ συστρατεῦσαι τοὺς Ἐπείους καὶ συνανελεῖν
  αὐτῷ τόν τε Αὐγέαν καὶ τὴν Ἦλιν (Hekat. ap. Strab. viii. p. 341).

It has already been mentioned that Ætôlus, son of Endymiôn, quitted
Peloponnêsus in consequence of having slain Apis.[327] The country
on the north of the Corinthian gulf, between the rivers Euênus and
Achelôus, received from him the name of Ætôlia instead of that of
Kurêtis: he acquired possession of it after having slain Dôrus,
Laodokus and Polypœtes, sons of Apollo and Phthia, by whom he had been
well received. He had by his wife Pronoê (the daughter of Phorbas) two
sons, Pleurôn and Kalydôn, and from them the two chief towns in Ætôlia
were named.[328] Pleurôn married Xanthippê, daughter of Dôrus, and
had for his son Agênôr, from whom sprang Portheus, or Porthaôn, and
Demonikê: Euênos and Thestius were children of the latter by the god
Arês.[329]

  [327] Ephorus said that Ætôlus had been expelled by Salmôneus
  king of the Epeians and Pisatæ (ap. Strabo. viii. p. 357):
  he must have had before him a different story and different
  genealogy from that which is given in the text.

  [328] Apollodôr. i. 7, 6. Dôrus, son of Apollo and Phthia,
  killed by Ætôlus, after having hospitably received him, is here
  mentioned. Nothing at all is known of this; but the conjunction
  of names is such as to render it probable that there was some
  legend connected with them: possibly the assistance given by
  Apollo to the Kurêtes against the Ætôlians, and the death of
  Meleager by the hand of Apollo, related both in the Eoiai and the
  Minyas (Pausan. x. 31, 2), may have been grounded upon it. The
  story connects itself with what is stated by Apollodôrus about
  Dôrus son of Hellên (see _supra_, p. 136).

  [329] According to the ancient genealogical poet Asius, Thestius
  was son of Agênôr the son of Pleurôn (Asii Fragm. 6, p. 413,
  ed. Marktsch.). Compare the genealogy of Ætôlia and the general
  remarks upon it, in Brandstäter, Geschichte des Ætol. Landes,
  etc., Berlin, 1844, p. 23 _seq._

Portheus had three sons, Agrius, Melas and Œneus: among the offspring
of Thestius were Althæa and Lêda,[330]—names which bring us to a period
of interest in the legendary history. Lêda marries Tyndareus and
becomes mother of Helena and the Dioskuri: Althæa marries Œneus, and
has, among other children, Meleager and Deianeira; the latter being
begotten by the god Dionysus, and the former by Arês.[331] Tydeus also
is his son, the father of Diomêdês: warlike eminence goes hand in hand
with tragic calamity among the members of this memorable family.

  [330] Respecting Lêda, see the statements of Ibykus, Pherekydês,
  Hellanikus, etc. (Schol. Apollôn. Rhod. i. 146). The reference to
  the Corinthiaca of Eumêlus is curious: it is a specimen of the
  matters upon which these old genealogical poems dwelt.

  [331] Apollodôr. i. 8, 1; Euripidês, Meleager, Frag. 1. The three
  sons of Portheus are named in the Iliad (xiv. 116) as living at
  Pleurôn and Kalydôn. The name Œneus doubtless brings Dionysus
  into the legend.

We are fortunate enough to find the legend of Althæa and Meleager set
forth at considerable length in the Iliad, in the speech addressed by
Phœnix to appease the wrath of Achilles. Œneus, king of Kalydôn, in the
vintage sacrifices which he offered to the gods, omitted to include
Artemis: the misguided man either forgot her or cared not for her;[332]
and the goddess, provoked by such an insult, sent against the vineyards
of Œneus a wild boar, of vast size and strength, who tore up the trees
by the root and laid prostrate all their fruit. So terrible was this
boar, that nothing less than a numerous body of men could venture to
attack him: Meleager, the son of Œneus, however, having got together a
considerable number of companions, partly from the Kurêtes of Pleurôn,
at length slew him. But the anger of Artemis was not yet appeased, and
she raised a dispute among the combatants respecting the possession of
the boar’s head and hide,—the trophies of victory. In this dispute,
Meleager slew the brother of his mother Althæa, prince of the Kurêtes
of Pleurôn: these Kurêtes attacked the Ætôlians of Kalydôn in order
to avenge their chief. So long as Meleager contended in the field
the Ætôlians had the superiority. But he presently refused to come
forth, indignant at the curses imprecated upon him by his mother: for
Althæa, wrung with sorrow for the death of her brother, flung herself
upon the ground in tears, beat the earth violently with her hands,
and implored Hadês and Persephonê to inflict death upon Meleager,—a
prayer which the unrelenting Erinnys in Erebus heard but too well.
So keenly did the hero resent this behavior of his mother, that he
kept aloof from the war; and the Kurêtes not only drove the Ætôlians
from the field, but assailed the walls and gates of Kalydôn, and were
on the point of overwhelming its dismayed inhabitants. There was no
hope of safety except in the arm of Meleager; but Meleager lay in his
chamber by the side of his beautiful wife Kleopatra, the daughter of
Idas, and heeded not the necessity. While the shouts of expected
victory were heard from the assailants at the gates, the ancient men
of Ætôlia and the priests of the gods earnestly besought Meleager to
come forth,[333] offering him his choice of the fattest land in the
plain of Kalydôn. His dearest friends, his father Œneus, his sisters,
and even his mother herself added their supplications, but he remained
inflexible. At length the Kurêtes penetrated into the town and began to
burn it: at this last moment, Kleopatra his wife addressed to him her
pathetic appeal, to avert from her and from his family the desperate
horrors impending over them all. Meleager could no longer resist: he
put on his armor, went forth from his chamber, and repelled the enemy.
But when the danger was over, his countrymen withheld from him the
splendid presents which they had promised, because he had rejected
their prayers, and had come forth only when his own haughty caprice
dictated.[334]

  [332] Ἢ λάθετ᾽, ἢ οὐκ ἐνόησεν· ἀάσατο δὲ μέγα θυμῷ. (Iliad,
  ix. 533). The destructive influence of Atê is mentioned
  before, v. 502. The piety of Xenophôn reproduces this ancient
  circumstance,—Οἴνεως δ᾽ ἐν γήρᾳ ἐπιλαθομένου τῆς θεοῦ, etc. (De
  Venat. c. i.)

  [333] These priests formed the Chorus in the Meleager of
  Sophoklês (Schol. ad Iliad. _ib._ 575).

  [334] Iliad, ix. 525-595.

Such is the legend of Meleager in the Iliad: a verse in the second book
mentions simply the death of Meleager, without farther details, as a
reason why Thoas appeared in command of the Ætôlians before Troy.[335]
Though the circumstance is indicated only indirectly, there seems
little doubt that Homer must have conceived the death of the hero as
brought about by the maternal curse: the unrelenting Erinnys executed
to the letter the invocations of Althæa, though she herself must have
been willing to retract them.

  [335] Iliad, ii. 642.

Later poets both enlarged and altered the fable. The Hesiodic Eoiai,
as well as the old poem called the Minyas, represented Meleager as
having been slain by Apollo, who aided the Kurêtes in the war; and
the incident of the burning brand, though quite at variance with
Homer, is at least as old as the tragic poet Phrynichus, earlier than
Æschylus.[336] The Mœræ, or Fates, presenting themselves to Althæa
shortly after the birth of Meleager, predicted that the child would
die so soon as the brand then burning on the fire near at hand should
be consumed. Althæa snatched it from the flames and extinguished it,
preserving it with the utmost care, until she became incensed against
Meleager for the death of her brother. She then cast it into the fire,
and as soon as it was consumed the life of Meleager was brought to a
close.

  [336] Pausan. x. 31. 2. The Πλευρώνιαι, a lost tragedy of
  Phrynichus.

We know, from the sharp censure of Pliny, that Sophoklês heightened
the pathos of this subject by his account of the mournful death
of Meleager’s sisters, who perished from excess of grief. They
were changed into the birds called Meleagrides, and their
never-ceasing tears ran together into amber.[337] But in the hands
of Euripidês—whether originally through him or not,[338] we cannot
tell—Atalanta became the prominent figure and motive of the piece,
while the party convened to hunt the Kalydônian boar was made to
comprise all the distinguished heroes from every quarter of Greece.
In fact, as Heyne justly remarks, this event is one of the four
aggregate dramas of Grecian heroic life,[339] along with the Argonautic
expedition, the siege of Thêbes, and the Trojan war. To accomplish
the destruction of the terrific animal which Artemis in her wrath had
sent forth, Meleager assembled not merely the choice youth among the
Kurêtes and Ætôlians (as we find in the Iliad), but an illustrious
troop, including Kastôr and Pollux, Idas and Lynkeus, Pêleus and
Telamôn, Thêseus and Peirithous, Ankæus and Kêpheus, Jasôn, Amphiaraus,
Admêtus, Eurytiôn and others. Nestôr and Phœnix, who appear as old men
before the walls of Troy, exhibited their early prowess as auxiliaries
to the suffering Kalydônians.[340] Conspicuous amidst them all stood
the virgin Atalanta, daughter of the Arcadian Schœneus; beautiful
and matchless for swiftness of foot, but living in the forest as a
huntress and unacceptable to Aphroditê.[341] Several of the heroes were
slain by the boar, others escaped by various stratagems: at length
Atalanta first shot him in the back, next Amphiaraus in the eye, and,
lastly, Meleager killed him. Enamoured of the beauty of Atalanta,
Meleager made over to her the chief spoils of the animal, on the plea
that she had inflicted the first wound. But his uncles, the brothers
of Thestius, took them away from her, asserting their rights as next
of kin,[342] if Meleager declined to keep the prize for himself: the
latter, exasperated at this behavior, slew them. Althæa, in deep sorrow
for her brothers and wrath against her son, is impelled to produce the
fatal brand which she had so long treasured up, and consign it to the
flames.[343] The tragedy concludes with the voluntary death both of
Althæa and Kleopatra.

  [337] Plin. H. N. xxxvii. 2, 11.

  [338] There was a tragedy of Æschylus called Ἀταλάντη, of which
  nothing remains (Bothe, Æschyli Fragm. ix. p. 18).

  Of the more recent dramatic writers, several selected Atalanta as
  their subject (See Brandstäter, Geschichte Ætoliens, p. 65).

  [339] There was a poem of Stesichorus, Συόθηραι (Stesichor.
  Fragm. 15. p. 72).

  [340] The catalogue of these heroes is in Apollodôr. i. 8, 2;
  Ovid, Metamor. viii. 300; Hygin. fab. 173. Euripidês, in his play
  of Meleager, gave an enumeration and description of the heroes
  (see Fragm. 6 of that play, ed. Matth.). Nestôr, in this picture
  of Ovid, however, does not appear quite so invincible as in his
  own speeches in the Iliad. The mythographers thought it necessary
  to assign a reason why Hêraklês was _not_ present at the
  Kalydônian adventure: he was just at that time in servitude with
  Omphalê in Lydia (Apollod. ii. 6, 3). This seems to have been the
  idea of Ephorus, and it is much in his style of interpretation
  (see Ephor. Fragm. 9. ed. Didot.).

  [341] Euripid. Meleag. Fragm. vi. Matt.—

    Κύπριδος δὲ μίσημ᾽, Ἀρκὰς Ἀταλάντη, κύνας
    Καὶ τόξ᾽ ἔχουσα, etc.

  There was a drama “Meleager” both of Sophoklês and Euripidês: of
  the former hardly any fragments remain,—a few more of the latter.

  [342] Hyginus, fab. 229.

  [343] Diodôr, iv. 34. Apollôdorus (i. 8; 2-4) gives first the
  usual narrative, including Atalanta; next, the Homeric narrative
  with some additional circumstances, but not including either
  Atalanta or the fire-brand on which Meleager’s life depended. He
  prefaces the latter with the words οἱ δέ φασι, etc. Antoninus
  Liberalis gives this second narrative only, without Atalanta,
  from Nicander (Narrat. 2).

  The Latin scenic poet, Attius, had devoted one of his tragedies
  to this subject, taking the general story as given by Euripidês:
  “Remanet gloria apud me: exuvias dignavi Atalantæ dare,” seems
  to be the speech of Meleager. (Attii Fragm. 8, ap. Poet. Scen.
  Lat. ed. Bothe, p. 215). The readers of the Æneid will naturally
  think of the swift and warlike virgin Camilla, as the parallel of
  Atalanta.

Interesting as the Arcadian huntress, Atalanta, is in herself, she is
an intrusion, and not a very convenient intrusion, into the Homeric
story of the Kalydônian boar-hunt, wherein another female Kleopatra,
already occupied the foreground.[344] But the more recent version
became accredited throughout Greece, and was sustained by evidence
which few persons in those days felt any inclination to controvert.
For Atalanta carried away with her the spoils and head of the boar
into Arcadia; and there for successive centuries hung the identical
hide and the gigantic tusks of three feet in length, in the temple
of Athênê Alea at Tegea. Kallimachus mentions them as being there
preserved, in the third century before the Christian æra;[345] but the
extraordinary value set upon them is best proved by the fact that the
emperor Augustus took away the tusks from Tegea, along with the great
statue of Athênê Alea, and conveyed them to Rome, to be there preserved
among the public curiosities. Even a century and a half afterwards,
when Pausanias visited Greece, the skin worn out with age was shown
to him, while the robbery of the tusks had not been forgotten. Nor
were these relics of the boar the only memento preserved at Tegea of
the heroic enterprise. On the pediment of the temple of Athênê Alea,
unparalleled in Peloponnêsus for beauty and grandeur, the illustrious
statuary Skopas had executed one of his most finished reliefs,
representing the Kalydônian hunt. Atalanta and Meleager were placed in
the front rank of the assailants, and Ankæus, one of the Tegean heroes,
to whom the tusks of the boar had proved fatal,[346] was represented
as sinking under his death-wound into the arms of his brother Epochos.
And Pausanias observes, that the Tegeans, while they had manifested
the same honorable forwardness as other Arcadian communities in the
conquest of Troy, the repulse of Xerxês, and the battle of Dipæ against
Sparta—might fairly claim to themselves, through Ankæus and Atalanta,
that they alone amongst all Arcadians had participated in the glory of
the Kalydônian boar-hunt.[347] So entire and unsuspecting is the faith
both of the Tegeans and of Pausanias in the past historical reality
of this romantic adventure. Strabo indeed tries to transform the
romance into something which has the outward semblance of history, by
remarking that the quarrel respecting the boar’s head and hide cannot
have been the real cause of war between the Kurêtes and the Ætôlians;
the true ground of dispute (he contends) was probably the possession
of a portion of territory.[348] His remarks on this head are analogous
to those of Thucydidês and other critics, when they ascribe the Trojan
war, not to the rape of Helen, but to views of conquest or political
apprehensions. But he treats the general fact of the battle between
the Kurêtes and the Ætôlians, mentioned in the Iliad, as something
unquestionably real and historical—recapitulating at the same time a
variety of discrepancies on the part of different authors, but not
giving any decision of his own respecting their truth or falsehood.

  [344] The narrative of Apollodôrus reads awkwardly—Μελέαγρος
  ἔχων γυναῖκα Κλεοπάτραν, βουλόμενος δὲ καὶ ἐξ Ἀταλάντης
  τεκνοποιήσασθαι, etc. (i. 8, 2).

  [345] Kallimachus, Hymn. ad Dian. 217.—

    Οὔ μιν ἐπικλητοὶ Καλυδώνιοι ἀγρευτῆρες
    Μέμφονται κάπροιο· τὰ γὰρ σημήϊα νίκης
    Ἀρκαδίην εἰσῆλθεν, ἔχει δ᾽ ἔτι θηρὸς ὀδόντας.

  [346] See Pherekyd. Frag. 81, ed. Didot.

  [347] Pausan. viii. 45, 4; 46, 1-3; 47, 2. Lucian, adv. Indoctum,
  c. 14. t. iii. p. 111, Reiz.

  The officers placed in charge of the public curiosities or
  wonders at Rome (οἱ ἐπὶ τοῖς θαύμασιν) affirmed that one of the
  tusks had been accidentally broken in the voyage from Greece: the
  other was kept in the temple of Bacchus in the Imperial Gardens.

  It is numbered among the memorable exploits of Thêseus that he
  vanquished and killed a formidable and gigantic sow, in the
  territory of Krommyôn near Corinth. According to some critics,
  this Krommyônian sow was the mother of the Kalydônian boar
  (Strabo, viii. p. 380).

  [348] Strabo, x. p. 466. Πολέμου δ᾽ ἐμπεσόντος τοῖς Θεστιάδαις
  πρὸς Οἰνέα καὶ Μελέαγρον, ὁ μὲν Ποιητὴς, ἀμφὶ συὸς κεφαλῇ καὶ
  δέρματι, κατὰ τὴν περὶ τοῦ κάπρου μυθολογίαν· ὡς δὲ τὸ εἰκὸς,
  περὶ μέρους τῆς χώρας, etc. This remark is also similar to Mr.
  Payne Knight’s criticism on the true causes of the Trojan war,
  which were (he tells us) of a political character, independent of
  Helen and her abduction (Prolegom. ad Homer. c. 53).

In the same manner as Atalanta was intruded into the Kalydônian hunt,
so also she seems to have been introduced into the memorable funeral
games celebrated after the decease of Pelias at Iôlkos, in which she
had no place at the time when the works on the chest of Kypselus were
executed.[349] But her native and genuine locality is Arcadia; where
her race-course, near to the town of Methydrion, was shown even in
the days of Pausanias.[350] This race-course had been the scene of
destruction for more than one unsuccessful suitor. For Atalanta,
averse to marriage, had proclaimed that her hand should only be won
by the competitor who could surpass her in running: all who tried and
failed were condemned to die, and many were the persons to whom her
beauty and swiftness, alike unparalleled, had proved fatal. At length
Meilaniôn, who had vainly tried to win her affections by assiduous
services in her hunting excursions, ventured to enter the perilous
lists. Aware that he could not hope to outrun her except by stratagem,
he had obtained by the kindness of Aphroditê, three golden apples from
the garden of the Hesperides, which he successively let fall near
to her while engaged in the race. The maiden could not resist the
temptation of picking them up, and was thus overcome: she became the
wife of Meilaniôn and the mother of the Arcadian Parthenopæus, one of
the seven chiefs who perished in the siege of Thêbes.[351]

  [349] Compare Apollodôr. iii. 9, 2, and Pausan. v. 17, 4. She is
  made to _wrestle_ with Pêleus at these funeral games, which seems
  foreign to her character.

  [350] Pausan. viii. 35, 8.

  [351] Respecting the varieties in this interesting story,
  see Apollod. iii. 9, 2; Hygin. f. 185; Ovid, Metam. x.
  560-700; Propert. i. 1, 20; Ælian, V. H. xiii. i. Μειλανίωνος
  σωφρονέστερος. Aristophan. Lysistrat. 786 and Schol. In the
  ancient representation on the chest of Kypselus (Paus. v. 19, 1),
  Meilaniôn was exhibited standing near Atalanta, who was holding a
  fawn: no match or competition in running was indicated.

  There is great discrepancy in the naming and patronymic
  description of the parties in the story. Three different persons
  are announced as fathers of Atalanta, Schœneus, Jasus and
  Mænalos; the successful lover in Ovid (and seemingly in Euripidês
  also) is called Hippomenês, not Meilaniôn. In the Hesiodic poems
  Atalanta was daughter of Schœneus; Hellanikus called her daughter
  of Jasus. See Apollodôr. _l. c._; Kallimach. Hymn to Dian. 214,
  with the note of Spanheim; Schol. Eurip. Phœniss. 150; Schol.
  Theocr. Idyll. iii. 40; also the ample commentary of Bachet de
  Meziriac, Sur les Epîtres d’Ovide, vol. i. p. 366. Servius (ad
  Virg. Eclog. vi. 61; Æneid, iii. 113) calls Atalanta a native of
  Scyros.

  Both the ancient scholiasts (see Schol. Apoll. Rhod. i. 769) and
  the modern commentators, Spanheim and Heyne, seek to escape this
  difficulty by supposing two Atalantas,—an Arcadian and a Bœôtian:
  assuming the principle of their conjecture to be admissible, they
  ought to suppose at least three.

  Certainly, if personages of the Grecian mythes are to be treated
  as historically real, and their adventures as so many exaggerated
  and miscolored facts, it will be necessary to repeat the process
  of multiplying entities to an infinite extent. And this is one
  among the many reasons for rejecting the fundamental supposition.

  But when we consider these personages as purely legendary, so
  that an historical basis can neither be affirmed nor denied
  respecting them, we escape the necessity of such inconvenient
  stratagems. The test of identity is then to be sought in the
  attributes, not in the legal description,—in the predicates, not
  in the subject. Atalanta, whether born of one father or another,
  whether belonging to one place or another, is beautiful, cold,
  repulsive, daring, swift of foot and skilful with the bow,—these
  attributes constitute her identity. The Scholiast on Theocritus
  (iii. 40), in vindicating his supposition that there were two
  Atalantas, draws a distinction founded upon this very principle:
  he says that the Bœôtian Atalanta was τοξοτὶς, and the Arcadian
  Atalanta δρομαία. But this seems an over-refinement: both the
  shooting and the running go to constitute an accomplished
  huntress.

  In respect to Parthenopæus, called by Euripidês and by so many
  others the son of Atalanta, it is of some importance to add,
  that Apollodôrus, Aristarchus, and Antimachus, the author of the
  Thebaid, assigned to him a pedigree entirely different,—making
  him an Argeian, the son of Talaos and Lysimachê, and brother of
  Adrastus. (Apollodôr. i. 9, 13; Aristarch. ap. Schol. Soph. Œd.
  Col. 1320; Antimachus ap. Schol. Æschyl. Sep. Theb. 532; and
  Schol. Supplem. ad Eurip. Phœniss. t. viii. p. 461, ed. Matth.
  Apollodôrus is in fact inconsistent with himself in another
  passage).

We have yet another female in the family of Œneus, whose name the
legend has immortalized. His daughter Deianeira was sought in marriage
by the river Achelôus, who presented himself in various shapes,
first as a serpent and afterwards as a bull. From the importunity of
this hateful suitor she was rescued by the arrival of Hêraklês, who
encountered Achelôus, vanquished him and broke off one of his horns,
which Achelôus ransomed by surrendering to him the horn of Amaltheia,
endued with the miraculous property of supplying the possessor with
abundance of any food or drink which he desired. Hêraklês was rewarded
for his prowess by the possession of Deianeira, and he made over the
horn of Amaltheia as his marriage-present to Œneus.[352] Compelled
to leave the residence of Œneus in consequence of having in a fit of
anger struck the youthful attendant Eunomus, and involuntarily killed
him,[353] Hêraklês retired to Trachin, crossing the river Euênus at the
place where the Centaur Nessus was accustomed to carry over passengers
for hire. Nessus carried over Deianeira, but when he had arrived on
the other side, began to treat her with rudeness, upon which Hêraklês
slew him with an arrow tinged by the poison of the Lernæan hydra. The
dying Centaur advised Deianeira to preserve the poisoned blood which
flowed from his wound, telling her that it would operate as a philtre
to regain for her the affections of Hêraklês, in case she should ever
be threatened by a rival. Some time afterwards the hero saw and loved
the beautiful Iolê, daughter of Eurytos, king of Œchalia: he stormed
the town, killed Eurytos, and made Iolê his captive. The misguided
Deianeira now had recourse to her supposed philtre: she sent as a
present to Hêraklês a splendid tunic, imbued secretly with the poisoned
blood of the Centaur. Hêraklês adorned himself with the tunic on the
occasion of offering a solemn sacrifice to Zeus on the promontory of
Kênæon in Eubœa: but the fatal garment, when once put on, clung to him
indissolubly, burnt his skin and flesh, and occasioned an agony of pain
from which he was only relieved by death. Deianeira slew herself in
despair at this disastrous catastrophe.[354]

  [352] Sophokl. Trachin. 7. The horn of Amaltheia was described by
  Pherekydês (Apollod. ii. 7, 5); see also Strabo, x. p. 458 and
  Diodôr. iv. 35, who cites an interpretation of the fables (οἱ
  εἰκάζοντες ἐξ αὐτῶν τἀληθές) to the effect that it was symbolical
  of an embankment of the unruly river by Hêraklês, and consequent
  recovery of very fertile land.

  [353] Hellanikus (ap. Athen. ix. p. 410) mentioning this
  incident, in two different works, called the attendant by two
  different names.

  [354] The beautiful drama of the Trachiniæ has rendered this
  story familiar: compare Apollod. ii. 7, 7. Hygin. f. 36. Diodôr.
  iv. 36-37.

  The capture of Œchalia (Οἰχαλίας ἅλωσις) was celebrated in a
  very ancient epic poem by Kreophylos, of the Homeric and not of
  the Hesiodic character: it passed with many as the work of Homer
  himself. (See Düntzer, Fragm. Epic. Græcor. p. 8. Welcker, Der
  Epische Cyclus, p. 229). The same subject was also treated in
  the Hesiodic Catalogue, or in the Eoiai (see Hesiod, Fragm. 129,
  ed. Marktsch.): the number of the children of Eurytos was there
  enumerated.

  This exploit seems constantly mentioned as the last performed by
  Hêraklês, and as immediately preceding his death or apotheosis on
  Mount Œta: but whether the legend of Deianeira and the poisoned
  tunic be very old, we cannot tell.

  The tale of the death of Iphitos, son of Eurytos, by Hêraklês, is
  as ancient as the Odyssey (xxi. 19-40): but it is there stated,
  that Eurytos dying left his memorable bow to his son Iphitos (the
  bow is given afterwards by Iphitos to Odysseus, and is the weapon
  so fatal to the suitors),—a statement not very consistent with
  the story that Œchalia was taken and Eurytos slain by Hêraklês.
  It is plain that these were distinct and contradictory legends.
  Compare Soph. Trachin. 260-285 (where Iphitos dies before
  Eurytos), not only with the passage just cited from the Odyssey,
  but also with Pherekydês, Fragm. 34, Didot.

  Hyginus (f. 33) differs altogether in the parentage of Deianeira:
  he calls her daughter of Dexamenos: his account of her marriage
  with Hêraklês is in every respect at variance with Apollodôrus.
  In the latter, Mnêsimachê is the daughter of Dexamenos; Hêraklês
  rescues her from the importunities of the Centaur Eurytiôn (ii.
  5, 5).

We have not yet exhausted the eventful career of Œneus and his
family—ennobled among the Ætôlians especially, both by religious
worship and by poetical eulogy—and favorite themes not merely in some
of the Hesiodic poems, but also in other ancient epic productions,
the Alkmæênis and the Cyclic Thêbais.[355] By another marriage, Œneus
had for his son Tydeus, whose poetical celebrity is attested by the
many different accounts given both of the name and condition of his
mother. Tydeus, having slain his cousins, the sons of Melas, who were
conspiring against Œneus, was forced to become an exile, and took
refuge at Argos with Adrastus, whose daughter Deipylê he married. The
issue of this marriage was Diomêdês, whose brilliant exploits in the
siege of Troy were not less celebrated than those of his father at the
siege of Thêbes. After the departure of Tydeus, Œneus was deposed by
the sons of Agrios, and fell into extreme poverty and wretchedness,
from which he was only rescued by his grandson Diomêdês, after the
conquest of Troy.[356] The sufferings of this ancient warrior, and
the final restoration and revenge by Diomêdês, were the subject of a
lost tragedy of Euripidês, which even the ridicule of Aristophanês
demonstrates to have been eminently pathetic.[357]

  [355] See the references in Apollod. i, 8, 4-5. Pindar, Isthm.
  iv. 32. Μελέταν δὲ σοφισταῖς Διὸς ἕκατι πρόσβαλον σεβιζόμενοι Ἐν
  μὲν Αἰτωλῶν θυσίαισι φαενναῖς Οἰνεΐδαι κρατεροὶ, etc.

  [356] Hekat. Fragm. 341, Didot. In this story Œneus is connected
  with the first discovery of the vine and the making of wine
  (οἶνος): compare Hygin. f. 129, and Servius ad Virgil. Georgic.
  i. 9.

  [357] See Welcker (Griechisch. Tragöd. ii. p. 583) on the lost
  tragedy called Œneus.

Though the genealogy just given of Œneus is in part Homeric, and
seems to have been followed generally by the mythographers, yet
we find another totally at variance with it in Hekatæus, which he
doubtless borrowed from some of the old poets: the simplicity of the
story annexed to it seems to attest its antiquity. Orestheus, son of
Deukaliôn, first passed into Ætôlia, and acquired the kingdom: he
was father of Phytios, who was father of Œneus. Ætôlus was son of
Œneus.[358]

  [358] Timoklês, Comic. ap. Athenæ. vii. p. 223.—

    Γέρων τις ἀτυχεῖ; κατέμαθεν τὸν Οἰνέα.

  Ovid. Heroid. ix. 153.—

    “Heu! devota domus! Solio sedet Agrios alto
      Œnea desertum nuda senecta premit.”

  The account here given is in Hyginus (f. 175): but it is in many
  points different both from Apollodôrus (i. 8, 6; Pausan. ii.
  25) and Pherekydês (Fragm. 83, Didot). It seems to be borrowed
  from the lost tragedy of Euripidês. Compare Schol. ad Aristoph.
  Acharn. 417. Antonin. Liberal. c. 37. In the Iliad, Œneus is dead
  before the Trojan war (ii. 641).

  The account of Ephorus again is different (ap. Strabo. x. p.
  462); he joins Alkmæôn with Diomêdês: but his narrative has the
  air of a tissue of quasi-historical conjectures, intended to
  explain the circumstance that the Ætôlian Diomêdês is king of
  Argos during the Trojan war.

  Pausanias and Apollodôrus affirm that Œneus was buried at Œnoê
  between Argos and Mantineia, and they connect the name of this
  place with him. But it seems more reasonable to consider him as
  the eponymous hero of Œniadæ in Ætôlia.

The original migration of Ætolus from Elis to Ætôlia—and the
subsequent establishment in Elis of Oxylus, his descendant in the
tenth generation, along with the Dôrian invaders of Peloponnêsus—were
commemorated by two inscriptions, one in the agora of Elis, the other
in that of the Ætôlian chief town, Thermum, engraved upon the statues
of Ætôlus and Oxylus,[359] respectively.

  [359] Ephor. Fragm. 29. Didot ap. Strab. x.




CHAPTER VII.

THE PELOPIDS.


Among the ancient legendary genealogies, there was none which figured
with greater splendor, or which attracted to itself a higher degree
of poetical interest and pathos, than that of the Pelopids—Tantalus,
Pelops, Atreus and Thyestês, Agamemnôn and Menelaus and Ægisthus, Helen
and Klytæmnêstra, Orestês and Elektra and Hermionê. Each of these
characters is a star of the first magnitude in the Grecian hemisphere:
each name suggests the idea of some interesting romance or some
harrowing tragedy: the curse which taints the family from the beginning
inflicts multiplied wounds at every successive generation. So, at
least, the story of the Pelopids presents itself, after it had been
successively expanded and decorated by epic, lyric and tragic poets. It
will be sufficient to touch briefly upon events with which every reader
of Grecian poetry is more or less familiar, and to offer some remarks
upon the way in which they were colored and modified by different
Grecian authors.

Pelops is the eponym or name-giver of the Peloponnêsus: to find an
eponym for every conspicuous local name was the invariable turn of
Grecian retrospective fancy. The name Peloponnêsus is not to be found
either in the Iliad or the Odyssey, nor any other denomination which
can be attached distinctly and specially to the entire peninsula. But
we meet with the name in one of the most ancient post-Homeric poems of
which any fragments have been preserved—the Cyprian Verses—a poem which
many (seemingly most persons) even of the contemporaries of Herodotus
ascribed to the author of the Iliad, though Herodotus contradicts the
opinion.[360] The attributes by which the Pelopid Agamemnôn and his
house are marked out and distinguished from the other heroes of the
Iliad, are precisely those which Grecian imagination would naturally
seek in an eponymus—superior wealth, power, splendor and regality.
Not only Agamemnôn himself, but his brother Menelaus, is “more of a
king” even than Nestôr or Diomêdês. The gods have not given to the
king of the “much-golden” Mykênæ greater courage, or strength, or
ability, than to various other chiefs; but they have conferred upon
him a marked superiority in riches, power and dignity, and have thus
singled him out as the appropriate leader of the forces.[361] He enjoys
this preëminence as belonging to a privileged family and as inheriting
the heaven-descended sceptre of Pelops, the transmission of which is
described by Homer in a very remarkable way. The sceptre was made “by
Hêphæstos, who presented it to Zeus; Zeus gave it to Hermês, Hermês
to the charioteer Pelops; Pelops gave it to Atreus, the ruler of
men; Atreus at his death left it to Thyestês, the rich cattle-owner;
Thyestês in his turn left it to his nephew Agamemnôn to carry, that he
might hold dominion over many islands and over all Argos.”[362]

  [360] Hesiod. ii. 117. Fragment. Epicc. Græc. Düntzer, ix.
  Κύπρια, 8.—

                Αἶψα τε Λυγκεὺς
    Ταΰγετον προσέβαινε ποσὶν ταχέεσσι πεποιθὼς,
    Ἀκρότατον δ᾽ ἀναβὰς διεδέρκετο νῆσον ἅπασαν
    Τανταλίδεω Πέλοπος.

  Also the Homeric Hymn. Apoll. 419, 430, and Tyrtæus, Fragm. 1.—

    (~Εὐνομία~)—Εὐρεῖαν Πέλοπος νῆσον ἀφικόμεθα.

  The Schol. ad Iliad, ix. 246, intimates that the name
  Πελοπόννησος occurred in one or more of the Hesiodic epics.

  [361] Iliad, ix. 37. Compare ii. 580. Diomêdês addresses
  Agamemnôn—

    Σοὶ δὲ διάνδιχα δῶκε Κρόνου παῖς ἀγκυλομήτεω·
    Σκήπτρῳ μέν τοι δῶκε τετιμῆσθαι περὶ πάντων·
    Ἀλκὴν δ᾽ οὔ τοι δῶκεν, ὅ,τε κράτος ἐστὶ μέγιστον.

  A similar contrast is drawn by Nestôr (Il. i. 280) between
  Agamemnôn and Achilles. Nestôr says to Agamemnôn (Il. ix. 69)—

    Ἀτρείδη, σὺ μὲν ἄρχε· σὺ γὰρ ~βασιλεύτατός~ ἐσσι.

  And this attribute attaches to Menelaus as well as to his
  brother. For when Diomêdês is about to choose his companion
  for the night expedition into the Trojan camp, Agamemnôn thus
  addresses him (x. 232):

    Τὸν μὲν δὴ ἕταρόν γ᾽ αἱρήσεαι, ὅν κ᾽ ἐθέλῃσθα
    Φαινομένων τὸν ἄριστον, ἐπεὶ μεμάασί γε πολλοί·
    Μηδὲ σύ γ᾽ αἰδόμενος σῇσι φρεσὶ, τὸν μὲν ἀρείω
    Καλλείπειν, σὺ δὲ χείρον᾽ ὀπάσσεαι αἰδοῖ εἴκων
    Ἐς γενεὴν ὁρόων, εἰ καὶ βασιλεύτερός ἐστιν.
    Ὡς ἔφατ᾽, ἔδδεισε δὲ περὶ ξανθῷ Μενελάῳ.

  [362] Iliad, ii. 101.

We have here the unrivalled wealth and power of the “king of men,
Agamemnôn,” traced up to his descent from Pelops, and accounted for,
in harmony with the recognized epical agencies, by the present of
the special sceptre of Zeus through the hands of Hermês; the latter
being the wealth-giving god, whose blessing is most efficacious
in furthering the process of acquisition, whether by theft or by
accelerated multiplication of flocks and herds.[363] The wealth and
princely character of the Atreids were proverbial among the ancient
epic poets. Paris not only carries away Hellen, but much property along
with her:[364] the house of Menelaus, when Têlemachus visits it in the
Odyssey, is so resplendent with gold and silver and rare ornament,[365]
as to strike the beholder with astonishment and admiration. The
attributes assigned to Tantalus, the father of Pelops, are in
conformity with the general idea of the family—superhuman abundance
and enjoyments, and intimate converse with the gods, to such a degree
that his head is turned, and he commits inexpiable sin. But though
Tantalus himself is mentioned, in one of the most suspicious passages
of the Odyssey (as suffering punishment in the under-world), he is
not announced, nor is any one else announced, as father of Pelops,
unless we are to construe the lines in the Iliad as implying that the
latter was son of Hermês. In the conception of the author of the Iliad,
the Pelopids are, if not of divine origin, at least a mortal breed
specially favored and ennobled by the gods—beginning with Pelops, and
localized at Mykênæ. No allusion is made to any connection of Pelops
either with Pisa or with Lydia.

  [363] Iliad, xiv. 491. Hesiod. Theog. 444. Homer, Hymn. Mercur.
  526-568, Ὄλβου καὶ πλούτου δώσω περικάλλεα ῥάβδον. Compare
  Eustath. ad Iliad. xvi. 182.

  [364] Iliad, iii. 72; vii. 363. In the Hesiodic Eoiai was the
  following couplet (Fragm. 55. p. 43, Düntzer):—

    Ἁλκὴν μὲν γὰρ ἔδωκεν Ὀλύμπιος Αἰακίδῃσιν,
    Νοῦν δ᾽ Ἀμυθαονίδαις, πλοῦτον δ᾽ ἔπορ᾽ Ἀτρείδῃσι.

  Again, Tyrtæus, Fragm. 9, 4.—

    Οὐδ᾽ εἰ Τανταλίδεω Πέλοπος βασιλεύτερος εἴη, etc.

  [365] Odyss. iv. 45-71.

The legend which connected Tantalus and Pelops with Mount Sipylus
may probably have grown out of the Æolic settlements at Magnêsia and
Kymê. Both the Lydian origin and the Pisatic sovereignty of Pelops
are adapted to times later than the Iliad, when the Olympic games had
acquired to themselves the general reverence of Greece, and had come to
serve as the religious and recreative centre of the Peloponnêsus—and
when the Lydian and Phrygian heroic names, Midas and Gygês, were the
types of wealth and luxury, as well as of chariot driving, in the
imagination of a Greek. The inconsiderable villages of the Pisatid
derived their whole importance from the vicinity of Olympia: they are
not deemed worthy of notice in the Catalogue of Homer. Nor could the
genealogy which connected the eponym of the entire peninsula with
Pisa have obtained currency in Greece unless it had been sustained
by preëstablished veneration for the locality of Olympia. But if the
sovereign of the humble Pisa was to be recognized as forerunner of the
thrice-wealthy princes of Mykênæ, it became necessary to assign some
explanatory cause of his riches. Hence the supposition of his being
an immigrant, son of a wealthy Lydian named Tantalus, who was the
offspring of Zeus and Ploutô. Lydian wealth and Lydian chariot-driving
rendered Pelops a fit person to occupy his place in the legend, both
as ruler of Pisa and progenitor of the Mykenæan Atreids. Even with the
admission of these two circumstances there is considerable difficulty,
for those who wish to read the legends as consecutive history, in
making the Pelopids pass smoothly and plausibly from Pisa to Mykênæ.

I shall briefly recount the legends of this great heroic family as they
came to stand in their full and ultimate growth, after the localization
of Pelops at Pisa had been tacked on as a preface to Homer’s version of
the Pelopid genealogy.

Tantalus, residing near Mount Sipylus in Lydia, had two children,
Pelops and Niobê. He was a man of immense possessions and preëminent
happiness, above the lot of humanity: the gods communicated with him
freely, received him at their banquets, and accepted of his hospitality
in return. Intoxicated with such prosperity, Tantalus became guilty of
gross wickedness. He stole nectar and ambrosia from the table of the
gods, and revealed their secrets to mankind: he killed and served up to
them at a feast his own son Pelops. The gods were horror-struck when
they discovered the meal prepared for them: Zeus restored the mangled
youth to life, and as Dêmêtêr, then absorbed in grief for the loss
of her daughter Persephonê, had eaten a portion of the shoulder, he
supplied an ivory shoulder in place of it. Tantalus expiated his guilt
by exemplary punishment. He was placed in the under-world, with fruit
and water seemingly close to him, yet eluding his touch as often as
he tried to grasp them and leaving his hunger and thirst incessant and
unappeased.[366] Pindar, in a very remarkable passage, finds this old
legend revolting to his feelings: he rejects the tale of the flesh of
Pelops having been served up and eaten, as altogether unworthy of the
gods.[367]

  [366] Diodôr. iv. 77. Hom. Odyss. xi. 582. Pindar gives a
  different version of the punishment inflicted on Tantalus: a vast
  stone was perpetually impending over his head, and threatening to
  fall (Olymp. i. 56; Isthm. vii. 20).

  [367] Pindar, Olymp. i. 45. Compare the sentiment of Iphigeneia
  in Euripidês, Iph. Taur. 387.

Niobê, the daughter of Tantalus, was married to Amphiôn, and had a
numerous and flourishing offspring of seven sons and seven daughters.
Though accepted as the intimate friend and companion of Lêtô, the
mother of Apollo and Artemas,[368] she was presumptuous enough to
triumph over that goddess, and to place herself on a footing of higher
dignity, on account of the superior number of her children. Apollo
and Artemas avenged this insult by killing all the sons and all the
daughters: Niobê, thus left a childless and disconsolate mother, wept
herself to death, and was turned into a rock, which the later Greeks
continued always to identify on Mount Sipylus.[369]

  [368] Sapphô (Fragm. 82, Schneidewin)—

    Λατὼ καὶ Νιόβα μάλα μὲν φίλαι ἦσαν ἑταῖραι.

  Sapphô assigned to Niobê eighteen children (Aul. Gell. N. A. iv.
  Δ. xx. 7); Hesiod gave twenty; Homer twelve (Apollod. iii. 5).

  The Lydian historian Xanthus gave a totally different version
  both of the genealogy and of the misfortunes of Niobê (Parthen.
  Narr. 33).

  [369] Ovid, Metam. vi. 164-311. Pausan. i. 21, 5; viii. 2, 3.

Some authors represented Pelops as not being a Lydian, but a king
of Paphlagônia; by others it was said that Tantalus, having become
detested from his impieties, had been expelled from Asia by Ilus the
king of Troy,—an incident which served the double purpose of explaining
the transit of Pelops to Greece, and of imparting to the siege of
Troy by Agamemnôn the character of retribution for wrongs done to his
ancestor.[370] When Pelops came over to Greece, he found Œnomaus,
son of the god Arês and Harpinna, in possession of the principality
of Pisa, immediately bordering on the district of Olympia. Œnomaus,
having been apprized by an oracle that death would overtake him if he
permitted his daughter Hippodameia to marry, refused to give her in
marriage except to some suitor who should beat him in a chariot-race
from Olympia to the isthmus of Corinth;[371] the ground here selected
for the legendary victory of Pelops deserves attention, inasmuch as
it is a line drawn from the assumed centre of Peloponnêsus to its
extremity, and thus comprises the whole territory with which Pelops
is connected as eponym. Any suitor overmatched in the race was doomed
to forfeit his life; and the fleetness of the Pisan horses, combined
with the skill of the charioteer Myrtilus, had already caused thirteen
unsuccessful competitors to perish by the lance of Œnomaus.[372] Pelops
entered the lists as a suitor: his prayers moved the god Poseidôn to
supply him with a golden chariot and winged horses; or according to
another story, he captivated the affections of Hippodameia herself,
who persuaded the charioteer Myrtilus to loosen the wheels of Œnomaus
before he started, so that the latter was overturned and perished in
the race. Having thus won the hand of Hippodameia, Pelops became Prince
of Pisa.[373] He put to death the charioteer Myrtilus, either from
indignation at his treachery to Œnomaus,[374] or from jealousy on the
score of Hippodameia: but Myrtilus was the son of Hermês, and though
Pelops erected a temple in the vain attempt to propitiate that god,
he left a curse upon his race which future calamities were destined
painfully to work out.[375]

  [370] Apollôn. Rhod. ii. 358, and Schol.; Ister. Fragment. 59,
  Dindorf; Diodôr. iv. 74.

  [371] Diodôr. iv. 74.

  [372] Pausanias (vi. 21, 7) had read their names in the Hesiodic
  Eoiai.

  [373] Pindar, Olymp. i. 140. The chariot race of Pelops and
  Œnomaus was represented on the chest of Kypselus at Olympia: the
  horses of the former were given as having wings (Pausan. v. 17,
  4). Pherekydês gave the same story (ap. Schol. ad Soph. Elect.
  504).

  [374] It is noted by Herodotus and others as a remarkable fact,
  that no mules were ever bred in the Eleian territory: an Eleian
  who wished to breed a mule sent his mare for the time out of
  the region. The Eleians themselves ascribed this phænomenon to
  a disability brought on the land by a curse from the lips of
  Œnomaus (Herod. iv. 30; Plutarch, Quæst. Græc. p. 303).

  [375] Paus. v. 1, 1; Sophok. Elektr. 508; Eurip. Orest. 985, with
  Schol., Plato, Kratyl. p. 395.

Pelops had a numerous issue by Hippodameia: Pittheus, Trœzen and
Epidaurus, the eponyms of the two Argolic cities so called, are said
to have been among them: Atreus and Thyestês were also his sons, and
his daughter Nikippê married Sthenelus of Mykênæ, and became the
mother of Eurystheus.[376] We hear nothing of the principality of
Pisa afterwards: the Pisatid villages became absorbed into the larger
aggregate of Elis, after a vain struggle to maintain their separate
right of presidency over the Olympic festival. But the legend ran that
Pelops left his name to the whole peninsula: according to Thucydidês,
he was enabled to do this because of the great wealth which he had
brought with him from Lydia into a poor territory. The historian leaves
out all the romantic interest of the genuine legends—preserving only
this one circumstance, which, without being better attested than the
rest, carries with it, from its commonplace and prosaic character, a
pretended historical plausibility.[377]

  [376] Apollod. ii. 4, 5. Pausan. ii. 30, 8; 26, 3; v. 8, 1.
  Hesiod. ap. Schol. ad Iliad. xx. 116.

  [377] Thucyd. i. 5.

Besides his numerous issue by Hippodameia, Pelops had an illegitimate
son named Chrysippus, of singular grace and beauty, towards whom he
displayed so much affection as to rouse the jealousy of Hippodameia
and her sons. Atreus and Thyestês conspired together to put Chrysippus
to death, for which they were banished by Pelops and retired to
Mykênæ,[378]—an event which brings us into the track of the Homeric
legend. For Thucydidês, having found in the death of Chrysippus a
suitable ground for the secession of Atreus from Pelops, conducts him
at once to Mykênæ, and shows a train of plausible circumstances to
account for his having mounted the throne. Eurystheus, king of Mykênæ,
was the maternal nephew of Atreus: when he engaged in any foreign
expedition, he naturally entrusted the regency to his uncle; the
people of Mykênæ thus became accustomed to be governed by him, and he
on his part made efforts to conciliate them, so that when Eurystheus
was defeated and slain in Attica, the Mykênæan people, apprehensive
of an invasion from the Hêrakleids, chose Atreus as at once the most
powerful and most acceptable person for his successor.[379] Such
was the tale which Thucydidês derived “from those who had learnt
ancient Peloponnêsian matters most clearly from their forefathers.”
The introduction of so much sober and quasi-political history,
unfortunately unauthenticated, contrasts strikingly with the highly
poetical legends of Pelops and Atreus, which precede and follow it.

  [378] We find two distinct legends respecting Chrysippus: his
  abduction by Laius king of Thêbes, on which the lost drama
  of Euripidês called Chrysippus turned (see Welcker, Griech.
  Tragödien, ii. p. 536), and his death by the hands of his
  half-brothers. Hyginus (f. 85) blends the two together.

  [379] Thucyd. i. 9. λέγουσι δὲ οἱ τὰ Πελοποννησίων σαφέστατα
  μνήμῃ παρὰ τῶν πρότερον δεδεγμένοι. According to Hellanikus,
  Atreus the elder son returns to Pisa after the death of Pelops
  with a great army, and makes himself master of his father’s
  principality (Hellanik. ap Schol. ad Iliad, ii. 105). Hellanikus
  does not seem to have been so solicitous as Thucydidês to
  bring the story into conformity with Homer. The circumstantial
  genealogy given in Schol. ad Eurip. Orest. 5. makes Atreus and
  Thyestês reside during their banishment at Makestus in Triphylia:
  it is given without any special authority, but may perhaps come
  from Hellanikus.

Atreus and Thyestês are known in the Iliad only as successive
possessors of the sceptre of Zeus, which Thyestês at his death
bequeathes to Agamemnôn. The family dissensions among this fated
race commence, in the Odyssey, with Agamemnôn the son of Atreus, and
Ægisthus the son of Thyestês. But subsequent poets dwelt upon an
implacable quarrel between the two fathers. The cause of the bitterness
was differently represented: some alleged that Thyestês had intrigued
with the Krêtan Aeropê, the wife of his brother; other narratives
mentioned that Thyestês procured for himself surreptitiously the
possession of a lamb with a golden fleece, which had been designedly
introduced among the flocks of Atreus by the anger of Hermês, as a
cause of enmity and ruin to the whole family.[380] Atreus, after a
violent burst of indignation, pretended to be reconciled, and invited
Thyestês to a banquet, in which he served up to him the limbs of his
own son, and the father ignorantly partook of the fatal meal. Even the
all-seeing Hêlios is said to have turned back his chariot to the east
in order that he might escape the shocking spectacle of this Thyestêan
banquet: yet the tale of Thyestêan revenge—the murder of Atreus
perpetrated by Ægisthus, the incestuous offspring of Thyestês by his
daughter Pelopia—is no less replete with horrors.[381]

  [380] Æschyl. Agamem. 1204, 1253, 1608; Hygin. 86; Attii Fragm.
  19. This was the story of the old poem entitled Alkmæônis;
  seemingly also of Pherekydês, though the latter rejected the
  story that Hermês had produced the golden lamb with the special
  view of exciting discord between the two brothers, in order to
  avenge the death of Myrtilus by Pelops (see Schol. ad Eurip.
  Orest. 996).

  A different legend, alluded to in Soph. Aj. 1295 (see Schol. _ad
  loc._), recounted that Aeropê had been detected by her father
  Katreus in unchaste commerce with a low-born person; he entrusted
  her in his anger to Nauplius, with directions to throw her into
  the sea: Nauplius however not only spared her life, but betrothed
  her to Pleisthenês, father of Agamemnôn and son of Atreus.

  The tragedy entitled _Atreus_ of the Latin poet Attius, seems to
  have brought out, with painful fidelity, the harsh and savage
  features of this family legend (see Aul. Gell. xiii. 2, and the
  fragments of Attius now remaining, together with the tragedy
  called Thyestês, of Seneca).

  [381] Hygin. fab. 87-88.

Homeric legend is never thus revolting. Agamemnôn and Menelaus are
known to us chiefly with their Homeric attributes, which have not been
so darkly overlaid by subsequent poets as those of Atreus and Thyestês.
Agamemnôn and Menelaus are affectionate brothers: they marry two
sisters, the daughters of Tyndareus king of Sparta, Klytæmnêstra and
Helen; for Helen, the real offspring of Zeus, passes as the daughter of
Tyndareus.[382] The “king of men” reigns at Mykênæ; Menelaus succeeds
Tyndareus at Sparta. Of the rape of Helen, and the siege of Troy
consequent upon it, I shall speak elsewhere: I now touch only upon
the family legends of the Atreids. Menelaus, on his return from Troy
with the recovered Helen, is driven by storms far away to the distant
regions of Phœnicia and Egypt, and is exposed to a thousand dangers
and hardships before he again sets foot in Peloponnêsus. But at length
he reaches Sparta, resumes his kingdom, and passes the rest of his
days in uninterrupted happiness and splendor: being moreover husband
of the godlike Helen and son-in-law of Zeus, he is even spared the
pangs of death. When the fulness of his days is past he is transported
to the Elysian fields, there to dwell along with “the golden-haired
Rhadamanthus” in a delicious climate and in undisturbed repose.[383]

  [382] So we must say, in conformity to the ideas of antiquity:
  compare Homer, Iliad, xvi. 176 and Herodot. vi. 53.

  [383] Hom. Odyss. iii. 280-300; iv. 83-560.

Far different is the fate of the king of men, Agamemnôn. During his
absence, the unwarlike Ægisthus, son of Thyestês, had seduced his
wife Klytæmnêstra, in spite of the special warning of the gods, who,
watchful over this privileged family, had sent their messenger Hermês
expressly to deter him from the attempt.[384] A venerable bard had been
left by Agamemnôn as the companion and monitor of his wife, and so long
as that guardian was at hand, Ægisthus pressed his suit in vain. But he
got rid of the bard by sending him to perish in a desert island, and
then won without difficulty the undefended Klytæmnêstra. Ignorant of
what had passed, Agamemnôn returned from Troy victorious and full of
hope to his native country; but he had scarcely landed when Ægisthus
invited him to a banquet, and there with the aid of the treacherous
Klytæmnêstra, in the very hall of festivity and congratulation,
slaughtered him and his companions “like oxen tied to the manger.” His
concubine Kassandra, the prophetic daughter of Priam, perished along
with him by the hand of Klytæmnêstra herself.[385] The boy Orestês, the
only male offspring of Agamemnôn, was stolen away by his nurse, and
placed in safety at the residence of the Phokian Strophius.

  [384] Odyss. i. 38; iii. 310.—ἀνάλκιδος Αἰγίσθοιο.

  [385] Odyss. iii. 260-275; iv. 512-537; xi. 408. Deinias in his
  Argolica, and other historians of that territory, fixed the
  precise day of the murder of Agamemnôn,—the thirteenth of the
  month Gamêliôn (Schol. ad Sophokl. Elektr. 275).

For seven years Ægisthus and Klytæmnêstra reigned in tranquillity at
Mykênæ on the throne of the murdered Agamemnôn. But in the eighth year
the retribution announced by the gods overtook them: Orestês, grown to
manhood, returned and avenged his father by killing Ægisthus, according
to Homer; subsequent poets add, his mother also. He recovered the
kingdom of Mykênæ, and succeeded Menelaus in that of Sparta. Hermionê,
the only daughter of Menelaus and Helen, was sent into the realm of the
Myrmidons in Thessaly, as the bride of Neoptolemus, son of Achilles,
according to the promise made by her father during the siege of
Troy.[386]

  [386] Odyss. iii. 306; iv. 9

  Here ends the Homeric legend of the Pelopids, the final act
  of Orestês being cited as one of unexampled glory.[387] Later
  poets made many additions: they dwelt upon his remorse and
  hardly-earned pardon for the murder of his mother, and upon his
  devoted friendship for Pylades; they wove many interesting tales,
  too, respecting his sisters Iphigeneia and Elektra and his cousin
  Hermionê,—names which have become naturalized in every climate
  and incorporated with every form of poetry.

  [387] Odyss. i. 299.

These poets did not at all scruple to depart from Homer, and to give
other genealogies of their own, with respect to the chief persons
of the Pelopid family. In the Iliad and Odyssey, Agamemnôn is son
of Atreus: in the Hesiodic Eoiai and in Stesichorus, he is son of
Pleisthenês the son of Atreus.[388] In Homer, he is specially marked
as reigning at Mykênæ; but Stesichorus, Simonidês and Pindar[389]
represented him as having both resided and perished at Sparta or at
Amyklæ. According to the ancient Cyprian Verses, Helen was represented
as the daughter of Zeus and Nemesis: in one of the Hesiodic poems
she was introduced as an Oceanic nymph, daughter of Oceanus and
Têthys.[390] The genealogical discrepancies, even as to the persons of
the principal heroes and heroines, are far too numerous to be cited,
nor is it necessary to advert to them, except as they bear upon the
unavailing attempt to convert such legendary parentage into a basis of
historical record or chronological calculation.

  [388] Hesiod. Fragm. 60. p. 44, ed. Düntzer; Stesichor. Fragm.
  44, Kleine. The Scholiast ad Soph. Elektr. 539, in reference to
  another discrepancy between Homer and the Hesiodic poems about
  the children of Helen, remarks that we ought not to divert our
  attention from that which is moral and salutary to ourselves in
  the poets (τὰ ἠθικὰ καὶ χρήσιμα ἡμῖν τοῖς ἐντυγχάνουσι), in order
  to cavil at their genealogical contradictions.

  Welcker in vain endeavors to show that Pleisthenês was originally
  introduced as the father of Atreus, not as his son (Griech.
  Tragöd. p. 678).

  [389] Schol. ad Eurip. Orest. 46. Ὅμηρος ἐν Μυκήναις φησὶ τὰ
  βασιλεῖα τοῦ Ἀγαμέμνονος· Στησίχορος δὲ καὶ Σιμωνίδης, ἐν
  Λακεδαιμονίᾳ. Pindar, Pyth. xi. 31; Nem. viii. 21. Stêsichorus
  had composed an Ὀρέστεια, copied in many points from a still more
  ancient lyric Oresteia by Xanthus: compare Athen. xii. p. 513,
  and Ælian, V. H. iv. 26.

  [390] Hesiod, ap. Schol. ad Pindar, Nem. x. 150.

The Homeric poems probably represent that form of the legend,
respecting Agamemnôn and Orestês, which was current and popular among
the Æolic colonists. Orestês was the great heroic chief of the Æolic
emigration; he, or his sons, or his descendants, are supposed to have
conducted the Achæans to seek a new home, when they were no longer
able to make head against the invading Dôrians: the great families at
Tenedos and other Æolic cities even during the historical æra, gloried
in tracing back their pedigrees to this illustrious source.[391] The
legends connected with the heroic worship of these mythical ancestors
form the basis of the character and attributes of Agamemnôn and his
family, as depicted in Homer, in which Mykênæ appears as the first
place in Peloponnêsus, and Sparta only as the second: the former the
special residence of “the king of men;” the latter that of his younger
and inferior brother, yet still the seat of a member of the princely
Pelopids, and moreover the birth-place of the divine Helen. Sparta,
Argos and Mykênæ are all three designated in the Iliad by the goddess
Hêrê as her favorite cities;[392] yet the connection of Mykênæ with
Argos, though the two towns were only ten miles distant, is far less
intimate than the connection of Mykênæ with Sparta. When we reflect
upon the very peculiar manner in which Homer identifies Hêrê with the
Grecian host and its leader,—for she watches over the Greeks with the
active solicitude of a mother, and her antipathy against the Trojans
is implacable to a degree which Zeus cannot comprehend,[393]—and when
we combine this with the ancient and venerated Hêræon, or temple of
Hêrê, near Mykênæ, we may partly explain to ourselves the preëminence
conferred upon Mykênæ in the Iliad and Odyssey. The Hêræon was situated
between Argos and Mykênæ; in later times its priestesses were named and
its affairs administered by the Argeians: but as it was much nearer
to Mykênæ than to Argos, we may with probability conclude that it
originally belonged to the former, and that the increasing power of
the latter enabled them to usurp to themselves a religious privilege
which was always an object of envy and contention among the Grecian
communities. The Æolic colonists doubtless took out with them in their
emigration the divine and heroic legends, as well as the worship and
ceremonial rites, of the Hêræon; and in those legends the most exalted
rank would be assigned to the close-adjoining and administering city.

  [391] See the ode of Pindar addressed to Aristagoras of Tenedos
  (Nem. xi. 35; Strabo, xiii. p. 582). There were Penthilids at
  Mitylênê, from Penthilus, son of Orestês (Aristot. Polit v. 8,
  13, Schneid.).

  [392] Iliad, iv. 52. Compare Euripid. Hêrakleid. 350

  [393] Iliad, iv. 31. Zeus says to Hêrê,—

    Δαιμονίη, τί νύ σε Πρίαμος, Πριάμοιό τε παῖδες
    Τόσσα κακὰ ῥέζεσκον ὅτ᾽ ἀσπερχὲς μενεαίνεις
    Ἰλίου ἐξαλάπαξαι ἐϋκτίμενον πτολίεθρον;
    Εἰ δὲ σύ γ᾽, εἰσελθοῦσα πύλας καὶ τείχεα μακρὰ,
    Ὠμὸν βεβρώθοις Πρίαμον Πριάμοιό τε παῖδας,
    Ἄλλους τε Τρῶας, τότε κεν χόλον ἐξακέσαιο.

  Again, xviii. 358,—

                    ἦ ῥά νυ σεῖο
    Ἐξ αὐτῆς ἐγένοντο καρηκομόωντες Ἀχαιοί.

Mykênæ maintained its independence even down to the Persian invasion.
Eighty of its heavy-armed citizens, in the ranks of Leonidas at
Thermopylæ, and a number not inferior at Platæa, upheld the splendid
heroic celebrity of their city during a season of peril, when the more
powerful Argos disgraced itself by a treacherous neutrality. Very
shortly afterwards Mykênæ was enslaved and its inhabitants expelled by
the Argeians. Though this city so long maintained a separate existence,
its importance had latterly sunk to nothing, while that of the Dôrian
Argos was augmented very much, and that of the Dôrian Sparta still more.

The name of Mykênæ is imperishably enthroned in the Iliad and Odyssey;
but all the subsequent fluctuations of the legend tend to exalt
the glory of other cities at its expense. The recognition of the
Olympic games as the grand religious festival of Peloponnêsus gave
vogue to that genealogy which connected Pelops with Pisa or Elis and
withdrew him from Mykênæ. Moreover, in the poems of the great Athenian
tragedians, Mykênæ is constantly confounded and treated as one with
Argos. If any one of the citizens of the former, expelled at the time
of its final subjugation by the Argeians, had witnessed at Athens a
drama of Æschylus, Sophoklês, or Euripidês, or the recital of an ode
of Pindar, he would have heard with grief and indignation the city of
his oppressors made a partner in the heroic glories of his own.[394]
But the great political ascendency acquired by Sparta contributed still
farther to degrade Mykênæ, by disposing subsequent poets to treat the
chief of the Grecian armament against Troy as having been a Spartan.
It has been already mentioned that Stêsichorus, Simonidês and Pindar
adopted this version of the legend: we know that Zeus Agamemnôn, as
well as the hero Menelaus, was worshipped at the Dôrian Sparta,[395]
and the feeling of intimate identity, as well as of patriotic pride,
which had grown up in the minds of the Spartans connected with the
name of Agamemnôn, is forcibly evinced by the reply of the Spartan
Syagrus to Gelôn of Syracuse at the time of the Persian invasion of
Greece. Gelôn was solicited to lend his aid in the imminent danger of
Greece before the battle of Salamis: he offered to furnish an immense
auxiliary force, on condition that the supreme command should be
allotted to him. “Loudly indeed would the Pelopid Agamemnôn cry out
(exclaimed Syagrus in rejecting this application), if he were to learn
that the Spartans had been deprived of the headship by Gelôn and the
Syracusans.”[396] Nearly a century before this event, in obedience to
the injunctions of the Delphian oracle, the Spartans had brought back
from Tegea to Sparta the bones of “the Lacônian Orestês,” as Pindar
denominates him:[397] the recovery of these bones was announced to them
as the means of reversing a course of ill-fortune, and of procuring
victory in their war against Tegea.[398] The value which they set upon
this acquisition, and the decisive results ascribed to it, exhibit a
precise analogy with the recovery of the bones of Theseus from Skyros
by the Athenian Cimôn shortly after the Persian invasion.[399] The
remains sought were those of a hero properly belonging to their own
soil, but who had died in a foreign land, and of whose protection and
assistance they were for that reason deprived. And the superhuman
magnitude of the bones, which were contained in a coffin seven cubits
long, is well suited to the legendary grandeur of the son of Agamemnôn.

  [394] See the preface of Dissen to the tenth Nem. of Pindar.

  [395] Clemens Alexandr. Admonit. ad Gent. p. 24. Ἀγαμέμνονα γοῦν
  τινα Δία ἐν Σπάρτῃ τιμᾶσθαι Στάφυλος ἱστορεῖ. See also Œnomaus
  ap. Euseb. Præparat. Evangel. v. 28.

  [396] Herodot. vii. 159. Ἦ κε μέγ᾽ οἰμώξειεν ὁ Πελοπίδης
  Ἀγαμέμνων, πυθόμενος Σπαρτιήτας ἀπαραιρῆσθαι τὴν ἡγεμονίαν ὑπὸ
  Γέλωνός τε καὶ τῶν Συρακουσίων: compare Homer, Iliad, vii. 125.
  See what appears to be an imitation of the same passage in
  Josephus, De Bello Judaico, iii. 8, 4. Ἦ μεγάλα γ᾽ ἂν στενάξειαν
  οἱ πάτριοι νόμοι, etc.

  [397] Pindar. Pyth. xi. 16.

  [398] Herodot. i 68.

  [399] Plutarch. Thêseus, c. 36, Cimôn, c. 8; Pausan. iii. 3, 6.




CHAPTER VIII.

LACONIAN AND MESSENIAN GENEALOGIES.


The earliest names in Lacônian genealogy are, an autochthonous Lelex
and a Naiad nymph Kleochareia. From this pair sprung a son Eurôtas,
and from him a daughter Sparta, who became the wife of Lacedæmôn, son
of Zeus and Taygetê, daughter of Atlas. Amyklas, son of Lacedæmôn,
had two sons, Kynortas and Hyacinthus—the latter a beautiful youth,
the favorite of Apollo, by whose hand he was accidentally killed
while playing at quoits: the festival of the Hyacinthia, which the
Lacedæmônians generally, and the Amyklæans with special solemnity,
celebrated throughout the historical ages, was traced back to this
legend. Kynortas was succeeded by his son Periêrês, who married
Gorgophonê, daughter of Perseus, and had a numerous issue—Tyndareus,
Ikarius, Aphareus, Leukippus, and Hippokoon. Some authors gave the
genealogy differently, making Periêrês, son of Æolus, to be the father
of Kynortas, and Œbalus son of Kynortas, from whom sprung Tyndareus,
Ikarius and Hippokoon.[400]

  [400] Compare Apollod. iii. 10, 4. Pausan. iii. 1, 4.

Both Tyndareus and Ikarius, expelled by their brother Hippokoon, were
forced to seek shelter at the residence of Thestius, king of Kalydôn,
whose daughter, Lêda, Tyndareus espoused. It is numbered among the
exploits of the omnipresent Hêraklês, that he slew Hippokoon and his
sons, and restored Tyndareus to his kingdom, thus creating for the
subsequent Hêrakleidan kings a mythical title to the throne. Tyndareus,
as well as his brothers, are persons of interest in legendary
narrative: he is the father of Kastôr, of Timandra, married to Echemus,
the hero of Tegea,[401] and of Klytæmnêstra, married to Agamemnôn.
Pollux and the ever-memorable Helen are the offspring of Lêda by Zeus.
Ikarius is the father of Penelopê, wife of Odysseus: the contrast
between her behavior and that of Klytæmnêstra and Helen became the more
striking in consequence of their being so nearly related. Aphareus is
the father of Idas and Lynkeus, while Leukippus has for his daughters,
Phœbê and Ilaëira. According to one of the Hesiodic poems, Kastôr and
Pollux were both sons of Zeus by Lêda, while Helen was neither daughter
of Zeus nor of Tyndareus, but of Oceanus and Têthys.[402]

  [401] Hesiod. ap Schol. Pindar, Olymp. xi. 79.

  [402] Hesiod. ap. Schol. Pindar, Nem. x. 150. Fragm. Hesiod.
  Düntzer, 58. p. 44. Tyndareus was worshipped as a god at
  Lacedæmôn (Varro ap. Serv. ad Virgil. Æneid. viii. 275).

The brothers Kastôr and (Polydeukês, or) Pollux are no less
celebrated for their fraternal affection than for their great bodily
accomplishments: Kastôr, the great charioteer and horse-master; Pollux,
the first of pugilists. They are enrolled both among the hunters of the
Kalydônian boar and among the heroes of the Argonautic expedition, in
which Pollux represses the insolence of Amykus, king of the Bebrykes,
on the coast of Asiatic Thrace—the latter, a gigantic pugilist, from
whom no rival has ever escaped, challenges Pollux, but is vanquished
and killed in the fight.[403]

  [403] Apollôn. Rhod. ii. 1-96. Apollod. i. 9, 20. Theocrit. xxii.
  26-133. In the account of Apollônius and Apollôdorus, Amykus is
  slain in the contest; in that of Theocritus he is only conquered
  and forced to give in, with a promise to renounce for the future
  his brutal conduct; there were several different narratives. See
  Schol. Apollôn. Rhod. ii. 106.

The two brothers also undertook an expedition into Attica, for the
purpose of recovering their sister Helen, who had been carried off by
Thêseus in her early youth, and deposited by him at Aphidna, while
he accompanied Perithous to the under-world, in order to assist his
friend in carrying off Persephonê. The force of Kastôr and Pollux was
irresistible, and when they re-demanded their sister, the people of
Attica were anxious to restore her: but no one knew where Thêseus had
deposited his prize. The invaders, not believing in the sincerity of
this denial, proceeded to ravage the country, which would have been
utterly ruined, had not Dekelus, the eponymus of Dekeleia, been able to
indicate Aphidna as the place of concealment. The autochthonous Titakus
betrayed Aphidna to Kastôr and Pollux, and Helen was recovered: the
brothers in evacuating Attica, carried away into captivity Æthra,
the mother of Thêseus. In after-days, when Kastôr and Pollux, under
the title of the Dioskuri, had come to be worshipped as powerful
gods, and when the Athenians were greatly ashamed of this act of
Thêseus—the revelation made by Dekelus was considered as entitling him
to the lasting gratitude of his country, as well as to the favorable
remembrance of the Lacedæmônians, who maintained the Dekeleians in
the constant enjoyment of certain honorary privileges at Sparta,[404]
and even spared that dême in all their invasions of Attica. Nor is
it improbable that the existence of this legend had some weight in
determining the Lacedæmônians to select Dekelia as the place of their
occupation during the Peleponnêsian war.

  [404] Diodôr. ix. 63. Herod. iv. 73. Δεκελέων δὲ τῶν τότε
  ἐργασαμένων ἔργον χρήσιμον ἐς τὸν πάντα χρόνον, ὡς αὐτοὶ Ἀθηναῖοι
  λέγουσι. According to other authors, it was Akadêmus who made the
  revelation, and the spot called Akadêmia, near Athens, which the
  Lacedæmônians spared in consideration of this service (Plutarch,
  Thêseus, 31, 32, 33, where he gives several different versions of
  this tale by Attic writers, framed with the view of exonerating
  Thêseus). The recovery of Helen and the captivity of Æthra were
  represented on the ancient chest of Kypselus, with the following
  curious inscription:

    Τυνδαρίδα Ἑλέναν φέρετον, Αἴθραν δ᾽ Ἀθέναθεν
    Ἕλκετον.

      Pausan. v. 19, 1.

The fatal combat between Kastôr and Polydeukês on the one side, and
Idas and Lynkeus on the other, for the possession of the daughters of
Leukippus, was celebrated by more than one ancient poet, and forms the
subject of one of the yet remaining Idylls of Theocritus. Leukippus
had formally betrothed his daughters to Idas and Lynkeus; but the
Tyndarids, becoming enamored of them, outbid their rivals in the value
of the customary nuptial gifts, persuaded the father to violate his
promise, and carried off Phoebê and Ilaëira as their brides. Idas and
Lynkeus pursued them and remonstrated against the injustice: according
to Theocritus, this was the cause of the combat. But there was another
tale, which seems the older, and which assigns a different cause to the
quarrel. The four had jointly made a predatory incursion into Arcadia,
and had driven off some cattle, but did not agree about the partition
of the booty—Idas carried off into Messênia a portion of it which the
Tyndarids claimed as their own. To revenge and reimburse themselves,
the Tyndarids invaded Messênia, placing themselves in ambush in the
hollow of an ancient oak. But Lynkeus, endued with preternatural powers
of vision, mounted to the top of Taygetus, from whence, as he could see
over the whole Peleponnêsus, he detected them in their chosen place
of concealment. Such was the narrative of the ancient Cyprian Verses.
Kastôr perished by the hand of Idas, Lynkeus by that of Pollux. Idas,
seizing a stone pillar from the tomb of his father Aphareus, hurled
it at Pollux, knocked him down and stunned him; but Zeus, interposing
at the critical moment for the protection of his son, killed Idas
with a thunderbolt. Zeus would have conferred upon Pollux the gift
of immortality, but the latter could not endure existence without
his brother: he entreated permission to share the gift with Kastôr,
and both were accordingly permitted to live, but only on every other
day.[405]

  [405] Cypria Carm. Fragm. 8. p. 13, Düntzer. Lycophrôn, 538-566
  with Schol. Apollod. iii. 11, 1. Pindar, Nem. x. 55-90.
  ἑτερήμερον ἀθανασίαν: also Homer, Odyss. xi. 302, with the
  Commentary of Nitzsch, vol. iii. p. 245.

  The combat thus ends more favorably to the Tyndarids; but
  probably the account least favorable to them is the oldest, since
  their dignity went on continually increasing, until at last they
  became great deities.

The Dioskuri, or sons of Zeus,—as the two Spartan heroes, Kastôr and
Pollux, were denominated,—were recognized in the historical days of
Greece as gods, and received divine honors. This is even noticed
in a passage of the Odyssey,[406] which is at any rate a very old
interpolation, as well as in one of the Homeric hymns. What is yet more
remarkable is, that they were invoked during storms at sea, as the
special and all-powerful protectors of the endangered mariner, although
their attributes and their celebrity seem to be of a character so
dissimilar. They were worshipped throughout most parts of Greece, but
with preëminent sanctity at Sparta.

  [406] Odyss. xxi. 15. Diodôr. xv. 66.

Kastôr and Pollux being removed, the Spartan genealogy passes from
Tyndareus to Menelaus, and from him to Orestês.

Originally it appears that Messênê was a name for the western portion
of Lacônia, bordering on what was called Pylos: it is so represented
in the Odyssey, and Ephorus seems to have included it amongst the
possessions of Orestês and his descendants. Throughout the whole
duration of the Messênico-Dôrian kingdom, there never was any town
called Messênê: the town was first founded by Epameinondas, after the
battle of Leuctra. The heroic genealogy of Messênia starts from the
same name as that of Lacônia—from the autochthonous Lelex: his younger
son, Polykaôn, marries Messênê, daughter of the Argeian Triopas, and
settles the country. Pausanias tells us that the posterity of this pair
occupied the country for five generations; but he in vain searched the
ancient genealogical poems to find the names of their descendants.[407]
To them succeeded Periêrês, son of Æolus; and Aphareus and Leukippus,
according to Pausanias, were sons of Periêrês. Idas and Lynkeus are
the only heroes, distinguished for personal exploits and memorable
attributes, belonging to Messênia proper. They are the counterpart of
the Dioskuri, and were interesting persons in the old legendary poems.
Marpêssa was the daughter of Euênus, and wooed by Apollo: nevertheless
Idas[408] carried her off by the aid of a winged chariot which he
had received from Poseidôn, Euênus pursued them, and when he arrived
at the river Lykormas, he found himself unable to overtake them: his
grief caused him to throw himself into the river, which ever afterwards
bore his name. Idas brought Marpêssa safe to Messênia, and even when
Apollo there claimed her of him, he did not fear to risk a combat with
the god. But Zeus interfered as mediator, and permitted the maiden to
choose which of the two she preferred. She attached herself to Idas,
being apprehensive that Apollo would desert her in her old age: on the
death of her husband she slew herself. Both Idas and Lynkeus took part
in the Argonautic expedition and in the Kalydônian boar-hunt.[409]

  [407] Pausan. iv. 2, 1.

  [408] Iliad, ix. 553. Simonidês had handled this story in
  detail (Schol. Ven. II. ix. p. 553). Bacchylidês (ap. Schol.
  Pindar. Isthm. iv. 92) celebrated in one of his poems the
  competition among many eager suitors for the hand of Marpêssa,
  under circumstances similar to the competition for Hippodameia,
  daughter of Œnomaus. Many unsuccessful suitors perished by the
  hand of Euênus: their skulls were affixed to the wall of the
  temple of Poseidôn.

  [409] Apollod. i. 7, 9. Pausan. iv. 2, 5. Apollônius Rhodius
  describes Idas as full of boast and self-confidence, heedless of
  the necessity of divine aid. Probably this was the character of
  the brothers in the old legend, as the enemies of the Dioskuri.

  The wrath of the Dioskuri against Messênia was treated, even in
  the historical times, as the grand cause of the subjection of the
  Messênians by the Spartans: that wrath had been appeased at the
  time when Epameinondas reconstituted Messênê (Pausan. iv. 27, 1).

Aphareus, after the death of his sons, founded the town of Arênê, and
made over most part of his dominions to his kinsman Nêleus, with whom
we pass into the Pylian genealogy.




CHAPTER IX.

ARCADIAN GENEALOGY.


The Arcadian divine or heroic pedigree begins with Pelasgus, whom both
Hesiod and Asius considered as an indigenous man, though Akusilaus the
Argeian represented him as brother of Argos and son of Zeus by Niobê,
daughter of Phorôneus: this logographer wished to establish a community
of origin between the Argeians and the Arcadians.

Lykaôn son of Pelasgus and king of Arcadia, had, by different wives,
fifty sons, the most savage, impious and wicked of mankind: Mænalus
was the eldest of them. Zeus, in order that he might himself become
a witness of their misdeeds, presented himself to them in disguise.
They killed a child and served it up to him for a meal; but the god
overturned the table and struck dead with thunder Lykaôn and all his
fifty sons, with the single exception of Nyktimus, the youngest, whom
he spared at the earnest intercession of the goddess Gæa (the Earth).
The town near which the table was overturned received the name of
Trapezus (Tabletown).

This singular legend (framed on the same etymological type as that of
the ants in Ægina, recounted elsewhere) seems ancient, and may probably
belong to the Hesiodic Catalogue. But Pausanias tells us a story in
many respects different, which was represented to him in Arcadia as
the primitive local account, and which becomes the more interesting,
as he tells us that he himself fully believes it. Both tales indeed go
to illustrate the same point—the ferocity of Lykaôn’s character, as
well as the cruel rites which he practised. The latter was the first
who established the worship and solemn games of Zeus Lykæus: he offered
up a child to Zeus, and made libations with the blood upon the altar.
Immediately after having perpetrated this act, he was changed into a
wolf.[410]

  [410] Apollodôr. iii. 8, 1. Hygin. fab. 176. Eratosthen.
  Catasterism. 8. Pausan. viii. 2, 2-3. A different story
  respecting the immolation of the child is in Nikolaus Damask.
  Frag. p. 41, Orelli. Lykaôn is mentioned as the first founder
  of the temple of Zeus Lykæus in Schol. Eurip. Orest. 1662;
  but nothing is there said about the human sacrifice or its
  consequences. In the historical times, the festival and
  solemnities of the Lykæa do not seem to have been distinguished
  materially from the other agônes of Greece (Pindar, Olymp. xiii.
  104; Nem. x. 46): Xenias the Arcadian, one of the generals in the
  army of Cyrus the younger, celebrated the solemnity with great
  magnificence in the march through Asia Minor (Xen. Anab. i. 2,
  10). But the fable of the human sacrifice, and the subsequent
  transmutation of the person who had eaten human food into a wolf,
  continued to be told in connection with them (Plato, de Republic.
  viii. c. 15. p. 417). Compare Pliny, H. N. viii. 34. This passage
  of Plato seems to afford distinct indication that the practice
  of offering human victims at the altar of the Lykæan Zeus was
  neither prevalent nor recent, but at most only traditional and
  antiquated; and it therefore limits the sense or invalidates the
  authority of the Pseudo-Platonic dialogue, Minos, c. 5.

“Of the truth of this narrative (observes Pausanias) I feel persuaded:
it has been repeated by the Arcadians from old times, and it carries
probability along with it. For the men of that day, from their justice
and piety, were guests and companions at table with the gods, who
manifested towards them approbation when they were good, and anger
if they behaved ill, in a palpable manner: indeed at that time there
were some, who having once been men, became gods, and who yet retain
their privileges as such—Aristæus, the Krêtan Britomartis, Hêraklês
son of Alkmêna, Amphiaraus the son of Oiklês, and Pollux and Kastôr
besides. We may therefore believe that Lykaôn became a wild beast, and
that Niobê, the daughter of Tantalus, became a stone. But in my time,
wickedness having enormously increased, so as to overrun the whole
earth and all the cities in it, there are no farther examples of men
exalted into gods, except by mere title and from adulation towards the
powerful: moreover the anger of the gods falls tardily upon the wicked,
and is reserved for them after their departure from hence.”

Pausanias then proceeds to censure those who, by multiplying false
miracles in more recent times, tended to rob the old and genuine
miracles of their legitimate credit and esteem. The passage illustrates
forcibly the views which a religious and instructed pagan took of his
past time—how inseparably he blended together in it gods and men, and
how little he either recognized or expected to find in it the naked
phænomena and historical laws of connection which belonged to the world
before him. He treats the past as the province of legend, the present
as that of history; and in doing this he is more sceptical than the
persons with whom he conversed, who believed not only in the ancient,
but even in the recent and falsely reported miracles. It is true that
Pausanias does not always proceed consistently with this position: he
often rationalizes the stories of the past, as if he expected to find
historical threads of connection; and sometimes, though more rarely,
accepts the miracles of the present. But in the present instance he
draws a broad line of distinction between present and past, or rather
between what is recent and what is ancient: his criticism is, in the
main, analogous to that of Arrian in regard to the Amazons—denying
their existence during times of recorded history, but admitting it
during the early and unrecorded ages.

In the narrative of Pausanias, the sons of Lykaôn, instead of perishing
by thunder from Zeus, become the founders of the various towns in
Arcadia. And as that region was subdivided into a great number of small
and independent townships, each having its own eponym, so the Arcadian
heroic genealogy appears broken up and subdivided. Pallas, Orestheus,
Phigalus, Trapezeus, Mænalus, Mantinêus, and Tegeatês, are all numbered
among the sons of Lykaôn, and are all eponyms of various Arcadian
towns.[411]

  [411] Paus. viii. 3. Hygin. fab. 177.

The legend respecting Kallistô and Arkas, the eponym of Arcadia
generally, seems to have been originally quite independent of and
distinct from that of Lykaôn. Eumêlus, indeed, and some other poets
made Kallistô daughter of Lykaôn; but neither Hesiod, nor Asius, nor
Pherekydês, acknowledged any relationship between them.[412] The
beautiful Kallistô, companion of Artemis in the chase, had bound
herself by a vow of chastity. Zeus, either by persuasion or by force,
obtained a violation of the vow, to the grievous displeasure both of
Hêrê and Artemis. The former changed Kallistô into a bear, the latter
when she was in that shape killed her with an arrow. Zeus gave to the
unfortunate Kallistô a place among the stars, as the constellation of
the Bear: he also preserved the child Arkas, of which she was pregnant
by him, and gave it to the Atlantid nymph Maia to bring up.[413]

  [412] Apollod. iii. 8, 2.

  [413] Pausan. viii. 3, 2. Apollod. iii. 8, 2. Hesiod. apud
  Eratosthen. Catasterism. 1. Fragm. 182, Marktsch. Hygin. f. 177.

Arkas, when he became king, obtained from Triptolemus and communicated
to his people the first rudiments of agriculture; he also taught them
to make bread, to spin, and to weave. He had three sons—Azan, Apheidas,
and Elatus: the first was the eponym of Azania, the northern region
of Arcadia; the second was one of the heroes of Tegea; the third was
father of Ischys (rival of Apollo for the affections of Korônis), as
well as of Æpytus and Kyllên: the name of Æpytus among the heroes of
Arcadia is as old as the Catalogue in the Iliad.[414]

  [414] Homer, Iliad, ii. 604. Pind. Olymp. vi. 44-63.

  The tomb of Æpytus, mentioned in the Iliad, was shown to
  Pausanias between Pheneus and Stymphalus (Pausan. viii. 16, 2).
  Æpytus was a cognomen of Hermês (Pausan. viii. 47, 3).

  The hero Arkas was worshipped at Mantineia, under the special
  injunction of the Delphian oracle (Pausan. viii. 9, 2).

Aleus, son of Apheidas and king of Tegea, was the founder of the
celebrated temple and worship of Athênê Alea in that town. Lykurgus and
Kêpheus were his sons, Augê his daughter, who was seduced by Hêraklês,
and secretly bore to him a child: the father, discovering what had
happened, sent Augê to Nauplius to be sold into slavery: Teuthras,
king of Mysia in Asia Minor, purchased her and made her his wife: her
tomb was shown at Pergamus on the river Kaïkus even in the time of
Pausanias.[415]

  [415] Pausan. viii. 4, 6. Apollod. iii. 9, 1. Diodôr. iv. 33.

  A separate legend respecting Augê and the birth of Têlephus was
  current at Tegea, attached to the temple, statue, and cognomen of
  Eileithyia in the Tegeatic agora (Pausan. viii. 48, 5).

  Hekatæus seems to have narrated in detail the adventures of Augê
  (Pausan. viii. 4, 4; 47, 3. Hekatæ. Fragm. 345, Didot.).

  Euripidês followed a different story about Augê and the birth of
  Têlephus in his lost tragedy called Augê (See Strabo, xiii. p.
  615). Respecting the Μυσοὶ of Æschylus, and the two lost dramas,
  Ἀλεαδαὶ and Μυσοὶ of Sophoklês, little can be made out. (See
  Welcker, Griechisch. Tragöd. p. 53, 408-414).

The child Têlephus, exposed on Mount Parthenius, was wonderfully
sustained by the milk of a doe: the herdsmen of Korythus brought him
up, and he was directed by the Delphian oracle to go and find his
parents in Mysia. Teuthras adopted him, and he succeeded to the throne:
in the first attempt of the army of Agamemnôn against Troy, on which
occasion they mistook their point and landed in Mysia, his valor
signally contributed to the repulse of the Greeks, though he was at
last vanquished and desperately wounded by the spear of Achilles—by
whom however he was afterwards healed, under the injunction of the
oracle, and became the guide of the Greeks in their renewed attack upon
the Trojans.[416]

  [416] Têlephus and his exploits were much dwelt upon in the lost
  old epic poem, the Cyprian Verses. See argument of that poem ap.
  Düntzer, Ep. Fragm. p. 10. His exploits were also celebrated by
  Pindar (Olymp. ix. 70-79); he is enumerated along with Hectôr,
  Cycnus, Memnôn, the most distinguished opponents of Achilles
  (Isthm. iv. 46). His birth, as well as his adventures, became
  subjects with most of the great Attic tragedians.

From Lykurgus,[417] the son of Aleus and brother of Augê, we pass to
his son Ankæus, numbered among the Argonauts, finally killed in the
chase of the Kalydônian boar, and father of Agapenôr, who leads the
Arcadian contingent against Troy,—(the adventurers of his niece, the
Tegeatic huntress Atalanta, have already been touched upon),—then
to Echemus, son of Aëropus and grandson of the brother of Lykurgus,
Kêpheus. Echemus is the chief heroic ornament of Tegea. When Hyllus,
the son of Hêraklês, conducted the Hêrakleids on their first expedition
against Peloponnêsus, Echemus commanded the Tegean troops who assembled
along with the other Peloponnêsians at the isthmus of Corinth to repel
the invasion: it was agreed that the dispute should be determined
by single combat, and Echemus, as the champion of Peloponnêsus,
encountered and killed Hyllus. Pursuant to the stipulation by which
they had bound themselves, the Hêrakleids retired, and abstained for
three generations from pressing their claim upon Peloponnêsus. This
valorous exploit of their great martial hero was cited and appealed to
by the Tegeates before the battle of Platæa, as the principal evidence
of their claim to the second post in the combined army, next in point
of honor to that of the Lacedæmônians, and superior to that of the
Athenians: the latter replied to them by producing as counter-evidence
the splendid heroic deeds of Athens,—the protection of the Hêrakleids
against Eurystheus, the victory over the Kadmeians of Thêbes, and the
complete defeat of the Amazons in Attica.[418] Nor can there be any
doubt that these legendary glories were both recited by the speakers,
and heard by the listeners, with profound and undoubting faith, as well
as with heart-stirring admiration.

  [417] There were other local genealogies of Tegea deduced from
  Lykurgus: Bôtachus, eponym of the Dême Bôtachidæ at that place,
  was his grandson (Nicolaus ap. Steph. Byz. v. Βωταχίδαι).

  [418] Herodot. ix. 27. Echemus is described by Pindar (Ol. xi.
  69) as gaining the prize of wrestling in the fabulous Olympic
  games, on their first establishment by Hêraklês. He also found
  a place in the Hesiodic Catalogue as husband of Timandra, the
  sister of Helen and Klytæmnêstra (Hesiod, Fragm. 105, p. 318,
  Marktscheff.).

One other person there is—Ischys, son of Elatus and grandson of
Arkas—in the fabulous genealogy of Arcadia whom it would be improper to
pass over, inasmuch as his name and adventures are connected with the
genesis of the memorable god or hero Æsculapius, or Asklêpius. Korônis,
daughter of Phlegyas, and resident near the lake Bœbëis in Thessaly,
was beloved by Apollo and became pregnant by him: unfaithful to the
god, she listened to the propositions of Ischys son of Elatus, and
consented to wed him: a raven brought to Apollo the fatal news, which
so incensed him that he changed the color of the bird from white, as it
previously had been, into black.[419] Artemis, to avenge the wounded
dignity of her brother, put Korônis to death; but Apollo preserved the
male child of which she was about to be delivered, and consigned it to
the Centaur Cheirôn to be brought up. The child was named Asklêpius or
Æsculapius, and acquired, partly from the teaching of the beneficent
leech Cheirôn, partly from inborn and superhuman aptitude, a knowledge
of the virtues of herbs and a mastery of medicine and surgery, such
as had never before been witnessed. He not only cured the sick, the
wounded, and the dying, but even restored the dead to life. Kapaneus,
Eriphylê, Hippolytus, Tyndareus and Glaukus were all affirmed by
different poets and logographers to have been endued by him with a new
life.[420] But Zeus now found himself under the necessity of taking
precautions lest mankind, thus unexpectedly protected against sickness
and death, should no longer stand in need of the immortal gods: he
smote Asklêpius with thunder and killed him. Apollo was so exasperated
by this slaughter of his highly-gifted son, that he killed the Cyclôpes
who had fabricated the thunder, and Zeus was about to condemn him to
Tartarus for doing so; but on the intercession of Latôna he relented,
and was satisfied with imposing upon him a temporary servitude in the
house of Admêtus at Pheræ.

  [419] Apollodôr. iii. 10, 3; Hesiod, Fragm. 141-142,
  Marktscheff.; Strab. ix. p. 442; Pherekydês, Fragm. 8; Akusilaus,
  Fragm. 25, Didot.

    Τῷ μὲν ἄρ᾽ ἄγγελος ἦλθε κόραξ, ἱερῆς ἀπὸ δαιτὸς
    Πυθὼ ἐς ἠγαθέην, καὶ ῥ᾽ ἔφρασεν ἔργ᾽ ἀΐδηλα
    Φοίβῳ ἀκερσεκόμῃ, ὅτι Ἴσχυς γῆμε Κόρωνιν
    Εἰλατίδης, Φλεγύαο διογνήτοιο θύγατρα. (Hesiod, Fr.)

  The change of the color of the crow is noticed both in Ovid,
  Metamorph. ii. 632, in Antonin. Liberal. c. 20, and in Servius ad
  Virgil. Æneid. vii. 761, though the name “_Corvo_ custode ejus”
  is there printed with a capital letter, as if it were a man named
  _Corvus_.

  [420] Schol. Eurip. Alkêst. 1; Diodôr. iv. 71; Apollodôr. iii.
  10, 3; Pindar, Pyth. iii. 59; Sextus Empiric. adv. Grammatic. i.
  12. p. 271. Stesichorus named Eriphylê—the Naupaktian verses,
  Hippolytus—(compare Servius ad Virgil. Æneid. vii. 761);
  Panyasis, Tyndareus; a proof of the popularity of this tale among
  the poets. Pindar says that Æsculapius was “tempted by gold” to
  raise a man from the dead, and Plato (Legg. iii. p. 408) copies
  him: this seems intended to afford some color for the subsequent
  punishment. “Mercede id captum (observes Boeckh. ad Pindar. _l.
  c._) Æsculapium fecisse recentior est fictio; Pindari fortasse
  ipsius, quem tragici secuti sunt: haud dubie a medicorum avaris
  moribus profecta, qui Græcorum medicis nostrisque communes
  sunt.” The rapacity of the physicians (granting it to be ever
  so well-founded, both then and now) appears to me less likely
  to have operated upon the mind of Pindar, than the disposition
  to extenuate the cruelty of Zeus, by imputing guilty and sordid
  views to Asklêpius. Compare the citation from Dikæarchus,
  _infrà_, p. 249, note 1.

Asklêpius was worshipped with very great solemnity at Trikka, at Kôs,
at Knidus, and in many different parts of Greece, but especially at
Epidaurus, so that more than one legend had grown up respecting the
details of his birth and adventures: in particular, his mother was
by some called Arsinoê. But a formal application had been made on
this subject (so the Epidaurians told Pausanias) to the oracle of
Delphi, and the god in reply acknowledged that Asklêpius was his son
by Korônis.[421] The tale above recounted seems to have been both the
oldest and the most current. It is adorned by Pindar in a noble ode,
wherein however he omits all mention of the raven as messenger—not
specifying who or what the spy was from whom Apollo learnt the
infidelity of Korônis. By many this was considered as an improvement in
respect of poetical effect, but it illustrates the mode in which the
characteristic details and simplicity of the old fables[422] came to be
exchanged for dignified generalities, adapted to the altered taste of
society.

  [421] Pausan. ii. 26, where several distinct stories are
  mentioned, each springing up at some one or other of the
  sanctuaries of the god: quite enough to justify the idea of these
  Æsculapii (Cicero, N. D. iii. 22).

  Homer, Hymn ad Æsculap. 2. The tale briefly alluded to in the
  Homeric Hymn. ad Apollin. 209. is evidently different: Ischys is
  there the companion of Apollo, and Korônis is an Arcadian damsel.

  Aristidês, the fervent worshipper of Asklêpius, adopted the story
  of Korônis, and composed hymns on the γάμον Κορωνίδος καὶ γένεσιν
  τοῦ θεοῦ (Orat. 23. p. 463, Dind.).

  [422] See Pindar, Pyth. iii. The Scholiast puts a construction
  upon Pindar’s words which is at any rate far-fetched, if indeed
  it be at all admissible: he supposes that Apollo knew the fact
  from his own omniscience, without any informant, and he praises
  Pindar for having thus transformed the old fable. But the words
  οὐδ᾽ ἔλαθε σκόπον seem certainly to imply some informant: to
  suppose that σκόπον means the god’s own mind, is a strained
  interpretation.

Machaôn and Podaleirius, the two sons of Asklêpius, command the
contingent from Trikka, in the north-west region of Thessaly, at
the siege of Troy by Agamemnôn.[423] They are the leeches of the
Grecian army, highly prized and consulted by all the wounded chiefs.
Their medical renown was further prolonged in the subsequent poem
of Arktinus, the Iliu-Persis, wherein the one was represented as
unrivalled in surgical operations, the other as sagacious in detecting
and appreciating morbid symptoms. It was Podaleirius who first noticed
the glaring eyes and disturbed deportment which preceded the suicide
of Ajax.[424]

  [423] Iliad, ii. 730. The Messênians laid claim to the sons of
  Asklêpius as their heroes, and tried to justify the pretension by
  a forced construction of Homer (Pausan. iii. 4, 2).

  [424] Arktinus, Epicc. Græc. Fragm. 2. p. 22, Düntzer. The
  Ilias Minor mentioned the death of Machaôn by Eurypylus, son of
  Têlephus (Fragm. 5. p. 19, Düntzer).

Galen appears uncertain whether Asklêpius (as well as Dionysus) was
originally a god, or whether he was first a man and then became
afterwards a god;[425] but Apollodôrus professed to fix the exact
date of his apotheosis.[426] Throughout all the historical ages the
descendants of Asklêpius were numerous and widely diffused. The many
families or gentes called Asklêpiads, who devoted themselves to the
study and practice of medicine, and who principally dwelt near the
temples of Asklêpius whither sick and suffering men came to obtain
relief—all recognized the god not merely as the object of their common
worship, but also as their actual progenitor. Like Solôn, who reckoned
Nêleus and Poseidôn as his ancestors, or the Milêsian Hekatæus, who
traced his origin through fifteen successive links to a god—like
the privileged gens at Pêlion in Thessaly,[427] who considered the
wise Centaur Cheirôn as their progenitor, and who inherited from him
their precious secrets respecting the medicinal herbs of which their
neighborhood was full,—Asklêpiads, even of the later times, numbered
and specified all the intermediate links which separated them from
their primitive divine parent. One of these genealogies has been
preserved to us, and we may be sure that there were many such, as
the Asklêpiads were found in many different places.[428] Among them
were enrolled highly instructed and accomplished men, such as the
great Hippocratês and the historian Ktêsias, who prided themselves on
the divine origin of themselves and their gens[429]—so much did the
legendary element pervade even the most philosophical and positive
minds of historical Greece. Nor can there be any doubt that their
means of medical observation must have been largely extended by
their vicinity to a temple so much frequented by the sick, who came
in confident hopes of divine relief, and who, whilst they offered
up sacrifice and prayer to Æsculapius, and slept in his temple in
order to be favored with healing suggestions in their dreams, might,
in case the god withheld his supernatural aid, consult his living
descendants.[430] The sick visitors at Kôs, or Trikka, or Epidaurus,
were numerous and constant, and the tablets usually hung up to record
the particulars of their maladies, the remedies resorted to, and the
cures operated by the god, formed both an interesting decoration of the
sacred ground and an instructive memorial to the Asklêpiads.[431]

  [425] Ἀσκληπιός γέ τοι καὶ Διόνυσος, εἴτ᾽ ἄνθρωποι πρότερον ἤστην
  εἴτε καὶ ἀρχῆθεν θεοί (Galen, Protreptic. 9. t. 1. p. 22, Kühn.).
  Pausanias considers him as θεὸς ἐξ ἀρχῆς (ii. 26, 7). In the
  important temple at Smyrna he was worshipped as Ζεὺς Ἀσκληπιός
  (Aristidês, Or. 6. p. 64; Or. 23. p. 456, Dind.).

  [426] Apollodôr. ap. Clem. Alex. Strom. i. p. 381; see Heyne,
  Fragment. Apollodôr. p. 410. According to Apollodôrus, the
  apotheosis of Hêraklês and of Æsculapius took place at the same
  time, thirty-eight years after Hêraklês began to reign at Argos.

  [427] About Hekatæus, Herodot. ii. 143; about Solôn, Diogen.
  Laërt. Vit. Platon. init.

  A curious fragment, preserved from the lost works of Dikæarchus,
  tells us of the descendants of the Centaur Cheirôn at the town of
  Pêlion, or perhaps at the neighboring town of Dêmêtrias,—it is
  not quite certain which, perhaps at both (see Dikæarch. Fragment.
  ed. Fuhr, p. 408). Ταύτην δὲ τὴν δύναμιν ἓν τῶν πολιτῶν οἶδε
  γένος, ὁ δὴ λέγεται Χείρωνος ἀπόγονον εἶναι· παραδίδωσι δὲ καὶ
  δείκνυσι πατὴρ υἱῷ, καὶ οὕτως ἡ δύναμις φυλάσσεται, ὡς οὐδεὶς
  ἄλλος οἶδε τῶν πολιτῶν· οὐχ ὅσιον δὲ τοὺς ἐπισταμένους τὰ φάρμακα
  μισθοῦ τοῖς καμνοῦσι βοηθεῖν, ἀλλὰ προῖκα.

  Plato, de Republ. iii. 4 (p. 391). Ἀχιλλεὺς ὑπὸ τῷ σοφωτάτῳ
  Χείρωνι τεθραμμένος. Compare Xenophôn, De Venat. c. 1.

  [428] See the genealogy at length in Le Clerc, Historie de
  la Médecine, lib. ii. c. 2. p. 78, also p. 287; also Littré,
  Introduction aux Œuvres Complètes d’Hippocrate, t. i. p. 35.
  Hippocratês was the seventeenth from Æsculapius.

  Theopompus the historian went at considerable length into the
  pedigree of the Asklêpiads of Kôs and Knidus, tracing them up
  to Podaleirius and his first settlement at Syrnus in Karia (see
  Theopomp. Fragm. 111, Didot): Polyanthus of Kyrênê composed
  a special treatise περὶ τῆς τῶν Ἀσκληπιαδῶν γενέσεως (Sextus
  Empiric. adv. Grammat. i. 12. p. 271); see Stephan. Byz. v. Κῶς,
  and especially Aristidês, Orat. vii. _Asclêpiadæ_. The Asklêpiads
  were even reckoned among the Ἀρχηγέται of Rhodes, jointly with
  the Hêrakleids (Aristidês, Or. 44, ad Rhod. p. 839, Dind.).

  In the extensive sacred enclosure at Epidaurus stood the
  statues of Asklêpius and his wife Epionê (Pausan. ii. 29, 1):
  two daughters are coupled with him by Aristophanês, and he was
  considered especially εὔπαις (Plutus, 654); Jaso, Panakeia and
  Hygieia are named by Aristidês.

  [429] Plato, Protagor. c. 6 (p. 311). Ἱπποκράτη τὸν Κῶον, τὸν
  τῶν Ἀσκληπιαδῶν; also Phædr. c. 121. (p. 270). About Ktêsias,
  Galen, Opp. t. v. p. 652, Basil.; and Bahrt, Fragm. Ktêsiæ, p. 20.
  Aristotle (see Stahr. Aristotelia, i. p. 32) and Xenophôn, the
  physician of the emperor Claudius, were both Asklêpiads (Tacit.
  Annal. xii. 61). Plato, de Republ. iii. 405, calls them τοὺς
  κομψοὺς Ἀσκληπιάδας.

  Pausanias, a distinguished physician at Gela in Sicily, and
  contemporary of the philosopher Empedoklês, was also an
  Asklêpiad: see the verses of Empedoklês upon him, Diogen. Laërt.
  viii. 61.

  [430] Strabo, viii. p. 374; Aristophan. Vesp. 122; Plutus,
  635-750; where the visit to the temple of Æsculapius is described
  in great detail, though with a broad farcical coloring.

  During the last illness of Alexander the Great, several of his
  principal officers slept in the temple of Serapis, in the hope
  that remedies would be suggested to them in their dreams (Arrian,
  vii. 26).

  Pausanias, in describing the various temples of Asklêpius which
  he saw, announces as a fact quite notorious and well-understood,
  “Here cures are wrought by the god” (ii. 36, 1; iii. 26, 7; vii.
  27, 4): see Suidas, v. Ἀρίσταρχος. The Orations of Aristidês,
  especially the 6th and 7th, _Asklêpius and the Asklêpiadæ_,
  are the most striking manifestations of faith and thanksgiving
  towards Æsculapius, as well as attestations of his extensive
  working throughout the Grecian world; also Orat. 23 and 25, Ἱερῶν
  Λόγος, 1 and 3; and Or. 45 (De Rhetoricâ, p. 22. Dind.), αἵ τ᾽ ἐν
  Ἀσκληπιοῦ τῶν ἀεὶ διατριβόντων ἀγελαὶ, etc.

  [431] Pausan. ii. 27, 3; 36, 1. Ταύταις ἐγγεγραμμένα ἐστὶ καὶ
  ἀνδρῶν καὶ γυναικῶν ὀνόματα ἀκεσθέντων ὑπὸ τοῦ Ἀσκληπιοῦ, πρόσετι
  δὲ καὶ νόσημα, ὅ,τι ἕκαστος ἐνόσησε, καὶ ὅπως ἰάθη,—the cures are
  wrought by the god himself.

The genealogical descent of Hippocratês and the other Asklêpiads
from the god Asklêpius is not only analogous to that of Hekatæus and
Solôn from their respective ancestoral gods, but also to that of the
Lacedæmônian kings from Hêraklês, upon the basis of which the whole
supposed chronology of the ante-historical times has been built, from
Eratosthenês and Apollodôrus down to the chronologers of the present
century.[432] I shall revert to this hereafter.

  [432] “Apollodôrus ætatem Herculis pro cardine chronologiæ
  habuit” (Heyne, ad Apollodôr. Fragm. p. 410).




CHAPTER X.

ÆAKUS AND HIS DESCENDANTS.—ÆGINA, SALAMIS, AND PHTHIA.


The memorable heroic genealogy of the Æakids establishes a fabulous
connection between Ægina, Salamis, and Phthia, which we can only
recognize as a fact, without being able to trace its origin.

Æakus was the son of Zeus, born of Ægina, daughter of Asôpus, whom the
god had carried off and brought into the island to which he gave her
name: she was afterwards married to Aktôr, and had by him Menœtius,
father of Patroclus. As there were two rivers named Asôpus, one
between Phlius and Sikyôn, and another between Thêbes and Platæa—so
the Æginêtan heroic genealogy was connected both with that of Thêbes
and with that of Phlius: and this belief led to practical consequences
in the minds of those who accepted the legends as genuine history. For
when the Thêbans, in the 68th Olympiad, were hard-pressed in war by
Athens, they were directed by the Delphian oracle to ask assistance
of their next of kin: recollecting that Thêbê and Ægina had been
sisters, common daughters of Asôpus, they were induced to apply to the
Æginêtans as their next of kin, and the Æginêtans gave them aid, first
by sending to them their common heroes, the Æakids, next by actual
armed force.[433] Pindar dwells emphatically on the heroic brotherhood
between Thêbes, his native city, and Ægina.[434]

  [433] Herodot. v. 81.

  [434] Nem. iv. 22. Isthm. vii. 16.

Æakus was alone in Ægina: to relieve him from this solitude, Zeus
changed all the ants in the island into men, and thus provided him
with a numerous population, who, from their origin, were called
Myrmidons.[435] By his wife Endêis, daughter of Cheirôn, Æakus had for
his sons Pêleus and Telamôn: by the Nereid Psamathê, he had Phôkus. A
monstrous crime had then recently been committed by Pelops, in killing
the Arcadian prince, Stymphalus, under a simulation of friendship and
hospitality: for this the gods had smitten all Greece with famine
and barrenness. The oracles affirmed that nothing could relieve
Greece from this intolerable misery except the prayers of Æakus, the
most pious of mankind. Accordingly envoys from all quarters flocked
to Ægina, to prevail upon Æakus to put up prayers for them: on his
supplications the gods relented, and the suffering immediately ceased.
The grateful Greeks established in Ægina the temple and worship of Zeus
Panhellênius, one of the lasting monuments and institutions of the
island, on the spot where Æakus had offered up his prayer. The statues
of the envoys who had come to solicit him were yet to be seen in the
Æakeion, or sacred edifice of Æakus, in the time of Pausanias: and the
Athenian Isokratês, in his eulogy of Evagoras, the despot of Salamis in
Cyprus (who traced his descent through Teukrus to Æakus), enlarges upon
this signal miracle, recounted and believed by other Greeks as well as
by the Æginêtans, as a proof both of the great qualities and of the
divine favor and patronage displayed in the career of the Æakids.[436]
Æakus was also employed to aid Poseidôn and Apollo in building the
walls of Troy.[437]

  [435] This tale, respecting the transformation of the ants into
  men, is as old as the Hesiodic Catalogue of Women. See Düntzer,
  Fragm. Epicc. 21. p. 34; evidently an etymological tale from the
  name Myrmidones. Pausanias throws aside both the etymology and
  the details of the miracle: he says that Zeus raised men from the
  earth, at the prayer of Æakus (ii. 29, 2): other authors retained
  the etymology of Myrmidons from μύρμηκες, but gave a different
  explanation (Kallimachus, Fragm. 114, Düntzer). Μυρμιδόνων ἐσσῆνα
  (Strabo, viii. p. 375). Ἐσσὴν, ὁ οἰκιστής (Hygin. fab. 52).

  According to the Thessalian legend, Myrmidôn was the son of
  Zeus by Eurymedusa, daughter of Kletor; Zeus having assumed the
  disguise of an ant (Clemens Alex. Admon. ad Gent. p. 25. Sylb.).

  [436] Apollod. iii. 12, 6. Isokrat. Evagor. Encom. vol. ii. p.
  278, Auger. Pausan. i. 45, 13; ii. 29, 6. Schol. Aristoph. Equit.
  1253.

  So in the 106th Psalm, respecting the Israelites and Phinees, v.
  29, “They provoked the Lord to anger by their inventions, and the
  plague was great among them;” “Then stood up Phinees and prayed,
  and so the plague ceased;” “And that was counted unto him for
  righteousness, among all posterities for evermore.”

  [437] Pindar, Olymp. viii. 41, with the Scholia. Didymus did not
  find this story in any other poet older than Pindar.

Pêleus and Telamôn, the sons of Æakus, contracting a jealousy of
their bastard brother, Phôkus, in consequence of his eminent skill in
gymnastic contests, conspired to put him to death. Telamôn flung his
quoit at him while they were playing together, and Pêleus despatched
him by a blow with his hatchet in the back. They then concealed the
dead body in a wood, but Æakus, having discovered both the act and the
agents, banished the brothers from the island.[438] For both of them
eminent destinies were in store.

  [438] Apollod. iii. 12, 6, who relates the tale somewhat
  differently; but the old epic poem Alkmæônis gave the details
  (ap. Schol. Eurip. Andromach. 685)—

    Ἔνθα μὲν ἀντίθεος Τελαμὼν τροχοειδέϊ δίσκῳ
    Πλῆξε κάρη· Πηλεὺς δὲ θοῶς ἀνὰ χεῖρα τανύσσας
    Ἀξίνην ἐΰχαλκον ἐπεπλήγει μετὰ νῶτα.

While we notice the indifference to the moral quality of actions
implied in the old Hesiodic legend, when it imputes distinctly and
nakedly this proceeding to two of the most admired persons of the
heroic world—it is not less instructive to witness the change of
feeling which had taken place in the age of Pindar. That warm eulogist
of the great Æakid race hangs down his head with shame, and declines
to recount, though he is obliged darkly to glance at the cause which
forced the pious Æakus to banish his sons from Ægina. It appears that
Kallimachus, if we may judge by a short fragment, manifested the same
repugnance to mention it.[439]

  [439] Pindar, Nem. v. 15, with Scholia, and Kallimach. Frag. 136.
  Apollônius Rhodius represents the fratricide as inadvertent and
  unintentional (i. 92); one instance amongst many of the tendency
  to soften down and moralize the ancient tales.

  Pindar, however, seems to forget this incident when he speaks
  in other places of the general character of Pêleus (Olymp. ii.
  75-86. Isthm. vii. 40).

Telamôn retired to Salamis, then ruled by Kychreus, the son of Poseidôn
and Salamis, who had recently rescued the island from the plague of
a terrible serpent. This animal, expelled from Salamis, retired to
Eleusis in Attica, where it was received and harbored by the goddess
Dêmêtêr in her sacred domicile.[440] Kychreus dying childless left his
dominion to Telamôn, who, marrying Peribœa, daughter of Alkathoos,
and grand-daughter of Pelops, had for his son the celebrated Ajax.
Telamôn took part both in the chase of the Kalydônian boar and in the
Argonautic expedition: he was also the intimate friend and companion of
Hêraklês, whom he accompanied in his enterprise against the Amazons,
and in the attack made with only six ships upon Laomedôn, king of Troy.
This last enterprise having proved completely successful, Telamôn was
rewarded by Hêraklês with the possession of the daughter of Laomedôn,
Hêsionê—who bore to him Teukros, the most distinguished archer amidst
the host of Agamemnôn, and the founder of Salamis in Cyprus.[441]

  [440] Apollod. iii. 12, 7. Euphoriôn, Fragm. 5, Düntzer, p.
  43, Epicc. Græc. There may have been a tutelary serpent in the
  temple at Eleusis, as there was in that of Athênê Polias at
  Athens (Herodot viii. 41. Photius, v. Οἰκοῦρον ὄφιν. Aristophan.
  Lysistr. 759, with the Schol.).

  [441] Apollod. iii. 12, 7. Hesiod. ap. Strab. ix. p. 393.

  The libation and prayer of Hêraklês, prior to the birth of Ajax,
  and his fixing the name of the yet unborn child, from an eagle
  (αἰετὸς) which appeared in response to his words, was detailed in
  the Hesiodic Eoia, and is celebrated by Pindar (Isthm. v. 30-54).
  See also the Scholia.

Pêleus went to Phthia, where he married the daughter of Eurytiôn,
son of Aktôr, and received from him the third part of his dominions.
Taking part in the Kalydônian boar-hunt, he unintentionally killed his
father-in-law Eurytiôn, and was obliged to flee to Iôlkos, where he
received purification from Akastus, son of Pelias: the danger to which
he became exposed by the calumnious accusations of the enamoured wife
of Akastus has already been touched upon in a previous section. Pêleus
also was among the Argonauts; the most memorable event in his life
however was his marriage with the sea-goddess Thetis. Zeus and Poseidôn
had both conceived a violent passion for Thetis. But the former, having
been forewarned by Promêtheus that Thetis was destined to give birth to
a son more powerful than his father, compelled her, much against her
own will, to marry Pêleus; who, instructed by the intimations of the
wise Cheirôn, was enabled to seize her on the coast called Sêpias in
the southern region of Thessaly. She changed her form several times,
but Pêleus held her fast until she resumed her original appearance,
and she was then no longer able to resist. All the gods were present,
and brought splendid gifts to these memorable nuptials: Apollo sang
with his harp, Poseidôn gave to Pêleus the immortal horses Xanthos and
Balios, and Cheirôn presented a formidable spear, cut from an ash-tree
on Mount Pêlion. We shall have reason hereafter to recognize the value
of both these gifts in the exploits of Achilles.[442]

  [442] Apollodôr. iii. 13, 5. Homer, Iliad, xviii. 434; xxiv. 62.
  Pindar, Nem. iv. 50-68; Isthm. vii. 27-50. Herodot. vii. 192.
  Catullus, Carm. 64. Epithal. Pel. et Thetidos, with the prefatory
  remarks of Dœring.

  The nuptials of Pêleus and Thetis were much celebrated in the
  Hesiodic Catalogue, or perhaps in the Eoiai (Düntzer, Epic. Græc.
  Frag. 36. p. 39), and Ægimius—see Schol. ad Apollôn. Rhod. iv.
  869—where there is a curious attempt of Staphylus to rationalize
  the marriage of Pêleus and Thetis.

  There was a town, seemingly near Pharsalus in Thessaly, called
  Thetideium. Thetis is said to have been carried by Pêleus to both
  these places: probably it grew up round a temple and sanctuary of
  this goddess (Pherekyd. Frag. 16, Didot; Hellank. ap. Steph. Byz.
  Θεστιδεῖον).

The prominent part assigned to Thetis in the Iliad is well known,
and the post-Homeric poets of the Legend of Troy introduced her as
actively concurring first to promote the glory, finally to bewail the
death of her distinguished son.[443] Pêleus, having survived both his
son Achilles and his grandson Neoptolemus, is ultimately directed to
place himself on the very spot where he had originally seized Thetis,
and thither the goddess comes herself to fetch him away, in order
that he may exchange the desertion and decrepitude of age for a life
of immortality along with the Nêreids.[444] The spot was indicated to
Xerxês when he marched into Greece by the Iônians who accompanied him,
and his magi offered solemn sacrifices to her as well as to the other
Nêreids, as the presiding goddesses and mistresses of the coast.[445]

  [443] See the arguments of the lost poems, the Cypria and the
  Æthiopis, as given by Proclus, in Düntzer, Fragm. Epic. Gr. p.
  11-16; also Schol. ad Iliad. xvi. 140; and the extract from the
  lost Ψυχοστασία of Æschylus, ap. Plato. de Republic. ii. c. 21
  (p. 382, St.).

  [444] Eurip. Androm. 1242-1260; Pindar, Olymp. ii. 86.

  [445] Herodot. vii. 198.

Neoptolemus or Pyrrhus, the son of Achilles, too young to engage in the
commencement of the siege of Troy, comes on the stage after the death
of his father as the indispensable and prominent agent in the final
capture of the city. He returns victor from Troy, not to Phthia, but to
Epirus, bringing with him the captive Andromachê, widow of Hectôr, by
whom Molossus is born to him. He himself perishes in the full vigor of
life at Delphi by the machinations of Orestês, son of Agamemnôn. But
his son Molossus—like Fleance, the son of Banquo, in Macbeth—becomes
the father of the powerful race of Molossian kings, who played so
conspicuous a part during the declining vigor of the Grecian cities,
and to whom the title and parentage of Æakids was a source of peculiar
pride, identifying them by community of heroic origin with genuine and
undisputed Hellênes.[446]

  [446] Plutarch, Pyrrh. 1; Justin, xi. 3; Eurip. Androm. 1253;
  Arrian, Exp. Alexand. i. 11.

The glories of Ajax, the second grandson of Æakus, before Troy, are
surpassed only by those of Achilles. He perishes by his own hand,
the victim of an insupportable feeling of humiliation, because a
less worthy claimant is allowed to carry off from him the arms of
the departed Achilles. His son Philæus receives the citizenship of
Athens, and the gens or dême called Philaidæ traced up to him its name
and its origin: moreover the distinguished Athenians, Miltiadês and
Thucydidês, were regarded as members of this heroic progeny.[447]

  [447] Pherekydês and Hellanikus ap. Marcellin. Vit. Thucydid.
  init.; Pausan. ii. 29, 4; Plutarch, Solôn, 10. According to
  Apollodôrus, however, Pherekydês said that Telamôn was only the
  friend of Pêleus, not his brother,—not the son of Æakus (iii. 12,
  7): this seems an inconsistency. There was however a warm dispute
  between the Athenians and the Megarians respecting the title to
  the hero Ajax, who was claimed by both (see Pausan. i. 42, 4;
  Plutarch, _l. c._): the Megarians accused Peisistratus of having
  interpolated a line into the Catalogue in the Iliad (Strabo, ix.
  p. 394).

Teukrus escaped from the perils of the siege of Troy as well as from
those of the voyage homeward, and reached Salamis in safety. But
his father Telamôn, indignant at his having returned without Ajax,
refused to receive him, and compelled him to expatriate. He conducted
his followers to Cyprus, where he founded the city of Salamis: his
descendant Evagoras was recognized as a Teukrid and as an Æakid even in
the time of Isokratês.[448]

  [448] Herodot. vii. 90; Isokrat. Enc. Evag. _ut sup._; Sophokl.
  Ajax, 984-995; Vellei. Patercul. i. 1; Æschyl. Pers. 891, and
  Schol. The return from Troy of Teukrus, his banishment by
  Telamôn, and his settlement in Cyprus, formed the subject of the
  Τεῦκρος of Sophoklês, and of a tragedy under a similar title by
  Pacuvius (Cicero de Orat. i. 58; ii. 46); Sophokl. Ajax, 892;
  Pacuvii Fragm. Teucr. 15.—

    “Te repudio, nec recipio, natum abdico,
    Facesse.”

  The legend of Teukros was connected in Attic archæology with the
  peculiar functions and formalities of the judicature, ἐν Φρεαττοῖ
  (Pausan. i. 28, 12; ii. 29, 7).

Such was the splendid heroic genealogy of the Æakids,—a family renowned
for military excellence. The Æakeion at Ægina, in which prayer and
sacrifice were offered to Æakus, remained in undiminished dignity down
to the time of Pausanias.[449] This genealogy connects together various
eminent gentes in Achaia Phthiôtis, in Ægina, in Salamis, in Cyprus,
and amongst the Epirotic Molossians. Whether we are entitled to infer
from it that the island of Ægina was originally peopled by Myrmidones
from Achaia Phthiôtis, as O. Müller imagines,[450] I will not pretend
to affirm. These mythical pedigrees seem to unite together special
clans or gentes, rather than the bulk of any community—just as we know
that the Athenians generally had no part in the Æakid genealogy, though
certain particular Athenian families laid claim to it. The intimate
friendship between Achilles and the Opuntian hero Patroclus—and the
community of name and frequent conjunction between the Locrian Ajax,
son of Oïleus, and Ajax, son of Telamôn—connect the Æakids with Opus
and the Opuntian Locrians, in a manner which we have no farther means
of explaining. Pindar too represents Menœtius, father of Patroclus, as
son of Aktôr and Ægina, and therefore maternal brother of Æakus.[451]

  [449] Hesiod, Fragm. Düntz. Eoiai, 55, y. 43.—

    Ἀλκὴν μὲν γὰρ ἔδωκεν Ὀλύμπιος Αἰακίδαισι,
    Νοῦν δ᾽ Ἀμυθαονίδαις, πλοῦτον δ᾽ ἔπορ᾽ Ἀτρείδῃσι.

  Polyb. v. 2.—

    Αἰακίδας, πολέμῳ κεχαρηότας ἠΰτε δαιτί.

  [450] See his Æginetica, p. 14, his earliest work.

  [451] Pindar, Olymp. ix. 74. The hero Ajax, son of Oïleus, was
  especially worshipped at Opus; solemn festivals and games were
  celebrated in his honor.




CHAPTER XI.

ATTIC LEGENDS AND GENEALOGIES.


The most ancient name in Attic archæology, as far as our means of
information reach, is that of Erechtheus, who is mentioned both in
the Catalogue of the Iliad and in a brief allusion of the Odyssey.
Born of the Earth, he is brought up by the goddess Athênê, adopted
by her as her ward, and installed in her temple at Athens, where the
Athenians offer to him annual sacrifices. The Athenians are styled in
the Iliad, “the people of Erechtheus.”[452] This is the most ancient
testimony concerning Erechtheus, exhibiting him as a divine or heroic,
certainly a superhuman person, and identifying him with the primitive
germination (if I may use a term, the Grecian equivalent of which would
have pleased an Athenian ear) of Attic man. And he was recognized in
this same character, even at the close of the fourth century before the
Christian æra, by the Butadæ, one of the most ancient and important
Gentes at Athens, who boasted of him as their original ancestor: the
genealogy of the great Athenian orator Lykurgus, a member of this
family, drawn up by his son Abrôn, and painted on a public tablet in
the Erechtheion, contained as its first and highest name, Erechtheus,
son of Hêphæstos and the Earth. In the Erechtheion, Erechtheus
was worshipped conjointly with Athênê: he was identified with the
god Poseidôn, and bore the denomination of Poseidôn Erechtheus:
one of the family of the Butadæ, chosen among themselves by lot,
enjoyed the privilege and performed the functions of his hereditary
priest.[453] Herodotus also assigns the same earth-born origin to
Erechtheus:[454] but Pindar, the old poem called the Danais, Euripidês
and Apollodôrus—all name Erichthonius, son of Hêphæstos and the Earth,
as the being who was thus adopted and made the temple-companion of
Athênê, while Apollodôrus in another place identifies Erichthonius
with Poseidôn.[455] The Homeric scholiast treated Erechtheus and
Erichthonius as the same person under two names:[456] and since, in
regard to such mythical persons, there exists no other test of identity
of the subject except perfect similarity of the attributes, this seems
the reasonable conclusion.

  [452] Iliad, ii. 546. Odyss. vii. 81.—

      Οἳ δ᾽ ἄρ᾽ Ἀθήνας εἶχον ...
    Δῆμον Ἐρεχθῆος μεγαλήτορος, ὅν ποτ᾽ Ἀθήνη
    Θρέψε, Διὸς θυγάτηρ, τέκε δὲ ζείδωρος Ἄρουρα,
    Κὰδ δ᾽ ἐν Ἀθήνῃσ᾽ εἷσεν ἑῷ ἐνὶ πίονι νηῷ,
    Ἐνθάδε μιν ταύροισι καὶ ἀρνειοῖς ἱλάονται
    Κοῦροι Ἀθηναίων, περιτελλομένων ἐνιαυτῶν.

  [453] See the Life of Lykurgus, in Plutarch’s (I call it by
  that name, as it is always printed with his works) Lives of the
  Ten Orators, tom. iv. p. 382-384, Wytt. Κατῆγον δὲ τὸ γένος ἀπὸ
  τούτων καὶ Ἐρεχθέως τοῦ Γῆς καὶ Ἡφαίστου ... καὶ ἐστὶν αὐτὴ ἡ
  καταγωγὴ τοῦ γένους τῶν ἱερασαμένων του Ποσειδῶνος, etc. Ὃς τὴν
  ἱερωσύνην Ποσειδῶνος Ἐρεχθέως εἶχε (pp. 382, 383). Erechtheus
  Πάρεδρος of Athênê—Aristidês, Panathenaic. p. 184, with the
  Scholia of Frommel.

  Butês, the eponymus of the Butadæ, is the first priest of
  Poseidôn Erichthonius: Apollod. iii. 15, 1. So Kallias (Xenoph.
  Sympos. viii. 40), ἱερεὺς θεῶν τῶν ἀπ᾽ Ἐρεχθέως.

  [454] Herodot. viii. 55.

  [455] Harpokration, v. Αὐτοχθών. Ὁ δὲ Πίνδαρος καὶ ὁ τὴν Δαναΐδα
  πεποιηκὼς φασιν, Ἐριχθόνιον ἐξ Ἡφαίστου καὶ Γῆς φανῆναι.
  Euripidês, Ion. 21. Apollod. iii. 14, 6; 15, 1. Compare Plato,
  Timæus, c. 6.

  [456] Schol. ad Iliad, ii. 546, where he cites also Kallimachus
  for the story of Erichthonius. Etymologicon Magn. Ἐρεχθεύς.
  Plato (Kritias, c. 4) employs vague and general language to
  describe the agency of Hêphæstos and Athênê, which the old fable
  in Apollodôrus (iii. 14, 6) details in coarser terms. See Ovid,
  Metam. ii. 757.

We may presume, from the testimony of Homer, that the first and oldest
conception of Athens and its sacred acropolis places it under the
special protection, and represents it as the settlement and favorite
abode of Athênê, jointly with Poseidôn; the latter being the inferior,
though the chosen companion of the former, and therefore exchanging
his divine appellation for the cognomen of Erechtheus. But the country
called Attica, which, during the historical ages, forms one social and
political aggregate with Athens, was originally distributed into many
independent dêmes or cantons, and included, besides, various religious
clans or hereditary sects (if the expression may be permitted); that
is, a multitude of persons not necessarily living together in the same
locality, but bound together by an hereditary communion of sacred
rites, and claiming privileges, as well as performing obligations,
founded upon the traditional authority of divine persons for whom
they had a common veneration. Even down to the beginning of the
Peloponnesian war, the demots of the various Attic dêmes, though long
since embodied in the larger political union of Attica, and having no
wish for separation, still retained the recollection of their original
political autonomy. They lived in their own separate localities,
resorted habitually to their own temples, and visited Athens only
occasionally for private or political business, or for the great public
festivals. Each of these aggregates, political as well as religious,
had its own eponymous god or hero, with a genealogy more or less
extended, and a train of mythical incidents more or less copious,
attached to his name, according to the fancy of the local exegetes and
poets. The eponymous heroes Marathôn, Dekelus, Kolônus, or Phlius, had
each their own title to worship, and their own position as themes of
legendary narrative, independent of Erechtheus, or Poseidôn, or Athênê,
the patrons of the acropolis common to all of them.

But neither the archæology of Attica, nor that of its various component
fractions, was much dwelt upon by the ancient epic poets of Greece.
Theseus is noticed both in the Iliad and Odyssey as having carried
off from Krête Ariadnê, the daughter of Minos—thus commencing that
connection between the Krêtan and Athenian legends which we afterwards
find so largely amplified—and the sons of Thêseus take part in the
Trojan war.[457] The chief collectors and narrators of the Attic mythes
were, the prose logographers, authors of the many compositions called
Atthides, or works on Attic archæology. These writers—Hellanikus,
the contemporary of Herodotus, is the earliest composer of an
Atthis expressly named, though Pherekydês also touched upon the
Attic fables—these writers, I say, interwove into one chronological
series the legends which either greatly occupied their own fancy, or
commanded the most general reverence among their countrymen. In this
way the religious and political legends of Eleusis, a town originally
independent of Athens, but incorporated with it before the historical
age, were worked into one continuous sequence along with those of the
Erechtheids. In this way, Kekrops, the eponymous hero of the portion of
Attica called Kekropia, came to be placed in the mythical chronology at
a higher point even than the primitive god or hero Erechtheus.

  [457] Æthra, mother of Theseus, is also mentioned (Homer, Iliad,
  iii. 144).

Ogygês is said to have reigned in Attica[458] 1020 years before the
first Olympiad, or 1796 years B. C. In his time happened the deluge
of Deukaliôn, which destroyed most of the inhabitants of the country:
after a long interval, Kekrops, an indigenous person, half man and
half serpent, is given to us by Apollodôrus as the first king of the
country: he bestowed upon the land, which had before been called
Actê, the name of Kekropia. In his day there ensued a dispute between
Athênê and Poseidôn respecting the possession of the acropolis at
Athens, which each of them coveted. First, Poseidôn struck the rock
with his trident, and produced the well of salt water which existed
in it, called the Erechthêis: next came Athênê, who planted the
sacred olive-tree ever afterwards seen and venerated in the portion
of Erechtheion called the cell of Pandrosus. The twelve gods decided
the dispute; and Kekrops having testified before them that Athênê
had rendered this inestimable service, they adjudged the spot to
her in preference to Poseidôn. Both the ancient olive-tree and the
well produced by Poseidôn were seen on the acropolis, in the temple
consecrated jointly to Athênê and Erechtheus, throughout the historical
ages. Poseidôn, as a mark of his wrath for the preference given to
Athênê, inundated the Thriasian plain with water.[459]

  [458] Hellanikus, Fragm. 62; Philochor. Fragm. 8, ap. Euseb.
  Præp. Evang. x. 10. p. 489. Larcher (Chronologie d’Hérodote, ch.
  ix. s. 1. p. 278) treats both the historical personality and the
  date of Ogygês as perfectly well authenticated.

  It is not probable that Philochorus should have given any
  calculation of time having reference to Olympiads; and hardly
  conceivable that Hellanikus should have done so. Justin Martyr
  quotes Hellanikus and Philochorus as having mentioned Moses,—ὡς
  σφόδρα ἀρχαίου καὶ παλαιοῦ τῶν Ἰουδαίων ἄρχοντος Μωϋσέως
  μέμνηνται—which is still more incredible even than the assertion
  of Eusebius about their having fixed the date of Ogygês by
  Olympiads (see Philochor. Fragm. 9).

  [459] Apollod. iii. 14, 1; Herodot. viii. 55; Ovid. Metam. vi.
  72. The story current among the Athenians represented Kekrops as
  the judge of this controversy (Xenoph. Memor. iii. 5, 10).

  The impressions of the trident of Poseidôn were still shown
  upon the rock in the time of Pausanias (Pausan. i. 26, 4). For
  the sanctity of the ancient olive-tree, see the narrative of
  Herodotus (_l. c._), relating what happened to it when Xerxes
  occupied the acropolis. As this tale seems to have attached
  itself specially to the local peculiarities of the Erechtheion,
  the part which Poseidôn plays in it is somewhat mean: that god
  appears to greater advantage in the neighborhood of the Ἱπποτὴς
  Κολωνὸς, as described in the beautiful Chorus of Sophoklês (Œdip.
  Colon. 690-712).

  A curious rationalization of the monstrous form ascribed to
  Kekrops διφυὴς in Plutarch (Sera Num. Vindict. p. 551).

During the reign of Kekrops, Attica was laid waste by Karian pirates
on the coast, and by invasions of the Aônian inhabitants from Bœôtia.
Kekrops distributed the inhabitants of Attica into twelve local
sections—Kekropia, Tetrapolis, Epakria, Dekeleia, Eleusis, Aphidna,
Thorikus, Braurôn, Kythêrus, Sphêttus, Kêphisius, Phalerus. Wishing to
ascertain the number of inhabitants, he commanded each man to cast a
single stone into a general heap: the number of stones was counted, and
it was found that there were twenty thousand.[460]

  [460] Philochor. ap. Strabo. ix. p. 397.

Kekrops married the daughter of Aktæus, who (according to Pausanias’s
version) had been king of the country before him, and had called it by
the name of Aktæa.[461] By her he had three daughters, Aglaurus, Ersê
and Pandrosus, and a son, Erysichthôn. Kekrops is called by Pausanias
contemporary of the Arcadian Lykaôn, and is favorably contrasted with
that savage prince in respect of his piety and humanity.[462] Though
he has been often designated in modern histories as an immigrant from
Egypt into Attica, yet the far greater number of ancient authorities
represent him as indigenous or earth-born.[463]

  [461] The Parian chronological marble designates Aktæus as an
  autochthonous person. Marmor Parium, Epoch. 3. Pausan. i. 2, 5.
  Philochorus treated Aktæus as a fictitious name (Fragm. 8, _ut
  sup._).

  [462] Pausan. viii. 2. 2. The three daughters of Kekrops were
  not unnoticed in the mythes (Ovid, Metam. ii. 739): the tale of
  Kephalus, son of Hersê by Hermês, who was stolen away by the
  goddess Eôs or Hêmera in consequence of his surpassing beauty,
  was told in more than one of the Hesiodic poems (Pausan. i. 3, 1;
  Hesiod. Theog. 986). See also Eurip. Ion. 269.

  [463] Jul. Africanus also (ap. Euseb. x. 9. p. 486-488) calls
  Kekrops γηγενὴς and αὐτοχθών.

Erysichthôn died without issue, and Kranaus succeeded him,—another
autochthonous person and another eponymus,—for the name Kranai was
an old denomination of the inhabitants of Attica.[464] Kranaus was
dethroned by Amphiktyôn, by some called an autochthonous man; by
others, a son of Deukaliôn: Amphiktyôn in his turn was expelled
by Erichthonius, son of Hêphæstos and the Earth,—the same person
apparently as Erechtheus, but inserted by Apollodôrus at this point of
the series. Erichthonius, the pupil and favored companion of Athênê,
placed in the acropolis the original Palladium or wooden statue of
that goddess, said to have dropped from heaven: he was moreover the
first to celebrate the festival of the Panathenæa. He married the nymph
Pasithea, and had for his son and successor Pandiôn.[465] Erichthonius
was the first person who taught the art of breaking in horses to the
yoke, and who drove a chariot and four.[466]

  [464] Herod. viii. 44. Κρανααὶ Ἀθῆναι, Pindar.

  [465] Apollod. iii. 14. Pausan. i. 26, 7.

  [466] Virgil, Georgic iii. 114.

In the time of Pandiôn, who succeeded to Erichthonius, Dionysus and
Dêmêtêr both came into Attica: the latter was received by Keleos at
Eleusis.[467] Pandiôn married the nymph Zeuxippê, and had twin sons,
Erechtheus and Butês, and two daughters, Proknê and Philomêla. The two
latter are the subjects of a memorable and well-known legend. Pandiôn
having received aid in repelling the Thêbans from Têreus, king of
Thrace, gave him his daughter Proknê in marriage, by whom he had a son,
Itys. The beautiful Philomêla, going to visit her sister, inspired
the barbarous Thracian with an irresistible passion: he violated her
person, confined her in a distant pastoral hut, and pretended that she
was dead, cutting out her tongue to prevent her from revealing the
truth. After a long interval, Philomêla found means to acquaint her
sister of the cruel deed which had been perpetrated; she wove into a
garment words describing her melancholy condition, and despatched it
by a trusty messenger. Proknê, overwhelmed with sorrow and anger, took
advantage of the free egress enjoyed by women during the Bacchanalian
festival to go and release her sister: the two sisters then revenged
themselves upon Têreus by killing the boy Itys, and serving him up
for his father to eat: after the meal had been finished, the horrid
truth was revealed to him. Têreus snatched a hatchet to put Proknê to
death: she fled, along with Philomêla, and all the three were changed
into birds—Proknê became a swallow, Philomêla a nightingale, and
Têreus an hoopoe.[468] This tale, so popular with the poets, and so
illustrative of the general character of Grecian legend, is not less
remarkable in another point of view—that the great historian Thucydidês
seems to allude to it as an historical fact,[469] not however directly
mentioning the final metamorphosis.

  [467] The mythe of the visit of Dêmêtêr to Eleusis, on which
  occasion she vouchsafed to teach her holy rites to the leading
  Eleusinians, is more fully touched upon in a previous chapter
  (see _ante_, p. 50).

  [468] Apollod. iii. 14, 8; Æsch. Supplic. 61; Soph. Elektr. 107;
  Ovid, Metamorph. vi. 425-670. Hyginus gives the fable with some
  additional circumstances, fab. 45. Antoninus Liberalis (Narr.
  11), or Bœus, from whom he copies, has composed a new narrative
  by combining together the names of Pandareos and Aêdôn, as given
  in the Odyssey, xix. 523, and the adventures of the old Attic
  fable. The hoopoe still continued the habit of chasing the
  nightingale; it was to the Athenians a present fact. See Schol.
  Aristoph. Aves, 212.

  [469] Thucyd. ii. 29. He makes express mention of the nightingale
  in connection with the story, though not of the metamorphosis.
  See below, chap. xvi. p. 544, note 2. So also does Pausanias
  mention and reason upon it as a real incident: he founds upon it
  several moral reflections (i. 5, 4; x. 4, 5): the author of the
  Λόγος Ἐπιτάφιος, ascribed to Demosthenês, treats it in the same
  manner, as a fact ennobling the tribe Pandionis, of which Pandiôn
  was the eponymus. The same author, in touching upon Kekrops, the
  eponymus of the Kekropis tribe, cannot believe literally the
  story of his being half man and half serpent: he rationalizes it
  by saying that Kekrops was so called because in wisdom he was
  like a man, in strength like a serpent (Demosth. p. 1397, 1398,
  Reiske). Hesiod glances at the fable (Opp. Di. 566), ὀρθογόη
  Πανδιονὶς ὦρτο χελιδών; see also Ælian., V. H. xii. 20. The
  subject was handled by Sophoklês in his lost Têreus.

After the death of Pandiôn, Erechtheus succeeded to the kingdom,
and his brother, Butês, became priest of Poseidôn Erichthonius, a
function which his descendants ever afterwards exercised, the Butadæ
or Eteobutadæ. Erechtheus seems to appear in three characters in the
fabulous history of Athens—as a god, Poseidôn Erechtheus[470]—as a
hero, Erechtheus, son of the Earth—and now, as a king, son of Pandiôn:
so much did the ideas of divine and human rule become confounded and
blended together in the imagination of the Greeks in reviewing their
early times.

  [470] Poseidôn is sometimes spoken of under the name of
  Erechtheus simply (Lycophrôn, 158). See Hesychius, v. Ἐρεχθεύς.

The daughters of Erechtheus were not less celebrated in Athenian legend
than those of Pandiôn. Prokris, one of them, is among the heroines seen
by Odysseus in Hadês: she became the wife of Kephalus, son of Deionês,
and lived in the Attic dême of Thorikus. Kephalus tried her fidelity
by pretending that he was going away for a long period; but shortly
returned, disguising his person and bringing with him a splendid
necklace. He presented himself to Prokris without being recognized, and
succeeded in triumphing over her chastity. Having accomplished this
object, he revealed to her his true character: she earnestly besought
his forgiveness, and prevailed upon him to grant it. Nevertheless he
became shortly afterwards the unintentional author of her death: for
he was fond of hunting, and staid out a long time on his excursions,
so that Prokris suspected him of visiting some rival. She determined
to watch him by concealing herself in a thicket near the place of his
midday repose; and when Kephalus implored the presence of Nephelê (a
cloud) to protect him from the sun’s rays, she suddenly started from
her hiding-place: Kephalus, thus disturbed, cast his hunting-spear
unknowingly into the thicket and slew his wife. Erechtheus interred her
with great magnificence, and Kephalus was tried for the act before the
court of Areopagus, which condemned him to exile.[471]

  [471] Pherekydês, Fragm. 77, Didot; ap. Schol. ad Odyss. xi. 320;
  Hellanikus Fr. 82; ap. Schol. Eurip. Orest. 1648. Apollodôrus
  (iii. 15, 1) gives the story differently.

Kreüsa, another daughter of Erechtheus, seduced by Apollo, becomes the
mother of Iôn, whom she exposes immediately after his birth in the cave
north of the acropolis, concealing the fact from every one. Apollo
prevails upon Hermês to convey the new-born child to Delphi, where he
is brought up as a servant of the temple, without knowing his parents.
Kreüsa marries Xuthus, son of Æolus, but continuing childless, she goes
with Xuthus to the Delphian oracle to inquire for a remedy. The god
presents to them Iôn, and desires them to adopt him as their son: their
son Achæus is afterwards born to them, and Iôn and Achæus become the
eponyms of the Iônians and Achæans.[472]

  [472] Upon this story of Iôn is founded the tragedy of Euripidês
  which bears that name. I conceive many of the points of that
  tragedy to be of the invention of Euripidês himself: but to
  represent Iôn as son of Apollo, not of Xuthus, seems a genuine
  Attic legend. Respecting this drama, see O. Müller, Hist. of
  Dorians, ii. 2. 13-15. I doubt however the distinction which he
  draws between the Ionians and the other population of Attica.

Oreithyia, the third daughter of Erechtheus, was stolen away by the god
Boreas while amusing herself on the banks of the Ilissus, and carried
to his residence in Thrace. The two sons of this marriage, Zêtês
and Kalaïs, were born with wings: they took part in the Argonautic
expedition, and engaged in the pursuit of the Harpies: they were slain
at Tênos by Hêraklês. Kleopatra, the daughter of Boreas and Oreithyia,
was married to Phineus, and had two sons, Plexippus and Pandiôn; but
Phineus afterwards espoused a second wife, Idæa, the daughter of
Dardanus, who, detesting the two sons of the former bed, accused them
falsely of attempting her chastity, and persuaded Phineus in his wrath
to put out the eyes of both. For this cruel proceeding he was punished
by the Argonauts in the course of their voyage.[473]

  [473] Apollodôr. iii. 15, 2; Plato, Phædr. c. 3; Sophok. Antig.
  984; also the copious Scholion on Apollôn. Rhod. i. 212.

  The tale of Phineus is told very differently in the Argonautic
  expedition as given by Apollônius Rhodius, ii. 180. From
  Sophoklês we learn that this was the Attic version.

  The two winged sons of Boreas and their chase of the Harpies were
  noticed in the Hesiodic Catalogue (see Schol. Apollôn. Rhod. ii.
  296). But whether the Attic legend of Oreithyia was recognized in
  the Hesiodic poems seems not certain.

  Both Æschylus and Sophoklês composed dramas on the subject of
  Oreithyia (Longin. de Sublimit. c. 3). “Orithyia Atheniensis,
  filia Terrigenæ, et a Borea in Thraciam rapta.” (Servius ad Virg.
  Æneid. xii. 83). Terrigenæ is the γηγενὴς Ἐρεχθεύς. Philochorus
  (Fragm. 30) rationalized the story, and said that it alluded to
  the effects of a violent wind.

On more than one occasion the Athenians derived, or at least
believed themselves to have derived, important benefits from this
marriage of Boreas with the daughter of their primæval hero: one
inestimable service, rendered at a juncture highly critical for
Grecian independence, deserves to be specified.[474] At the time of
the invasion of Greece by Xerxês, the Grecian fleet was assembled at
Chalcis and Artemision in Eubœa, awaiting the approach of the Persian
force, so overwhelming in its numbers as well by sea as on land. The
Persian fleet had reached the coast of Magnêsia and the south-eastern
corner of Thessaly without any material damage, when the Athenians
were instructed by an oracle “to invoke the aid of their son-in-law.”
Understanding the advice to point to Boreas, they supplicated his
aid and that of Oreithyia, most earnestly, as well by prayer as by
sacrifice,[475] and the event corresponded to their wishes. A furious
north-easterly wind immediately arose, and continued for three days
to afflict the Persian fleet as it lay on an unprotected coast: the
number of ships driven ashore, both vessels of war and of provision,
was immense, and the injury done to the armament was never thoroughly
repaired. Such was the powerful succor which the Athenians derived, at
a time of their utmost need, from their son-in-law Boreas; and their
gratitude was shown by consecrating to him a new temple on the banks of
the Ilissus.

  [474] Herodot. vii. 189. Οἱ δ᾽ ὦν Ἀθηναῖοί σφι λέγουσι βοηθήσαντα
  τὸν Βορῆν πρότερον, καὶ τότε ἐκεῖνα κατεργάσασθαι· καὶ ἱρὸν
  ἀπελθόντες Βορέω ἱδρύσαντο παρὰ ποταμὸν Ἴλισσον.

  [475] Herodot. _l. c._ Ἀθηναῖοι τὸν Βορῆν ἐκ θεοπροπίου
  ἐπεκαλέσαντο, ἐλθόντος σφι ἄλλου χρηστηρίου, τὸν γαμβρὸν
  ἐπίκουρον καλέσασθαι. Βορῆς δὲ, κατὰ τὸν Ἑλλήνων λόγον ἔχει
  γυναῖκα Ἀττικὴν, Ὠρειθυίην τὴν Ἐρεχθῆος. Κατὰ δὴ τὸ κῆδος τοῦτο,
  οἱ Ἀθηναῖοι, συμβαλλεόμενοί σφι τὸν Βορῆν γαμβρὸν εἶναι, etc.

The three remaining daughters of Erechtheus—he had six in all[476]—were
in Athenian legend yet more venerated than their sisters, on account of
having voluntarily devoted themselves to death for the safety of their
country. Eumolpus of Eleusis was the son of Poseidôn and the eponymous
hero of the sacred gens called the Eumolpids, in whom the principal
functions, appertaining to the mysterious rites of Dêmêtêr at Eleusis,
were vested by hereditary privilege: he made war upon Erechtheus and
the Athenians, with the aid of a body of Thracian allies; indeed it
appears that the legends of Athens, originally foreign and unfriendly
to those of Eleusis, represented him as having been himself a Thracian
born and an immigrant into Attica.[477] Respecting Eumolpus however
and his parentage, the discrepancies much exceed even the measure of
license usual in the legendary genealogies, and some critics, both
ancient and modern, have sought to reconcile these contradictions
by the usual stratagem of supposing two or three different persons
of the same name. Even Pausanias, so familiar with this class of
unsworn witnesses, complains of the want of native Eleusinian
genealogists,[478] and of the extreme license of fiction in which other
authors had indulged.

  [476] Suidas and Photius, v. Πάρθενοι: Protogeneia and Pandôra
  are given as the names of two of them. The sacrifice of Pandôra,
  in the Iambi of Hippônax (Hippônact. Fragm. xxi. Welck. ap.
  Athen. ix. p. 370), seems to allude to this daughter of
  Erechtheus.

  [477] Apollodôr. iii. 15, 3; Thucyd. ii. 15; Isokratês (Panegyr.
  t. i. p. 206; Panathenaic. t. ii. p. 560, Auger), Lykurgus,
  cont. Leocrat. p. 201, Reiske, Pausan. i. 38, 3; Euripid.
  Erechth. Fragm. The Schol. ad. Soph. Œd. Col. 1048 gives valuable
  citations from Ister, Akestodôrus and Androtiôn: we see that
  the inquirers of antiquity found it difficult to explain how
  the Eumolpids could have acquired their ascendant privileges in
  the management of the Eleusinia, seeing that Eumolpus himself
  was a foreigner.—Ζητεῖται, τί δήποτε οἱ Εὐμολπίδαι τῶν τελετῶν
  ἐξάρχουσι, ξένοι ὄντες. Thucydidês does _not_ call Eumolpus
  a Thracian: Strabo’s language is very large and vague (vii.
  p. 321): Isokratês says that he assailed Athens in order to
  vindicate the rights of his father Poseidôn to the sovereign
  patronage of the city. Hyginus copies this (fab. 46).

  [478] Pausan. i. 38. 3. Ἐλευσίνιοί τε ἀρχαῖοι, ἅτε οὐ προσόντων
  σφισι γενεαλόγων, ἄλλα τε πλάσασθαι δεδώκασι καὶ μάλιστα ἐς τὰ
  γένη τῶν ἡρώων. See Heyne ad Apollodôr. iii. 15, 4. “Eumolpi
  nomen modo communicatum pluribus, modo plurium hominum res et
  facta cumulata in unum. Is ad quem Hercules venisse dicitur,
  serior ætate fuit: antiquior est is de quo hoc loco agitur ...
  antecessisse tamen hunc debet alius, qui cum Triptolemo vixit,”
  etc. See the learned and valuable comments of Lobeck in his
  Aglaophamus, tom. i. p. 206-213: in regard to the discrepancies
  of this narrative he observes, I think, with great justice (p.
  211), “quo uno exemplo ex innumerabilibus delecto, arguitur
  eorum temeritas, qui ex variis discordibusque poetarum et
  mythographorum narratiunculis, antiquæ famæ formam et quasi
  lineamenta recognosci posse sperant.”

In the Homeric Hymn to Dêmêtêr, the most ancient testimony before
us,—composed, to all appearance, earlier than the complete
incorporation of Eleusis with Athens,—Eumolpus appears (to repeat
briefly what has been stated in a previous chapter) as one of the
native chiefs or princes of Eleusis, along with Triptolemus, Dioklês,
Polyxeinus and Dolichus: Keleos is the king, or principal among these
chiefs, the son or lineal descendant of the eponymous Eleusis himself.
To these chiefs, and to the three daughters of Keleos, the goddess
Dêmêtêr comes in her sorrow for the loss of her daughter Persephonê:
being hospitably entertained by Keleos she reveals her true character,
commands that a temple shall be built to her at Eleusis, and prescribes
to them the rites according to which they are to worship her.[479] Such
seems to have been the ancient story of the Eleusinians respecting
their own religious antiquities: Keleos, with Metaneira his wife, and
the other chiefs here mentioned, were worshipped at Eleusis, and from
thence transferred to Athens as local gods or heroes.[480] Eleusis
became incorporated with Athens, apparently not very long before the
time of Solôn; and the Eleusinian worship of Dêmêtêr was then received
into the great religious solemnities of the Athenian state, to which
it owes its remarkable subsequent extension and commanding influence.
In the Atticized worship of the Eleusinian Dêmêtêr, the Eumolpids and
the Kêrŷkes were the principal hereditary functionaries: Eumolpus, the
eponym of this great family, came thus to play the principal part in
the Athenian legendary version of the war between Athens and Eleusis.
An oracle had pronounced that Athens could only be rescued from his
attack by the death of the three daughters of Erechtheus; their
generous patriotism consented to the sacrifice, and their father put
them to death. He then went forth confidently to the battle, totally
vanquished the enemy, and killed Eumolpus with his own hand.[481]
Erechtheus was worshipped as a god, and his daughters as goddesses,
at Athens.[482] Their names and their exalted devotion were cited
along with those of the warriors of Marathôn, in the public assembly
of Athens, by orators who sought to arouse the languid patriot, or to
denounce the cowardly deserter; and the people listened both to one and
the other with analogous feelings of grateful veneration, as well as
with equally unsuspecting faith in the matter of fact.[483]

  [479] Homer, Hymn, ad Cerer. 153-475.—

         ... Ἡ δὲ κίουσα θεμιστοπόλοις βασιλεῦσι
    Δεῖξεν Τριπτολέμῳ τε, Διόκλεΐ τε πληξίππῳ,
    Εὐμόλπου τε βίῃ, Κελέῳ θ᾽ ἡγήτορι λαῶν,
    Δρησμοσύνην ἱερῶν.

  Also v. 105.

    Τὴν δὲ ἴδον Κελέοιο Ἐλευσινίδαο θύγατρες.

  The hero Eleusis is mentioned in Pausanias, i. 38, 7: some said
  that he was the son of Hermês, others that he was the son of
  Ogygus. Compare Hygin. f. 147.

  [480] Keleos and Metaneira were worshipped by the Athenians
  with divine honors (Athenagoras, Legat. p. 53, ed. Oxon.):
  perhaps he confounds divine and heroic honors, as the Christian
  controversialists against Paganism were disposed to do.
  Triptolemus had a temple at Eleusis (Pausan. i. 38, 6).

  [481] Apollodôr. iii. 15, 4. Some said that Immaradus, son of
  Eumolpus, had been killed by Erechtheus (Pausan. i. 5, 2);
  others, that both Eumolpus and his son had experienced this fate
  (Schol. ad Eurip. Phœniss. 854). But we learn from Pausanias
  himself what the story in the interior of the Erechtheion
  was,—that Erechtheus killed Eumolpus (i. 27, 3).

  [482] Cicero, Nat. Deor. iii. 19; Philochor. ap. Schol. Œdip.
  Col. 100. Three daughters of Erechtheus perished, and three
  daughters were worshipped (Apollodôr. iii. 15, 4; Hesychius,
  Ζεῦγος τριπάρθενον; Eurip. Erechtheus, Fragm. 3, Dindorf); but
  both Euripidês and Apollodôrus said that Erechtheus was only
  required to sacrifice, and only did sacrifice, _one_,—the other
  two slew themselves voluntarily, from affection for their sister.
  I cannot but think (in spite of the opinion of Welcker to the
  contrary, Griechisch. Tragöd. ii. p. 722) that the genuine legend
  represented Erechtheus as having sacrificed all three, as appears
  in the Iôn of Euripidês (276):—

       IÔN. Πατὴρ Ἐρεχθεὺς σὰς ἔθυσε συγγόνους;
    CREÜSA. Ἔτλη πρὸ γαίας σφάγια παρθένους κτανεῖν.
       IÔN. Σὺ δ᾽ ἐξεσώθης πῶς κασιγνήτων μόνη;
    CREÜSA. Βρέφος νέογνον μητρὸς ἦν ἐν ἀγκάλαις.

  Compare with this passage, Demosthen. Λόγος Ἐπιταφ. p. 1397,
  Reisk. Just before, the death of the three daughters of Kekrops,
  for infringing the commands of Athênê, had been mentioned.
  Euripidês modified this in his Erechtheus, for he there
  introduced the mother Praxithea consenting to the immolation
  of one daughter, for the rescue of the country from a foreign
  invader: to propose to a mother the immolation of three daughters
  at once, would have been too revolting. In most instances we find
  the strongly marked features, the distinct and glaring incidents
  as well as the dark contrasts, belong to the Hesiodic or old
  Post-Homeric legend; the changes made afterwards go to soften,
  dilute, and to complicate, in proportion as the feelings of the
  public become milder and more humane; sometimes however the later
  poets add new horrors.

  [483] See the striking evidence contained in the oration of
  Lykurgus against Leocratês (p. 201-204. Reiske; Demosthen. Λόγ.
  Ἐπιταφ. _l. c._; and Xenophon, Memor. iii. 5, 9): from the two
  latter passages we see that the Athenian story represented the
  invasion under Eumolpus as a combined assault from the western
  continent.

Though Erechtheus gained the victory over Eumolpus, yet the story
represents Poseidôn as having put an end to the life and reign of
Erechtheus, who was (it seems) slain in the battle. He was succeeded
by his son Kekrops II., and the latter again by his son Pandiôn
II.,[484]—two names unmarked by any incidents, and which appear to be
mere duplication of the former Kekrops and Pandiôn, placed there by
the genealogizers for the purpose of filling up what seemed to them a
chronological chasm. The Attic legends were associated chiefly with
a few names of respected eponymous personages; and if the persons
called the children of Pandiôn were too numerous to admit of their
being conveniently ascribed to one father, there was no difficulty in
supposing a second prince of the same name.

  [484] Apollodôr. iii. 15, 5; Eurip. Iôn, 282; Erechth. Fragm. 20,
  Dindorf.

Apollodôrus passes at once from Erechtheus to his son Kekrops II., then
to Pandiôn II., next to the four sons of the latter, Ægeus, Pallas,
Nisus and Lykus. But the tragedians here insert the story of Xuthus,
Kreüsa and Iôn; the latter being the son of Kreüsa by Apollo, but given
by the god to Xuthus, and adopted by the latter as his own. Iôn becomes
the successor of Erechtheus, and his sons Teleon, Hoplês, Argadês and
Aigikorês become the eponyms of the four ancient tribes of Athens,
which subsisted until the revolution of Kleisthenês. Iôn himself is
the eponym of the Iônic race both in Asia, in Europe, and in the Ægean
islands: Dôrus and Achæus are the sons of Kreüsa by Xuthus, so that Iôn
is distinguished from both of them by being of divine parentage.[485]
According to the story given by Philochorus, Iôn rendered such
essential service in rescuing the Athenians from the attack of the
Thracians under Eumolpus, that he was afterwards made king of the
country, and distributed all the inhabitants into four tribes or
castes, corresponding to different modes of life,—soldiers, husbandmen,
goatherds, and artisans.[486] And it seems that the legend explanatory
of the origin of the festival Boëdromia, originally important enough to
furnish a name to one of the Athenian months, was attached to the aid
thus rendered by Iôn.[487]

  [485] Eurip. Iôn. 1570-1595. The Kreüsa of Sophoklês, a lost
  tragedy, seems to have related to the same subject.

  Pausanias (vii. 1, 2) tells us that Xuthus was chosen to
  arbitrate between the contending claims of the sons of Erechtheus.

  [486] Philochor. ap. Harpocrat. v. Βοηδρόμια; Strabo, viii. p.
  383.

  [487] Philochor. ap. Harpocrat. v. Βοηδρόμια.

We pass from Iôn to persons of far greater mythical dignity and
interest,—Ægeus and his son Thêseus.

Pandiôn had four sons, Ægeus, Nisus, Lykus, and Pallas, between whom
he divided his dominions. Nisus received the territory of Megaris,
which had been under the sway of Pandiôn, and there founded the seaport
of Nisæa. Lykus was made king of the eastern coast, but a dispute
afterwards ensued, and he quitted the country altogether, to establish
himself on the southern coast of Asia Minor among the Termilæ, to
whom he gave the name of Lykians.[488] Ægeus, as the eldest of the
four, became king of Athens; but Pallas received a portion both of the
south-western coast and the interior, and he as well as his children
appear as frequent enemies both to Ægeus and to Thêseus. Pallas is
the eponym of the dême Pallênê, and the stories respecting him and
his sons seem to be connected with old and standing feuds among the
different dêmes of Attica, originally independent communities. These
feuds penetrated into the legend, and explain the story which we find
that Ægeus and Thêseus were not genuine Erechtheids, the former being
denominated a supposititious child to Pandiôn.[489]

  [488] Sophokl. ap. Strab. ix. p. 392; Herodot. i. 173; Strabo,
  xii. p. 573.

  [489] Plutarch, Thêseus, c. 13. Αἰγεὺς θετὸς γενόμενος Πανδίονι,
  καὶ μηδὲν τοῖς Ἐρεχθείδαις προσήκων. Apollodôr. iii. 15, 6.

Ægeus[490] has little importance in the mythical history except as
the father of Thêseus: it may even be doubted whether his name is
anything more than a mere cognomen of the god Poseidôn, who was (as we
are told) the real father of this great Attic Hêraklês. As I pretend
only to give a very brief outline of the general territory of Grecian
legend, I cannot permit myself to recount in detail the chivalrous
career of Thêseus, who is found both in the Kalydônian boar-hunt and
in the Argonautic expedition—his personal and victorious encounters
with the robbers Sinnis, Procrustês, Periphêtês, Scirôn and others—his
valuable service in ridding his country of the Krommyonian sow and
the Marathônian bull—his conquest of the Minotaur in Krête, and his
escape from the dangers of the labyrinth by the aid of Ariadnê, whom he
subsequently carries off and abandons—his many amorous adventures, and
his expeditions both against the Amazons and into the under-world along
with Peirithous.[491]

  [490] Ægeus had by Mêdea (who took refuge at Athens after her
  flight from Corinth) a son named Mêdus, who passed into Asia,
  and was considered as the eponymus and progenitor of the Median
  people. Datis, the general who commanded the invading Persian
  army at the battle of Marathôn, sent a formal communication to
  the Athenians announcing himself as the descendant of Mêdus, and
  requiring to be admitted as king of Attica: such is the statement
  of Diodôrus (Exc. Vatic. vii.-x. 48: see also Schol. Aristophan.
  Pac. 289).

  [491] Ovid, Metamorph. vii. 433.—

                 ... “Te, maxime Theseu,
    Mirata est Marathon Cretæi sanguine Tauri:
    Quodque Suis securus arat Cromyona colonus,
    Munus opusque tuum est. Tellus Epidauria per te
    Clavigeram vidit Vulcani occumbere prolem:
    Vidit et immanem Cephisias ora Procrustem.
    Cercyonis letum vidit Cerealis Eleusin.
    Occidit ille Sinis,” etc.

  Respecting the amours of Thêseus, Ister especially seems to have
  entered into great details; but some of them were noticed both
  in the Hesiodic poems and by Kekrops, not to mention Pherekydês
  (Athen. xiii. p. 557). Peirithous, the intimate friend and
  companion of Thêseus, is the eponymous hero of the Attic dême or
  gens Perithoidæ (Ephorus ap. Photium, v. Περιθοῖδαι).

Thucydidês delineates the character of Thêseus as a man who combined
sagacity with political power, and who conferred upon his country the
inestimable benefit of uniting all the separate and self-governing
dêmes of Attica into one common political society.[492] From the
well-earned reverence attached to the assertion of Thucydidês, it
has been customary to reason upon this assertion as if it were
historically authentic, and to treat the romantic attributes which we
find in Plutarch and Diodôrus as if they were fiction superinduced
upon this basis of fact. Such a view of the case is in my judgment
erroneous. The athletic and amorous knight-errant is the old version
of the character—the profound and long-sighted politician is a
subsequent correction, introduced indeed by men of superior mind, but
destitute of historical warranty, and arising out of their desire to
find reasons of their own for concurring in the veneration which the
general public paid more easily and heartily to their national hero.
Thêseus, in the Iliad and Odyssey, fights with the Lapithæ against the
Centaurs: Thêseus, in the Hesiodic poems, is misguided by his passion
for the beautiful Æglê, daughter of Panopeus:[493] and the Thêseus
described in Plutarch’s biography is in great part a continuation
and expansion of these same or similar attributes, mingled with many
local legends, explaining, like the Fasti of Ovid, or the lost Aitia
of Kallimachus, the original genesis of prevalent religious and social
customs.[494] Plutarch has doubtless greatly softened down and modified
the adventures which he found in the Attic logographers as well as
in the poetical epics called Thêsêis. For in his preface to the life
of Thêseus, after having emphatically declared that he is about to
transcend the boundary both of the known and the knowable, but that the
temptation of comparing the founder of Athens with the founder of Rome
is irresistible, he concludes with the following remarkable words: “I
pray that this fabulous matter may be so far obedient to my endeavors
as to receive, when purified by reason, the aspect of history: in those
cases where it haughtily scorns plausibility and will admit no alliance
with what is probable, I shall beg for indulgent hearers, willing to
receive antique narrative in a mild spirit.”[495] We see here that
Plutarch sat down, not to recount the old fables as he found them, but
to purify them by reason and to impart to them the aspect of history.
We have to thank him for having retained, after this purification,
so much of what is romantic and marvellous; but we may be sure that
the sources from which he borrowed were more romantic and marvellous
still. It was the tendency of the enlightened men of Athens, from
the days of Solôn downwards, to refine and politicize the character
of Thêseus:[496] even Peisistratus expunged from one of the Hesiodic
poems the line which described the violent passion of the hero for the
fair Æglê:[497] and the tragic poets found it more congenial to the
feelings of their audience to exhibit him as a dignified and liberal
sovereign, rather than as an adventurous single-handed fighter. But the
logographers and the Alexandrine poets remained more faithful to the
old fables. The story of Hekalê, the hospitable old woman who received
and blessed Thêseus when he went against the Marathônian bull, and whom
he found dead when he came back to recount the news of his success, was
treated by Kallimachus:[498] and Virgil must have had his mind full of
the unrefined legends when he numbered this Attic Hêraklês among the
unhappy sufferers condemned to endless penance in the under-world.[499]

  [492] Thuc. ii. 15. Ἐπειδὴ δὲ Θησεὺς ἐβασίλευσε, γενόμενος μετὰ
  τοῦ ξυνετοῦ καὶ δυνατὸς, τά τε ἄλλα διεκόσμησε τὴν χώραν, καὶ
  κατάλυσας τῶν ἄλλων πόλεων τά τε βουλευτήρια καὶ τὰς ἀρχὰς, ἐς
  τὴν νῦν πόλιν ... ξυνῴκισε πάντας.

  [493] Iliad, i. 265; Odyss. xi. 321. I do not notice the
  suspected line, Odyss. xi. 630.

  [494] Diodôrus also, from his disposition to assimilate Thêseus
  to Hêraklês, has given us his chivalrous as well as his political
  attributes (iv. 61).

  [495] Plutarch, Thêseus, i. Εἴη μὲν οὖν ἡμῖν, ἐκκαθαιρόμενον
  λόγῳ τὸ μυθῶδες ὑπακοῦσαι καὶ λαβεῖν ἱστορίας ὄψιν· ὅπου δ᾽ ἂν
  αὐθαδῶς τοῦ πιθανοῦ περιφρονῇ, καὶ μὴ δέχηται ~τὴν πρὸς τὸ εἰκὸς
  μίξιν~, εὐγνωμόνων ἀκροατῶν δεησόμεθα, καὶ πρᾴως τὴν ἀρχαιολογίαν
  προσδεχομένων.

  [496] See Isokratês, Panathenaic. (t. ii. p. 510-512, Auger);
  Xenoph. Memor. iii. 5, 10. In the Helenæ Encomium, Isokratês
  enlarges more upon the personal exploits of Thêseus in
  conjunction with his great political merits (t. ii. p. 342-350,
  Auger).

  [497] Plutarch, Thêseus, 20.

  [498] See the epigram of Krinagoras, Antholog. Pal. vol. ii. p.
  144; ep. xv. ed. Brunck. and Kallimach. Frag. 40.

    Ἀείδει δ᾽ (Kallimachus) Ἑκάλης τε φιλοξείνοιο καλιὴν,
      Καὶ Θησεῖ Μαραθὼν οὓς ἐπέθηκε πόνους.

  Some beautiful lines are preserved by Suidas, v. Ἐπαύλια,
  περὶ Ἑκάλης θανούσης (probably spoken by Thêseus himself, see
  Plutarch, Theseus, c. 14).

                  Ἴθι, πρηεῖα γυναικῶν,
    Τὴν ὁδὸν, ἣν ἀνίαι θυμαλγέες οὐ περόωσιν·
    Πόλλακι σεῖ᾽, ὦ μαῖα, φιλοξείνοιο καλιῆς
    Μνησόμεθα· ξυνὸν γὰρ ἐπαύλιον ἔσκεν ἅπασι.

  [499] Virgil, Æneid, vi. 617. “Sedet æternumque sedebit Infelix
  Thêseus.”

Two however among the Thêseian fables cannot be dismissed without
some special notice,—the war against the Amazons, and the expedition
against Krête. The former strikingly illustrates the facility as well
as the tenacity of Grecian legendary faith; the latter embraces the
story of Dædalus and Minos, two of the most eminent among Grecian
ante-historical personages.

The Amazons, daughters of Arês and Harmonia,[500] are both early
creations and frequent reproductions of the ancient epic—which
was indeed, we may generally remark, largely occupied both with
the exploits and sufferings of women, or heroines, the wives and
daughters of the Grecian heroes—and which recognized in Pallas Athênê
the finished type of an irresistible female warrior. A nation of
courageous, hardy and indefatigable women, dwelling apart from men,
permitting only a short temporary intercourse for the purpose of
renovating their numbers, and burning out their right breast with
a view of enabling themselves to draw the bow freely,—this was at
once a general type stimulating to the fancy of the poet and a theme
eminently popular with his hearers. Nor was it at all repugnant to
the faith of the latter—who had no recorded facts to guide them, and
no other standard of credibility as to the past except such poetical
narratives themselves—to conceive communities of Amazons as having
actually existed in anterior time. Accordingly we find these warlike
females constantly reappearing in the ancient poems, and universally
accepted as past realities. In the Iliad, when Priam wishes to
illustrate emphatically the most numerous host in which he ever found
himself included, he tells us that it was assembled in Phyrgia, on the
banks of the Sangarius, for the purpose of resisting the formidable
Amazons. When Bellerophôn is to be employed on a deadly and perilous
undertaking,[501] by those who indirectly wish to procure his death,
he is despatched against the Amazons. In the Æthiopis of Arktinus,
describing the post-Homeric war of Troy, Penthesileia, queen of the
Amazons, appears as the most effective ally of the besieged city,
and as the most formidable enemy of the Greeks, succumbing only to
the invincible might of Achilles.[502] The Argonautic heroes find
the Amazons on the river Thermôdon, in their expedition along the
southern coast of the Euxine. To the same spot Hêraclês goes to attack
them, in the performance of the ninth labor imposed upon him by
Eurystheus, for the purpose of procuring the girdle of the Amazonian
queen, Hippolytê;[503] and we are told that they had not yet recovered
from the losses sustained in this severe aggression when Thêseus also
assaulted and defeated them, carrying off their queen, Antiopê.[504]
This injury they avenged by invading Attica,—an undertaking (as
Plutarch justly observes) “neither trifling nor feminine,” especially
if according to the statement of Hellanikus, they crossed the Cimmerian
Bosporus on the winter ice, beginning their march from the Asiatic
side of the Paulus Mæotis.[505] They overcame all the resistances
and difficulties of this prodigious march, and penetrated even into
Athens itself, where the final battle, hard-fought and at one time
doubtful, by which Theseus crushed them, was fought—in the very heart
of the city. Attic antiquaries confidently pointed out the exact
position of the two contending armies: the left wing of the Amazons
rested upon the spot occupied by the commemorative monument called the
Amazoneion; the right wing touched the Pnyx, the place in which the
public assemblies of the Athenian democracy were afterwards held. The
details and fluctuations of the combat, as well as the final triumph
and consequent truce, were recounted by these authors with as complete
faith and as much circumstantiality as those of the battle of Platæa
by Herodotus. The sepulchral edifice called the Amazoneion, the tomb
or pillar of Antiopê near the western gate of the city—the spot called
the Horkomosion near the temple of Thêseus—even the hill of Areiopagus
itself, and the sacrifices which it was customary to offer to the
Amazons at the periodical festival of the Thêseia—were all so many
religious mementos of this victory;[506] which was moreover a favorite
subject of art both with the sculptor and the painter, at Athens as
well as in other parts of Greece.

  [500] Pherekyd. Fragm. 25, Didot.

  [501] Iliad, iii. 186; vi. 152.

  [502] See Proclus’s Argument of the lost Æthiopis (Fragm. Epicor.
  Græcor. ed. Düntzer, p. 16). We are reduced to the first book of
  Quintus Smyrnæus for some idea of the valor of Penthesileia; it
  is supposed to be copied more or less closely from the Æthiopis.
  See Tychsen’s Dissertation prefixed to his edition of Quintus,
  sections 5 and 12. Compare Dio. Chrysostom. Or. xi. p. 350,
  Reiske. Philostratus (Heroica, c. 19, p. 751) gives a strange
  transformation of this old epical narrative into a descent of
  Amazons upon the island sacred to Achilles.

  [503] Apollôn. Rhod. ii. 966, 1004; Apollod. ii. 5-9; Diodôr.
  ii. 46; iv. 16. The Amazons were supposed to speak the Thracian
  language (Schol. Apoll Rhod. ii. 953), though some authors
  asserted them to be natives of Libyia, others of Æthiopia (_ib._
  965).

  Hellanikus (Frag. 33, ap. Schol. Pindar. Nem. iii. 65) said that
  all the Argonauts had assisted Hêraklês in this expedition: the
  fragment of the old epic poem (perhaps the Ἀμαζόνια) there quoted
  mentions Telamôn specially.

  [504] The many diversities in the story respecting Thêseus and
  the Amazon Antiopê are well set forth in Bachet de Meziriac
  (Commentaires sur Ovide, t. i. p. 317).

  Welcker (Der Epische Cyclus, p. 313) supposes that the ancient
  epic poem called by Suidas Ἀμαζόνια, related to the invasion of
  Attica by the Amazons, and that this poem is the same, under
  another title, as the Ἀτθὶς of Hegesinous cited by Pausanias: I
  cannot say that he establishes this conjecture satisfactorily,
  but the chapter is well worth consulting. The epic Thêsêis seems
  to have given a version of the Amazonian contest in many respects
  different from that which Plutarch has put together out of the
  logographers (see Plut. Thês. 28): it contained a narrative of
  many unconnected exploits belonging to Thêseus, and Aristotle
  censures it on that account as ill-constructed (Poetic. c. 17).

  The Ἀμαζονὶς or Ἀμαζονικὰ of Onasus can hardly have been (as
  Heyne supposes, ad Apollod. ii. 5, 9) an epic poem: we may infer
  from the rationalizing tendency of the citation from it (Schol.
  ad Theocrit. xiii. 46, and Schol. Apollôn. Rhod. i. 1207) that it
  was a work in prose. There was an Ἀμαζονὶς by Possis of Magnêsia
  (Athenæus, vii. p. 296).

  [505] Plutarch, Thêseus, 27. Pindar (Olymp. xiii. 84) represents
  the Amazons as having come from the extreme north, when
  Bellerophôn conquers them.

  [506] Plutarch, Thêseus, 27-28; Pausan. i. 2, 4; Plato, Axiochus,
  c. 2; Harpocratiôn, v. Ἀμαζονεῖον; Aristophan. Lysistrat. 678,
  with the Scholia. Æschyl. (Eumenid. 685) says that the Amazons
  assaulted the citadel from the Areiopagus:—

    Πάγον τ᾽ Ἄρειον τόνδ᾽, Ἀμαζόνων ἕδραν
    Σκηνάς τ᾽, ὅτ᾽ ἦλθον Θησέως κατὰ φθόνον
    Στρατηλατοῦσαι, καὶ πόλιν νεόπτολιν
    Τήνδ᾽ ὑψίπυργον ἀντεπύργωσάν ποτε.

No portion of the ante-historical epic appears to have been more deeply
worked into the national mind of Greece than this invasion and defeat
of the Amazons. It was not only a constant theme of the logographers,
but was also familiarly appealed to by the popular orators along
with Marathôn and Salamis, among those antique exploits of which
their fellow-citizens might justly be proud. It formed a part of the
retrospective faith of Herodotus, Lysias, Plato and Isokratês,[507] and
the exact date of the event was settled by the chronologists.[508] Nor
did the Athenians stand alone in such a belief. Throughout many other
regions of Greece, both European and Asiatic, traditions and memorials
of the Amazons were found. At Megara, at Trœzen, in Laconia near Cape
Tænarus, at Chæroneia in Bœôtia, and in more than one part of Thessaly,
sepulchres or monuments of the Amazons were preserved. The warlike
women (it was said), on their way to Attica, had not traversed those
countries, without leaving some evidences of their passage.[509]

  [507] Herodot. ix. 27, Lysias (Epitaph, c. 3) represents the
  Amazons as ἄρχουσαι πολλῶν ἔθνων: the whole race, according to
  him, was nearly extinguished in their unsuccessful and calamitous
  invasion of Attica. Isokratês (Panegyric. t. i. p. 206, Auger)
  says the same; also Panathênaic. t. iii. p. 560, Auger; Demosth.
  Epitaph, p. 1391. Reisk. Pausanias quotes Pindar’s notice of the
  invasion, and with the fullest belief of its historical reality
  (vii. 2, 4) Plato mentions the invasion of Attica by the Amazons
  in the Menexenus (c. 9), but the passage in the treatise De Legg.
  c. ii. p. 804,—ἀκούων γὰρ δὴ μύθους παλαιοὺς πέπεισμαι, etc.—is
  even a stronger evidence of his own belief. And Xenophon in the
  Anabasis, when he compares the quiver and the hatchet of his
  barbarous enemies to “those which the Amazons carry,” evidently
  believed himself to be speaking of real persons, though he could
  have seen only the costumes and armature of those painted by
  Mikôn and others (Anabas. iv. 4, 10; compare Æschyl. Supplic.
  293, and Aristophan. Lysistr. 678; Lucian. Anachars, c. 34. v.
  iii. p. 318).

  How copiously the tale was enlarged upon by the authors of the
  Atthides, we see in Plutarch, Thêseus, 27-28.

  Hekatæus (ap. Steph. Byz. Ἀμαζονεῖον; also Fragm. 350, 351, 352,
  Didot) and Xanthus (ap. Hesychium, v. Βουλεψίη) both treated
  of the Amazons: the latter passage ought to be added to the
  collection of the Fragments of Xanthus by Didot.

  [508] Clemens Alexandr. Stromat, i. p. 336; Marmor Parium, Epoch.
  21.

  [509] Plutarch, Thês. 27-28. Steph. Byz. v. Ἀμαζονεῖον. Pausan.
  ii. 32, 8; iii. 25, 2.

Amongst the Asiatic Greeks the supposed traces of the Amazons were yet
more numerous. Their proper territory was asserted to be the town and
plain of Themiskyra, near the Grecian colony of Amisus, on the river
Thermôdôn, a region called, after their name by Roman historians and
geographers.[510] But they were believed to have conquered and occupied
in early times a much wider range of territory, extending even to the
coast of Iônia and Æolis. Ephesus, Smyrna, Kymê, Myrina, Paphos and
Sinopê were affirmed to have been founded and denominated by them.[511]
Some authors placed them in Libya or Ethiopia; and when the Pontic
Greeks on the north-western shore of the Euxine had become acquainted
with the hardy and daring character of the Sarmatian maidens,—who were
obliged to have slain each an enemy in battle as the condition of
obtaining a husband, and who artificially prevented the growth of the
right breast during childhood,—they could imagine no more satisfactory
mode of accounting for such attributes than by deducing the Sarmatians
from a colony of vagrant Amazons, expelled by the Grecian heroes from
their territory on the Thermôdôn.[512] Pindar ascribed the first
establishment of the memorable temple of Artemis at Ephesus to the
Amazons. And Pausanias explains in part the preëminence which this
temple enjoyed over every other in Greece by the widely diffused renown
of its female founders,[513] respecting whom he observes (with perfect
truth, if we admit the historical character of the old epic), that
women possess an unparalleled force of resolution in resisting adverse
events, since the Amazons, after having been first roughly handled
by Hêraklês and then completely defeated by Thêseus, could yet find
courage to play so conspicuous a part in the defence of Troy against
the Grecian besiegers.[514]

  [510] Pherekydês ap. Schol. Apollôn. Rh. ii. 373-992; Justin, ii.
  4; Strabo, xii. p. 547, Θεμίσκυραν, τὸ τῶν Ἀμαζόνων οἰκητήριον;
  Diodôr. ii. 45-46; Sallust ap. Serv. ad Virgil. Æneid. xi. 659;
  Pompon. Mela, i. 19; Plin. H. N. vi. 4. The geography of Quintus
  Curtius (vi. 4) and of Philostratus (Heroic c. 19) is on this
  point indefinite, and even inconsistent.

  [511] Ephor. Fragm. 87, Didot. Strabo, xi. p. 505; xiii p. 573;
  xiii. p. 622. Pausan. iv. 31, 6; vii. 2. 4. Tacit. Ann. iii. 61.
  Schol. Apollôn. Rhod. ii. 965.

  The derivation of the name Sinopê from an Amazon was given by
  Hekatæus (Fragm. 352). Themiskyra also had one of the Amazons for
  its eponymus (Appian, Bell. Mithridat. 78).

  Some of the most venerated religious legends at Sinopê were
  attached to the expedition of Hêraklês against the Amazons:
  Autolykus, the oracle-giving hero, worshipped with great
  solemnity even at the time when the town was besieged by
  Lucullus, was the companion of Hêraclês (Appian, _ib._ c. 83).
  Even a small mountain village in the territory of Ephesus, called
  Latoreia, derived its name from one of the Amazons (Athenæ. i. p.
  31).

  [512] Herodot. iv. 108-117, where he gives the long tale,
  imagined by the Pontic Greeks, of the origin of the Sarmatian
  nation. Compare Hippokratês, De Aëre, Locis et Aquis, c. 17;
  Ephorus, Fragm. 103; Skymn. Chius, v. 102; Plato, Legg. vii. p.
  804; Diodôr. ii. 34.

  The testimony of Hippokrates certifies the practice of the
  Sarmatian women to check the growth of the right breast: Τὸν
  δέξιον δὲ μαζὸν οὐκ ἔχουσιν. Παιδίοισι γὰρ ἐοῦσιν ἔτι νηπίοισιν
  αἱ μητέρες χαλκεῖον τετεχνήμενον ἐπ᾽ αὐτέῳ τούτῳ διάπυρον
  ποιέουσαι, πρὸς τὸν μαζὸν τιθέασι τὸν δέξιον· καὶ ἐπικαίεται,
  ὥστε τὴν αὔξησιν φθείρεσθαι, ἐς δὲ τὸν δέξιον ὦμον καὶ βραχίονα
  πᾶσαν τὴν ἴσχυν καὶ τὸ πλῆθος ἐκδιδόναι.

  Ktêsias also compares a warlike Sakian woman to the Amazons
  (Fragm. Persic. ii. pp. 221, 449, Bähr).

  [513] Pausan. iv. 31, 6; vii. 2, 4. Dionys. Periêgêt. 828.

  [514] Pausan. i. 15, 2.

It is thus that in what is called early Grecian history, as the
Greeks themselves looked back upon it, the Amazons were among the
most prominent and undisputed personages. Nor will the circumstance
appear wonderful if we reflect, that the belief in them was first
established at a time when the Grecian mind was fed with nothing else
but religious legend and epic poetry, and that the incidents of the
supposed past, as received from these sources, were addressed to their
faith and feelings, without being required to adapt themselves to any
canons of credibility drawn from present experience. But the time came
when the historians of Alexander the Great audaciously abused this
ancient credence. Amongst other tales calculated to exalt the dignity
of that monarch, they affirmed that after his conquest and subjugation
of the Persian empire, he had been visited in Hyrcania by Thalestris,
queen of the Amazons, who admiring his warlike prowess, was anxious to
be enabled to return into her own country in a condition to produce
offspring of a breed so invincible.[515] But the Greeks had now been
accustomed for a century and a half to historical and philosophical
criticism—and that uninquiring faith, which was readily accorded to the
wonders of the past, could no longer be invoked for them when tendered
as present reality. For the fable of the Amazons was here reproduced in
its naked simplicity, without being rationalized or painted over with
historical colors.

  [515] Arrian, Exped. Alex. vii. 13; compare iv. 15; Quint. Curt.
  vi. 4; Justin, xlii. 4. The note of Freinshemius on the above
  passage of Quintus Curtius is full of valuable references on the
  subject of the Amazons.

Some literary men indeed, among whom were Dêmêtrius of Skepsis, and
the Mitylenæan Theophanês, the companion of Pompey in his expeditions,
still continued their belief both in Amazons present and Amazons past;
and when it becomes notorious that at least there were none such on the
banks of the Thermôdôn, these authors supposed them to have migrated
from their original locality, and to have settled in the unvisited
regions north of Mount Caucasus.[516] Strabo, on the contrary, feeling
that the grounds of disbelief applied with equal force to the ancient
stories and to the modern, rejected both the one and the other. But
he remarks at the same time, not without some surprise, that it was
usual with most persons to adopt a middle course,—to retain the Amazons
as historical phænomena of the remote past, but to disallow them as
realities of the present, and to maintain that the breed had died
out.[517] The accomplished intellect of Julius Cæsar did not scruple
to acknowledge them as having once conquered and held in dominion
a large portion of Asia;[518] and the compromise between early,
traditional, and religious faith on the one hand, and established
habits of critical research on the other, adopted by the historian
Arrian, deserves to be transcribed in his own words, as illustrating
strikingly the powerful sway of the old legends even over the most
positive-minded Greeks:—“Neither Aristobulus nor Ptolemy (he observes),
nor any other competent witness, has recounted this (visit of the
Amazons and their queen to Alexander): nor does it seem to me that the
race of the Amazons was preserved down to that time, nor have they been
noticed either by any one before Alexander, or by Xenophôn, though he
mentions both the Phasians and the Kolchians, and the other barbarous
nations which the Greeks saw both before and after their arrival at
Trapezus, in which marches they must have met with the Amazons, if the
latter had been still in existence. Yet _it is incredible to me_ that
this race of women, celebrated as they have been by authors so many
and so commanding, _should never have existed at all_. The story tells
of Hêraklês, that he set out from Greece and brought back with him the
girdle of their queen Hippolytê; also of Thêseus and the Athenians,
that they were the first who defeated in battle and repelled these
women in their invasion of Europe; and the combat of the Athenians
with the Amazons has been painted by Mikôn, not less than that between
the Athenians and the Persians. Moreover Herodotus has spoken in many
places of these women, and those Athenian orators who have pronounced
panegyrics on the citizens slain in battle, have dwelt upon the victory
over the Amazons as among the most memorable of Athenian exploits. If
the satrap of Media sent any equestrian women at all to Alexander, I
think that they must have come from some of the neighboring tribes,
practised in riding and equipped in the costume generally called
Amazonian.”[519]

  [516] Strabo, xi. p. 503-504; Appian, Bell. Mithridat. c. 103;
  Plutarch, Pompeius, c. 35. Plin. N. H. vi. 7. Plutarch still
  retains the old description of Amazons from the mountains near
  the Thermôdôn. Appian keeps clear of this geographical error,
  probably copying more exactly the language of Theophanês, who
  must have been well aware that when Lucullus besieged Themiskyra,
  he did not find it defended by the Amazons (see Appian, Bell.
  Mithridat. c. 78). Ptolemy (v. 9) places the Amazons in the
  imperfectly known regions of Asiatic Sarmatia, north of the
  Caspian and near the river Rha (Volga). “This fabulous community
  of women (observes Forbiger, Handbuch der alten Geographie,
  ii. 77, p. 457) was a phænomenon much too interesting for the
  geographers easily to relinquish.”

  [517] Strabo, xi. p. 505. Ἴδιον δέ τι συμβέβηκε τῷ λόγῳ τῷ περὶ
  τῶν Ἀμαζόνων. Οἱ μὲν γὰρ ἄλλοι τὸ μυθῶδες καὶ τὸ ἱστορικὸν
  διωρίσμενον ἔχουσι· τὰ γὰρ παλαιὰ καὶ ψευδῆ καὶ τερατώδη, μῦθοι
  καλοῦνται· [_Note._ Strabo does not always speak of the μῦθοι in
  this disrespectful tone; he is sometimes much displeased with
  those who dispute the existence of an historical kernel in the
  inside, especially with regard to Homer.] ἡ δ᾽ ἱστορία βούλεται
  τἀληθὲς, ἄντε παλαιὸν, ἄντε νέον· καὶ τὸ τερατῶδες ἢ οὐκ ἔχει, ἢ
  σπάνιον. Περὶ δὲ τῶν Ἀμαζόνων τὰ αὐτὰ λέγεται καὶ νῦν καὶ παλαὶ,
  τερατώδη τ᾽ ὄντα, καὶ πίστεως πόῤῥω. Τίς γὰρ ἂν πιστεύσειεν,
  ὡς γυναικῶν στράτος, ἢ πόλις, ἢ ἔθνος, συσταίη ἂν πότε χωρὶς
  ἀνδρῶν; καὶ οὐ μόνον συσταίη, ἀλλὰ καὶ ἐφόδους ποιήσαιτο ἐπὶ τὴν
  ἀλλοτρίαν, καὶ κρατήσειεν οὐ τῶν ἐγγὺς μόνον, ὥστε καὶ μέχρι τῆς
  νῦν Ἰωνίας προελθεῖν, ἀλλὰ καὶ διαπόντιον στείλαιτο στρατίαν
  μέχρι τῆς Ἀττικῆς; Ἀλλὰ μὴν ταῦτά γε αὐτὰ καὶ νῦν λέγεται περὶ
  αὐτῶν· ~ἐπιτείνει δὲ τὴν ἰδιότητα καὶ τὸ πιστεύεσθαι τὰ παλαιὰ
  μᾶλλον ἢ τὰ νῦν~. There are however, other passages in which he
  speaks of the Amazons as realities.

  Justin (ii. 4) recognizes the great power and extensive conquests
  of the Amazons in very early times, but says that they gradually
  declined down to the reign of Alexander, in whose time there
  were _just a few remaining_; the queen with these few visited
  Alexander, but shortly afterwards the whole breed became extinct.
  This hypothesis has the merit of convenience, perhaps of
  ingenuity.

  [518] Suetonius, Jul. Cæsar, c. 22. “In Syriâ quoque regnasse
  Semiramin (Julius Cæsar said this), magnamque Asiæ partem
  Amazonas tenuisse quondam.”

  In the splendid triumph of the emperor Aurelian at Rome after the
  defeat of Zenobia, a few Gothic women who had been taken in arms
  were exhibited among the prisoners; the official placard carried
  along with them announced them as _Amazons_ (Vopiscus Aurel. in
  Histor. August. Scrip. p. 260, ed. Paris).

  [519] Arrian, Expedit. Alexand. vii. 13.

There cannot be a more striking evidence of the indelible force with
which these ancient legends were worked into the national faith and
feelings of the Greeks, than these remarks of a judicious historian
upon the fable of the Amazons. Probably if any plausible mode of
rationalizing it, and of transforming it into a quasi-political event,
had been offered to Arrian, he would have been better pleased to
adopt such a middle term, and would have rested comfortably in the
supposition that he believed the legend in its true meaning, while his
less inquiring countrymen were imposed upon by the exaggerations of
poets. But as the story was presented to him plain and unvarnished,
either for acceptance or rejection, his feelings as a patriot
and a religious man prevented him from applying to the past such
tests of credibility as his untrammelled reason acknowledged to be
paramount in regard to the present. When we see moreover how much
his belief was strengthened, and all tendency to scepticism shut out
by the familiarity of his eye and memory with sculptured or painted
Amazons[520]—we may calculate the irresistible force of this sensible
demonstration on the convictions of the unlettered public, at once
more deeply retentive of passive impressions, and unaccustomed to the
countervailing habit of rational investigation into evidence. Had the
march of an army of warlike women, from the Thermôdôn or the Tanais
into the heart of Attica, been recounted to Arrian as an incident
belonging to the time of Alexander the Great, he would have rejected
it no less emphatically than Strabô; but cast back as it was into an
undefined past, it took rank among the hallowed traditions of divine
or heroic antiquity,—gratifying to extol by rhetoric, but repulsive to
scrutinize in argument.[521]

  [520] Ktêsias described as real animals, existing in wild and
  distant regions, the heterogeneous and fantastic combinations
  which he saw sculptured in the East (see this stated and
  illustrated in Bähr, Preface to the Fragm. of Ktêsias, pp. 58,
  59).

  [521] Heyne observes (Apollodôr. ii. 5, 9) with respect to the
  fable of the Amazons, “In his historiarum fidem aut vestigia nemo
  quæsiverit.” Admitting the wisdom of this counsel (and I think
  it indisputable), why are we required to presume, in the absence
  of all proof, an historical basis for each of those _other_
  narratives, such as the Kalydônian boar-hunt, the Argonautic
  expedition, or the siege of Troy, which go to make up, along
  with the story of the Amazons, the aggregate matter of Grecian
  legendary faith? If the tale of the Amazons could gain currency
  without any such support, why not other portions of the ancient
  epic?

  An author of easy belief, Dr. F. Nagel, vindicates the historical
  reality of the Amazons (Geschichte der Amazonen, Stutgart,
  1838). I subjoin here a different explanation of the Amazonian
  tale, proceeding from another author who rejects the historical
  basis, and contained in a work of learning and value (Guhl,
  _Ephesiaca_, Berlin, 1843. p. 132):—

  “Id tantum monendum videtur, Amazonas nequaquam historice
  accipiendas esse, sed e contrario totas ad mythologiam pertinere.
  Earum enim fabulas quum ex frequentium hierodularum gregibus in
  cultibus et sacris Asiaticis ortas esse ingeniose ostenderit
  Tolken, jam _inter omnes mythologiæ peritos constat_, Amazonibus
  nihil fere nisi peregrini cujusdam cultus notionem expressum
  esse, ejusque cum Græcorum religione certamen frequentibus istis
  pugnis designatum esse, quas cum Amazonibus tot Græcorum heroes
  habuisse credebantur, Hercules, Bellerophon, Theseus, Achilles,
  et vel ipse, quem Ephesi cultum fuisse supra ostendimus,
  Dionysus. Quæ Amazonum notio primaria, quum paulatim Euemeristicâ
  (ut ita dicam) ratione ita transformaretur, ut Amazones pro vero
  feminarum populo haberentur, necesse quoque erat, ut omnibus
  fere locis, ubi ejusmodi religionum certamina locum habuerunt,
  Amazones habitasse, vel eo usque processisse, crederentur.
  Quod cum nusquam manifestius fuerit, quam in Asiâ minore, et
  potissimum in eâ parte quæ Græciam versus vergit, haud mirandum
  est omnes fere ejus oræ urbes ab Amazonibus conditas putari.”

  I do not know the evidence upon which this conjectural
  interpretation rests, but the statement of it, though it boasts
  so many supporters among mythological critics, carries no
  appearance of probability to my mind. Priam fights against the
  Amazons as well as the Grecian heroes.




CHAPTER XII.

KRETAN LEGENDS.—MINOS AND HIS FAMILY.


To understand the adventures of Thêseus in Krête, it will be necessary
to touch briefly upon Minôs and the Krêtan heroic genealogy.

Minôs and Rhadamanthus, according to Homer, are sons of Zeus, by
Europê,[522] daughter of the widely-celebrated Phœnix, born in Krête.
Minôs is the father of Deukaliôn, whose son Idomeneus, in conjunction
with Mêrionês, conducts the Krêtan troops to the host of Agamemnôn
before Troy. Minôs is ruler of Knossus, and familiar companion of
the great Zeus. He is spoken of as holding guardianship in Krête—not
necessarily meaning the whole of the island: he is farther decorated
with a golden sceptre, and constituted judge over the dead in the
under-world to settle their disputes, in which function Odysseus finds
him—this however by a passage of comparatively late interpolation into
the Odyssey. He also had a daughter named Ariadnê, for whom the artist
Dædalus fabricated in the town of Knossus the representation of a
complicated dance, and who was ultimately carried off by Thêseus: she
died in the island of Dia, deserted by Thêseus and betrayed by Dionysos
to the fatal wrath of Artemis. Rhadamanthus seems to approach to Minôs
both in judicial functions and posthumous dignity. He is conveyed
expressly to Eubœa, by the semi-divine sea-carriers the Phæacians, to
inspect the gigantic corpse of the earth-born Tityus—the longest voyage
they ever undertook. He is moreover after death promoted to an abode
of undisturbed bliss in the Elysian plain at the extremity of the
earth.[523]

  [522] Europê was worshipped with very peculiar solemnity in the
  island of Krête (see Dictys Cretensis, De Bello Trojano, i. c. 2).

  The venerable plane-tree, under which Zeus and Europê had
  reposed, was still shown, hard by a fountain at Gortyn in
  Krête, in the time of Theophrastus: it was said to be the only
  plane-tree in the neighborhood which never cast its leaves
  (Theophrast. Hist. Plant. i. 9).

  [523] Homer, Iliad, xiii. 249, 450; xiv. 321. Odyss. xi. 322-568;
  xix. 179; iv. 564-vii. 321.

  The Homeric Minôs in the under-world is not a judge of the
  previous lives of the dead, so as to determine whether they
  deserve reward or punishment for their conduct on earth: such
  functions are not assigned to him earlier than the time of Plato.
  He administers justice _among_ the dead, who are conceived as a
  sort of society, requiring some presiding judge: θεμιστεύοντα
  νεκύεσσι, with regard to Minôs, is said very much like (Odyss.
  xi. 484) νῦν αὖτε μέγα κρατέεις νεκύεσσι with regard to Achilles.
  See this matter partially illustrated in Heyne’s Excursus xi. to
  the sixth book of the Æneid of Virgil.

According to poets later than Homer, Europê is brought over by Zeus
from Phœnicia to Krête, where she bears to him three sons, Minôs,
Rhadamanthus and Sarpêdôn. The latter leaves Krête and settles in
Lykia, the population of which, as well as that of many other portions
of Asia Minor, is connected by various mythical genealogies with
Krête, though the Sarpêdôn of the Iliad has no connection with Krête,
and is not the son of Europê. Sarpêdôn having become king of Lykia,
was favored by his father, Zeus, with permission to live for three
generations.[524] At the same time the youthful Milêtus, a favorite
of Sarpêdôn, quitted Krête, and established the city which bore his
name on the coast of Asia Minor. Rhadamanthus became sovereign of
and lawgiver among the islands in the Ægean: he subsequently went to
Bœôtia, where he married the widowed Alkmênê, mother of Hêraklês.

  [524] Apollodôr. iii. 1, 2. Καὶ αὐτῷ δίδωσι Ζεὺς ἐπὶ τρεῖς γενεὰς
  ζῇν. This circumstance is evidently imagined by the logographers
  to account for the appearance of Sarpêdôn in the Trojan war,
  fighting against Idomeneus, the grandson of Minôs. Nisus is the
  eponymus of Nisæa, the port of the town of Megara: his tomb
  was shown at Athens (Pausan. i. 19, 5). Minôs is the eponym of
  the island of Minoa (opposite the port of Nisæa), where it was
  affirmed that the fleet of Minôs was stationed (Pausan. i. 44, 5).

Europê finds in Krête a king Astêrius, who marries her and adopts her
children by Zeus: this Astêrius is the son of Krês, the eponym of the
island, or (according to another genealogy by which it was attempted to
be made out that Minôs was of Dôrian race) he was a son of the daughter
of Krês by Tektamus, the son of Dôrus, who had migrated into the island
from Greece.

Minôs married Pasiphaê, daughter of the god Hêlios and Perseïs, by
whom he had Katreus, Deukaliôn, Glaukus, Androgeos, names marked in
the legendary narrative,—together with several daughters, among whom
were Ariadnê and Phædra. He offended Poseidôn by neglecting to fulfil a
solemnly-made vow, and the displeased god afflicted his wife Pasiphaê
with a monstrous passion for a bull. The great artist Dædalus, son of
Eupalamus, a fugitive from Athens, became the confidant of this amour,
from which sprang the Minôtaur, a creature half man and half bull.[525]
This Minôtaur was imprisoned by Minôs in the labyrinth, an inextricable
inclosure constructed by Dædalus for that express purpose, by order of
Minôs.

  [525] Apollodôr iii. 1, 2.

Minôs acquired great nautical power, and expelled the Karian
inhabitants from many of the islands of the Ægean, which he placed
under the government of his sons on the footing of tributaries. He
undertook several expeditions against various places on the coast—one
against Nisus, the son of Pandiôn, king of Megara, who had amongst
the hair of his head one peculiar lock of a purple color: an oracle
had pronounced that his life and reign would never be in danger so
long as he preserved this precious lock. The city would have remained
inexpugnable, if Scylla, the daughter of Nisus, had not conceived a
violent passion for Minôs. While her father was asleep, she cut off
the lock on which his safety hung, so that the Krêtan king soon became
victorious. Instead of performing his promise to carry Scylla away
with him to Krête, he cast her from the stern of his vessel into the
sea:[526] both Scylla and Nisus were changed into birds.

  [526] Apollodôr. iii. 15, 8. See the Ciris of Virgil, a juvenile
  poem on the subject of this fable; also Hyginus, f. 198; Schol.
  Eurip. Hippol. 1200. Propertius (iii. 19, 21) gives the features
  of the story with tolerable fidelity; Ovid takes considerable
  liberties with it (Metam. viii. 5-150).

Androgeos, son of Minôs having displayed such rare qualities as to
vanquish all his competitors at the Panathenaic festival in Athens,
was sent by Ægeus the Athenian king to contend against the bull of
Marathôn,—an enterprise in which he perished, and Minôs made war upon
Athens to avenge his death. He was for a long time unable to take the
city: at length he prayed to his father Zeus to aid him in obtaining
redress from the Athenians, and Zeus sent upon them pestilence and
famine. In vain did they endeavor to avert these calamities by offering
up as propitiatory sacrifices the four daughters of Hyacinthus. Their
sufferings still continued, and the oracle directed them to submit to
any terms which Minôs might exact. He required that they should send
to Krête a tribute of seven youths and seven maidens, periodically,
to be devoured by the Minôtaur,[527]—offered to him in a labyrinth
constructed by Dædalus, including countless different passages, out of
which no person could escape.

  [527] Apollodôr. iii. 15, 8.

Every ninth year this offering was to be despatched. The more common
story was, that the youths and maidens thus destined to destruction
were selected by lot—but the logographer Hellanikus said that Minôs
came to Athens and chose them himself.[528] The third period for
despatching the victims had arrived, and Athens was plunged in the
deepest affliction, when Thêseus determined to devote himself as one
of them, and either to terminate the sanguinary tribute or to perish.
He prayed to Poseidôn for help, and the Delphian god assured him that
Aphroditê would sustain and extricate him. On arriving at Knossus
he was fortunate enough to captivate the affections of Ariadnê, the
daughter of Minôs, who supplied him with a sword and a clue of thread.
With the former he contrived to kill the Minôtaur, the latter served to
guide his footsteps in escaping from the labyrinth. Having accomplished
this triumph, he left Krête with his ship and companions unhurt,
carrying off Ariadnê, whom however he soon abandoned on the island of
Naxos. On his way home to Athens, he stopped at Dêlos, where he offered
a grateful sacrifice to Apollo for his escape, and danced along with
the young men and maidens whom he had rescued from the Minôtaur, a
dance called the Geranus, imitated from the twists and convolutions
of the Krêtan labyrinth. It had been concerted with his father Ægeus,
that if he succeeded in his enterprise against the Minôtaur, he should
on his return hoist white sails in his ship in place of the black
canvas which she habitually carried when employed on this mournful
embassy. But Thêseus forgot to make the change of sails; so that Ægeus,
seeing the ship return with her equipment of mourning unaltered, was
impressed with the sorrowful conviction that his son had perished,
and cast himself into the sea. The ship which made this voyage was
preserved by the Athenians with careful solicitude, being constantly
repaired with new timbers, down to the time of the Phalerian Dêmêtrius:
every year she was sent from Athens to Dêlos with a solemn sacrifice
and specially-nominated envoys. The priest of Apollo decked her stern
with garlands before she quitted the port, and during the time which
elapsed until her return, the city was understood to abstain from all
acts carrying with them public impurity, so that it was unlawful to put
to death any person even under formal sentence by the dikastery. This
accidental circumstance becomes especially memorable, from its having
postponed for thirty days the death of the lamented Socratês.[529]

  [528] See, on the subject of Thêseus and the Minôtaur, Eckermann,
  Lehrbuch der Religions Geschichte und Mythologie, vol. ii. ch.
  xiii. p. 133. He maintains that the tribute of these human
  victims paid by Athens to Minôs is an historical fact. Upon what
  this belief is grounded, I confess I do not see.

  [529] Plato, Phædon, c. 2, 3; Xenoph. Memor. iv. 8. 2. Plato
  especially noticed τοὺς δὶς ἕπτα ἐκείνους, the seven youths and
  the seven maidens whom Thêseus conveyed to Krête and brought
  back safely: this number seems an old and constant feature in
  the legend, maintained by Sappho and Bacchylidês as well as by
  Euripidês (Herc. Fur. 1318). See Servius ad Virgil Æneid. vi. 21.

The legend respecting Thêseus, and his heroic rescue of the seven
noble youths and maidens from the jaws of the Minôtaur, was thus both
commemorated and certified to the Athenian public, by the annual holy
ceremony and by the unquestioned identity of the vessel employed in it.
There were indeed many varieties in the mode of narrating the incident;
and some of the Attic logographers tried to rationalize the fable by
transforming the Minôtaur into a general or a powerful athlete, named
Taurus, whom Thêseus vanquished in Krête.[530] But this altered version
never overbore the old fanciful character of the tale as maintained by
the poets. A great number of other religious ceremonies and customs,
as well as several chapels or sacred enclosures in honor of different
heroes, were connected with different acts and special ordinances
of Thêseus. To every Athênian who took part in the festivals of
the Oschophoria, the Pyanepsia, or the Kybernêsia, the name of this
great hero was familiar, and the motives for offering to him solemn
worship at his own special festival of the Thêseia, became evident and
impressive.

  [530] For the general narrative and its discrepancies, see
  Plutarch, Thês. c. 15-19; Diodôr. iv. 60-62; Pausan. i. 17, 3;
  Ovid, Epist. Ariadn. Thês. 104. In that other portion of the
  work of Diodôrus which relates more especially to Krête, and is
  borrowed from Kretan logographers and historians (v. 64-80), he
  mentions nothing at all respecting the war of Minôs with Athens.

  In the drama of Euripidês called Thêseus, the genuine story
  of the youths and maidens about to be offered as food to the
  Minotaur was introduced (Schol. ad Aristoph. Vesp. 312).

  Ariadnê figures in the Odyssey along with Thêseus: she is the
  daughter of Minôs, carried off by Thêseus from Krête, and killed
  by Artemis in the way home: there is no allusion to Minôtaur, or
  tribute, or self-devotion of Thêseus (Odyss. xi. 324). This is
  probably the oldest and simplest form of the legend—one of the
  many amorous (compare Theognis, 1232) adventures of Thêseus: the
  rest is added by post-Homeric poets.

  The respect of Aristotle for Minôs induces him to adopt the
  hypothesis that the Athenian youths and maidens were not put
  to death in Krête, but grew old in servitude (Aristot. Fragm.
  Βοττιαίων Πολιτεία, p. 106. ed. Neumann. of the Fragments of the
  treatise Περὶ Πολιτειῶν, Plutarch, Quæst. Græc. p. 298).

The same Athenian legends which ennobled and decorated the character
of Thêseus, painted in repulsive colors the attributes of Minôs; and
the traits of the old Homeric comrade of Zeus were buried under those
of the conqueror and oppressor of Athens. His history like that of
the other legendary personages of Greece, consists almost entirely of
a string of family romances and tragedies. His son Katreus, father
of Aëropê, wife of Atreus, was apprized by an oracle that he would
perish by the hand of one of his own children: he accordingly sent them
out of the island, and Althæmenês, his son, established himself in
Rhodes. Katreus having become old, and fancying that he had outlived
the warning of the oracle, went over to Rhodes to see Althæmenês.
In an accidental dispute which arose between his attendants and the
islanders, Althæmenês inadvertently took part and slew his father
without knowing him. Glaukus, the youngest son of Minôs, pursuing a
mouse, fell into a reservoir of honey and was drowned. No one knew
what had become of him, and his father was inconsolable; at length the
Argeian Polyeidus, a prophet wonderfully endowed by the gods, both
discovered the boy and restored him to life, to the exceeding joy of
Minôs.[531]

  [531] Apollodôr. iii. cap. 2-3.

The latter at last found his death in an eager attempt to overtake and
punish Dædalus. This great artist, the eponymous hero of the Attic gens
or dême called the Dædalidæ, and the descendant of Erechtheus through
Mêtion, had been tried at the tribunal of Areiopagus and banished for
killing his nephew Talos, whose rapidly improving skill excited his
envy.[532] He took refuge in Krête, where he acquired the confidence of
Minôs, and was employed (as has been already mentioned) in constructing
the labyrinth; subsequently however he fell under the displeasure
of Minôs, and was confined as a close prisoner in the inextricable
windings of his own edifice. His unrivalled skill and resource however
did not forsake him. He manufactured wings both for himself and for
his son Ikarus, with which they flew over the sea: the father arrived
safely in Sicily at Kamikus, the residence of the Sikanian king
Kokalus, but the son, disdaining paternal example and admonition, flew
so high that his wings were melted by the sun and he fell into the sea,
which from him was called the Ikarian sea.[533]

  [532] Pherekyd. Fragm. 105; Hellanik. Fragm. 82 (Didot); Pausan.
  vii. 4, 5.

  [533] Diodôr. iv. 79; Ovid, Metamorph. viii. 181. Both Ephorus
  and Philistus mentioned the coming of Dædalus to Kokalus in
  Sicily (Ephor. Fr. 99; Philist. Fragm. 1, Didot): probably
  Antiochus noticed it also (Diodôr. xii. 71). Kokalus was the
  point of commencement for the Sicilian historians.

Dædalus remained for some time in Sicily, leaving in various parts of
the island many prodigious evidences of mechanical and architectural
skill.[534] At length Minôs bent upon regaining possession of his
person, undertook an expedition against Kokalus with a numerous fleet
and army. Kokalus affecting readiness to deliver up the fugitive, and
receiving Minôs with apparent friendship, ordered a bath to be prepared
for him by his three daughters, who, eager to protect Dædalus at any
price, drowned the Krêtan king in the bath with hot water.[535] Many
of the Krêtans who had accompanied him remained in Sicily and founded
the town of Minoa, which they denominated after him. But not long
afterwards Zeus roused all the inhabitants of Krête (except the towns
of Polichna and Præsus) to undertake with one accord an expedition
against Kamikus for the purpose of avenging the death of Minôs. They
besieged Kamikus in vain for five years, until at last famine compelled
them to return. On their way along the coast of Italy, in the Gulf
of Tarentum, a terrible storm destroyed their fleet and obliged them
to settle permanently in the country: they founded Hyria with other
cities, and became Messapian Iapygians. Other settlers, for the most
part Greeks, immigrated into Krête to the spots which this movement
had left vacant, and in the second generation after Minôs occurred
the Trojan war. The departed Minôs was exceedingly offended with the
Krêtans for coöperating in avenging the injury to Menelaus, since the
Greeks generally had lent no aid to the Krêtans in their expedition
against the town of Kamikus. He sent upon Krête, after the return
of Idomeneus from Troy, such terrible visitations of famine and
pestilence, that the population again died out or expatriated, and was
again renovated by fresh immigrations. The intolerable suffering[536]
thus brought upon the Krêtans by the anger of Minôs, for having
coöperated in the general Grecian aid to Menelaus, was urged by them to
the Greeks as the reason why they could take no part in resisting the
invasion of Xerxês; and it is even pretended that they were advised and
encouraged to adopt this ground of excuse by the Delphian oracle.[537]

  [534] Diodôr. iv. 80.

  [535] Pausan. vii. 4, 5; Schol. Pindar. Nem. iv. 95; Hygin. fab.
  44; Conon, Narr. 25; Ovid, Ibis, 291.—

    “Vel tua maturet, sicut Minoia fata,
        Per caput infusæ fervidus humor aquæ.”

  This story formed the subject of a lost drama of Sophoklês,
  Καμίκιοι or Μίνως; it was also told by Kallimachus, ἐν Αἰτίοις,
  as well as by Philostephanus (Schol. Iliad, ii. 145).

  [536] This curious and very characteristic narrative is given by
  Herodot. vii. 169-171.

  [537] Herodot. vii. 169. The answer ascribed to the Delphian
  oracle, on the question being put by the Krêtan envoys whether it
  would be better for them to aid the Greeks against Xerxês or not,
  is highly emphatic and poetical: Ὦ νήπιοι, ἐπιμέμφεσθε ὅσα ὑμῖν
  ἐκ τῶν Μενελέω τιμωρημάτων Μίνως ἔπεμψε μηνίων δακρύματα, ὅτι
  οἱ μὲν οὐ ξυνεξεπρήξαντο αὐτῷ τὸν ἐν Καμίκῳ θάνατον γενόμενον,
  ὑμεῖς δὲ κείνοισι τὴν ἐκ Σπάρτης ἁρπασθεῖσαν ὑπ᾽ ἀνδρὸς βαρβάρου
  γυναῖκα.

  If such an answer was ever returned at all, I cannot but think
  that it must have been from some oracle in Krête itself, not from
  Delphi. The Delphian oracle could never have so far forgotten
  its obligations to the general cause of Greece, at that critical
  moment, which involved moreover the safety of all its own
  treasures, as to deter the Krêtans from giving assistance.

Such is the Minôs of the poets and logographers, with his legendary and
romantic attributes: the familiar comrade of the great Zeus,—the judge
among the dead in Hadês,—the husband of Pasiphaê, daughter of the god
Hêlios,—the father of the goddess Ariadnê, as well as of Androgeos,
who perishes and is worshipped at Athens,[538] and of the boy Glaukus,
who is miraculously restored to life by a prophet,—the person
beloved by Scylla, and the amorous pursuer of the nymph or goddess
Britomartis,[539]—the proprietor of the Labyrinth and of the Minôtaur,
and the exacter of a periodical tribute of youths and maidens from
Athens as food for this monster,—lastly, the follower of the fugitive
artist Dædalus to Kamikus, and the victim of the three ill-disposed
daughters of Kokalus in a bath. With this strongly-marked portrait,
the Minôs of Thucydidês and Aristotle has scarcely anything in common
except the name. He is the first to acquire _Thalassokraty_, or command
of the Ægean sea: he expels the Karian inhabitants from the Cyclades
islands, and sends thither fresh colonists under his own sons; he
puts down piracy, in order that he may receive his tribute regularly;
lastly, he attempts to conquer Sicily, but fails in the enterprise and
perishes.[540] Here we have conjectures, derived from the analogy of
the Athenian maritime empire in the historical times, substituted in
place of the fabulous incidents, and attached to the name of Minôs.

  [538] Hesiod, Theogon. 949; Pausan. i. 1, 4.

  [539] Kallimach. Hymn. ad Dian. 189. Strabo (x. p. 476) dwells
  also upon the strange contradiction of the legends concerning
  Minôs: I agree with Hoeckh (Kreta, ii. p. 93) that δασμόλογος in
  this passage refers to the tribute exacted from Athens for the
  Minôtaur.

  [540] Thucyd. i. 4. Μίνως γὰρ, παλαίτατος ὧν ἀκοῇ ἴσμεν,
  ναυτικὸν ἐκτήσατο, καὶ τῆς νῦν Ἑλληνικῆς θαλάσσης ἐπὶ πλεῖστον
  ἐκράτησε, καὶ τῶν Κυκλάδων νήσων ἦρξέ τε καὶ οἰκιστὴς αὐτὸς τῶν
  πλείστων ἐγένετο, Κᾶρας ἐξελάσας καὶ τοὺς ἑαυτοῦ παῖδας ἡγεμόνας
  ἐγκαταστήσας· τό τε λῃστικὸν, ὡς εἰκὸς, καθῄρει ἐκ τῆς θαλάσσης,
  ἐφ᾽ ὅσον ἠδύνατο, τοῦ τὰς προσόδους μᾶλλον ἰέναι αὐτῷ. See also
  c. 8.

  Aristot. Polit. ii. 7, 2, Δοκεῖ δ᾽ ἡ νῆσος καὶ πρὸς τὴν ἀρχὴν
  τὴν Ἑλληνικὴν πεφυκέναι καὶ κεῖσθαι καλῶς ... διὸ καὶ τὴν
  τῆς θαλάσσης ἀρχὴν κατέσχεν ὁ Μίνως, καὶ τὰς νήσους τὰς μὲν
  ἐχειρώσατο, τὰς δὲ ᾤκισε· τέλος δ᾽ ἐπιθέμενος τῇ Σικελίᾳ τὸν βίον
  ἐτελεύτησεν ἐκεῖ περὶ Κάμικον.

  Ephorus (ap. Skymn. Chi. 542) repeated the same statement: he
  mentioned also the autochthonous king Krês.

In the fable, a tribute of seven youths and seven maidens is paid
to him periodically by the Athenians; in the historicized narrative
this character of a tribute-collector is preserved, but the tribute
is money collected from dependent islands;[541] and Aristotle points
out to us how conveniently Krête is situated to exercise empire over
the Ægean. The expedition against Kamikus, instead of being directed
to the recovery of the fugitive Dædalus, is an attempt on the part of
the great thalassokrat to conquer Sicily. Herodotus gives us generally
the same view of the character of Minôs as a great maritime king, but
his notice of the expedition against Kamikus includes the mention of
Dædalus as the intended object of it.[542] Ephorus, while he described
Minôs as a commanding and comprehensive lawgiver imposing his commands
under the sanction of Zeus, represented him as the imitator of an
earlier lawgiver named Rhadamanthus, and also as an immigrant into
Krête from the Æolic Mount Ida, along with the priests or sacred
companions of Zeus called the Idæi Dactyli. Aristotle too points him
out as the author of the Syssitia, or public meals common in Krête as
well as at Sparta,—other divergences in a new direction from the spirit
of the old fables.[543]

  [541] It is curious that Herodotus expressly denies this, and in
  language which shows that he had made special inquiries about it:
  he says that the Karians or Leleges in the islands (who were,
  according to Thucydidês, expelled by Minôs) paid no tribute to
  Minôs, but manned his navy, _i. e._ they stood to Minôs much in
  the same relation as Chios and Lesbos stood to Athens (Herodot.
  i. 171). One may trace here the influence of those discussions
  which must have been prevalent at that time respecting the
  maritime empire of Athens.

  [542] Herodot. vii. 170. Λέγεται γὰρ Μίνω κατὰ ζήτησιν Δαιδάλου
  ἀπικόμενον ἐς Σικανίην, τὴν νῦν Σικελίην καλουμένην, ἀποθανεῖν
  βιαίῳ θανάτῳ. Ἀνὰ δὲ χρόνον Κρῆτας, θεοῦ σφὶ ἐποτρύνοντος, etc.

  [543] Aristot. Polit. ii. 7, 1; vii. 9, 2. Ephorus, Fragm. 63,
  64, 65. He set aside altogether the Homeric genealogy of Minôs,
  which makes him brother of Rhadamanthus and born in Krête.

  Strabo, in pointing out the many contradictions respecting Minôs,
  remarks, Ἔστι δὲ καὶ ἄλλος λόγος οὐχ ὁμολογούμενος, τῶν μὲν ξένον
  τῆς νήσου τὸν Μίνω λεγόντων, τῶν δὲ ἐπιχώριον.. By the former he
  doubtless means Ephorus, though he has not here specified him (x.
  p. 477).

The contradictory attributes ascribed to Minôs, together with the
perplexities experienced by those who wished to introduce a regular
chronological arrangement into these legendary events, has led both
in ancient and in modern times to the supposition of two kings
named Minôs, one the grandson of the other,—Minôs I., the son of
Zeus, lawgiver and judge,—Minôs II., the thalassokrat,—a gratuitous
conjecture, which, without solving the problem required, only adds
one to the numerous artifices employed for imparting the semblance of
history to the disparate matter of legend. The Krêtans were at all
times, from Homer downward, expert and practised seamen. But that they
were ever united under one government, or ever exercised maritime
dominion in the Ægean is a fact which we are neither able to affirm
nor to deny. The Odyssey, in so far as it justifies any inference at
all, points against such a supposition, since it recognizes a great
diversity both of inhabitants and of languages in the island, and
designates Minôs as king specially of Knôssus: it refutes still more
positively the idea that Minôs put down piracy, which the Homeric
Krêtans as well as others continue to practise without scruple.

Herodotus, though he in some places speaks of Minôs as a person
historically cognizable, yet in one passage severs him pointedly from
the generation of man. The Samian despot “Polykratês (he tells us) was
the first person who aspired to nautical dominion, excepting Minôs
of Knôssus, and others before him (if any such there ever were) who
may have ruled the sea; but Polykratês is the first of that which is
called _the generation of man_ who aspired with much chance of success
to govern Iônia and the islands of the Ægean.”[544] Here we find it
manifestly intimated that Minôs did not belong to the generation of
man, and the tale given by the historian respecting the tremendous
calamities which the wrath of the departed Minôs inflicted on Krête
confirms the impression. The king of Knôssus is a god or a hero, but
not a man; he belongs to legend, not to history. He is the son as
well as the familiar companion of Zeus; he marries the daughter of
Hêlios, and Ariadnê is numbered among his offspring. To this superhuman
person are ascribed the oldest and most revered institutions of the
island, religious and political, together with a period of supposed
ante-historical dominion. That there is much of Krêtan religious ideas
and practice embodied in the fables concerning Minôs can hardly be
doubted: nor is it improbable that the tale of the youths and maidens
sent from Athens may be based in some expiatory offerings rendered to
a Krêtan divinity. The orgiastic worship of Zeus, solemnized by the
armed priests with impassioned motions and violent excitement, was
of ancient date in that island, as well as the connection with the
worship of Apollo both at Delphi and at Dêlos. To analyze the fables
and to elicit from them any trustworthy particular facts, appears to
me a fruitless attempt. The religious recollections, the romantic
invention, and the items of matter of fact, if any such there be, must
forever remain indissolubly amalgamated as the poet originally blended
them, for the amusement or edification of his auditors. Hoeckh, in his
instructive and learned collection of facts respecting ancient Krête,
construes the mythical genealogy of Minôs to denote a combination
of the orgiastic worship of Zeus, indigenous among the Eteokrêtes,
with the worship of the moon imported from Phœnicia, and signified by
the names Europê, Pasiphaê, and Ariadnê.[545] This is specious as a
conjecture, but I do not venture to speak of it in terms of greater
confidence.

  [544] Herodot. iii. 122. Πολυκράτης γὰρ ἐστὶ πρῶτος τῶν ἡμεῖς
  ἴδμεν Ἑλλήνων, ὃς θαλασσοκρατέειν ἐπενοήθη, παρὲξ Μίνωός τε τοῦ
  Κνωσσίου, καὶ εἰ δή τις ἄλλος πρότερος τούτου ἦρξε τῆς θαλάττης·
  ~τῆς δὲ ἀνθρωπηΐης λεγομένης γενεῆς~ Πολυκράτης ἐστὶ πρῶτος
  ἐλπίδας πολλὰς ἔχων Ἰωνίης τε καὶ νήσων ἄρξειν.

  The expression exactly corresponds to that of Pausanias, ix. 5,
  1, ἐπὶ τῶν καλουμένων Ἡρώων, for the age preceding the ἀνθρωπηΐη
  γενέη; also viii. 2. 1, ἐς τὰ ἀνωτέρω τοῦ ἀνθρώπων γένους.

  [545] Hoeckh, Kreta, vol. ii. pp. 56-67. K. O. Müller also
  (Dorier. ii. 2, 14) puts a religious interpretation upon these
  Kreto-Attic legends, but he explains them in a manner totally
  different from Hoeckh.

From the connection of religious worship and legendary tales between
Krête and various parts of Asia Minor,—the Trôad, the coast of Milêtus
and Lykia, especially between Mount Ida in Krête and Mount Ida in
Æôlis,—it seems reasonable to infer an ethnographical kindred or
relationship between the inhabitants anterior to the period of Hellenic
occupation. The tales of Krêtan settlement at Minoa and Engyiôn on
the south-western coast of Sicily, and in Iapygia on the Gulf of
Tarentum, conduct us to a similar presumption, though the want of
evidence forbids our tracing it farther. In the time of Herodotus, the
Eteokrêtes, or aboriginal inhabitants of the island, were confined to
Polichna and Præsus; but in earlier times, prior to the encroachments
of the Hellênes, they had occupied the larger portion, if not the whole
of the island. Minôs was originally their hero, subsequently adopted by
the immigrant Hellênes,—at least Herodotus considers him as barbarian,
not Hellenic.[546]

  [546] Herodot. i. 173.




CHAPTER XIII.

ARGONAUTIC EXPEDITION.


The ship Argô was the theme of many songs during the oldest periods
of the Grecian epic, even earlier than the Odyssey. The king Æêtês,
from whom she is departing, the hero Jasôn, who commands her, and the
goddess Hêrê, who watches over him, enabling the Argô to traverse
distances and to escape dangers which no ship had ever before
encountered, are all circumstances briefly glanced at by Odysseus in
his narrative to Alkinous. Moreover, Eunêus, the son of Jasôn and
Hypsipylê, governs Lemnos during the siege of Troy by Agamemnôn, and
carries on a friendly traffic with the Grecian camp, purchasing from
them their Trojan prisoners.[547]

  [547] Odyss. xii. 69.—

    Οἴη δὴ κείνη γε παρέπλει ποντόπορος νῆυς,
    Ἀργὼ πασιμέλουσα, παρ᾽ Αἰήταο πλέουσα·
    Καὶ νύ κε τὴν ἔνθ᾽ ὦκα βάλεν μεγάλας ποτὶ πέτρας,
    Ἀλλ᾽ Ἥρη παρέπεμψεν, ἐπεὶ φίλος ἦεν Ἰήσων.

  See also Iliad, vii. 470.

The legend of Halus in Achaia Phthiôtis, respecting the religious
solemnities connected with the family of Athamas and Phryxus (related
in a previous chapter), is also interwoven with the voyage of the
Argonauts; and both the legend and the solemnities seem evidently of
great antiquity. We know further, that the adventures of the Argô were
narrated not only by Hesiod and in the Hesiodic poems, but also by
Eumêlus and the author of the Naupactian verses—by the latter seemingly
at considerable length.[548] But these poems are unfortunately lost,
nor have we any means of determining what the original story was; for
the narrative, as we have it, borrowed from later sources, is enlarged
by local tales from the subsequent Greek colonies—Kyzikus, Heraklêia,
Sinopê, and others.

  [548] See Hesiod, Fragm. _Catalog._ Fr. 6. p. 33, Düntz.;
  _Eoiai_, Frag. 36. p. 39; Frag. 72. p. 47. Compare Schol. ad
  Apollôn. Rhod. i. 45; ii. 178-297, 1125; iv. 254-284. Other
  poetical sources—

  The old epic poem _Ægimius_, Frag. 5. p. 57, Düntz.

  _Kinæthôn_ in the _Hêraklêia_ touched upon the death of Hylas
  near Kius in Mysia (Schol. Apollôn. Rhod. i. 1357).

  The epic poem _Naupactia_, Frag. 1 to 6, Düntz. p. 61.

  _Eumêlus_, Frag. 2, 3, 5, p. 65, Düntz.

  _Epimenidês_, the Krêtan prophet and poet, composed a poem in
  6500 lines, Ἀργοῦς ναυπηγίαν τε, καὶ Ἰάσονος εἰς Κόλχους ἀποπλοῦν
  (Diogen. Laër. i. 10, 5), which is noticed more than once in the
  Scholia on Apollônius, on subjects connected with the poem (ii.
  1125; iii. 42). See Mimnerm. Frag. 10, Schneidewin, p. 15.

  _Antimachus_, in his poem _Lydê_, touched upon the Argonautic
  expedition, and has been partially copied by Apollônius Rhod.
  (Schol. Ap. Rh. i. 1290; ii. 296: iii. 410; iv. 1153).

  The logographers Pherekydês and Hekatæus seem to have related the
  expedition at considerable length.

  The Bibliothek der alten Literatur und Kunst (Göttingen, 1786,
  2tes Stück, p. 61) contains an instructive Dissertation by
  Groddeck, Ueber die Argonautika, a summary of the various
  authorities respecting this expedition.

Jasôn, commanded by Pelias to depart in quest of the golden fleece
belonging to the speaking ram which had carried away Phryxus and
Hellê, was encouraged by the oracle to invite the noblest youth
of Greece to his aid, and fifty of the most distinguished amongst
them obeyed the call. Hêraklês, Thêseus, Telamôn and Pêleus, Kastôr
and Pollux, Idas and Lynkeus—Zêtês and Kalaïs, the winged sons of
Boreas—Meleager, Amphiaraus, Kêpheus, Laertês, Autolykus, Menœtius,
Aktôr, Erginus, Euphêmus, Ankæus, Pœas, Periklymenus, Augeas, Eurytus,
Admêtus, Akastus, Kæneus, Euryalus, Pêneleôs and Lêitus, Askalaphus and
Ialmenus, were among them. Argus the son of Phryxus, directed by the
promptings of Athênê, built the ship, inserting in the prow a piece of
timber from the celebrated oak of Dodona, which was endued with the
faculty of speech:[549] Tiphys was the steersman, Idmôn the son of
Apollo and Mopsus accompanied them as prophets, while Orpheus came to
amuse their weariness and reconcile their quarrels with his harp.[550]

  [549] Apollôn. Rhod. i. 525; iv. 580. Apollodôr. i. 9, 16.
  Valerius Flaccus (i. 300) softens down the speech of the ship
  Argô into a dream of Jasôn. Alexander Polyhistor explained what
  wood was used (Plin. H. N. xiii. 22).

  [550] Apollônius Rhodius, Apollodôrus, Valerius Flaccus, the
  Orphic Argonautica, and Hyginus, have all given Catalogues of
  the Argonautic heroes (there was one also in the lost tragedy
  called Λήμνιαι of Sophoklês, see Welcker Gr. Trag. i. 327):
  the discrepancies among them are numerous and irreconcilable.
  Burmann, in the Catalogus Argonautarum, prefixed to his edition
  of Valerius Flaccus, has discussed them copiously. I transcribe
  one or two of the remarks of this conscientious and laborious
  critic, out of many of a similar tenor, on the impracticability
  of a fabulous chronology. Immediately before the first article,
  _Acastus_—“Neque enim in ætatibus Argonautarum ullam rationem
  temporum constare, neque in stirpe et stemmate deducenda ordinem
  ipsum naturæ congruere videbam. Nam et huic militiæ adscribi
  videbam Heroas, qui per naturæ leges et ordinem fati eo usque
  vitam extrahere non potuêre, ut aliis ab hac expeditione
  remotis Heroum militiis nomina dedisse narrari deberent a
  Poetis et Mythologis. In idem etiam tempus avos et Nepotes
  conjici, consanguineos ætate longe inferiores prioribus ut
  æquales adjungi, concoquere vix posse videtur.”—Art. _Ancæus_:
  “Scio objici posse, si seriem illam majorem respiciamus,
  hunc Ancæum simul cum proavo suo Talao in eandem profectum
  fuisse expeditionem. Sed similia exempla in aliis occurrent,
  et in fabulis rationem temporum non semper accuratam licet
  deducere.”—Art. _Jasôn_: “Herculi enim jam provectâ ætate
  adhæsit Theseus juvenis, et in Amazoniâ expeditione socius fuit,
  interfuit huic expeditioni, venatui apri Calydonii, et rapuit
  Helenam, quæ circa Trojanum bellum maxime floruit: quæ omnia si
  Theseus tot temporum intervallis distincta egit, secula duo vel
  tria vixisse debuit. Certe Jason Hypsipylem neptem Ariadnes, nec
  videre, nec Lemni cognoscere potuit.”—Art. _Meleager_: “Unum
  est quod alicui longum ordinem majorum recensenti scrupulum
  movere possit: nimis longum intervallum inter Æolum et Meleagrum
  intercedere, ut potuerit interfuisse huic expeditioni: cum
  nonus fere numeretur ab Æolo, et plurimi ut Jason, Argus, et
  alii tertiâ tantum ab Æolo generatione distent. Sed sæpe jam
  notavimus, frustra temporum concordiam in fabulis quæri.”

  Read also the articles _Castôr and Pollux_, _Nestôr Pêleus_,
  _Staphylus_, etc.

  We may stand excused for keeping clear of a chronology which is
  fertile only in difficulties, and ends in nothing but illusions.

First they touched at the island of Lêmnos, in which at that time there
were no men; for the women, infuriated by jealousy and ill-treatment,
had put to death their fathers, husbands and brothers. The Argonauts,
after some difficulty, were received with friendship, and even admitted
into the greatest intimacy. They staid some months, and the subsequent
population of the island was the fruit of their visit. Hypsipylê, the
queen of the island, bore to Jasôn two sons.[551]

  [551] Apollodôr. i. 9, 17; Apollôn. Rhod. i. 609-915; Herodot.
  iv. 145. Theocritus (Idyll, xiii. 29) omits all mention of
  Lêmnos, and represents the Argô as arriving on the third day
  from Iôlkos at the Hellespont. Diodôrus (iv. 41) also leaves out
  Lêmnos.

They then proceeded onward along the coast of Thrace, up the
Hellespont, to the southern coast of the Propontis, inhabited by the
Doliones and their king Kyzikus. Here they were kindly entertained, but
after their departure were driven back to the same spot by a storm; and
as they landed in the dark, the inhabitants did not know them. A battle
took place, in which the chief, Kyzikus, was killed by Jasôn; whereby
much grief was occasioned as soon as the real facts became known.
After Kyzikus had been interred with every demonstration of mourning
and solemnity, the Argonauts proceeded along the coast of Mysia.[552]
In this part of the voyage they left Hêraklês behind. For Hylas, his
favorite youthful companion, had been stolen away by the nymphs of a
fountain, and Hêraklês, wandering about in search of him, neglected
to return. At last he sorrowfully retired, exacting hostages from the
inhabitants of the neighboring town of Kius that they would persist in
the search.[553]

  [552] Apollôn. Rhod. 940-1020; Apollodôr. i. 9, 18.

  [553] Apollodôr. i. 9, 19. This was the religious legend,
  explanatory of a ceremony performed for many centuries by the
  people of Prusa: they ran round the lake Askanias shouting and
  clamoring for Hylas—“ut littus Hyla, Hyla omne sonaret.” (Virgil,
  Eclog.) ... “in cujus memoriam adhuc solemni cursatione lacum
  populus circuit et Hylam voce clamat.” Solinus, c. 42.

  There is endless discrepancy as to the concern of Hêraklês with
  the Argonautic expedition. A story is alluded to in Aristotle
  (Politic, iii. 9) that the ship Argô herself refused to take him
  on board, because he was so much superior in stature and power
  to all the other heroes—οὐ γὰρ ἐθέλειν αὐτὸν ἄγειν τὴν Ἀργὼ μετὰ
  τῶν ἄλλων, ὡς ὑπερβάλλοντα πολὺ τῶν πλωτήρων. This was the story
  of Pherekydês (Fr. 67, Didot) as well as of Antimachus (Schol.
  Apoll. Rhod. i. 1290): it is probably a very ancient portion of
  the legend, inasmuch as it ascribes to the ship sentient powers,
  in consonance with her other miraculous properties. The etymology
  of Aphetæ in Thessaly was connected with the tale of Hêraklês
  having there been put on shore from the Argô (Herodot. vii. 193):
  Ephorus said that he staid away voluntarily from fondness for
  Omphalê (Frag. 9, Didot). The old epic poet Kinæthôn said that
  Hêraklês had placed the Kian hostages at Trachin, and that the
  Kians ever afterwards maintained a respectful correspondence with
  that place (Schol. Ap. Rh. i. 1357). This is the explanatory
  legend connected with some existing custom, which we are unable
  further to unravel.

They next stopped in the country of the Bebrykians, where the
boxing contest took place between the king Amykus and the Argonaut
Pollux:[554] they then proceeded onward to Bithynia, the residence of
the blind prophet Phineus. His blindness had been inflicted by Poseidôn
as a punishment for having communicated to Phryxus the way to Kolchis.
The choice had been allowed to him between death and blindness, and he
had preferred the latter.[555] He was also tormented by the harpies,
winged monsters who came down from the clouds whenever his table
was set, snatched the food from his lips and imparted to it a foul
and unapproachable odor. In the midst of this misery, he hailed the
Argonauts as his deliverers—his prophetic powers having enabled him
to foresee their coming. The meal being prepared for him, the harpies
approached as usual, but Zêtês and Kalaïs, the winged sons of Boreas,
drove them away and pursued them. They put forth all their speed, and
prayed to Zeus to be enabled to overtake the monsters; when Hermês
appeared and directed them to desist, the harpies being forbidden
further to molest Phineus,[556] and retiring again to their native
cavern in Krête.[557]

  [554] See above, chap. viii. p. 169.

  [555] Such was the old narrative of the Hesiodic Catalogue and
  Eoiai. See Schol. Apollôn. Rhod. ii. 181-296.

  [556] This again was the old Hesiodic story (Schol. Apoll. Rhod.
  ii. 296),—

    Ἐνθ᾽ οἵγ᾽ εὔχεσθον Αἰνηΐῳ ὑψιμέδοντι.

  Apollodôrus (i. 9, 21), Apollônius (178-300), and Valerius Flacc.
  (iv. 428-530) agree in most of the circumstances.

  [557] Such was the fate of the harpies as given in the old
  Naupaktian Verses (See Fragm. Ep. Græc. Düntzer, Naupakt. Fr. 2.
  p. 61).

  The adventure of the Argonauts with Phineus is given by Diodôrus
  in a manner totally different (Diodôr. iv. 44): he seems to
  follow Dionysius of Mitylênê (see Schol. Apollôn. Rhod. ii. 207).

Phineus, grateful for the relief afforded to him by the Argonauts,
forewarned them of the dangers of their voyage and of the precautions
necessary for their safety; and through his suggestions they were
enabled to pass through the terrific rocks called Symplêgades. These
were two rocks which alternately opened and shut, with a swift and
violent collision, so that it was difficult even for a bird to fly
through during the short interval. When the Argô arrived at the
dangerous spot, Euphêmus let loose a dove. which flew through and just
escaped with the loss of a few feathers of her tail. This was a signal
to the Argonauts, according to the prediction of Phineus, that they
might attempt the passage with confidence. Accordingly they rowed with
all their might, and passed safely through: the closing rocks, held
for a moment asunder by the powerful arms of Athênê just crushed the
ornaments at the stern of their vessel. It had been decreed by the
gods, that so soon as any ship once got through, the passage should
forever afterwards be safe and easy to all. The rocks became fixed in
their separate places, and never again closed.[558]

  [558] Apollodôr. i. 9, 22. Apollôn. Rhod. ii. 310-615.

After again halting on the coast of the Mariandynians, where their
steersman Tiphys died, as well as in the country of the Amazons,
and after picking up the sons of Phryxus, who had been cast away by
Poseidôn in their attempt to return from Kolchis to Greece, they
arrived in safety at the river Phasis and the residence of Æêtês. In
passing by Mount Caucasus, they saw the eagle which gnawed the liver
of Prometheus nailed to the rock, and heard the groans of the sufferer
himself. The sons of Phryxus were cordially welcomed by their mother
Chalciopê.[559] Application was made to Æêtês, that he would grant to
the Argonauts, heroes of divine parentage and sent forth by the mandate
of the gods, possession of the golden fleece: their aid in return was
proffered to him against any or all of his enemies. But the king was
wroth, and peremptorily refused, except upon conditions which seemed
impracticable.[560] Hêphæstos had given him two ferocious and untamable
bulls, with brazen feet, which breathed fire from their nostrils:
Jasôn was invited, as a proof both of his illustrious descent and
of the sanction of the gods to his voyage, to harness these animals
to the yoke, so as to plough a large field and sow it with dragon’s
teeth.[561] Perilous as the condition was, each one of the heroes
volunteered to make the attempt. Idmôn especially encouraged Jasôn to
undertake it,[562] and the goddesses Hêrê and Aphroditê made straight
the way for him.[563] Mêdea, the daughter of Æêtês and Eidyia, having
seen the youthful hero in his interview with her father, had conceived
towards him a passion which disposed her to employ every means for his
salvation and success. She had received from Hekatê preëminent magical
powers, and she prepared for Jasôn the powerful Prometheian unguent,
extracted from an herb which had grown where the blood of Promêtheus
dropped. The body of Jasôn having been thus pre-medicated, became
invulnerable[564] either by fire or by warlike weapons. He undertook
the enterprise, yoked the bulls without suffering injury, and ploughed
the field: when he had sown the dragon’s teeth, armed men sprung out of
the furrows. But he had been forewarned by Mêdea to cast a vast rock
into the midst of them, upon which they began to fight with each other,
so that he was easily enabled to subdue them all.[565]

  [559] Apollodôr. i. 9, 23. Apollôn. Rhod. ii. 850-1257.

  [560] Apollôn. Rhod. iii. 320-385.

  [561] Apollôn. Rhod. iii. 410. Apollodôr. i. 9, 23.

  [562] This was the story of the Naupaktian Verses (Schol.
  Apollôn. Rhod. iii. 515-525): Apollônius and others altered it.
  Idmôn, according to them, died in the voyage before the arrival
  at Kolchis.

  [563] Apollôn. Rhod. iii. 50-200. Valer. Flacc. vi. 440-480.
  Hygin. fab. 22.

  [564] Apollôn. Rhod. iii. 835. Apollodôr. i. 9, 23. Valer. Flacc
  vii. 356 Ovid, Epist. xii. 15.

    “Isset anhelatos non præmedicatus in ignes
      Immemor Æsonides, oraque adunca boum.”

  [565] Apollôn. Rhod. iii. 1230-1400.

The task prescribed had thus been triumphantly performed. Yet Æêtês not
only refused to hand over the golden fleece, but even took measures for
secretly destroying the Argonauts and burning their vessel. He designed
to murder them during the night after a festal banquet; but Aphroditê,
watchful for the safety of Jasôn,[566] inspired the Kolchian king at
the critical moment with an irresistible inclination for his nuptial
bed. While he slept, the wise Idmôn counselled the Argonauts to make
their escape, and Mêdea agreed to accompany them.[567] She lulled to
sleep by a magic potion the dragon who guarded the golden fleece,
placed that much-desired prize on board the vessel, and accompanied
Jasôn with his companions in their flight, carrying along with her the
young Apsyrtus, her brother.[568]

  [566] The Naupaktian Verses stated this (see the Fragm. 6, ed.
  Düntzer, p. 61, ap. Schol. Apollôn. Rhod. iv. 59-86).

  [567] Such was the story of the Naupaktian Verses (See Fragm. 6.
  p 61 Düntzer ap. Schol. Apollôn. Rhod. iv. 59, 86, 87).

  [568] Apollodôr. i. 9, 23. Apollôn. Rhod. iv. 220.

  Pherekydês said that Jasôn killed the dragon (Fr. 74, Did.).

Æêtês, profoundly exasperated at the flight of the Argonauts with his
daughter, assembled his forces forthwith, and put to sea in pursuit
of them. So energetic were his efforts that he shortly overtook the
retreating vessel, when the Argonauts again owed their safety to the
stratagem of Mêdea. She killed her brother Apsyrtus, cut his body in
pieces and strewed the limbs round about in the sea. Æêtês on reaching
the spot found these sorrowful traces of his murdered son; but while he
tarried to collect the scattered fragments, and bestow upon the body
an honorable interment, the Argonauts escaped.[569] The spot on which
the unfortunate Apsyrtus was cut up received the name of Tomi.[570]
This fratricide of Mêdea, however, so deeply provoked the indignation
of Zeus, that he condemned the Argô and her crew to a trying voyage,
full of hardship and privation, before she was permitted to reach home.
The returning heroes traversed an immeasurable length both of sea and
of river: first up the river Phasis into the ocean which flows round
the earth—then following the course of that circumfluous stream until
its junction with the Nile,[571] they came down the Nile into Egypt,
from whence they carried the Argô on their shoulders by a fatiguing
land-journey to the lake Tritônis in Libya. Here they were rescued from
the extremity of want and exhaustion by the kindness of the local god
Tritôn, who treated them hospitably, and even presented to Euphêmus a
clod of earth, as a symbolical promise that his descendants should one
day found a city on the Libyan shore. The promise was amply redeemed
by the flourishing and powerful city of Kyrênê,[572] whose princes the
Battiads boasted themselves as lineal descendants of Euphêmus.

  [569] This is the story of Apollodôrus (i. 9, 24), who seems
  to follow Pherekydês (Fr. 73, Didot). Apollônius (iv. 225-480)
  and Valerius Flaccus (viii. 262 _seq._) give totally different
  circumstances respecting the death of Apsyrtus; but the narrative
  of Pherekydês seems the oldest: so revolting a story as that of
  the cutting up of the little boy cannot have been imagined in
  later times.

  Sophoklês composed two tragedies on the adventures of Jasôn and
  Mêdea, both lost—the Κολχίδες and the Σκύθαι. In the former he
  represented the murder of the child Apsyrtus as having taken
  place in the house of Æêtês: in the latter he introduced the
  mitigating circumstance, that Apsyrtus was the son of Æêtês by a
  different mother from Mêdea (Schol. Apollôn Rhod. iv. 223).

  [570] Apollodôr. i. 9, 24, τὸν τόπον προσηγόρευσε Τόμους. Ovid.
  Trist. iii. 9. The story that Apsyrtus was cut in pieces, is the
  etymological legend explanatory of the name Tomi.

  There was however a place called Apsarus, on the southern coast
  of the Euxine, west of Trapezus, where the tomb of Apsyrtus was
  shown, and where it was affirmed that he had been put to death.
  He was the eponymus of the town, which was said to have been once
  called Apsyrtus, and only corrupted by a barbarian pronunciation
  (Arrian. Periplus, Euxin. p. 6; Geogr. Min. v. 1). Compare
  Procop. Bell. Goth. iv. 2.

  Strabo connects the death of Apsyrtus with the Apsyrtides,
  islands off the coast of Illyria, in the Adriatic (vii p. 315).

  [571] The original narrative was, that the Argô returned by
  navigating the circumfluous ocean. This would be almost certain,
  even without positive testimony, from the early ideas entertained
  by the Greeks respecting geography; but we know further that
  it was the representation of the Hesiodic poems, as well as of
  Mimnermus, Hekatæus and Pindar, and even of Antimachus. Schol.
  Parisina Ap. Rhod. iv. 254. Ἑκαταῖος δὲ ὁ Μιλήσιος διὰ τοῦ
  Φάσιδος ἀνελθεῖν φησὶν αὐτοὺς εἰς τὸν Ὠκεανόν· διὰ δὲ τοῦ Ὠκεανοῦ
  κατελθεῖν εἰς τὸν Νεῖλον· ἐκ δὲ τοῦ Νείλου εἰς τὴν καθ᾽ ἡμᾶς
  θάλασσαν. Ἡσίοδος δὲ καὶ Πίνδαρος ἐν Πυθιονίκαις καὶ Ἀντίμαχος
  ἐν Λυδῇ διὰ τοῦ Ὠκεανοῦ φασὶν ἐλθεῖν αὐτοὺς εἰς Λιβύην· εἶτα
  βαστάσαντας τὴν Ἀργὼ εἰς τὸ ἡμέτερον ἀφικέσθαι πέλαγος. Compare
  the Schol. Edit. ad iv. 259.

  [572] See the fourth Pythian Ode of Pindar, and Apollôn. Rhod.
  iv. 1551-1756.

  The tripod of Jasôn was preserved by the Euesperitæ in Libya,
  Diod. iv. 56: but the legend, connecting the Argonauts with
  the lake Tritônis in Libya, is given with some considerable
  differences in Herodotus, iv. 179.

Refreshed by the hospitality of Tritôn, the Argonauts found themselves
again on the waters of the Mediterranean in their way homeward. But
before they arrived at Iôlkos they visited Circê, at the island of Ææa,
where Mêdea was purified for the murder of Apsyrtus: they also stopped
at Korkyra, then called Drepanê, where Alkinous received and protected
them. The cave in that island where the marriage of Mêdea with Jasôn
was consummated, was still shown in the time of the historian Timæus,
as well as the altars to Apollo which she had erected, and the rites
and sacrifices which she had first instituted.[573] After leaving
Korkyra, the Argô was overtaken by a perilous storm near the island of
Thêra. The heroes were saved from imminent peril by the supernatural
aid of Apollo, who, shooting from his golden bow an arrow which pierced
the waves like a track of light, caused a new island suddenly to spring
up in their track and present to them a port of refuge. The island was
called Anaphê; and the grateful Argonauts established upon it an altar
and sacrifices in honor of Apollo Æglêtês, which were ever afterwards
continued, and traced back by the inhabitants to this originating
adventure.[574]

  [573] Apollôn. Rhod. iv. 1153-1217. Timæus, Fr. 7-8, Didot.
  Τίμαιος ἐν Κερκύρᾳ λέγων γενέσθαι τοὺς γάμους, καὶ περὶ τῆς
  θυσίας ἱστορεῖ, ἔτι καὶ νῦν λέγων ἄγεσθαι αὐτὴν κατ᾽ ἐνιαυτὸν,
  Μηδείας πρῶτον θυσάσης ἐν τῷ τοῦ Απολλῶνος ἱερῷ. Καὶ Βωμοὺς δέ
  φησι μνημεῖα τῶν γάμων ἱδρύσασθαι συνεγγὺς μὲν τῆς θαλάσσης, οὐ
  μακρὰν δὲ τῆς πόλεως. Ὀνομάζουσι δὲ τὸν μὲν, Νυμφῶν· τὸν δὲ,
  Νηρηΐδων.

  [574] Apollodôr. i. 9, 25. Apollôn. Rhod. iv. 1700-1725.

On approaching the coast of Krête, the Argonauts were prevented
from landing by Talôs, a man of brass, fabricated by Hêphæstos, and
presented by him to Minôs for the protection of the island.[575] This
vigilant sentinel hurled against the approaching vessel fragments of
rock, and menaced the heroes with destruction. But Mêdea deceived
him by a stratagem and killed him; detecting and assailing the one
vulnerable point in his body. The Argonauts were thus enabled to land
and refresh themselves. They next proceeded onward to Ægina, where
however they again experienced resistance before they could obtain
water—then along the coast of Eubœa and Locris back to Iôlkos in the
gulf of Pagasæ, the place from whence they had started. The proceedings
of Pelias during their absence, and the signal revenge taken upon him
by Mêdea after their return, have already been narrated in a preceding
section.[576] The ship Argô herself, in which the chosen heroes of
Greece had performed so long a voyage and braved so many dangers, was
consecrated by Jasôn to Poseidôn at the isthmus of Corinth. According
to another account, she was translated to the stars by Athênê, and
became a constellation.[577]

  [575] Some called Talôs a remnant of the brazen race of men
  (Schol. Apoll. Rhod. iv. 1641).

  [576] Apollodôr. i. 9, 26. Apollôn. Rhod. iv. 1638.

  [577] Diodôr. iv. 53. Eratosth. Catasterism. c. 35.

Traces of the presence of the Argonauts were found not only in the
regions which lay between Iôlkos and Kolchis, but also in the western
portion of the Grecian world—distributed more or less over all the
spots visited by Grecian mariners or settled by Grecian colonists, and
scarcely less numerous than the wanderings of the dispersed Greeks
and Trojans after the capture of Troy. The number of Jasonia, or
temples for the heroic worship of Jasôn, was very great, from Abdêra
in Thrace,[578] eastward along the coast of the Euxine, to Armenia
and Media. The Argonauts had left their anchoring-stone on the coast
of Bebrykia, near Kyzikus, and there it was preserved during the
historical ages in the temple of the Jasonian Athênê.[579] They had
founded the great temple of the Idæan mother on the mountain Dindymon,
near Kyzikus, and the Hieron of Zeus Urios on the Asiatic point at the
mouth of the Euxine, near which was also the harbor of Phryxus.[580]
Idmôn, the prophet of the expedition, who was believed to have died
of a wound by a wild boar on the Mariandynian coast, was worshipped
by the inhabitants of the Pontic Hêrakleia with great solemnity, as
their Heros Poliuchus, and that too by the special direction of the
Delphian god. Autolykus, another companion of Jasôn, was worshipped
as Œkist by the inhabitants of Sinopê. Moreover, the historians of
Hêrakleia pointed out a temple of Hekatê in the neighboring country of
Paphlagonia, first erected by Mêdea;[581] and the important town of
Pantikapæon, on the European side of the Cimmerian Bosporus, ascribed
its first settlement to a son of Æêtês.[582] When the returning ten
thousand Greeks sailed along the coast, called the Jasonian shore,
from Sinopê to Hêrakleia, they were told that the grandson of Æêtês
was reigning king of the territory at the mouth of the Phasis, and the
anchoring-places where the Argô had stopped were specially pointed
out to them.[583] In the lofty regions of the Moschi, near Kolchis,
stood the temple of Leukothea, founded by Phryxus, which remained
both rich and respected down to the times of the kings of Pontus, and
where it was an inviolable rule not to offer up a ram.[584] The town
of Dioskurias, north of the river Phasis, was believed to have been
hallowed by the presence of Kastôr and Pollux in the Argô, and to have
received from them its appellation.[585] Even the interior of Mêdea and
Armenia was full of memorials of Jasôn and Mêdea and their son Mêdus,
or of Armenus the son of Jasôn, from whom the Greeks deduced not only
the name and foundation of the Medes and Armenians, but also the great
operation of cutting a channel through the mountains for the efflux
of the river Araxes, which they compared to that of the Peneius in
Thessaly.[586] And the Roman general Pompey, after having completed
the conquest and expulsion of Mithridatês, made long marches through
Kolchis into the regions of Caucasus, for the express purpose of
contemplating the spots which had been ennobled by the exploits of the
Argonauts, the Dioskuri and Hêraklês.[587]

  [578] Strabo. xi. p. 526-531.

  [579] Apollôn. Rhod. i. 955-960, and the Scholia.

  There was in Kyzikus a temple of Apollo under different
  ἐπικλήσεις; some called it the temple of the Jasonian Apollo.

  Another anchor however was preserved in the temple of Rhea on the
  banks of the Phasis, which was affirmed to be the anchor of the
  ship Argô. Arrian saw it there, but seems to have doubted its
  authenticity (Periplus, Euxin. Pont. p. 9. Geogr. Min. v. 1).

  [580] Neanthês ap. Strabo. i. p. 45. Apollôn. Rhod. i. 1125, and
  Schol. Steph. Byz. v. Φρίξος.

  Apollônius mentions the fountain called Jasoneæ, on the hill
  of Dindymon. Apollôn. Rhod. ii. 532, and the citations from
  Timosthenês and Herodôrus in the Scholia. See also Appian.
  Syriac. c. 63.

  [581] See the historians of Hêrakleia, Nymphis and Promathidas,
  Fragm. Orelli, pp. 99, 100-104. Schol. ad Apollôn. Rhod. iv. 247.
  Strabo, xii. p. 546. Autolykus, whom he calls companion of Jasôn,
  was, according to another legend, comrade of Hêraklês in his
  expedition against the Amazons.

  [582] Stephan. Byz. v. Παντικαπαῖον, Eustath. ad Dionys.
  Periêgêt. 311.

  [583] Xenophôn, Anabas. vi. 2, 1; v. 7, 37.

  [584] Strabo, xi. p. 499.

  [585] Appian, Mithridatic. c. 101.

  [586] Strabo, xi. p. 499, 503, 526, 531; i. p. 45-48. Justin,
  xlii. 3, whose statements illustrate the way in which men found a
  present home and application for the old fables,—“Jason, primus
  humanorum post Herculem et Liberum, qui reges Orientis fuisse
  traduntur, eam cœli plagam domuisse dicitur. Cum Albanis fœdus
  percussit, qui Herculem ex Italiâ ab Albano monte, cum, Geryone
  extincto, armenta ejus per Italiam duceret, secuti dicuntur;
  quique, memores Italicæ originis, exercitum Cn. Pompeii bello
  Mithridatico fratres consalutavêre. Itaque Jasoni totus fere
  Oriens, ut conditori, divinos honores templaque constituit; quæ
  Parmenio, dux Alexandri Magni, post multos annos dirui jussit, ne
  cujusquam nomen in Oriente venerabilius quam Alexandri esset.”

  The Thessalian companions of Alexander the Great, placed by his
  victories in possession of rich acquisitions in these regions,
  pleased themselves by vivifying and multiplying all these
  old fables, proving an ancient kindred between the Medes and
  Thessalians. See Strabo, xi. p. 530. The temples of Jasôn were
  τιμώμενα σφόδρα ὑπὸ τῶν βαρβάρων (_ib._ p. 526).

  The able and inquisitive geographer Eratosthenês was among those
  who fully believed that Jasôn had left his ships in the Phasis,
  and had undertaken a land expedition into the interior country,
  in which he had conquered Media and Armenia (Strabo, i. p. 48).

  [587] Appian, Mithridatic. 103: τοὺς Κόλχους ἐπήει, καθ᾽ ἰστορίαν
  τῆς Ἀργοναυτῶν καὶ Διοσκούρων καὶ Ἡρακλέους ἐπιδημίας, καὶ
  μάλιστα τὸ πάθος ἰδεῖν ἐθέλων, ὃ Προμηθεῖ φασὶ γενέσθαι περὶ τὸ
  Καύκασον ὄρος. The lofty crag of Caucasus called Strobilus, to
  which Promêtheus had been attached, was pointed out to Arrian
  himself in his Periplus (p. 12. Geogr. Minor vol. i.).

In the west, memorials either of the Argonauts or of the pursuing
Kolchians were pointed out in Korkyra, in Krête, in Epirus near the
Akrokeraunian mountains, in the islands called Apsyrtides near the
Illyrian coast, at the bay of Caieta as well as at Poseidônia on the
southern coast of Italy, in the island of Æthalia or Elba, and in
Libya.[588]

  [588] Strabo, i. pp. 21, 45, 46; v. 224-252. Pompon. Mel. ii. 3.
  Diodôr. iv. 56. Apollôn. Rhod. iv. 656. Lycophron, 1273.—

    Τύρσιν μακεδνὰς ἀμφὶ Κιρκαίου νάπας
    Ἀργοῦς τε κλεινὸν ὅρμον Αἰήτην μέγαν.

Such is a brief outline of the Argonautic expedition, one of the most
celebrated and widely-diffused among the ancient tales of Greece. Since
so many able men have treated it as an undisputed reality, and even
made it the pivot of systematic chronological calculations, I may here
repeat the opinion long ago expressed by Heyne, and even indicated by
Burmann, that the process of dissecting the story, in search of a basis
of fact, is one altogether fruitless.[589] Not only are we unable to
assign the date or identify the crew, or decipher the log-book, of the
Argô, but we have no means of settling even the preliminary question,
whether the voyage be matter of fact badly reported, or legend from
the beginning. The widely-distant spots in which the monuments of the
voyage were shown, no less than the incidents of the voyage itself,
suggests no other parentage than epical fancy. The supernatural and the
romantic not only constitute an inseparable portion of the narrative,
but even embrace all the prominent and characteristic features; if they
do not comprise the whole, and if there be intermingled along with
them any sprinkling of historical or geographical fact,—a question
to us indeterminable,—there is at least no solvent by which it can
be disengaged, and no test by which it can be recognized. Wherever
the Grecian mariner sailed, he carried his religious and patriotic
mythes along with him. His fancy and his faith were alike full of
the long wanderings of Jasôn, Odysseus, Perseus, Hêraklês, Dionysus,
Triptolemus or Iô; it was pleasing to him in success, and consoling
to him in difficulty, to believe that their journeys had brought them
over the ground which he was himself traversing. There was no tale
amidst the wide range of the Grecian epic more calculated to be popular
with the seaman, than the history of the primæval ship Argô and her
distinguished crew, comprising heroes from all parts of Greece, and
especially the Tyndarids Kastôr and Pollux, the heavenly protectors
invoked during storm and peril. He localized the legend anew wherever
he went, often with some fresh circumstances suggested either by his
own adventures or by the scene before him. He took a sort of religious
possession of the spot, connecting it by a bond of faith with his
native land, and erecting in it a temple or an altar with appropriate
commemorative solemnities. The Jasonium thus established, and indeed
every visible object called after the name of the hero, not only served
to keep alive the legend of the Argô in the minds of future comers or
inhabitants, but was accepted as an obvious and satisfactory proof that
this marvellous vessel had actually touched there in her voyage.

  [589] Heyne, Observ. ad Apollodôr. i. 9, 16. p. 72. “Mirum in
  modum fallitur, qui in his commentis certum fundum historicum
  vel geographicum aut exquirere studet, aut se reperisse, atque
  historicam vel geographicam aliquam doctrinam, systema nos
  dicimus, inde procudi posse, putat,” etc.

  See also the observations interspersed in Burmann’s Catalogus
  Argonautarum, prefixed to his edition of Valerius Flaccus.

  The Persian antiquarians whom Herodotus cites at the beginning
  of his history (i. 2-4—it is much to be regretted that Herodotus
  did not inform us who they were, and whether they were the same
  as those who said that Perseus was an Assyrian by birth and
  had become a Greek, vi. 54), joined together the abductions of
  Iô and of Eurôpê, of Mêdea and of Helen, as pairs of connected
  proceedings, the second injury being a retaliation for the
  first,—they drew up a debtor and creditor account of abductions
  between Asia and Europe. The Kolchian king (they said) had sent a
  herald to Greece to ask for his satisfaction for the wrong done
  to him by Jasôn and to re-demand his daughter Mêdea; but he was
  told in reply that the Greeks had received no satisfaction for
  the previous rape of Iô.

  There was some ingenuity in thus binding together the old
  fables, so as to represent the invasions of Greece by Darius and
  Xerxês as retaliations for the unexpiated destruction wrought by
  Agamemnôn.

The epic poets, building both on the general love of fabulous incident
and on the easy faith of the people, dealt with distant and unknown
space in the same manner as with past and unrecorded time. They created
a mythical geography for the former, and a mythical history for the
latter. But there was this material difference between the two: that
while the unrecorded time was beyond the reach of verification, the
unknown space gradually became trodden and examined. In proportion as
authentic local knowledge was enlarged, it became necessary to modify
the geography, or shift the scene of action, of the old mythes; and
this perplexing problem was undertaken by some of the ablest historians
and geographers of antiquity,—for it was painful to them to abandon any
portion of the old epic, as if it were destitute of an ascertainable
basis of truth.

Many of these fabulous localities are to be found in Homer and Hesiod,
and the other Greek poets and logographers,—Erytheia, the garden of
the Hesperides, the garden of Phœbus,[590] to which Boreas transported
the Attic maiden Orithyia, the delicious country of the Hyperboreans,
the Elysian plain,[591] the fleeting island of Æolus, Thrinakia,
the country of the Æthiopians, the Læstrygones, the Kyklôpes, the
Lotophagi, the Sirens, the Cimmerians and the Gorgons,[592] etc. These
are places which (to use the expression of Pindar respecting the
Hyperboreans) you cannot approach either by sea or by land:[593] the
wings of the poet alone can carry you thither. They were not introduced
into the Greek mind by incorrect geographical reports, but, on the
contrary, had their origin in the legend, and passed from thence
into the realities of geography,[594] which they contributed much to
pervert and confuse. For the navigator or emigrant, starting with an
unsuspicious faith in their real existence, looked out for them in
his distant voyages, and constantly fancied that he had seen or heard
of them, so as to be able to identify their exact situation. The most
contradictory accounts indeed, as might be expected, were often given
respecting the latitude and longitude of such fanciful spots, but this
did not put an end to the general belief in their real existence.

  [590] Sophokl. ap. Strabo. vii. p. 295.—

    Ὑπέρ τε πόντον πάντ᾽ ἐπ᾽ ἔσχατα χθονὸς,
    Νυκτός τε πηγὰς οὐρανοῦ τ᾽ ἀναπτυχὰς,
    Φοίβου τε παλαιὸν κῆπον.

  [591] Odyss. iv. 562. The Islands of the Blessed, in Hesiod, are
  near the ocean (Opp. Di. 169).

  [592] Hesiod, Theogon. 275-290. Homer, Iliad, i. 423. Odyss. i.
  23; ix 86-206; x 4-83; xii. 135. Mimnerm. Fragm. 13, Schneidewin.

  [593] Pindar, Pyth. x. 29.—

    Ναυσὶ δ᾽ οὔτε πεζὸς ἰὼν ἂν εὕροις
    Ἐς Ὑπερβορέων ἀγῶνα θαυματὰν ὁδόν.
    Παρ᾽ οἷς ποτε Περσεὺς ἐδαίσατο λαγετὰς, etc.

  Hesiod, and the old epic poem called the Epigoni, both mentioned
  the Hyperboreans (Herod. iv. 32-34).

  [594] This idea is well stated and sustained by Völcker
  (Mythische Geographie der Griechen und Römer, cap. i. p. 11),
  and by Nitzsch in his Comments on the Odyssey—Introduct. Remarks
  to b. ix. p. xii.-xxxiii. The twelfth and thirteenth chapters of
  the History of Orchomenos, by O. Müller, are also full of good
  remarks on the geography of the Argonautic voyage (pp. 274-299).

  The most striking evidence of this disposition of the Greeks is
  to be found in the legendary discoveries of Alexander and his
  companions, when they marched over the untrodden regions in the
  east of the Persian empire (see Arrian, Hist. Al. v. 3: compare
  Lucian. Dialog. Mortuor. xiv. vol. i. p. 212. Tauch) because
  these ideas were first broached at a time when geographical
  science was sufficiently advanced to canvass and criticize them.
  The early settlers in Italy, Sicily and the Euxine, indulged
  their fanciful vision without the fear of any such monitor: there
  was no such thing as a map before the days of Anaximander, the
  disciple of Thalês.

In the present advanced state of geographical knowledge, the story of
that man who after reading Gulliver’s Travels went to look in his
map for Lilliput, appears an absurdity. But those who fixed the exact
locality of the floating island of Æolus or the rocks of the Sirens
did much the same,[595] and, with their ignorance of geography and
imperfect appreciation of historical evidence, the error was hardly to
be avoided. The ancient belief which fixed the Sirens on the islands
of Sirenusæ off the coast of Naples—the Kyklôpes, Erytheia, and the
Læstrygones in Sicily—the Lotophagi on the island of Mêninx[596] near
the Lesser Syrtis—the Phæakians at Korkyra—and the goddess Circê at the
promontory of Circeium—took its rise at a time when these regions were
first Hellenized and comparatively little visited. Once embodied in the
local legends, and attested by visible monuments and ceremonies, it
continued for a long time unassailed; and Thucydidês seems to adopt it,
in reference to Korkyra and Sicily before the Hellenic colonization, as
matter of fact generally unquestionable,[597] though little avouched as
to details. But when geographical knowledge became extended, and the
criticism upon the ancient epic was more or less systematized by the
literary men of Alexandria and Pergamus, it appeared to many of them
impossible that Odysseus could have seen so many wonders, or undergone
such monstrous dangers, within limits so narrow, and in the familiar
track between the Nile and the Tiber. The scene of his weather-driven
course was then shifted further westward. Many convincing evidences
were discovered, especially by Asklepiadês of Myrlea, of his having
visited various places in Ibêria:[598] several critics imagined that
he had wandered about in the Atlantic Ocean outside of the Strait of
Gibraltar,[599] and they recognized a section of Lotophagi on the
coast of Mauritania, over and above those who dwelt on the island of
Mêninx.[600] On the other hand, Eratosthenês and Apollodôrus treated
the places visited by Odysseus as altogether unreal, for which
scepticism they incurred much reproach.[601]

  [595] See Mr. Payne Knight, Prolegg. ad Homer. c. 49. Compare
  Spohn—“de extremâ Odysseæ parte”—p. 97.

  [596] Strabo. xvii. p. 834. An altar of Odysseus was shown upon
  this island, as well as some other evidences (σύμβολα) of his
  visit to the place.

  Apollônius Rhodius copies the Odyssey in speaking of the island
  of Thrinakia and the cattle of Helios (iv. 965, with Schol.). He
  conceives Sicily as Thrinakia, a name afterwards exchanged for
  Trinakria. The Scholiast ad Apoll. (1. c.) speaks of Trinax king
  of Sicily. Compare iv. 291 with the Scholia.

  [597] Thucyd. i. 25-vi. 2. These local legends appear in the eyes
  of Strabo convincing evidence (i. p. 23-26),—the tomb of the
  siren Parthenopê at Naples, the stories at Cumæ and Dikæarchia
  about the νεκυομαντεῖον of Avernus, and the existence of places
  named after Baius and Misênus, the companions of Odysseus, etc.

  [598] Strabo, iii. p. 150-157. Οὐ γὰρ μόνον οἱ κατὰ τὴν
  Ἰταλίαν καὶ Σικελίαν τόποι καὶ ἄλλοι τινὲς τῶν τοιούτων
  σημεῖα ὑπογράφουσιν, ἀλλὰ καὶ ἐν τῇ Ἰβηρίᾳ Ὀδύσσεια πόλις
  δείκνυται, καὶ Ἀθηνᾶς ἱερὸν, καὶ ἄλλα μύρια ἴχνη τῆς τε ἐκείνου
  πλάνης, καὶ ἄλλων τῶν ἐκ τοῦ Τρωϊκοῦ πολέμου περιγενομένων (I
  adopt Grosskurd’s correction of the text from γενομένων to
  περιγενομένων, in the note to his German translation of Strabo).

  Asklepiadês (of Myrlea in Bithynia, about 170 B. C.) resided
  some time in Turditania, the south-western region of Spain along
  the Guadalquivir, as a teacher of Greek literature (παιδεύσας τὰ
  γραμματικὰ), and composed a periegesis of the Iberian tribes,
  which unfortunately has not been preserved. He made various
  discoveries in archæology, and successfully connected his old
  legends with several portions of the territory before him. His
  discoveries were,—1. In the temple of Athênê, at this Iberian
  town of Odysseia, there were shields and beaks of ships affixed
  to the walls, monuments of the visit of Odysseus himself. 2.
  Among the Kallæki, in the northern part of Portugal, several of
  the companions of Teukros had settled and left descendants: there
  were in that region two Grecian cities, one called Hellenês,
  the other called Amphilochi; for Amphilochus also, the son of
  Amphiaraus, had died in Iberia, and many of his soldiers had
  taken up their permanent residence in the interior. 3. Many new
  inhabitants had come into Iberia with the expedition of Hêraklês;
  some also after the conquest of Messênê by the Lacedæmonians. 4.
  In Cantabria, on the north. coast of Spain, there was a town and
  region of Lacedæmonian colonists. 5. In the same portion of the
  country there was the town of Opsikella, founded by Opsikellas,
  one of the companions of Antenôr in his emigration from Troy
  (Strabo, iii. p. 157).

  This is a specimen of the manner in which the seeds of Grecian
  mythus came to be distributed over so large a surface. To an
  ordinary Greek reader, these legendary discoveries of Asklepiadês
  would probably be more interesting than the positive facts
  which he communicated respecting the Iberian tribes; and his
  Turditanian auditors would be delighted to hear—while he was
  reciting and explaining to them the animated passage of the
  Iliad, in which Agamemnôn extols the inestimable value of the bow
  of Teukros (viii. 281)—that the heroic archer and his companions
  had actually set foot in the Iberian peninsula.

  [599] This was the opinion of Kratês of Mallus, one of the
  most distinguished of the critics on Homer: it was the subject
  of an animated controversy between him and Aristarchus (Aulus
  Gellius, N. A. xiv. 6; Strabo, iii. p. 157). See the instructive
  treatise of Lehrs, De Aristarchi Studiis, c. v. § 4. p. 251.
  Much controversy also took place among the critics respecting
  the ground which Menelaus went over in his wanderings (Odyss.
  iv.). Kratês affirmed that he had circumnavigated the southern
  extremity of Africa and gone to India: the critic Aristonikus,
  Strabo’s contemporary, enumerated all the different opinions
  (Strabo, i. p. 38).

  [600] Strabo, iii. p. 157.

  [601] Strabo, i. p. 22-44; vii. p. 299.

The fabulous island of Erytheia,—the residence of the three headed
Geryôn with his magnificent herd of oxen, under the custody of the
two-headed dog Orthrus, and described by Hesiod, like the garden of
the Hesperides, as extra-terrestrial, on the farther side of the
circumfluous ocean;—this island was supposed by the interpreters of
Stesichorus the poet to be named by him off the south-western region
of Spain called Tartêssus, and in the immediate vicinity of Gadês. But
the historian Hekatæus, in his anxiety to historicize the old fable,
took upon himself to remove Erytheia from Spain nearer home to Epirus.
He thought it incredible that Hêraklês should have traversed Europe
from east to west, for the purpose of bringing the cattle of Geryôn to
Eurystheus at Mykênæ, and he pronounced Geryôn to have been a king of
Epirus, near the Gulf of Ambrakia. The oxen reared in that neighborhood
were proverbially magnificent, and to get them even from thence and
bring them to Mykênæ (he contended) was no inconsiderable task. Arrian,
who cites this passage from Hekatæus, concurs in the same view,—an
illustration of the license with which ancient authors fitted on their
fabulous geographical names to the real earth, and brought down the
ethereal matter of legend to the lower atmosphere of history.[602]

  [602] Stesichori Fragm. ed. Kleine; Geryonis, Fr. 5. p. 60; ap.
  Strabo. iii. p. 148; Herodot. iv. 8. It seems very doubtful
  whether Stesichorus meant to indicate any neighboring island as
  Erytheia, if we compare Fragm. 10. p. 67 of the Geryonis, and the
  passages of Athenæus and Eustathius there cited. He seems to have
  adhered to the old fable, placing Erytheia on the opposite side
  of the ocean-stream, for Hêraklês crosses the ocean to get to it.

  Hekatæus, ap. Arrian. Histor. Alex. ii. 16. Skylax places
  Erytheia, “whither Geryôn is said to have come to feed his oxen,”
  in the Kastid territory near the Greek city of Apollônia on the
  Ionic Gulf, northward of the Keraunian mountains. There were
  splendid cattle consecrated to Hêlios near Apollônia, watched
  by the citizens of the place with great care (Herodot. ix. 93;
  Skylax, c. 26).

  About Erytheia, Cellerius observes (Geogr. Ant. ii. 1, 227),
  “Insula Erytheia, quam veteres adjungunt Gadibus, vel demersa
  est, vel in scopulis quærenda, vel pars est ipsarum Gadium, neque
  hodie ejus formæ aliqua, uti descripta est, fertur superesse.” To
  make the disjunctive catalogue complete, he ought to have added,
  “or it never really existed,”—not the least probable supposition
  of all.

Both the track and the terminus of the Argonautic voyage appear in the
most ancient epic as little within the conditions of reality, as the
speaking timbers or the semi-divine crew of the vessel. In the Odyssey,
Æêtês and Circê (Hesiod names Mêdea also) are brother and sister,
offspring of Hêlios. The Ææan island, adjoining the circumfluous ocean,
“where the house and dancing-ground of Eôs are situated, and where
Hêlios rises,” is both the residence of Circê and of Æêtês, inasmuch
as Odysseus, in returning from the former, follows the same course as
the Argô had previously taken in returning from the latter.[603] Even
in the conception of Mimnermus, about 600 B. C., Æa still retained
its fabulous attributes in conjunction with the ocean and Hêlios,
without having been yet identified with any known portion of the
solid earth;[604] and it was justly remarked by Dêmêtrius of Skêpsis
in antiquity[605] (though Strabo vainly tries to refute him), that
neither Homer nor Mimnermus designates Kolchis either as the residence
of Æêtês, or as the terminus of the Argonautic voyage. Hesiod carried
the returning Argonauts through the river Phasis into the ocean. But
some of the poems ascribed to Eumêlus were the first which mentioned
Æêtês and Kolchis, and interwove both of them into the Corinthian
mythical genealogy.[606] These poems seem to have been composed
subsequent to the foundation of Sinopê, and to the commencement of
Grecian settlement on the Borysthenês, between the years 600 and 500 B.
C. The Greek mariners who explored and colonized the southern coast of
the Euxine, found at the extremity of their voyage the river Phasis and
its barbarous inhabitants: it was the easternmost point which Grecian
navigation (previous to the time of Alexander the Great) ever attained,
and it was within sight of the impassable barrier of Caucasus.[607]
They believed, not unnaturally, that they had here found “the house of
Eôs (the morning) and the rising place of the sun,” and that the river
Phasis, if they could follow it to its unknown beginning, would conduct
them to the circumfluous ocean. They gave to the spot the name of Æa,
and the fabulous and real title gradually became associated together
into one compound appellation,—the Kolchian Æa, or Æa of Kolchis.[608]
While Kolchis was thus entered on the map as a fit representative for
the Homeric “house of the morning,” the narrow strait of the Thracian
Bosporus attracted to itself the poetical fancy of the Symplêgades,
or colliding rocks, through which the heaven-protected Argô had been
the first to pass. The powerful Greek cities of Kyzikus, Hêrakleia and
Sinopê, each fertile in local legends, still farther contributed to
give this direction to the voyage; so that in the time of Hekatæus it
had become the established belief that the Argô had started from Iôlkos
and gone to Kolchis.

  [603] Hesiod, Theogon. 956-992; Homer, Odyss. xii. 3-69.—

    Νῆσον ἐς Αἰαίην, ὅθι τ᾽ Ἠοῦς ἠριγενείης
    Οἴκια καὶ χόροι εἰσὶ, καὶ ἀντολαὶ ἠελίοιο.

  [604] Mimnerm. Fragm. 10-11, Schneidewin; Athenæ. vii. p. 277.—

    Οὐδέ κοτ᾽ ἂν μέγα κῶας ἀνήγαγεν αὐτὸς Ἰήσων
      Ἐξ Αἴης τελέσας ἀλγινόεσσαν ὁδὸν,
    Ὑβρίστῃ Πελίῃ τελέων χαλεπῆρες ἄεθλον,
      Οὐδ᾽ ἂν ἐπ᾽ Ὠκεανοῦ καλὸν ἵκοντο ῥόον.
       ·   ·   ·   ·   ·
    Αἰήταο πόλιν, τόθι τ᾽ ὠκέος Ἠελίοιο
      Ἀκτῖνες χρυσέῳ κείαται ἐν θαλάμῳ,
    Ὠκεανοῦ παρὰ χείλεσ᾽, ἵν᾽ ὤχετο θεῖος Ἰήσων.

  [605] Strabo, i. p. 45-46. Δημήτριος ὁ Σκήψιος ... πρὸς Νεάνθη
  τὸν Κυζικηνὸν ~φιλοτιμοτέρως~ ἀντιλέγων, εἰπόντα, ὅτι οἱ
  Ἀργοναῦται πλέοντες εἰς Φᾶσιν τὸν ὑφ᾽ Ὁμήρου καὶ τῶν ἄλλων
  ὁμολογούμενον πλοῦν, ἱδρύσαντο τὰ τῆς Ἰδαίας μητρὸς ἱερὰ ἐπὶ
  Κύζικον ... ~ἀρχήν φησὶ μηδ᾽ εἰδέναι τὴν εἰς Φᾶσιν ἀποδημίαν τοῦ
  Ἰάσονος Ὅμηρον~. Again, p. 46, παραλαβὼν μάρτυρα Μίμνερμον, ὃς ἐν
  τῷ Ὠκεανῷ ποιήσας οἴκησιν Αἰήτου, etc.

  The adverb φιλοτιμοτέρως reveals to us the municipal rivalry
  and contention between the small town Skêpsis and its powerful
  neighbor Kyzikus, respecting points of comparative archæology.

  [606] Eumêlus, Fragm. Εὐρωπία 7, Κορινθιακὰ 2-5. pp. 63-68,
  Düntzer.

  [607] Arrian, Periplus Pont. Euxin. p. 12; ap. Geogr. Minor. vol.
  i. He saw the Caucasus from Dioskurias.

  [608] Herodot i. 2; vii. 193-197. Eurip. Mêd. 2. Valer. Flacc. v.
  51.

Æêtês thus received his home from the legendary faith and fancy
of the eastern Greek navigators: his sister Circê, originally his
fellow-resident, was localized by the western. The Hesiodic and other
poems, giving expression to the imaginative impulses of the inhabitants
of Cumæ and other early Grecian settlers in Italy and Sicily,[609] had
referred the wanderings of Odysseus to the western or Tyrrhenian sea,
and had planted the Cyclôpes, the Læstrygones, the floating island of
Æolus, the Lotophagi, the Phæacians, etc., about the coast of Sicily,
Italy, Libya, and Korkyra. In this way the Ææan island,—the residence
of Circê, and the extreme point of the wanderings of Odysseus, from
whence he passes only to the ocean and into Hadês—came to be placed in
the far west, while the Æa of Æêtês was in the far east,—not unlike our
East and West Indies. The Homeric brother and sister were separated and
sent to opposite extremities of the Grecian terrestrial horizon.[610]

  [609] Strabo, i. p. 23. Völcker (Ueber Homerische Geographie,
  v. 66) is instructive upon this point, as upon the geography
  of the Greek poets generally. He recognizes the purely
  mythical character of Æa in Homer and Hesiod, but he tries to
  prove—unsuccessfully, in my judgment—that Homer places Æêtês in
  the east, while Circê is in the west, and that Homer refers the
  Argonautic voyage to the Euxine Sea.

  [610] Strabo (or Polybius, whom he has just been citing) contends
  that Homer knew the existence of Æêtês in Kolchis, and of Circê
  at Circeium, as historical persons, as well as the voyage of
  Jasôn to Æa as an historical fact. Upon this he (Homer) built
  a superstructure of fiction (προσμύθευμα): he invented the
  brotherhood between them, and he placed both the one and the
  other in the exterior ocean (συγγενείας τε ἔπλασε τῶν οὕτω
  διῳκισμένων, καὶ ἐξωκεανισμὸν ἀμφοῖν, i. p. 20); perhaps also
  Jasôn might have wandered as far as Italy, as evidences (σημεῖά
  τινα) are shown that he did (_ib._).

  But the idea that Homer conceived Æêtês in the extreme east and
  Circê in the extreme west, is not reconcilable with the Odyssey.
  The supposition of Strabo is alike violent and unsatisfactory.

  Circê was worshipped as a goddess at Circeii (Cicero, Nat. Deor.
  iii. 19). Hesiod, in the Theogony, represents the two sons of
  Circê by Odysseus as reigning over all the warlike Tyrrhenians
  (Theog. 1012), an undefined western sovereignty. The great
  Mamilian gens at Tusculum traced their descent to Odysseus and
  Circê (Dionys. Hal. iv. 45).

The track from Iôlkos to Kolchis, however, though plausible as far as
it went, did not realize all the conditions of the genuine fabulous
voyage: it did not explain the evidences of the visit of these
maritime heroes which were to be found in Libya, in Krêtê in Anaphê,
in Korkyra, in the Adriatic Gulf, in Italy and in Æthalia. It became
necessary to devise another route for them in their return, and the
Hesiodic narrative was (as I have before observed), that they came back
by the circumfluous ocean; first going up the river Phasis into the
circumfluous ocean; following that deep and gentle stream until they
entered the Nile, and came down its course to the coast of Libya. This
seems also to have been the belief of Hekatæus.[611] But presently
several Greeks (and Herodotus among them) began to discard the idea of
a circumfluous ocean-stream, which had pervaded their old geographical
and astronomical fables, and which explained the supposed easy
communication between one extremity of the earth and another. Another
idea was then started for the returning voyage of the Argonauts.
It was supposed that the river Ister, or Danube, flowing from the
Rhipæan mountains in the north-west of Europe, divided itself into two
branches, one of which fell into the Euxine Sea, and the other into the
Adriatic.

  [611] See above, p. 239. There is an opinion cited from Hekatæus
  in Schol. Apoll. Rhod. iv. 284. contrary to this, which is given
  by the same scholiast on iv. 259. But, in spite of the remarks
  of Klausen (ad. Fragment. Hekatæi, 187. p. 98), I think that the
  Schol. ad. iv. 284 has made a mistake in citing Hekatæus; the
  more so as the scholiast, as printed from the Codex Parisinus,
  cites the same opinion without mentioning Hekatæus. According
  to the old Homeric idea, the ocean stream flowed all round the
  earth, and was the source of all the principal rivers which
  flowed into the great internal sea, or Mediterranean (see
  Hekatæus, Fr. 349; Klausen, ap. Arrian. ii. 16, where he speaks
  of the Mediterranean as the μεγάλη θάλασσα). Retaining this old
  idea of the ocean-stream, Hekatæus would naturally believe that
  the Phasis joined it: nor can I agree with Klausen (ad Fr. 187)
  that this implies a degree of ignorance too gross to impute to
  him.

The Argonauts, fleeing from the pursuit of Æêtês, had been obliged to
abandon their regular course homeward, and had gone from the Euxine
Sea up the Ister; then passing down the other branch of that river,
they had entered into the Adriatic, the Kolchian pursuers following
them. Such is the story given by Apollônius Rhodius from Timagêtus,
and accepted even by so able a geographer as Eratosthenês—who preceded
him by one generation, and who, though sceptical in regard to the
localities visited by Odysseus, seems to have been a firm believer in
the reality of the Argonautic voyage.[612] Other historians again,
among whom was Timæus, though they considered the ocean as an
outer sea, and no longer admitted the existence of the old Homeric
ocean-stream, yet imagined a story for the return-voyage of the
Argonauts somewhat resembling the old tale of Hesiod and Hekatæus.
They alleged that the Argô, after entering into the Palus Mæotis, had
followed the upward course of the river Tanais; that she had then been
carried overland and launched in a river which had its mouth in the
ocean or great outer sea. When in the ocean, she had coasted along
the north and west of Europe until she reached Gadês and the Strait
of Gibraltar, where she entered into the Mediterranean, and there
visited the many places specified in the fable. Of this long voyage,
in the outer sea to the north and west of Europe, many traces were
affirmed to exist along the coast of the ocean.[613] There was again
a third version, according to which the Argonauts came back as they
went, through the Thracian Bosporus and the Hellespont. In this way
geographical plausibility was indeed maintained, but a large portion of
the fabulous matter was thrown overboard.[614]

  [612] Apollôn. Rhod. iv. 287; Schol. ad iv. 284; Pindar, Pyth.
  iv. 447, with Schol.; Strabo, i. p. 46-57; Aristot. Mirabil.
  Auscult. c. 105. Altars were shown in the Adriatic, which had
  been erected both by Jasôn and by Mêdea (_ib._).

  Aristotle believed in the forked course of the Ister, with one
  embouchure in the Euxine and another in the Adriatic: he notices
  certain fishes called τρίχιαι, who entered the river (like the
  Argonauts) from the Euxine, went up it as far as the point of
  bifurcation and descended into the Adriatic (Histor. Animal.
  viii. 15). Compare Ukert, Geographie der Griech. und Römer, vol.
  iii. p. 145-147, about the supposed course of the Ister.

  [613] Diodôr. iv. 56; Timæus, Fragm. 53. Göller. Skymnus the
  geographer also adopted this opinion (Schol. Apoll. Rhod.
  284-287). The pseudo-Orpheus in the poem called Argonautica seems
  to give a jumble of all the different stories.

  [614] Diodôr. iv. 49. This was the tale both of Sophoklês and of
  Kallimachus (Schol. Apoll. Rhod. iv. 284).

  See the Dissertation of Ukert, Beylage iv. vol. i. part 2, p. 320
  of his Geographie der Griechen und Römer, which treats of the
  Argonautic voyage at some length; also J. H. Voss, Alte Weltkunde
  über die Gestalt der Erde, published in the second volume of the
  Kritische Blätter, pp. 162, 314-326; and Forbiger, Handbuch der
  Alten Geographie-Einleitung, p. 8.

Such were the various attempts made to reconcile the Argonautic
legend with enlarged geographical knowledge and improved historical
criticism. The problem remained unsolved, but the faith in the legend
did not the less continue. It was a faith originally generated at
a time when the unassisted narrative of the inspired poet sufficed
for the conviction of his hearers; it consecrated one among the
capital exploits of that heroic and superhuman race, whom the Greek
was accustomed at once to look back upon as his ancestors and to
worship conjointly with his gods: it lay too deep in his mind either
to require historical evidence for its support, or to be overthrown
by geographical difficulties as they were then appreciated. Supposed
traces of the past event, either preserved in the names of places, or
embodied in standing religious customs with their explanatory comments,
served as sufficient authentication in the eyes of the curious
inquirer. And even men trained in a more severe school of criticism
contented themselves with eliminating the palpable contradictions
and softening down the supernatural and romantic events, so as to
produce an Argonautic expedition of their own invention as the true
and accredited history. Strabo, though he can neither overlook nor
explain the geographical impossibilities of the narrative, supposes
himself to have discovered the basis of actual fact, which the original
poets had embellished or exaggerated. The golden fleece was typical
of the great wealth of Kolchis, arising from gold-dust washed down by
the rivers; and the voyage of Jasôn was in reality an expedition at
the head of a considerable army, with which he plundered this wealthy
country and made extensive conquests in the interior.[615] Strabo has
nowhere laid down what he supposes to have been the exact measure and
direction of Jasôn’s march, but he must have regarded it as very long,
since he classes Jasôn with Dionysus and Hêraklês, and emphatically
characterizes all the three as having traversed wider spaces of ground
than any moderns could equal.[616] Such was the compromise which a
mind like that of Strabo made with the ancient legends. He shaped or
cut them down to the level of his own credence, and in this waste of
historical criticism, without any positive evidence, he took to himself
the credit of greater penetration than the literal believers, while he
escaped the necessity of breaking formally with the bygone heroic world.

  [615] Strabo, i. p. 45. He speaks here of the voyage of Phryxus,
  as well as that of Jasôn, as having been a military undertaking
  (στρατεία): so again, iii. p. 149, he speaks of the military
  expedition of Odysseus—ἡ τοῦ Ὀδυσσέως στρατία, and ἡ Ἡρακλέους
  στρατία (_ib._). Again xi. p. 498. Οἱ μῦθοι, αἰνιττόμενοι τὴν
  Ἰάσονος στρατείαν προελθόντος μέχρι καὶ Μηδίας· ἔτι δὲ πρότερον
  τὴν Φρίξου. Compare also Justin, xlii. 2-3; Tacit. Annal. vi. 34.

  Strabo cannot speak of the old fables with literal fidelity: he
  unconsciously transforms them into quasi-historical incidents of
  his own imagination. Diodôrus gives a narrative of the same kind,
  with decent substitutes for the fabulous elements (iv. 40-47-56).

  [616] Strabo, i. p. 48. The far-extending expeditions undertaken
  in the eastern regions by Dionysus and Hêraklês were constantly
  present to the mind of Alexander the Great as subjects of
  comparison with himself: he imposed upon his followers perilous
  and trying marches, from anxiety to equal or surpass the alleged
  exploits of Semiramis, Cyrus, Perseus, and Hêraklês. (Arrian, v.
  2, 3; vi. 24, 3; vii. 10, 12. Strabo, iii. p. 171; xv. p. 686;
  xvii. p. 81).




CHAPTER XIV.

LEGENDS OF THEBES.


The Bœôtians generally, throughout the historical age, though well
endowed with bodily strength and courage,[617] are represented as
proverbially deficient in intelligence, taste and fancy. But the
legendary population of Thêbes, the Kadmeians, are rich in mythical
antiquities, divine as well as heroic. Both Dionysus and Hêraklês
recognize Thêbes as their natal city. Moreover, the two sieges of
Thêbes by Adrastus, even taken apart from Kadmus, Antiopê, Amphiôn and
Zethus, etc., are the most prominent and most characteristic exploits,
next to the siege of Troy, of that preëxisting race of heroes who lived
in the imagination of the historical Hellênes.

  [617] The eponym Bœôtus is son of Poseidôn and Arnê (Euphorion
  ap. Eustath. ad Iliad. ii. 507). It was from Arnê in Thessaly
  that the Bœôtians were said to have come, when they invaded
  and occupied Bœôtia. Euripidês made him son of Poseidôn and
  Melanippê. Another legend recited Bœôtus and Hellên as sons of
  Poseidôn and Antiopê (Hygin. f. 157-186).

  The Tanagræan poetess Korinna (the rival of Pindar, whose
  compositions in the Bœôtian dialect are unfortunately lost)
  appears to have dwelt upon this native Bœôtian genealogy: she
  derived the Ogygian gates of Thêbes from Ogygus, son of Bœôtus
  (Schol. Apollôn. Rhod. iii. 1178), also the Fragments of Korinna
  in Schneidewin’s edition, fr. 2. p. 432.

It is not Kadmus, but the brothers Amphiôn and Zethus, who are given
to us in the Odyssey as the first founders of Thêbes and the first
builders of its celebrated walls. They are the sons of Zeus by Antiopê,
daughter of Asôpus. The scholiasts who desire to reconcile this tale
with the more current account of the foundation of Thêbes by Kadmus,
tell us that after the death of Amphiôn and Zethus, Eurymachus, the
warlike king of the Phlegyæ, invaded and ruined the newly-settled
town, so that Kadmus on arriving was obliged to re-found it.[618] But
Apollodôrus, and seemingly the older logographers before him, placed
Kadmus at the top, and inserted the two brothers at a lower point
in the series. According to them, Bêlus and Agenôr were the sons of
Epaphus, son of the Argeian Iô, by Libya. Agenôr went to Phœnicia and
there became king: he had for his offspring Kadmus, Phœnix, Kilix,
and a daughter Eurôpa; though in the Iliad Eurôpa is called daughter
of Phœnix.[619] Zeus fell in love with Eurôpa, and assuming the shape
of a bull, carried her across the sea upon his back from Egypt to
Krête, where she bore to him Minôs, Rhadamanthus and Sarpêdôn. Two out
of the three sons sent out by Agenôr in search of their lost sister,
wearied out by a long-protracted as well as fruitless voyage, abandoned
the idea of returning home: Kilix settled in Kilikia, and Kadmus
in Thrace.[620] Thasus, the brother or nephew of Kadmus, who had
accompanied them in the voyage, settled and gave name to the island of
Phasus.

  [618] Homer, Odyss. xi. 262, and Eustath. _ad loc._ Compare
  Schol. ad Iliad. xiii. 301.

  [619] Iliad, xiv. 321. Iô is κερόεσσα προμάτωρ of the Thêbans.
  Eurip. Phœniss. 247-676.

  [620] Apollodôr. ii. 1, 3; iii. 1, 8. In the Hesiodic poems
  (ap. Schol. Apoll. Rhod. ii. 178), Phœnix was recognized as son
  of Agenôr. Pherekydês also described both Phœnix and Kadmus as
  sons of Agenôr (Pherekyd. Fragm. 40, Didot). Compare Servius
  ad. Virgil. Æneid. 1. 338. Pherekydês expressly mentioned
  Kilix (Apollod. _ib._). Besides the Εὐρώπεια of Stesichorus
  (see Stesichor. Fragm. xv. p. 73, ed. Kleine), there were
  several other ancient poems on the adventures of Europa; one in
  particular by Eumêlus (Schol. ad Iliad. vi. 138), which however
  can hardly be the same as the τὰ ἔπη τὰ εἰς Εὐρώπην alluded to by
  Pausanias (ix. 5, 4). See Wüllner de Cyclo Epico, p. 57 (Münster
  1825).

Both Herodotus and Euripidês represent Kadmus as an emigrant from
Phœnicia, conducting a body of followers in quest of Eurôpa. The
account of Apollodôrus describes him as having come originally from
Libya or Egypt to Phœnicia: we may presume that this was also the
statement of the earlier logographers Pherekydês and Hellanikus. Conôn,
who historicizes and politicizes the whole legend, seems to have found
two different accounts; one connecting Kadmus with Egypt, another
bringing him from Phœnicia. He tries to melt down the two into one, by
representing that the Phœnicians, who sent out Kadmus, had acquired
great power in Egypt—that the seat of their kingdom was the Egyptian
Thêbes—that Kadmus was despatched, under pretence indeed of finding
his lost sister, but really on a project of conquest—and that the name
Thêbes, which he gave to his new establishment in Bœôtia, was borrowed
from Thêbes in Egypt, his ancestorial seat.[621]

  [621] Conôn, Narrat. 37. Perhaps the most remarkable thing of all
  is the tone of unbounded self-confidence with which Conôn winds
  up this tissue of uncertified suppositions—περὶ μὲν Κάδμου καὶ
  Θηβῶν οἰκίσεως οὗτος ὁ ἀληθὴς λόγος· τὸ δὲ ἄλλο μῦθος καὶ γοητεία
  ἀκοῆς.

Kadmus went from Thrace to Delphi to procure information respecting
his sister Eurôpa, but the god directed him to take no further trouble
about her; he was to follow the guidance of a cow, and to found a
city on the spot where the animal should lie down. The condition was
realized on the site of Thêbes. The neighboring fountain Areia was
guarded by a fierce dragon, the offspring of Arês, who destroyed all
the persons sent to fetch water. Kadmus killed the dragon, and at
the suggestion of Athênê sowed his teeth in the earth:[622] there
sprang up at once the armed men called the Sparti, among whom he flung
stones, and they immediately began to assault each other until all
were slain except five. Arês, indignant at this slaughter, was about to
kill Kadmus; but Zeus appeased him, condemning Kadmus to an expiatory
servitude of eight years, after which he married Harmonia, the
daughter of Arês and Aphroditê—presenting to her the splendid necklace
fabricated by the hand of Hêphæstos, which had been given by Zeus to
Eurôpa.[623] All the gods came to the Kadmeia, the citadel of Thêbes,
to present congratulations and gifts at these nuptials, which seem to
have been hardly less celebrated in the mythical world than those of
Pêleus and Thetis. The issue of the marriage was one son, Polydôrus,
and four daughters, Autonoê, Inô, Semelê and Agavê.[624]

  [622] Stesichor. (Fragm. 16; Kleine) ap. Schol. Eurip. Phœniss.
  680. The place where the heifer had lain down was still shown in
  the time of Pausanias (ix. 12, 1).

  Lysimachus, a lost author who wrote Thebaïca, mentioned Eurôpa
  as having come with Kadmus to Thêbes, and told the story in many
  other respects very differently (Schol. Apoll. Rhod. iii. 1179).

  [623] Apollodôr. iii. 4, 1-3. Pherekydês gave this account of
  the necklace, which seems to imply that Kadmus must have found
  his sister Eurôpa. The narrative here given is from Hellanikus;
  that of Pherekydês differed from it in some respects: compare
  Hellanik. Fragm. 8 and 9, and Pherekyd. Frag. 44. The resemblance
  of this story with that of Jasôn and Æêtês (see above, chap.
  xiii. p. 237) will strike everyone. It is curious to observe how
  the old logographer Pherekydês explained this analogy in his
  narrative; he said that Athênê had given half the dragon’s teeth
  to Kadmus and half to Æêtês (see Schol. Pindar. Isthm. vi. 13).

  [624] Hesiod, Theogon. 976. Leukothea, the sea-goddess, daughter
  of Kadmus, is mentioned in the Odyssey, v. 334; Diodôr. iv. 2.

From the five who alone survived of the warriors sprung from the
dragon’s teeth, arose five great families or gentes in Thêbes; the
oldest and noblest of its inhabitants, coeval with the foundation of
the town. They were called Sparti, and their name seems to have given
rise, not only to the fable of the sowing of the teeth, but also to
other etymological narratives.[625]

  [625] Eurip. Phœniss. 680, with the Scholia; Pherekydês, Fragm.
  44; Androtiôn, ap. Schol. Pindar. Isthm. vi. 13. Dionysius (?)
  called the Sparti an ἔθνος Βοιωτίας (Schol. Phœniss. 1. c).

  Even in the days of Plutarch, there were persons living who
  traced their descent to the Sparti of Thêbes (Plutarch, Ser. Num.
  Vindict. p. 563).

All the four daughters of Kadmus are illustrious in fabulous history.
Inô, wife of Athamas, the son of Æolus, has already been included among
the legends of the Æolids. Semelê became the mistress of Zeus, and
inspired Hêrê with jealousy. Misguided by the malicious suggestions of
that goddess, she solicited Zeus to visit her with all the solemnity
and terrors which surrounded him when he approached Hêrê herself.
The god unwillingly consented, and came in his chariot in the midst
of thunder and lightning, under which awful accompaniments the mortal
frame of Semelê perished. Zeus, taking from her the child of which she
was pregnant, sewed it into his own thigh: after the proper interval
the child was brought out and born, and became the great god Dionysus
or Bacchus. Hermês took him to Inô and Athamas to receive their
protection. Afterwards, however, Zeus having transformed him into a kid
to conceal him from the persecution of Hêrê, the nymphs of the mountain
Nysa became his nurses.[626]

  [626] Apollodôr. iii. 4, 2-9; Diodôr. iv. 2.

Autonoê, the third daughter of Kadmus, married the pastoral hero or god
Aristæus, and was mother of Aktæôn, a devoted hunter and a favorite
companion of the goddess Artemis. She however became displeased with
him—either because he looked into a fountain while she was bathing
and saw her naked—or according to the legend set forth by the poet
Stesichorus, because he loved and courted Semelê—or according to
Euripidês, because he presumptuously vaunted himself as her superior in
the chase. She transformed him into a stag, so that his own dogs set
upon and devoured him. The rock upon which Aktæôn used to sleep when
fatigued with the chase, and the spring whose transparent waters had
too clearly revealed the form of the goddess, were shown to Pausanias
near Platæa, on the road to Megara.[627]

  [627] See Apollodôr. iii. 4, 3; Stesichor. Fragm. xvii. Kleine;
  Pausan. ix. 2, 3; Eurip. Bacch. 337; Diodôr. iv. 81. The old
  logographer Akusilaus copied Stesichorus.

  Upon this well-known story it is unnecessary to multiply
  references. I shall however briefly notice the remarks made upon
  it by Diodôrus and by Pausanias, as an illustration of the manner
  in which the literary Greeks of a later day dealt with their old
  national legends.

  Both of them appear implicitly to believe the fact, that Aktæôn
  was devoured by his own dogs, but they differ materially in the
  explanation of it.

  Diodôrus accepts and vindicates the miraculous interposition
  of the displeased goddess to punish Aktæôn, who, according
  to one story, had boasted of his superiority in the chase to
  Artemis,—according to another story, had presumed to solicit the
  goddess in marriage, emboldened by the great numbers of the feet
  of animals slain in the chase which he had hung up as offerings
  in her temple. “It is not improbable (observes Diodôrus) that
  the goddess was angry on both these accounts. For whether Aktæôn
  abused these hunting presents so far as to make them the means
  of gratifying his own desires towards one unapproachable in
  wedlock, or whether he presumed to call himself an abler hunter
  than her with whom the gods themselves will not compete in this
  department,—in either case the wrath of the goddess against him
  was just and legitimate (ὁμολογουμένην καὶ δικαίαν ὀργὴν ἔσχε
  πρὸς αὐτὸν ἡ θεός). With perfect propriety therefore (Καθόλου δὲ
  πιθανῶς) was he transformed into an animal such as those he had
  hunted, and torn to pieces by the very dogs who had killed them.”
  (Didot. iv. 80.)

  Pausanias, a man of exemplary piety, and generally less inclined
  to scepticism than Diodôrus, thinks the occasion unsuitable for
  a miracle or special interference. Having alluded to the two
  causes assigned for the displeasure of Artemis (they are the two
  first-mentioned in my text, and distinct from the two noticed
  by Diodôrus), he proceeds to say, “But I believe that the dogs
  of Aktæôn went mad, without the interference of the goddess: in
  this state of madness they would have torn in pieces without
  distinction any one whom they met (Paus. ix. 2, 3. ἐγὼ δὲ ἄνευ
  θεοῦ πείθομαι νόσον λύσσαν ἐπιβαλεῖν τοῦ Ἀκταίωνος τοὺς κύνας).”
  He retains the truth of the final catastrophe, but rationalizes
  it, excluding the special intervention of Artemis.

Agavê, the remaining daughter of Kadmus, married Echiôn, one of the
Sparti. The issue of these nuptials was Pentheus, who, when Kadmus
became old succeeded him as king of Thêbes. In his reign Dionysus
appeared as a god, the author or discoverer of the vine with all its
blessings. He had wandered over Asia, India and Thrace, at the head of
an excited troop of female enthusiasts—communicating and inculcating
everywhere the Bacchic ceremonies, and rousing in the minds of women
that impassioned religious emotion which led them to ramble in solitary
mountains at particular seasons, there to give vent to violent
fanatical excitement, apart from the men, clothed in fawn-skins and
armed with the thyrsus. The obtrusion of a male spectator upon these
solemnities was esteemed sacrilegious. Though the rites had been
rapidly disseminated and fervently welcomed in many parts of Thrace,
yet there were some places in which they had been obstinately resisted
and their votaries treated with rudeness; especially by Lykurgus, king
of the Edonian Thracians, upon whom a sharp and exemplary punishment
was inflicted by Dionysus.

Thêbes was the first city of Greece to which Dionysus came, at the
head of his Asiatic troop of females, to obtain divine honors and
to establish his peculiar rites in his native city. The venerable
Kadmus, together with his daughters and the prophet Teiresias, at once
acknowledged the divinity of the new god, and began to offer their
worship and praise to him along with the solemnities which he enjoined.
But Pentheus vehemently opposed the new ceremonies, reproving and
maltreating the god who introduced them: nor was his unbelief at all
softened by the miracles which Dionysus wrought for his own protection
and for that of his followers. His mother Agavê, with her sisters, and
a large body of other women from Thêbes, had gone out from Thêbes to
Mount Kithærôn to celebrate their solemnities under the influence of
the Bacchic frenzy. Thither Pentheus followed to watch them, and there
the punishment due to his impiety overtook him. The avenging touch of
the god having robbed him of his senses, he climbed a tall pine for the
purpose of overlooking the feminine multitude, who detected him in this
position, pulled down the tree, and tore him in pieces. Agavê, mad and
bereft of consciousness, made herself the foremost in this assault, and
carried back in triumph to Thêbes the head of her slaughtered son. The
aged Kadmus, with his wife Harmonia, retired among the Illyrians, and
at the end of their lives were changed into serpents, Zeus permitting
them to be transferred to the Elysian fields.[628]

  [628] Apollod. iii. 5, 3-4; Theocrit. Idyll. xxvi. Eurip. Bacch.
  _passim_. Such is the tragical plot of this memorable drama. It
  is a striking proof of the deep-seated reverence of the people
  of Athens for the sanctity of the Bacchic ceremonies, that they
  could have borne the spectacle of Agavê on the stage with her
  dead son’s head, and the expressions of triumphant sympathy in
  her action on the part of the Chorus (1168), Μάκαιρ᾽ Ἀγαύη! This
  drama, written near the close of the life of Euripidês, and
  exhibited by his son after his death (Schol. Aristoph. Ran. 67,),
  contains passages strongly inculcating the necessity of implicit
  deference to ancestorial authority in matters of religion, and
  favorably contrasting the uninquiring faith of the vulgar with
  the dissenting and inquisitive tendencies of superior minds: see
  v. 196; compare vv. 389 and 422.—

    Οὐδὲν σοφιζόμεσθα τοῖσι δαίμοσιν.
    Πατρίους παραδοχὰς, ἅς θ᾽ ὁμήλικας χρόνῳ
    Κεκτήμεθ᾽, οὐδεὶς αὐτὰ καταβαλεῖ λόγος,
    Οὐδ᾽ ἢν δι᾽ ἄκρων τὸ σοφὸν εὕρηται φρένων.

  Such reproofs “insanientis sapientiæ” certainly do not fall in
  with the plot of the drama itself, in which Pentheus appears as
  a Conservative, resisting the introduction of the new religious
  rites. Taken in conjunction with the emphatic and submissive
  piety which reigns through the drama, they countenance the
  supposition of Tyrwhitt, that Euripidês was anxious to repel the
  imputations, so often made against him, of commerce with the
  philosophers and participation in sundry heretical opinions.

  Pacuvius in his Pentheus seems to have closely copied Euripidês;
  see Servius ad Virg. Æneid. iv. 469.

  The old Thespis had composed a tragedy on the subject of
  Pentheus: Suidas, Θέσπις; also Æschylus; compare his Eumenidês,
  25.

  According to Apollodôrus (iii. 5, 5), Labdakus also perished in a
  similar way to Pentheus, and from the like impiety,—ἐκείνῳ φρονῶν
  παραπλήσια.

Polydôrus and Labdakus successively became kings of Thêbes: the latter
at his death left an infant son, Laius, who was deprived of his throne
by Lykus. And here we approach the legend of Antiopê, Zêthus and
Amphiôn, whom the fabulists insert at this point of the Thêban series.
Antiopê is here the daughter of Nykteus, the brother of Lykus. She is
deflowered by Zeus, and then, while pregnant, flies to Epôpeus king of
Sikyôn: Nykteus dying entreats his brother to avenge the injury, and
Lykus accordingly invades Sikyôn, defeats and kills Epôpeus, and brings
back Antiopê prisoner to Thêbes. In her way thither, in a cave near
Eleutheræ, which was shown to Pausanias,[629] she is delivered of the
twin sons of Zeus—Amphiôn and Zêthus—who, exposed to perish, are taken
up and nourished by a shepherd, and pass their youth amidst herdsmen,
ignorant of their lofty descent.

  [629] Pausan. i. 38, 9.

Antiopê is conveyed to Thêbes, where, after undergoing a long
persecution from Lykus and his cruel wife Dirkê, she at length escapes,
and takes refuge in the pastoral dwelling of her sons, now grown to
manhood. Dirkê pursues and requires her to be delivered up; but the
sons recognize and protect their mother, taking an ample revenge upon
her persecutors. Lykus is slain, and Dirkê is dragged to death, tied to
the horns of a bull.[630] Amphiôn and Zêthus, having banished Laius,
become kings of Thêbes. The former, taught by Hermês, and possessing
exquisite skill on the lyre, employs it in fortifying the city, the
stones of the walls arranging themselves spontaneously in obedience to
the rhythm of his song.[631]

  [630] For the adventures of Antiopê and her sons, see Apollodôr.
  iii. 5; Pausan. ii. 6, 2; ix. 5, 2.

  The narrative given respecting Epôpeus in the ancient Cyprian
  verses seems to have been very different from this, as far as we
  can judge from the brief notice in Proclus’s Argument,—ὡς Ἐπωπεὺς
  φθείρας τὴν Λυκούργου (Λύκου) γυναῖκα ἐξεπορθήθη: it approaches
  more nearly to the story given in the seventh fable of Hyginus,
  and followed by Propertius (iii. 15); the eighth fable of Hyginus
  contains the tale of Antiopê as given by Euripidês and Ennius.
  The story of Pausanias differs from both.

  The Scholiast ad Apollôn. Rhod. i. 735. says that there were
  two persons named Antiopê; one, daughter of Asôpus, the other,
  daughter of Nykteus. Pausanias is content with supposing one
  only, really the daughter of Nykteus, but there was a φήμη
  that she was daughter of Asôpus (ii. 6, 2). Asius made Antiopê
  daughter of Asôpus, and mother (both by Zeus and by Epôpeus: such
  a junction of divine and human paternity is of common occurrence
  in the Greek legends) of Zêthus and Amphiôn (ap. Paus. 1. c).

  The contradictory versions of the story are brought together,
  though not very perfectly, in Sterk’s Essay De Labdacidarum
  Historiâ, p. 38-43 (Leyden, 1829).

  [631] This story about the lyre of Amphiôn is not noticed in
  Homer, but it was narrated in the ancient ἔπη ἐς Εὐρώπην which
  Pausanias had read: the wild beasts as well as the stones were
  obedient to his strains (Paus. ix. 5, 4). Pherekydês also
  recounted it (Pherekyd. Fragm. 102, Didot). The tablet of
  inscription (Ἀναγραφὴ) at Sikyôn recognized Amphiôn as the first
  composer of poetry and harp-music (Plutarch, de Musicâ, c. 3. p.
  1132).

Zêthus marries Aêdôn, who, in the dark and under a fatal mistake, kills
her son Itylus: she is transformed into a nightingale, while Zêthus
dies of grief.[632] Amphiôn becomes the husband of Niobê, daughter
of Tantalus, and the father of a numerous offspring, the complete
extinction of which by the hands of Apollo and Artemis has already been
recounted in these pages.

  [632] The tale of the wife and son of Zêthus is as old as the
  Odyssey (xix. 525). Pausanias adds the statement that Zêthus died
  of grief (ix. 5, 5; Pherekydês, Fragm. 102, Did.). Pausanias,
  however, as well as Apollodôrus, tells us that Zêthus married
  Thêbê, from whom the name Thêbes was given to the city. To
  reconcile the conflicting pretensions of Zêthus and Amphiôn with
  those of Kadmus, as founders of Thêbes, Pausanias supposes that
  the latter was the original settler of the hill of the Kadmeia,
  while the two former extended the settlement to the lower city
  (ix. 5, 1-3).

Here ends the legend of the beautiful Antiopê and her twin sons—the
rude and unpolished, but energetic, Zêthus—and the refined and amiable,
but dreamy, Amphiôn. For so Euripidês, in the drama of Antiopê
unfortunately lost, presented the two brothers, in affectionate union
as well as in striking contrast.[633] It is evident that the whole
story stood originally quite apart from the Kadmeian family, and so
the rudiments of it yet stand in the Odyssey; but the logographers, by
their ordinary connecting artifices, have opened a vacant place for it
in the descending series of Thêban mythes. And they have here proceeded
in a manner not usual with them. For whereas they are generally fond
of multiplying entities, and supposing different historical personages
of the same name, in order to introduce an apparent smoothness in
the chronology—they have here blended into one person Amphiôn the
son of Antiopê and Amphiôn the father of Chlôris, who seem clearly
distinguished from each other in the Odyssey. They have further
assigned to the same person all the circumstances of the legend of
Niobê, which seems to have been originally framed quite apart from the
sons of Antiopê.

  [633] See Valckenaer. Diatribe in Eurip. Reliq. cap. 7, p. 58;
  Welcker, Griechisch. Tragöd. ii. p. 811. There is a striking
  resemblance between the Antiopê of Euripidês and the Tyrô of
  Sophoklês in many points.

  Plato in his Gorgias has preserved a few fragments, and a
  tolerably clear general idea of the characters of Zêthus and
  Amphiôn (Gorg. 90-92); see also Horat. Epist. i. 18, 42.

  Both Livius and Pacuvius had tragedies on the scheme of this of
  Euripidês, the former seemingly a translation.

Amphiôn and Zêthus being removed, Laius became king of Thêbes. With
him commences the ever-celebrated series of adventures of Œdipus and
his family. Laius forewarned by the oracle that any son whom he might
beget would kill him, caused Œdipus as soon as he was born to be
exposed on Mount Kithærôn. Here the herdsmen of Polybus king of Corinth
accidentally found him and conveyed him to their master, who brought
him up as his own child. In spite of the kindest treatment, however,
Œdipus when he grew up found himself exposed to taunts on the score of
his unknown parentage, and went to Delphi to inquire of the god the
name of his real father. He received for answer an admonition not to
go back to his country; if he did so, it was his destiny to kill his
father and become the husband of his mother. Knowing no other country
but Corinth, he accordingly determined to keep away from that city,
and quitted Delphi by the road towards Bœôtia and Phôkis. At the exact
spot where the roads leading to these two countries forked, he met
Laius in a chariot drawn by mules, when the insolence of one of the
attendants brought on an angry quarrel, in which Œdipus killed Laius,
not knowing him to be his father. The exact place where this event
happened, called the Divided Way[634], was memorable in the eyes of
all literary Greeks, and is specially adverted to by Pausanias in his
periegesis.

  [634] See the description of the locality in K. O. Müller
  (Orchomenos, c. i. p. 37).

  The tombs of Laius and his attendant were still seen there in the
  days of Pausanias (x. 5, 2).

On the death of Laius, Kreôn, the brother of Jokasta, succeeded
to the kingdom of Thêbes. At this time the country was under the
displeasure of the gods, and was vexed by a terrible monster, with
the face of a woman, the wings of a bird, and the tail of a lion,
called the Sphinx[635]—sent by the wrath of Hêrê and occupying the
neighboring mountain of Phikium. The Sphinx had learned from the
Muses a riddle, which she proposed to the Thêbans to resolve: on
every occasion of failure she took away one of the citizens and ate
him up. Still no person could solve the riddle; and so great was the
suffering occasioned, that Kreôn was obliged to offer both the crown
and the nuptials of his sister Jokasta to any one who could achieve
the salvation of the city. At this juncture Œdipus arrived and solved
the riddle: upon which the Sphinx immediately threw herself from the
acropolis and disappeared. As a recompense for this service, Œdipus was
made king of Thêbes, and married Jokasta, not aware that she was his
mother.

  [635] Apollodôr. iii. 5, 8. An author named Lykus, in his work
  entitled _Thêbaïca_, ascribed this visitation to the anger of
  Dionysus (Schol. Hesiod, Theogon. 326). The Sphinx (or _Phix_,
  from the Bœôtian Mount Phikium) is as old as the Hesiodic
  Theogony,—Φῖκ᾽ ὀλόην τέκε, Καδμείοισιν ὄλεθρον (Theog. 326).

These main tragical circumstances—that Œdipus had ignorantly killed
his father and married his mother—belong to the oldest form of the
legend as it stands in the Odyssey. The gods (it is added in that poem)
quickly made the facts known to mankind. Epikasta (so Jokasta is here
called) in an agony of sorrow hanged herself: Œdipus remained king of
the Kadmeians, but underwent many and great miseries, such as the
Erinnyes, who avenge an injured mother, inflict.[636] A passage in the
Iliad implies that he died at Thêbes, since it mentions the funeral
games which were celebrated there in honor of him. His misfortunes
were recounted by Nestôr, in the old Cyprian verses, among the stories
of aforetime.[637] A fatal curse hung both upon himself and upon his
children, Eteoklês, Polynikês, Antigonê and Ismênê. According to
that narrative which the Attic tragedians have rendered universally
current, they were his children by Jokasta, the disclosure of her true
relationship to him having been very long deferred. But the ancient
epic called Œdipodia, treading more closely in the footsteps of Homer,
represented him as having after her death married a second wife,
Euryganeia, by whom the four children were born to him: and the painter
Onatas adopted this story in preference to that of Sophoklês.[638]

  [636] Odyss. xi. 270. Odysseus, describing what he saw in the
  under-world, says,—

    Μητέρα τ᾽ Οἰδιπόδαο ἴδον, καλὴν Ἐπικάστην,
    Ἣ μέγα ἔργον ἔρεξεν αἰδρεΐῃσι νόοιο,
    Γημαμένη ᾧ υἱεῖ· ὁ δ᾽ ὃν πατέρ᾽ ἐξεναρίξας
    Γῆμεν· ἄφαρ δ᾽ ἀνάπυστα θεοὶ θέσαν ἀνθρώποισι.
    Ἀλλ᾽ ὁ μὲν ἐν Θήβῃ πολυηράτῳ ἄλγεα πάσχων,
    Καδμείων ἤνασσε, θεῶν ὀλοὰς διὰ βουλάς·
    Ἡ δ᾽ ἔβη εἰς Αἰδάο πυλάρταο κρατεροῖο
    Ἁψαμένη βρόχον αἰπὺν ἀφ᾽ ὑψήλοιο μελάθρου,
    Ὧ ἄχεϊ σχομένη· τῷ δ᾽ ἄλγεα κάλλιπ᾽ ὀπίσσω
    Πολλὰ μάλ᾽, ὅσσα τε μητρὸς Ἐριννύες ἐκτελέουσιν.

  [637] Iliad, xxiii. 680, with the scholiast who cites Hesiod.
  Proclus, Argum. ad Cypria, ap. Düntzer, Fragm. Epic. Græc. p. 10.
  Νέστωρ δὲ ἐν παρεκβάσει διηγεῖται ... καὶ τὰ περὶ Οἰδίπουν, etc.

  [638] Pausan. ix. 5, 5. Compare the narrative from Peisander in
  Schol. ad Eurip. Phœniss. 1773; where, however, the blindness
  of Œdipus seems to be unconsciously interpolated out of the
  tragedians. In the old narrative of the Cyclic Thêbaïs, Œdipus
  does not seem to be represented as blind (Leutsch, Thebaidis
  Cyclici Reliquiæ, Götting. 1830, p. 42).

  Pherekydês (ap. Schol. Eurip. Phœniss. 52) tells us that Œdipus
  had three children by Jokasta, who were all killed by Erginus
  and the Minyæ (this must refer to incidents in the old poems
  which we cannot now recover); then the four celebrated children
  by Euryganeia; lastly, that he married a third wife, Astymedusa.
  Apollodôrus follows the narrative of the tragedians, but alludes
  to the different version about Euryganeia,—εἰσὶ δ᾽ οἵ φασιν, etc.
  (iii. 5, 8).

  Hellanikus (ap. Schol. Eur. Phœniss. 59) mentioned the
  self-inflicted blindness of Œdipus; but it seems doubtful whether
  this circumstance was included in the narrative of Pherekydês.

The disputes of Eteoklês and Polynikês for the throne of their father
gave occasion not only to a series of tragical family incidents,
but also to one of the great quasi-historical events of legendary
Greece—the two sieges of Thêbes by Adrastus, king of Argos. The two
ancient epic poems called the Thêbaïs and the Epigoni (if indeed both
were not parts of one very comprehensive poem) detailed these events
at great length, and as it appears, with distinguished poetical merit;
for Pausanias pronounces the Cyclic Thêbaïs (so it was called by the
subsequent critics to distinguish it from the more modern Thêbaïs of
Antimachus) inferior only to the Iliad and Odyssey; and the ancient
elegiac poet Kallinus treated it as an Homeric composition.[639] Of
this once-valued poem we unfortunately possess nothing but a few
scanty fragments. The leading points of the legend are briefly glanced
at in the Iliad; but our knowledge of the details is chiefly derived
from the Attic tragedians, who transformed the narratives of their
predecessors at pleasure, and whose popularity constantly eclipsed and
obliterated the ancient version. Antimachus of Kolophôn, contemporary
with Euripidês, in his long epic, probably took no less liberties
with the old narrative. His Thêbaïd never became generally popular,
but it exhibited marks of study and elaboration which recommended it
to the esteem of the Alexandrine critics, and probably contributed to
discredit in their eyes the old cyclic poem.

  [639] Pausan. ix. 9. 3. Ἐποιήθη δὲ ἐς τὸν πόλεμον τοῦτον καὶ ἔπη,
  Θηβαΐς· τὰ δὲ ἔπη ταῦτα Καλλῖνος, ἀφικόμενος αὐτῶν ἐς μνήμην,
  ἔφησεν Ὅμηρον τὸν ποιήσαντα εἶναι. Καλλίνῳ δὲ πολλοί τε καὶ
  ἄξιοι λόγου κατὰ ταῦτα ἔγνωσαν· ἐγὼ δὲ τὴν ποίησιν ταύτην μετά
  γε Ἰλιάδα καὶ τὰ ἔπη τὰ ἐς Ὀδυσσέα ἐπαινῶ μάλιστα. The name in
  the text of Pausanias stands Καλαῖνος, an unknown person: most of
  the critics recognize the propriety of substituting Καλλῖνος, and
  Leutsch and Welcker have given very sufficient reasons for doing
  so.

  The Ἀμφιάρεω ἐξελασία ἐς Θέβας, alluded to in the
  pseudo-Herodotean life of Homer, seems to be the description of a
  special passage in this Thêbaïs.

The logographers, who gave a continuous history of this siege of
Thêbes, had at least three preëxisting epic poems—the Thêbaïs, the
Œdipodia, and the Alkmæônis,—from which they could borrow. The subject
was also handled in some of the Hesiodic poems, but we do not know to
what extent.[640] The Thêbaïs was composed more in honor of Argos than
of Thêbes, as the first line of it, one of the few fragments still
preserved, betokens.[641]

  [640] Hesiod, ap. Schol. Iliad. xxiii. 680, which passage does
  not seem to me so much at variance with the incidents stated in
  other poets as Leutsch imagines.

  [641] Ἄργος ἄειδε, θεὰ, πολυδίψιον, ἔνθεν ἄνακτες (see Leutsch,
  _ib._ c. 4. p. 29).


SIEGES OF THEBES.

The legend, about to recount fraternal dissension of the most
implacable kind, comprehending in its results not only the immediate
relations of the infuriated brothers, but many chosen companions of the
heroic race along with them, takes its start from the paternal curse of
Œdipus, which overhangs and determines all the gloomy sequel.

Œdipus, though king of Thêbes and father of four children by Euryganeia
(according to the Œdipodia), has become the devoted victim of the
Erinnyes, in consequence of the self-inflicted death of his mother,
which he has unconsciously caused, as well as of his unintentional
parricide. Though he had long forsworn the use of all the ornaments and
luxuries which his father had inherited from his kingly progenitors,
yet when through age he had come to be dependent upon his two sons,
Polynikês one day broke through this interdict, and set before him
the silver table and the splendid wine-cup of Kadmus, which Laius had
always been accustomed to employ. The old king had no sooner seen these
precious appendages of the regal life of his father, than his mind was
overrun by a calamitous phrenzy, and he imprecated terrible curses on
his sons, predicting that there would be bitter and endless warfare
between them. The goddess Erinnys heard and heeded him; and he repeated
the curse again on another occasion, when his sons, who had always
been accustomed to send to him the shoulder of the victims sacrificed
on the altar, caused the buttock to be served to him in place of
it.[642] He resented this as an insult, and prayed the gods that they
might perish each by the hand of the other. Throughout the tragedians
as well as in the old epic, the paternal curse, springing immediately
from the misguided Œdipus himself, but remotely from the parricide
and incest with which he has tainted his breed, is seen to domineer
over the course of events—the Erinnys who executes that curse being
the irresistible, though concealed, agent. Æschylus not only preserves
the fatal efficiency of the paternal curse, but even briefly glances
at the causes assigned for it in the Thêbaïs, without superadding any
new motives. In the judgment of Sophoklês, or of his audience, the
conception of a father cursing his sons upon such apparently trifling
grounds was odious; and that great poet introduced many aggravating
circumstances, describing the old blind father as having been
barbarously turned out of doors by his sons to wander abroad in exile
and poverty. Though by this change he rendered his poem more coherent
and self-justifying, yet he departed, from the spirit of the old
legend, according to which Œdipus has contracted by his unconscious
misdeeds an incurable taint destined to pass onward to his progeny.
His mind is alienated, and he curses them, not because he has suffered
seriously by their guilt, but because he is made the blind instrument
of an avenging Erinnys for the ruin of the house of Laius.[643]

  [642] Fragm. of the Thêbaïs, ap. Athenæ. xii. p. 465, ὅτι αὐτῷ
  παρέθηκαν ἐκπώματα ἃ ἀπηγορεύκει, λέγων οὕτως·

    Αὐτὰρ ὁ διογένης ἥρως ξανθὸς Πολυνείκης
    Πρῶτα μὲν Οἰδίποδι καλὴν παρέθηκε τράπεζαν
    Ἀργυρέην Κάδμοιο θεόφρονος· αὐταρ ἔπειτα
    Χρύσεον ἔμπλησεν καλὸν δέπας ἥδεος οἴνου·
    Αὐτὰρ ὅγ᾽ ὡς φράσθη παρακείμενα πατρὸς ἑοῖο
    Τιμήεντα γέρα, μέγα οἱ κακὸν ἔμπεσε θυμῷ.
    Αἶψα δὲ παισὶν ἑοῖσι μετ᾽ ἀμφοτέροισιν ἐπαρὰς
    Ἀργαλέας ἠρᾶτο· θεὸν δ᾽ οὐ λάνθαν᾽ Ἐριννύν·
    Ὡς οὐ οἱ πατρῷα γ᾽ ἐνὶ φιλότητι δάσαιντο,
    Εἶεν δ᾽ αμφοτέροις αἰεὶ πόλεμοί τε μάχαι τε.

  See Leutsch, Thebaid. Cycl. Reliq. p. 38.

  The other fragment from the same Thêbaïs is cited by the Schol.
  ad Soph. Œdip. Colon. 1378.—

    Ἴσχιον ὡς ἐνόησε, χαμαὶ βάλεν, εἶπέ τε μῦθον·
    Ὦ μοι ἐγὼ, παῖδές μοι ὀνειδείοντες ἔπεμψαν.
    Εὐκτο Διῒ βασιλῆϊ καὶ ἄλλοις ἀθανάτοισι,
    Χερσὶν ὑπ᾽ ἀλλήλων καταβήμεναι Ἄϊδος εἴσω.

  Τὰ δὲ παραπλήσια τῷ ἐποποιῷ καὶ Αἴσχυλος ἐν τοῖς Ἕπτα ἐπι Θήβας.
  In spite of the protest of Schutz, in his note, I think that the
  scholiast has understood the words ἐπίκοτος τροφᾶς (Sept. ad
  Theb. 787) in their plain and just meaning.

  [643] The curses of Œdipus are very frequently and emphatically
  dwelt upon both by Æschylus and Sophoklês (Sept. ad Theb. 70-586,
  655-697, etc.; Œdip. Colon. 1293-1378). The former continues the
  same point of view as the Thêbaïs, when he mentions—

                     ... Τὰς περιθύμους
    Κατάρας βλαψίφρονος Οἰδιπόδα (727);

  or, λόγου τ᾽ ἄνοια καὶ φρενῶν Ἐριννύς (Soph. Antig. 584).

  The Scholiast on Sophoklês (Œd. Col. 1378) treats the cause
  assigned by the ancient Thêbaïs for the curse vented by Œdipus as
  trivial and ludicrous.

  The Ægeids at Sparta, who traced their descent to Kadmus,
  suffered from terrible maladies which destroyed the lives of
  their children; an oracle directed them to appease the Erinnyes
  of Laius and Œdipus by erecting a temple, upon which the maladies
  speedily ceased (Herodot. iv.).

After the death of Œdipus and the celebration of his funeral games,
at which amongst others, Argeia, daughter of Adrastus (afterwards the
wife of Polynikês), was present,[644] his two sons soon quarrelled
respecting the succession. The circumstances are differently related;
but it appears that, according to the original narrative, the wrong and
injustice was on the part of Polynikês, who, however, was obliged to
leave Thêbes and to seek shelter with Adrastus, king of Argos. Here he
met Tydeus, a fugitive, at the same time, from Ætôlia: it was dark when
they arrived, and a broil ensued between the two exiles, but Adrastus
came out and parted them. He had been enjoined by an oracle to give his
two daughters in marriage to a lion and a boar, and he thought this
occasion had now arrived, inasmuch as one of the combatants carried on
his shield a lion, the other a boar. He accordingly gave Deipylê in
marriage to Tydeus, and Argeia to Polynikês: moreover, he resolved to
restore by armed resistance both his sons-in-law to their respective
countries.[645]

  [644] Hesiod. ap. Schol. Iliad. xxiii. 680.

  [645] Apollodôr. iii. 5, 9; Hygin. f. 69; Æschyl. Sept. ad Theb.
  573. Hyginus says that Polynikês came clothed in the skin of a
  lion, and Tydeus in that of a boar; perhaps after Antimachus, who
  said that Tydeus had been brought up by swineherds (Antimach.
  Fragm. 27, ed. Düntzer; ap. Schol. Iliad. iv. 400). Very
  probably, however, the old Thêbaïs compared Tydeus and Polynikês
  to a lion and a boar, on account of their courage and fierceness;
  a simile quite in the Homeric character. Mnaseas gave the words
  of the oracle (ap. Schol. Eurip. Phœniss. 411).

On proposing the expedition to the Argeian chiefs around him he
found most of them willing auxiliaries; but Amphiaräus—formerly his
bitter opponent, but now reconciled to him, and husband of his sister
Eriphylê—strongly opposed him.[646] He denounced the enterprise
as unjust and contrary to the will of the gods. Again, being of a
prophetic stock, descended from Melampus, he foretold the certain
death both of himself and of the principal leaders, should they
involve themselves as accomplices in the mad violence of Tydeus or
the criminal ambition of Polynikês. Amphiaräus, already distinguished
both in the Kalydônian boar-hunt and in the funeral games of Pelias,
was in the Thêban war the most conspicuous of all the heroes, and
absolutely indispensable to its success. But his reluctance to engage
in it was invincible, nor was it possible to prevail upon him except
through the influence of his wife Eriphylê. Polynikês, having brought
with him from Thêbes the splendid robe and necklace given by the
gods to Harmonia on her marriage with Kadmus, offered it as a bribe
to Eriphylê, on condition that she would influence the determination
of Amphiaräus. The sordid wife, seduced by so matchless a present,
betrayed the lurking-place of her husband, and involved him in the
fatal expedition.[647] Amphiaräus, reluctantly dragged forth, and
foreknowing the disastrous issue of the expedition both to himself and
to his associates, addressed his last injunctions, at the moment of
mounting his chariot, to his sons Alkmæôn and Amphilochus, commanding
Alkmæôn to avenge his approaching death by killing the venal Eriphylê,
and by undertaking a second expedition against Thêbes.

  [646] See Pindar, Nem. ix. 30, with the instructive Scholium.

  [647] Apollodôr. iii. 6, 2. The treachery of “the hateful
  Eriphylê” is noticed in the Odyssey, xi. 327: Odysseus sees her
  in the under-world along with the many wives and daughters of the
  heroes.

The Attic dramatists describe this expedition as having been conducted
by seven chiefs, one to each of the seven celebrated gates of Thêbes.
But the Cyclic Thêbaïs gave to it a much more comprehensive character,
mentioning auxiliaries from Arcadia, Messênê, and various parts of
Peloponnêsus;[648] and the application of Tydeus and Polynikês at
Mykênæ in the course of their circuit made to collect allies, is
mentioned in the Iliad. They were well received at Mykênæ; but the
warning signals given by the gods were so terrible that no Mykenæan
could venture to accompany them.[649] The seven principal chiefs
however were Adrastus, Amphiaräus, Kapaneus, Hippomedôn, Parthenopæus,
Tydeus and Polynikês.[650] When the army had advanced as far as the
river Asôpus, a halt was made for sacrifice and banquet; while Tydeus
was sent to Thêbes as envoy to demand the restoration of Polynikês to
his rights. His demand was refused; but finding the chief Kadmeians
assembled at the banquet in the house of Eteoklês, he challenged them
all to contend with him in boxing or wrestling. So efficacious was the
aid of the goddess Athênê that he overcame them all; and the Kadmeians
were so indignant at their defeat, that they placed an ambuscade of
fifty men to intercept him in his way back to the army. All of them
perished by the hand of this warrior, small in stature and of few
words, but desperate and irresistible in the fight. One alone was
spared, Mæon, in consequence of special signals from the gods.[651]

  [648] Pausan. ii. 20, 4; ix. 9, 1. His testimony to this, as he
  had read and admired the Cyclic Thêbaïs, seems quite sufficient,
  in spite of the opinion of Welcker to the contrary (Æschylische
  Trilogie. p. 375).

  [649] Iliad, iv. 376.

  [650] There are differences in respect to the names of the
  seven: Æschylus (Sept. ad Theb. 461) leaves out Adrastus as one
  of the seven, and includes Eteoklus instead of him; others left
  out Tydeus and Polynikês, and inserted Eteoklus and Mekisteus
  (Apollodôr. iii. 6, 3). Antimachus, in his poetical _Thêbaïs_,
  called Parthenopæus an Argeian, not an Arcadian (Schol. ad
  Æschyl. Sept. ad. Theb. 532).

  [651] Iliad, iv. 381-400, with the Schol. The first celebration
  of the Nemean games is connected with this march of the army
  of Adrastus against Thêbes; they were celebrated in honor of
  Archemorus, the infant son of Lykurgus, who had been killed by a
  serpent while his nurse Hypsipylê went to show the fountain to
  the thirsty Argeian chiefs (Apollod. iii. 6, 4; Schol. ad Pindar
  Nem. 1).

The Kadmeians, assisted by their allies the Phôkians and the Phlegyæ,
marched out to resist the invaders, and fought a battle near the
Ismênian hill, in which they were defeated and forced to retire within
the walls. The prophet Teiresias acquainted them that if Menœkeus,
son of Kreôn, would offer himself as a victim to Arês, victory would
be assured to Thêbes. The generous youth, as soon as he learnt that
his life was to be the price of safely to his country, went and slew
himself before the gates. The heroes along with Adrastus now commenced
a vigorous attack upon the town, each of the seven selecting one of the
gates to assault. The contest was long and strenuously maintained but
the devotion of Menœkeus had procured for the Thêbans the protection
of the gods. Parthenopæus was killed with a stone by Periklymenus;
and when the furious Kapaneus, having planted a scaling-ladder, had
mounted the walls, he was smitten by a thunderbolt from Zeus and
cast down dead upon the earth. This event struck terror into the
Argeians, and Adrastus called back his troops from the attack. The
Thêbans now sallied forth to pursue them, when Eteoklês, arresting
the battle, proposed to decide the controversy by single combat with
his brother. The challenge, eagerly accepted by Polynikês, was agreed
to by Adrastus: a single combat ensued between the two brothers, in
which both were exasperated to fury and both ultimately slain by each
other’s hand. This equal termination left the result of the general
contest still undetermined, and the bulk of the two armies renewed the
fight. In the sanguinary struggle which ensued the sons of Astakus on
the Thêban side displayed the most conspicuous and successful valor.
One of them,[652] Melanippus, mortally wounded Tydeus—while two others,
Leades and Amphidikus, killed Eteoklus and Hippomedôn. Amphiaräus
avenged Tydeus by killing Melanippus; but unable to arrest the rout of
the army, he fled with the rest, closely pursued by Periklymenus. The
latter was about to pierce him with his spear, when the beneficence
of Zeus rescued him from this disgrace—miraculously opening the earth
under him, so that Amphiaräus with his chariot and horses was received
unscathed into her bosom.[653] The exact spot where this memorable
incident happened was indicated by a sepulchral building, and shown by
the Thêbans down to the days of Pausanias—its sanctity being attested
by the fact, that no animal would consent to touch the herbage which
grew within the sacred inclosure. Amphiaräus, rendered immortal by
Zeus, was worshipped as a god at Argos, at Thêbes and at Orôpus—and for
many centuries gave answers at his oracle to the questions of the pious
applicant.[654]

  [652] The story recounted that the head of Melanippus was brought
  to Tydeus as he was about to expire of his wound, and that he
  gnawed it with his teeth, a story touched upon by Sophoklês (apud
  Herodian. in Rhetor. Græc. t. viii. p. 601, Walz.).

  The lyric poet Bacchylidês (ap. Schol. Aristoph. Aves, 1535)
  seems to have handled the story even earlier than Sophoklês.

  We find the same allegation embodied in charges against real
  historical men: the invective of Montanus against Aquilius
  Regulus, at the beginning of the reign of Vespasian, affirmed,
  “datam interfectori Pisonis pecuniam a Regulo, appetitumque morsu
  Pisonis caput” (Tacit. Hist. iv. 42).

  [653] Apollodôr. iii. 6, 8. Pindar, Olymp. vi. 11; Nem. ix.
  13-27. Pausan. ix. 8, 2; 18, 2-4.

  Euripidês, in the Phœnissæ (1122 _seqq._), describes the battle
  generally; see also Æsch. S. Th. 392. It appears by Pausanias
  that the Thêbans had poems or legends of their own, relative
  to this war: they dissented in various points from the Cyclic
  Thêbaïs (ix. 18, 4). The Thêbaïs said that Periklymenus had
  killed Parthenopæus; the Thêbans assigned this exploit to
  Asphodikus, a warrior not commemorated by any of the poets known
  to us.

  The village of Harma, between Tanagra and Mykalêssus, was
  affirmed by some to have been the spot where Amphiaräus closed
  his life (Strabo, ix. p. 404): Sophoklês placed the scene at the
  Amphiaræium near Orôpus (ap Strabon. ix. p. 399).

  [654] Pindar, Olymp. vi. 16. Ἕπτα δ᾽ ἔπειτα πυρᾶν νέκρων
  τελεσθέντων Ταλαϊονίδας Εἶπεν ἐν Θήβαισι τοιοῦτόν τι ἔπος· Ποθέω
  στρατιᾶς ὀφθαλμὸν ἐμᾶς Ἀμφότερον, μάντιν τ᾽ ἀγαθὸν καὶ δουρὶ
  μάχεσθαι.

  The scholiast affirms that these last expressions are borrowed by
  Pindar from the Cyclic Thêbaïs.

  The temple of Amphiaräus (Pausan. ii. 23, 2), his oracle,
  seems to have been inferior in estimation only to that of
  Delphi (Herodot. i. 52; Pausan. i. 34; Cicero, Divin. i. 40).
  Crœsus sent a rich present to Amphiaräus, πυθόμενος αὐτοῦ
  τήν τε ἀρετὴν καὶ τὴν πάθην (Herod. l. c); a striking proof
  how these interesting legends were recounted and believed as
  genuine historical facts. Other adventures of Amphiaräus in the
  expedition against Thêbes were commemorated in the carvings on
  the Thronus at Amyklæ (Pausan. iii. 18, 4).

  Æschylus (Sept. Theb. 611) seems to enter into the Thêban view,
  doubtless highly respectful towards Amphiaräus, when he places in
  the mouth of the Kadmeian king Eteoklês such high encomiums on
  Amphiaräus, and so marked a contrast with the other chiefs from
  Argos.

Adrastus, thus deprived of the prophet and warrior whom he regarded as
“the eye of his army,” and having seen the other chiefs killed in the
disastrous fight, was forced to take flight singly, and was preserved
by the matchless swiftness of his horse Areiôn, the offspring of
Poseidôn. He reached Argos on his return, bringing with him nothing
except “his garments of woe and his black-maned steed.”[655]

  [655] Pausan. viii. 25, 5, from the Cyclic Thêbaïs, Εἵματα λυγρὰ
  φέρων σὺν Ἀρείονι κυανοχαίτῃ; also Apollodôr. iii. 6, 8.

  The celebrity of the horse Areiôn was extolled in the Iliad
  (xxiii. 346), in the Cyclic Thêbaïs, and also in the Thêbaïs of
  Antimachus (Pausan. l. c.): by the Arcadians of Thelpusia he
  was said to be the offspring of Dêmêtêr by Poseidôn,—he, and
  a daughter whose name Pausanias will not communicate to the
  uninitiated (ἧς τὸ ὄνομα ἐς ἀτελέστους λέγειν οὐ νομίζουσι, _l.
  c._). A different story is in the Schol. Iliad, xxiii. 346; and
  in Antimachus, who affirmed that “Gæa herself had produced him,
  as a wonder to mortal men” (see Antimach. Frag. 16. p. 102; Epic.
  Græc. Frag. ed. Düntzer).

Kreôn, father of the heroic youth Menœkeus, succeeding to the
administration of Thêbes after the death of the two hostile brothers
and the repulse of Adrastus, caused Eteoklês to be buried with
distinguished honor, but cast out ignominiously the body of Polynikês
as a traitor to his country, forbidding every one on pain of death to
consign it to the tomb. He likewise refused permission to Adrastus to
inter the bodies of his fallen comrades. This proceeding, so offensive
to Grecian feeling, gave rise to two further tales; one of them at
least of the highest pathos and interest. Antigonê, the sister of
Polynikês, heard with indignation the revolting edict consigning her
brother’s body to the dogs and vultures, and depriving it of those
rites which were considered essential to the repose of the dead.
Unmoved by the dissuading counsel of an affectionate but timid sister,
and unable to procure assistance, she determined to brave the hazard
and to bury the body with her own hands. She was detected in the act;
and Kreôn, though forewarned by Teiresias of the consequences, gave
orders that she should be buried alive, as having deliberately set
at naught the solemn edict of the city. His son Hæmôn, to whom she
was engaged to be married, in vain interceded for her life. In an
agony of despair he slew himself in the sepulchre to which the living
Antigonê had been consigned; and his mother Eurydikê, the wife of
Kreôn, inconsolable for his death, perished by her own hand. And thus
the new light which seemed to be springing up over the last remaining
scion of the devoted family of Œdipus, is extinguished amidst gloom and
horrors—which overshadowed also the house and dynasty of Kreôn.[656]

  [656] Sophokl. Antigon. 581. Νῦν γὰρ ἐσχάτας ὑπὲρ Ῥίζας ἐτέτατο
  φάος ἐν Οἰδίπου δόμοις, etc.

  The pathetic tale here briefly recounted forms the subject of
  this beautiful tragedy of Sophoklês, the argument of which is
  supposed by Boeckh to have been borrowed in its primary rudiments
  from the Cyclic Thêbaïs or the Œdipodia (Boeckh, Dissertation
  appended to his translation of the Antigonê, c. x. p. 146); see
  Apollodôr. iii. 7, 1.

  Æschylus also touches upon the heroism of Antigonê (Sep. Theb.
  984).

The other tale stands more apart from the original legend, and seems to
have had its origin in the patriotic pride of the Athenians. Adrastus,
unable to obtain permission from the Thêbans to inter the fallen
chieftains, presented himself in suppliant guise, accompanied by their
disconsolate mothers, to Thêseus at Eleusis. He implored the Athenian
warrior to extort from the perverse Thêbans that last melancholy
privilege which no decent or pious Greeks ever thought of withholding,
and thus to stand forth as the champion of Grecian public morality in
one of its most essential points, not less than of the rights of the
subterranean gods. The Thêbans obstinately persisting in their refusal,
Thêseus undertook an expedition against their city, vanquished them in
the field, and compelled them by force of arms to permit the sepulture
of their fallen enemies. This chivalrous interposition, celebrated in
one of the preserved dramas of Euripidês, formed a subject of glorious
recollection to the Athenians throughout the historical age: their
orators dwelt upon it in terms of animated panegyric; and it seems
to have been accepted as a real fact of the past time, with not less
implicit conviction than the battle of Marathôn.[657] But the Thêbans,
though equally persuaded of the truth of the main story, dissented
from the Athenian version of it, maintaining that they had given up
the bodies for sepulture voluntarily and of their own accord. The
tomb of the chieftains was shown near Eleusis even in the days of
Pausanias.[658]

  [657] Apollodôr. iii. 7, 1; Eurip. Supp. _passim_; Herodot. ix.
  27; Plato, Menexen. c. 9; Lysias, Epitaph. c. 4; Isokrat. Orat.
  Panegyr. p. 196, Auger.

  [658] Pausan. i. 39, 2.

A large proportion both of the interesting persons and of the exalted
acts of legendary Greece belongs to the female sex. Nor can we on this
occasion pass over the name of Evadnê, the devoted widow of Kapaneus,
who cast herself on the funeral pile of her husband and perished.[659]

  [659] Eurip. Supplic. 1004-1110.

The defeat of the seven chiefs before Thêbes was amply avenged by
their sons, again under the guidance of Adrastus:—Ægialeus son of
Adrastus, Thersander son of Polynikês, Alkmæôn and Amphilochus, sons
of Amphiaräus, Diomêdês son of Tydeus, Sthenelus son of Kapaneus,
Promachus son of Parthenopæus, and Euryalus son of Mekistheus, joined
in this expedition. Though all these youthful warriors, called the
Epigoni, took part in the expedition, the grand and prominent place
appears to have been occupied by Alkmæôn, son of Amphiaräus. Assistance
was given to them from Corinth and Megara, as well as from Messênê and
Arcadia; while Zeus manifested his favorable dispositions by signals
not to be mistaken.[660] At the river Glisas the Epigoni were met by
the Thêbans in arms, and a battle took place in which the latter were
completely defeated. Laodamas, son of Eteoklês, killed Ægialeus, son of
Adrastus; but he and his army were routed and driven within the walls
by the valor and energy of Alkmæôn. The defeated Kadmeians consulted
the prophet Teiresias, who informed them that the gods had declared
for their enemies, and that there was no longer any hope of successful
resistance. By his advice they sent a herald to the assailants offering
to surrender the town, while they themselves conveyed away their
wives and children, and fled under the command of Laodamas to the
Illyrians,[661] upon which the Epigoni entered Thêbes, and established
Thersander, son of Polynikês, on the throne.

  [660] Homer, Iliad, iv. 406. Sthenelus, the companion of Diomêdês
  and one of the Epigoni, says to Agamemnôn,—

    Ἡμεῖς τοι πατέρων μέγ᾽ ἀμείνονες εὐχόμεθ᾽ εἶναι·
    Ἡμεῖς καὶ Θήβης ἕδος εἵλομεν ἑπταπύλοιο,
    Παυρότερον λαὸν ἀγαγόνθ᾽ ὑπὸ τεῖχος Ἄρειον,
    Πειθόμενοι τεράεσσι θεῶν καὶ Ζηνὸς ἀρωγῇ·
    Αὐτοὶ δὲ σφετέρῃσιν ἀτασθαλίῃσιν ὄλοντο.

  [661] Apollodôr. iii. 7, 4. Herodot. v. 57-61. Pausan. ix. 5, 7;
  9, 2. Diodôr. iv. 65-66.

  Pindar represents Adrastus as concerned in the second expedition
  against Thêbes (Pyth. viii. 40-58).

Adrastus, who in the former expedition had been the single survivor
amongst so many fallen companions, now found himself the only exception
to the general triumph and joy of the conquerors: he had lost his son
Ægialeus, and the violent sorrow arising from the event prematurely cut
short his life. His soft voice and persuasive eloquence were proverbial
in the ancient epic.[662] He was worshipped as a hero both at Argos and
at Sikyôn, but with especial solemnity in the last-mentioned place,
where his Herôum stood in the public agora, and where his exploits as
well as his sufferings were celebrated periodically in lyric tragedies.
Melanippus, son of Astakus, the brave defender of Thêbes, who had slain
both Tydeus and Mekistheus, was worshipped with no less solemnity
by the Thêbans.[663] The enmity of these two heroes rendered it
impossible for both of them to be worshipped close upon the same spot.
Accordingly it came to pass during the historical period, about the
time of the Solonian legislation at Athens, that Kleisthenês, despot of
Sikyôn, wishing to banish the hero Adrastus and abolish the religious
solemnities celebrated in honor of the latter by the Sikyonians, first
applied to the Delphian oracle for permission to carry this banishment
into effect directly and forcibly. That permission being refused, he
next sent to Thêbes an intimation that he was anxious to introduce
their hero Melanippus into Sikyôn. The Thêbans willingly consented, and
he assigned to the new hero a consecrated spot in the strongest and
most commanding portion of the Sikyonian prytaneium. He did this (says
the historian) “knowing that Adrastus would forthwith go away of his
own accord; since Melanippus was of all persons the most odious to
him, as having slain both his son-in-law and his brother.” Kleisthenês
moreover diverted the festivals and sacrifices which had been offered
to Adrastus, to the newly established hero Melanippus; and the lyric
tragedies from the worship of Adrastus to that of Dionysus. But his
dynasty did not long continue after his decease, and the Sikyonians
then reëstablished their ancient solemnities.[664]

  [662] Γλῶσσαν τ᾽ Ἀδρήστου μειλιχόγηρυν ἔχοι (Tyrtæus, Eleg. 9, 7,
  Schneidewin); compare Plato, Phædr. c. 118. “Adrasti pallentis
  imago” meets the eye of Æneas in the under-world (Æneid, vi. 480).

  [663] About Melanippus, see Pindar, Nem. x. 36. His sepulchre was
  shown near the Prœtid gates of Thêbes (Pausan. ix. 18, 1).

  [664] This very carious and illustrative story is contained
  in Herodot. v. 67. Ἐπεὶ δὲ ὁ θεὸς τοῦτο οὐ παρεδίδου, ἀπελθὼν
  ὀπίσω (Kleisthenês, returning from Delphi) ἐφρόντιζε μηχανὴν
  τῇ ~αὐτὸς ὁ Ἀδρήστος ἀπαλλάξεται~. Ὡς δὲ οἱ ἐξευρῆσθαι ἐδόκεε,
  πέμψας ἐς Θήβας τὰς Βοιωτίας, ἔφη θέλειν ἐπαγαγέσθαι Μελάνιππον
  τὸν Ἀστακοῦ· οἱ δὲ Θηβαῖοι ἔδοσαν. Ἐπηγάγετο δὲ τὸν Μελάνιππον
  ὁ Κλεισθένης, καὶ γὰρ τοῦτο δεῖ ἀπηγήσασθαι, ὡς ἔχθιστον ἐόντα
  Ἀδρήστῳ· ὃς τόν τε ἀδέλφεον Μηκιστέα ἀπεκτόνεε, καὶ τὸν γαμβρὸν
  Τυδέα.

  The Sikyonians (Herodotus says) τά τε δὴ ἄλλα ἐτίμων τὸν
  Ἄδρηστον, καὶ πρὸς τὰ πάθεα αὐτοῦ τραγικοῖσι χόροισι ἐγέραιρον·
  τὸν μὲν Διόνυσον οὐ τιμέωντες, τὸν δὲ Ἄδρηστον.

  Adrastus was worshipped as a hero at Megara as well as at Sikyôn:
  the Megarians affirmed that he had died there on his way back
  from Thêbes (Pausan. i. 43, 1; Dieuchidas, ap. Schol. ad Pindar.
  Nem. ix. 31). His house at Argos was still shown when Pausanias
  visited the town (ii. 23, 2).

Near the Prœtid gate of Thêbes were seen the tombs of two combatants
who had hated each other during life even more than Adrastus and
Melanippus—the two brothers Eteoklês and Polynikês. Even as heroes
and objects of worship, they still continued to manifest their
inextinguishable hostility: those who offered sacrifices to them
observed that the flame and the smoke from the two adjoining altars
abhorred all communion, and flew off in directions exactly opposite.
The Thêban exegetes assured Pausanias of this fact. And though he did
not himself witness it, yet having seen with his own eyes a miracle not
very dissimilar at Pioniæ in Mysia, he had no difficulty in crediting
their assertion.[665]

  [665] Pausan. ix. 18, 3. Τὰ ἐπ᾽ αὐτοῖς δρώμενα οὐ θεασάμενος
  πιστὰ ὅμως ὑπείληφα εἶναι. Compare Hygin. f. 68.

    “Et nova fraterno veniet concordia fumo,
      Quem vetus accensâ separat ira pyrâ.” (Ovid, Ibis, 35.)

  The tale was copied by Ovid from Kallimachus (Trist. v. 5, 38.)

Amphiaräus when forced into the first attack of Thêbes—against his own
foreknowledge and against the warnings of the gods—had enjoined his
sons Alkmæôn and Amphilochus not only to avenge his death upon the
Thêbans, but also to punish the treachery of their mother, “Eriphylê,
the destroyer of her husband.”[666] In obedience to this command, and
having obtained the sanction of the Delphian oracle, Alkmæôn slew his
mother;[667] but the awful Erinnys, the avenger of matricide, inflicted
on him a long and terrible punishment, depriving him of his reason,
and chasing him about from place to place without the possibility of
repose or peace of mind. He craved protection and cure from the god at
Delphi, who required him to dedicate at the temple, as an offering,
the precious necklace of Kadmus, that irresistible bribe which had
originally corrupted Eriphylê.[668] He further intimated to the unhappy
sufferer, that though the whole earth was tainted with his crime,
and had become uninhabitable for him, yet there was a spot of ground
which was not under the eye of the sun at the time when the matricide
was committed, and where therefore Alkmæôn yet might find a tranquil
shelter. The promise was realized at the mouth of the river Achelôus,
whose turbid stream was perpetually depositing new earth and forming
additional islands. Upon one of these, near Œniadæ, Alkmæôn settled,
permanently and in peace: he became the primitive hero of Akarnania,
to which his son Akarnan gave name.[669] The necklace was found among
the treasures of Delphi, together with that which had been given by
Aphroditê to Helen, by the Phôkian plunderers who stripped the temple
in the time of Philip of Macedôn. The Phôkian women quarrelled about
these valuable ornaments: and we are told that the necklace of Eriphylê
was allotted to a woman of gloomy and malignant disposition, who ended
by putting her husband to death; that of Helen to a beautiful but
volatile wife, who abandoned her husband from a preference for a young
Epirot.[670]

  [666] Ἀνδροδάμαντ᾽ Ἐριφύλην (Pindar, Nem. ix. 16). A poem
  _Eryphilê_ was included among the mythical compositions of
  Stesichorus: he mentioned in it that Asklêpius had restored
  Kapaneus to life, and that he was for that reason struck dead
  by thunder from Zeus (Stesichor. Fragm. Kleine, 18, p. 74). Two
  tragedies of Sophoklês once existed, _Epigoni_ and _Alkmæôn_
  (Welcker, Griechisch. Tragöd. i. p. 269): a few fragments also
  remain of the Latin _Epigoni_ and _Alphesibæa_ of Attius: Ennius
  and Attius both composed or translated from the Greek a Latin
  _Alkmæôn_ (Poet. Scenic. Latin. ed. Both. pp. 33, 164, 198).

  [667] Hyginus gives the fable briefly (f. 73; see also
  Asclepiadês, ap. Schol. Odyss. xi. 326). In like manner, in the
  case of the matricide of Orestês, Apollo not only sanctions,
  but enjoins the deed; but his protection against the avenging
  Erinnyês is very tardy, not taking effect until after Orestês has
  been long persecuted and tormented by them (see Æschyl. Eumen.
  76, 197 462).

  In the _Alkmæôn_ of the later tragic writer Thodektês, a
  distinction was drawn: the gods had decreed that Eriphylê should
  die, but not that Alkmæôn should kill her (Aristot. Rhetoric.
  ii. 24). Astydamas altered the story still more in his tragedy,
  and introduced Alkmæôn as killing his mother ignorantly and
  without being aware who she was (Aristot. Poetic. c. 27). The
  murder of Eriphylê by her son was one of the παρειλήμμενοι
  μῦθοι which could not be departed from; but interpretations
  and qualifications were resorted to, in order to prevent it
  from shocking the softened feelings of the spectators: see the
  criticism of Aristotle on the _Alkmæôn_ of Euripidês (Ethic.
  Nicom. iii. 1, 8).

  [668] Ephorus ap. Athenæ. vi. p. 232.

  [669] Thucyd. ii. 68-102.

  [670] Athenæ. _l. c._

There were several other legends respecting the distracted Alkmæôn,
either appropriated or invented by the Attic tragedians. He went to
Phêgeus, king of Psôphis in Arcadia, whose daughter Arsinoê he married,
giving as a nuptial present the necklace of Eriphylê. Being however
unable to remain there, in consequence of the unremitting persecutions
of the maternal Erinnys, he sought shelter at the residence of king
Achelôus, whose daughter Kallirhoê he made his wife, and on whose soil
he obtained repose.[671] But Kallirhoê would not be satisfied without
the possession of the necklace of Eriphylê, and Alkmæôn went back to
Psôphis to fetch it, where Phêgeus and his sons slew him. He had left
twin sons, infants, with Kallirhoê, who prayed fervently to Zeus that
they might be preternaturally invested with immediate manhood, in order
to revenge the murder of their father. Her prayer was granted, and
her sons Amphoterus and Akarnan, having instantaneously sprung up to
manhood, proceeded into Arcadia, slew the murderers of their father,
and brought away the necklace of Eriphylê, which they carried to
Delphi.[672]

  [671] Apollodôr. iii. 7, 5-6; Pausan. viii. 24, 4. These two
  authors have preserved the story of the Akarnanians and the old
  form of the legend, representing Alkmæôn as having found shelter
  at the abode of the person or king Achelôus, and married his
  daughter: Thucydidês omits the _personality_ of Achelôus, and
  merely announces the wanderer as having settled on certain new
  islands deposited by the river.

  I may remark that this is a singularly happy adaptation of a
  legend to an existing topographical fact. Generally speaking,
  before any such adaptation can be rendered plausible, the legend
  is of necessity much transformed; here it is taken exactly as it
  stands, and still fits on with great precision.

  Ephorus recounted the whole sequence of events as so much
  political history, divesting it altogether of the legendary
  character. Alkmæôn and Diomêdês, after having taken Thêbes with
  the other Epigoni, jointly undertook an expedition into Ætôlia
  and Akarnania: they first punished the enemies of the old Œneus,
  grandfather of Diomêdês, and established the latter as king in
  Kalydôn: next they conquered Akarnania for Alkmæôn. Alkmæôn,
  though invited by Agamemnôn to join in the Trojan war, would not
  consent to do so (Ephor. ap. Strabo. vii. p. 326; x. p. 462).

  [672] Apollodôr. iii. 7, 7; Pausan. viii. 24, 3-4. His remarks
  upon the mischievous longing of Kallirhoê for the necklace are
  curious: he ushers them in by saying, that “many men, and still
  more women, are given to fall into absurd desires,” etc. He
  recounts it with all the _bonne foi_ which belongs to the most
  assured matter of fact.

  A short allusion is in Ovid’s Metamorphoses (ix. 412).

Euripidês deviated still more widely from the ancient epic, by making
Alkmæôn the husband of Mantô, daughter of Teiresias, and the father of
Amphilochus. According to the Cyclic Thêbaïs, Mantô was consigned by
the victorious Epigoni as a special offering to the Delphian god; and
Amphilochus was son of Amphiaräus, not son of Alkmæôn.[673] He was the
eponymous hero of the town called the Amphilochian Argos, in Akarnania,
on the shore of the Gulf of Ambrakia. Thucydidês tells us that he went
thither on his return from the Trojan war, being dissatisfied with
the state of affairs which he found at the Peloponnêsian Argos.[674]
The Akarnanians were remarkable for the numerous prophets which they
supplied to the rest of Greece: their heroes were naturally drawn from
the great prophetic race of the Melampodids.

  [673] Thêbaïd, Cy. Reliqu. p. 70, Leutsch; Schol. Apollôn. Rhod.
  i. 408. The following lines cited in Athenæus (vii. p. 317) are
  supposed by Boeckh, with probable reason, to be taken from the
  Cyclic Thêbaïs; a portion of the advice of Amphiaräus to his sons
  at the time of setting out on his last expedition,—

    Πουλύποδός μοι, τέκνον, ἔχων νόον, Ἀμφίλοχ᾽ ἥρως,
    Τοῖσιν ἐφαρμόζου, τῶν ἂν κατὰ δῆμον ἵκηαι.

  There were two tragedies composed by Euripidês, under the
  title of Ἀλκμαίων, ὁ διὰ Ψωφῖδος, and Ἀλκμαίων, ὁ διὰ Κορίνθου
  (Dindorf, Fragm. Eurip. p. 77).

  [674] Apollodôr. iii. 7, 7; Thucyd. ii. 68.

Thus ends the legend of the two sieges of Thêbes; the greatest event,
except the siege of Troy, in the ancient epic; the greatest enterprise
of war, between Greeks and Greeks, during the time of those who are
called the Heroes.




CHAPTER XV.

LEGEND OF TROY.


We now arrive at the capital and culminating point of the Grecian
epic,—the two sieges and capture of Troy, with the destinies of the
dispersed heroes, Trojan as well as Grecian, after the second and most
celebrated capture and destruction of the city.

It would require a large volume to convey any tolerable idea of
the vast extent and expansion of this interesting fable, first
handled by so many poets, epic, lyric and tragic, with their endless
additions, transformations and contradictions,—then purged and
recast by historical inquirers, who under color of setting aside
the exaggerations of the poets, introduced a new vein of prosaic
invention,—lastly, moralized and allegorized by philosophers. In the
present brief outline of the general field of Grecian legend, or of
that which the Greeks believed to be their antiquities, the Trojan
war can be regarded as only one among a large number of incidents
upon which Hekatæus and Herodotus looked back as constituting their
fore-time. Taken as a special legendary event, it is indeed of wider
and larger interest than any other, but it is a mistake to single it
out from the rest as if it rested upon a different and more trustworthy
basis. I must therefore confine myself to an abridged narrative of
the current and leading facts; and amidst the numerous contradictory
statements which are to be found respecting every one of them, I
know no better ground of preference than comparative antiquity,
though even the oldest tales which we possess—those contained in the
Iliad—evidently presuppose others of prior date.

The primitive ancestor of the Trojan line of kings is Dardanus, son of
Zeus, founder and eponymus of Dardania:[675] in the account of later
authors, Dardanus was called the son of Zeus by Elektra, daughter of
Atlas, and was further said to have come from Samothrace, or from
Arcadia, or from Italy;[676] but of this Homer mentions nothing. The
first Dardanian town founded by him was in a lofty position on the
descent of Mount Ida; for he was not yet strong enough to establish
himself on the plain. But his son Erichthonius, by the favor of
Zeus, became the wealthiest of mankind. His flocks and herds having
multiplied, he had in his pastures three thousand mares, the offspring
of some of whom, by Boreas, produced horses of preternatural swiftness.
Trôs, the son of Erichthonius, and the eponym of the Trojans, had three
sons—Ilus, Assaracus, and the beautiful Ganymêdês, whom Zeus stole away
to become his cup-bearer in Olympus, giving to his father Trôs, as the
price of the youth, a team of immortal horses.[677]

  [675] Iliad, xx. 215.

  [676] Hellanik. Fragm. 129, Didot; Dionys. Hal. i. 50-61;
  Apollodôr. iii. 12, 1; Schol. Iliad. xviii. 486; Varro, ap.
  Servium ad Virgil. Æneid. iii. 167. Kephalôn. Gergithius ap.
  Steph. Byz. v. Ἀρίσβη.

  [677] Iliad, v. 265; Hellanik. Fr. 146; Apollod. ii. 5, 9.

From Ilus and Assaracus the Trojan and Dardanian lines diverge; the
former passing from Ilus to Laomedôn, Priam and Hectôr; the latter from
Assaracus to Capys, Anchisês and Æneas. Ilus founded in the plain of
Troy the holy city of Ilium; Assaracus and his descendants remained
sovereigns of Dardania.[678]

  [678] Iliad, xx. 236.

It was under the proud Laomedôn, son of Ilus, that Poseidôn and Apollo
underwent, by command of Zeus, a temporary servitude; the former
building the walls of the town, the latter tending the flocks and
herds. When their task was completed and the penal period had expired,
they claimed the stipulated reward; but Laomedôn angrily repudiated
their demand, and even threatened to cut off their ears, to tie them
hand and foot, and to sell them in some distant island as slaves.[679]
He was punished for this treachery by a sea-monster, whom Poseidôn
sent to ravage his fields and to destroy his subjects. Laomedôn
publicly offered the immortal horses given by Zeus to his father Trôs,
as a reward to any one who would destroy the monster. But an oracle
declared that a virgin of noble blood must be surrendered to him, and
the lot fell upon Hesionê, daughter of Laomedôn himself. Hêraklês
arriving at this critical moment, killed the monster by the aid of a
fort built for him by Athênê and the Trojans,[680] so as to rescue
both the exposed maiden and the people; but Laomedôn, by a second act
of perfidy, gave him mortal horses in place of the matchless animals
which had been promised. Thus defrauded of his due, Hêraklês equipped
six ships, attacked and captured Troy and killed Laomedôn,[681] giving
Hesionê to his friend and auxiliary Telamôn, to whom she bore the
celebrated archer Teukros.[682] A painful sense of this expedition was
preserved among the inhabitants of the historical town of Ilium, who
offered no worship to Hêraklês.[683]

  [679] Iliad, vii. 451; xxi. 456. Hesiod. ap. Schol. Lycophr. 393.

  [680] Iliad, xx. 145; Dionys. Hal. i. 52.

  [681] Iliad, v. 640. Meneklês (ap. Schol. Venet. _ad loc._)
  affirmed that this expedition of Hêraklês was a fiction; but
  Dikæarchus gave, besides, other exploits of the hero in the same
  neighborhood, at Thêbê Hypoplakiê (Schol. Iliad, vi. 396).

  [682] Diodôr. iv. 32-49. Compare Venet. Schol. ad Iliad. viii.
  284.

  [683] Strabo, xiii. p. 596.

Among all the sons of Laomedôn, Priam[684] was the only one who
had remonstrated against the refusal of the well-earned guerdon of
Hêraklês; for which the hero recompensed him by placing him on the
throne. Many and distinguished were his sons and daughters, as well by
his wife Hekabê, daughter of Kisseus, as by other women.[685] Among the
sons were Hectôr,[686] Paris, Dêiphobus, Helenus, Trôilus, Politês,
Polydôrus; among the daughters Laodikê, Kreüsa, Polyxena, and Kassandra.

  [684] As Dardanus, Trôs and Ilus are respectively eponyms of
  Dardania, Troy and Ilium, so Priam is eponym of the acropolis
  _Pergamum_. Πρίαμος is in the Æolic dialect Πέῤῥαμος (Hesychius):
  upon which Ahrens remarks, “Cæterum ex hac Æolicâ nominis formâ
  apparet, Priamum non minus arcis Περγάμων eponymum esse, quam
  Ilum urbis, Troem populi: Πέργαμα enim a Περίαμα natum est, ι
  in γ mutato.” (Ahrens, De Dialecto Æolicâ, 8, 7. p. 56: compare
  _ibid._ 28, 8. p. 150, πεῤῥ᾽ ἁπάλω).

  [685] Iliad, vi. 245; xxiv. 495.

  [686] Hectôr was affirmed, both by Stesichorus and Ibykus, to be
  the son of Apollo (Stesichorus, ap. Schol. Ven. ad Iliad. xxiv.
  259; Ibyki Fragm. xiv. ed. Schneidewin): both Euphoriôn (Fr. 125,
  Meineke) and Alexander Ætôlus follow the same idea. Stesichorus
  further stated, that after the siege Apollo had carried Hekabê
  away into Lykia to rescue her from captivity (Pausanias, x. 27,
  1): according to Euripidês, Apollo had promised that she should
  die in Troy (Troad. 427).

  By Sapphô, Hectôr was given as a surname of Zeus, Ζεὺς Ἕκτωρ
  (Hesychius, v. Ἕκτορες); a prince belonging to the regal family
  of Chios, anterior to the Ionic settlement, as mentioned by the
  Chian poet Iôn (Pausan. vii. 3, 3), was so called.

The birth of Paris was preceded by formidable presages; for Hekabê
dreamt that she was delivered of a firebrand, and Priam, on consulting
the soothsayers, was informed that the son about to be born would prove
fatal to him. Accordingly he directed the child to be exposed on Mount
Ida; but the inauspicious kindness of the gods preserved him, and he
grew up amidst the flocks and herds, active and beautiful, fair of hair
and symmetrical in person, and the special favorite of Aphroditê.[687]

  [687] Iliad, iii. 45-55; Schol. Iliad. iii. 325; Hygin. fab. 91;
  Apollodôr. iii. 12, 5.

It was to this youth, in his solitary shepherd’s walk on Mount Ida,
that the three goddesses Hêrê, Athênê, and Aphroditê were conducted,
in order that he might determine the dispute respecting their
comparative beauty, which had arisen at the nuptials of Pêleus and
Thetis,—a dispute brought about in pursuance of the arrangement, and in
accomplishment of the deep-laid designs, of Zeus. For Zeus, remarking
with pain the immoderate numbers of the then existing heroic race,
pitied the earth for the overwhelming burden which she was compelled
to bear, and determined to lighten it by exciting a destructive
and long-continued war.[688] Paris awarded the palm of beauty to
Aphroditê, who promised him in recompense the possession of Helena,
wife of the Spartan Menelaus,—the daughter of Zeus and the fairest of
living women. At the instance of Aphroditê, ships were built for him,
and he embarked on the enterprise so fraught with eventual disaster to
his native city, in spite of the menacing prophecies of his brother
Helenus, and the always neglected warnings of Kassandra.[689]

  [688] This was the motive assigned to Zeus by the old epic poem,
  the Cyprian Verses (Frag. 1. Düntz. p. 12; ap. Schol. ad Iliad.
  i. 4):—

  Ἡ δὲ ἱστορία παρὰ Στασίνῳ τῷ τὰ Κύπρια πεποιηκότι εἰπόντι οὕτως·

    Ἦν ὅτε μύρια φῦλα κατὰ χθόνα πλαζόμενα ...
    ... βαρυστέρνου πλάτος αἴης.
    Ζεὺς δὲ ἰδὼν ἐλέησε, καὶ ἐν πυκιναῖς πραπίδεσσι
    Σύνθετο κουφίσαι ἀνθρώπων παμβώτορα γαῖαν,
    Ῥιπίσας πολέμου μεγάλην ἔριν Ἰλιακοῖο,
    Ὄφρα κενώσειεν θάνατῳ βάρος· οἱ δ᾽ ἐνὶ Τροίῃ
    Ἥρωες κτείνοντο, Διὸς δ᾽ ἐτελείετο βουλή.

  The same motive is touched upon by Eurip. Orest, 1635; Helen.
  38; and seriously maintained, as it seems, by Chrysippus, ap.
  Plutarch. Stoic. Rep. p. 1049: but the poets do not commonly go
  back farther than the passion of Paris for Helen (Theognis, 1232;
  Simonid. Amorg. Fragm. 6, 118).

  The judgment of Paris was one of the scenes represented on the
  ancient chest of Kypselus at Olympia (Pausan. v. 19, 1).

  [689] Argument of the Ἔπη Κύπρια (ap. Düntzer, p. 10). These
  warnings of Kassandra form the subject of the obscure and
  affected poem of Lycophrôn.

Paris, on arriving at Sparta, was hospitably entertained by Menelaus
as well as by Kastôr and Pollux, and was enabled to present the rich
gifts which he had brought to Helen.[690] Menelaus then departed to
Krête, leaving Helen to entertain his Trojan guest—a favorable moment
which was employed by Aphroditê to bring about the intrigue and the
elopement. Paris carried away with him both Helen and a large sum
of money belonging to Menelaus—made a prosperous voyage to Troy—and
arrived there safely with his prize on the third day.[691]

  [690] According to the Cyprian Verses, Helena was daughter of
  Zeus by Nemesis, who had in vain tried to evade the connection
  (Athenæ. viii. 334). Hesiod (Schol. Pindar. Nem. x. 150)
  represented her as daughter of Oceanus and Têthys, an oceanic
  nymph: Sapphô (Fragm. 17, Schneidewin), Pausanias (i. 33, 7),
  Apollodôrus (iii. 10, 7), and Isokratês (Encom. Helen. v. ii. p.
  366, Auger) reconcile the pretensions of Lêda and Nemesis to a
  sort of joint maternity (see Heinrichsen, De Carminibus Cypriis,
  p. 45-46).

  [691] Herodot. ii. 117. He gives distinctly the assertion of
  the Cyprian Verses, which contradicts the argument of the poem
  as it appears in Proclus (Fragm. 1, 1), according to which
  latter, Paris is driven out of his course by a storm and captures
  the city of Sidôn. Homer (Iliad, vi. 293) seems however to
  countenance the statement in the argument.

  That Paris was guilty of robbery, as well as of the abduction of
  Helen, is several times mentioned in the Iliad (iii. 144; vii.
  350-363), also in the argument of the Cyprian Verses (see Æschyl.
  Agam. 534).

Menelaus, informed by Iris in Krête of the perfidious return made by
Paris for his hospitality, hastened home in grief and indignation
to consult with his brother Agamemnôn, as well as with the venerable
Nestôr, on the means of avenging the outrage. They made known the
event to the Greek chiefs around them, among whom they found universal
sympathy: Nestôr, Palamêdês and others went round to solicit aid in
a contemplated attack of Troy, under the command of Agamemnôn, to
whom each chief promised both obedience and unwearied exertion until
Helen should be recovered.[692] Ten years were spent in equipping the
expedition. The goddesses Hêrê and Athênê, incensed at the preference
given by Paris to Aphroditê, and animated by steady attachment to
Argos, Sparta and Mykênæ, took an active part in the cause; and the
horses of Hêrê were fatigued with her repeated visits to the different
parts of Greece.[693]

  [692] The ancient epic (Schol. ad Il. ii. 286-339) does not
  recognize the story of the numerous suitors of Helen, and
  the oath by which Tyndareus bound them all before he made
  the selection among them, that each should swear not only to
  acquiesce, but even to aid in maintaining undisturbed possession
  to the husband whom she should choose. This story seems to have
  been first told by Stesichorus (see Fragm. 20. ed. Kleine;
  Apollod. iii. 10, 8). Yet it was evidently one of the prominent
  features of the current legend in the time of Thucydidês (i. 9;
  Euripid. Iphig. Aul. 51-80; Soph. Ajax, 1100).

  The exact spot in which Tyndareus exacted this oath from the
  suitors, near Sparta, was pointed out even in the time of
  Pausanias (iii. 20, 9).

  [693] Iliad, iv. 27-55; xxiv. 765. Argument. Carm. Cypri. The
  point is emphatically touched upon by Dio Chrysostom (Orat.
  xi. p. 335-336) in his assault upon the old legend. Two years’
  preparation—in Dictys Cret. i. 16.

By such efforts a force was at length assembled at Aulis[694] in
Bœôtia, consisting of 1186 ships and more than 100,000 men,—a force
outnumbering by more than ten to one anything that the Trojans
themselves could oppose, and superior to the defenders of Troy even
with all her allies included.[695] It comprised heroes with their
followers from the extreme points of Greece—from the north-western
portions of Thessaly under Mount Olympus, as well as the western
islands of Dulichium and Ithaca, and the eastern islands of Krête
and Rhodes. Agamemnôn himself contributed 100 ships manned with the
subjects of his kingdom of Mykênæ, besides furnishing 60 ships to the
Arcadians, who possessed none of their own. Menelaus brought with him
60 ships, Nestôr from Pylus 90, Idomeneus from Krête and Diomêdês from
Argos 80 each. Forty ships were manned by the Eleians, under four
different chiefs; the like number under Megês from Dulichium and the
Echinades, and under Thoas from Kalydôn and the other Ætôlian towns.
Odysseus from Ithaca, and Ajax from Salamis, brought 12 ships each. The
Abantes from Eubœa, under Elephênôr, filled 40 vessels; the Bœôtians,
under Peneleôs and Lêitus, 50; the inhabitants of Orchomenos and
Aspledôn, 30; the light-armed Locrians, under Ajax son of Oileus,[696]
40; the Phôkians as many. The Athenians, under Menestheus, a chief
distinguished for his skill in marshalling an army, mustered 50 ships;
the Myrmidons from Phthia and Hellas, under Achilles, assembled in
50 ships; Protesilaus from Phylakê and Pyrasus, and Eurypylus from
Ormenium, each came with 40 ships; Machaôn and Podaleirius, from
Trikka, with 30; Admêtus, from Pheræ and the lake Bœbêis, with 11;
and Philoktêtês from Melibœa with 7: the Lapithæ, under Polypœtês,
son of Peirithous, filled 40 vessels; the Ænianes and Perrhæbians,
under Guneus,[697] 22; and the Magnêtês under Prothous, 40; these last
two were from the northernmost parts of Thessaly, near the mountains
Pêlion and Olympus. From Rhodes, under Tlêpolemus, son of Hêraklês,
appeared 9 ships; from Symê under the comely but effeminate Nireus,
3; from Kôs, Krapathus and the neighboring islands, 30, under the
orders of Pheidippus and Antiphus, sons of Thessalus and grandsons of
Hêraklês.[698]

  [694] The Spartan king Agesilaus, when about to start from Greece
  on his expedition into Asia Minor (396 B. C.) went to Aulis
  personally, in order that he too might sacrifice on the spot
  where Agamemnôn had sacrificed when he sailed for Troy (Xenoph.
  Hellen. iii. 4, 4).

  Skylax (c. 60) notices the ἱερὸν at Aulis, and nothing else: it
  seems to have been like the adjoining Delium, a temple with a
  small village grown up around it.

  Aulis is recognized as the port from which the expedition
  started, in the Hesiodic Works and Days (v. 650).

  [695] Iliad, ii. 128. Uschold (Geschichte des Trojanischen
  Kriegs, p. 9, Stutgart 1836) makes the total 135,000 men.

  [696] The Hesiodic Catalogue notices Oileus, or Ileus, with a
  singular etymology of his name (Fragm. 136, ed. Marktscheffel).

  [697] Γουνεὺς is the Heros Eponymus of the town of Gonnus in
  Thessaly; the duplication of the consonant and shortening of the
  vowel belong to the Æolic dialect (Ahrens, De Dialect. Æolic. 50,
  4. p. 220).

  [698] See the Catalogue in the second book of the Iliad. There
  must probably have been a Catalogue of the Greeks also in
  the Cyprian Verses; for a Catalogue of the allies of Troy is
  specially noticed in the Argument of Proclus (p. 12. Düntzer).

  Euripidês (Iphig. Aul. 165-300) devotes one of the songs of the
  Chorus to a partial Catalogue of the chief heroes.

  According to Dictys Cretensis, all the principal heroes engaged
  in the expedition were kinsmen, all Pelopids (i. 14): they take
  an oath not to lay down their arms until Helen shall have been
  recovered, and they receive from Agamemnôn a large sum of gold.

Among this band of heroes were included the distinguished warriors
Ajax and Diomêdês, and the sagacious Nestôr; while Agamemnôn himself,
scarcely inferior to either of them in prowess, brought with him a high
reputation for prudence in command. But the most marked and conspicuous
of all were Achilles and Odysseus; the former a beautiful youth born of
a divine mother, swift in the race, of fierce temper and irresistible
might; the latter not less efficient as an ally from his eloquence,
his untiring endurance, his inexhaustible resources under difficulty,
and the mixture of daring courage with deep-laid cunning which never
deserted him:[699] the blood of the arch-deceiver Sisyphus, through
an illicit connection with his mother Antikleia, was said to flow in
his veins,[700] and he was especially patronized and protected by
the goddess Athênê. Odysseus, unwilling at first to take part in the
expedition, had even simulated insanity; but Palamêdês, sent to Ithaca
to invite him, tested the reality of his madness by placing in the
furrow where Odysseus was ploughing, his infant son Telemachus. Thus
detected, Odysseus could not refuse to join the Achæan host, but the
prophet Halithersês predicted to him that twenty years would elapse
before he revisited his native land.[701] To Achilles the gods had
promised the full effulgence of heroic glory before the walls of Troy;
nor could the place be taken without both his coöperation and that of
his son after him. But they had forewarned him that this brilliant
career would be rapidly brought to a close; and that if he desired a
long life, he must remain tranquil and inglorious in his native land.
In spite of the reluctance of his mother Thetis, he preferred few years
with bright renown, and joined the Achæan host.[702] When Nestôr and
Odysseus came to Phthia to invite him, both he and his intimate friend
Patroclus eagerly obeyed the call.[703]

  [699] For the character of Odysseus, Iliad, iii. 202-220; x. 247.
  Odyss. xiii. 295.

  The Philoktêtês of Sophoklês carries out very justly the
  character of the Homeric Odysseus (see v. 1035)—more exactly than
  the Ajax of the same poet depicts it.

  [700] Sophokl. Philoktêt. 417, and Schol.—also Schol. ad Soph.
  Ajac. 190.

  [701] Homer, Odyss. xxiv. 115; Æschyl. Agam. 841; Sophokl.
  Philoktêt. 1011, with the Schol. Argument of the Cypria in
  Heinrichsen, De Carmin. Cypr. p. 23 (the sentence is left out in
  Düntzer, p. 11).

  A lost tragedy of Sophoklês, Ὀδυσσεὺς Μαινόμενος, handled this
  subject.

  Other Greek chiefs were not less reluctant than Odysseus to take
  part in the expedition: see the tale of Pœmandrus, forming a
  part of the temple-legend of the Achilleium at Tanagra in Bœôtia
  (Plutarch, Quæstion. Græc. p. 299).

  [702] Iliad, i. 352; ix. 411.

  [703] Iliad, xi. 782.

Agamemnôn and his powerful host set sail from Aulis; but being
ignorant of the locality and the direction, they landed by mistake in
Teuthrania, a part of Mysia near the river Kaïkus, and began to ravage
the country under the persuasion that it was the neighborhood of Troy.
Telephus, the king of the country,[704] opposed and repelled them, but
was ultimately defeated and severely wounded by Achilles. The Greeks
now, discovering their mistake, retired; but their fleet was dispersed
by a storm and driven back to Greece. Achilles attacked and took
Skyrus, and there married Deidamia, the daughter of Lycomêdês.[705]
Telephus, suffering from his wounds, was directed by the oracle to come
to Greece and present himself to Achilles to be healed, by applying
the scrapings of the spear with which the wound had been given: thus
restored, he became the guide of the Greeks when they were prepared to
renew their expedition.[706]

  [704] Telephus was the son of Augê, daughter of king Aleus
  of Tegea in Arcadia, by Hêraklês: respecting her romantic
  adventures, see the previous chapter on Arcadian legends—Strabo’s
  faith in the story (xii. p. 572).

  The spot called the Harbor of the Achæans, near Gryneium, was
  stated to be the place where Agamemnôn and the chiefs took
  counsel whether they should attack Telephus or not (Skylax, c.
  97; compare Strabo, xiv. p. 622).

  [705] Iliad, xi. 664; Argum. Cypr. p. 11, Düntzer; Diktys Cret.
  ii. 3-4.

  [706] Euripid. Telephus, Frag. 26, Dindorf; Hygin. f. 101;
  Diktys, ii. 10. Euripidês had treated the adventure of Telephus
  in this lost tragedy: he gave the miraculous cure with the dust
  of the spear, πριστοῖσι λογχῆς θέλγεται ῥινήμασι. Diktys softens
  down the prodigy: “Achilles cum Machaone et Podalirio adhibeutes
  curam vulneri,” etc. Pliny (xxxiv. 15) gives to the rust of brass
  or iron a place in the list of genuine remedies.

  “Longe omnino a Tiberi ad Caicum: quo in loco etiam Agamemnôn
  errasset, nisi ducem Telephum invenisset” (Cicero, Pro L. Flacco,
  c. 29). The portions of the Trojan legend treated in the lost
  epics and the tragedians, seem to have been just as familiar to
  Cicero as those noticed in the Iliad.

  Strabo pays comparatively little attention to any portion of the
  Trojan war except what appears in Homer. He even goes so far as
  to give a reason why the Amazons _did not_ come to the aid of
  Priam: they were at enmity with him, because Priam had aided
  the Phrygians against them (Iliad, iii. 188: in Strabo, τοῖς
  Ἰῶσιν must be a mistake for τοῖς Φρυξίν). Strabo can hardly have
  read, and never alludes to, Arktinus; in whose poem the brave
  and beautiful Penthesileia, at the head of her Amazons, forms a
  marked epoch and incident of the war (Strabo, xii. 552).

The armament was again assembled at Aulis, but the goddess Artemis,
displeased with the boastful language of Agamemnôn, prolonged the
duration of adverse winds, and the offending chief was compelled
to appease her by the well-known sacrifice of his daughter
Iphigeneia.[707] They then proceeded to Tenedos, from whence Odysseus
and Menelaus were despatched as envoys to Troy, to redemand Helen and
the stolen property. In spite of the prudent counsels of Antenôr,
who received the two Grecian chiefs with friendly hospitality, the
Trojans rejected the demand, and the attack was resolved upon. It was
foredoomed by the gods that the Greek who first landed should perish:
Protesilaus was generous enough to put himself upon this forlorn hope,
and accordingly fell by the hand of Hectôr.

  [707] Nothing occurs in Homer respecting the sacrifice of
  Iphigeneia (see Schol. Ven. ad Il. ix. 145).

Meanwhile the Trojans had assembled a large body of allies from various
parts of Asia Minor and Thrace: Dardanians under Æneas, Lykians under
Sarpedôn, Mysians, Karians, Mæonians, Alizonians,[708] Phrygians,
Thracians, and Pæonians.[709] But vain was the attempt to oppose
the landing of the Greeks: the Trojans were routed, and even the
invulnerable Cycnus,[710] son of Poseidôn, one of the great bulwarks of
the defence, was slain by Achilles. Having driven the Trojans within
their walls, Achilles attacked and stormed Lyrnêssus, Pêdasus, Lesbos
and other places in the neighborhood, twelve towns on the sea-coast
and eleven in the interior; he drove off the oxen of Æneas and pursued
the hero himself, who narrowly escaped with his life: he surprised
and killed the youthful Trôilus, son of Priam, and captured several
of the other sons, whom he sold as prisoners into the islands of the
Ægean.[711] He acquired as his captive the fair Brisêis, while Chrysêis
was awarded to Agamemnôn: he was moreover eager to see the divine
Helen, the prize and stimulus of this memorable struggle; and Aphroditê
and Thetis contrived to bring about an interview between them.[712]

  [708] No portion of the Homeric Catalogue gave more trouble
  to Dêmêtrius of Skêpsis and the other expositors than these
  Alizonians (Strabo, xii. p. 549; xiii. p. 603): a fictitious
  place called Alizonium, in the region of Ida, was got up to meet
  the difficulty (εἶτ᾽ Ἀλιζώνιον, τοῦτ᾽ ἤδη ~πεπλασμένον~ πρὸς τὴν
  τῶν Ἀλιζώνων ὑπόθεσιν, etc., Strabo, _l. c._).

  [709] See the Catalogue of the Trojans (Iliad, ii. 815-877).

  [710] Cycnus was said by later writers to be king of Kolônæ
  in the Troad (Strabo, xiii. p. 589-603; Aristotel. Rhetoric.
  ii. 23). Æschylus introduced upon the Attic stage both Cycnus
  and Memnôn in terrific equipments (Aristophan. Ran. 957. Οὐδ᾽
  ἐξέπληττον αὐτοὺς Κύκνους ἄγων καὶ Μέμνονας κωδωνοφαλαροπώλους).
  Compare Welcker, Æschyl. Trilogie, p. 433.

  [711] Iliad, xxiv. 752; Argument of the Cypria, pp. 11, 12,
  Düntzer. These desultory exploits of Achilles furnished much
  interesting romance to the later Greek poets (see Parthênius,
  Narrat. 21). See the neat summary of the principal events of the
  war in Quintus Smyrn. xiv. 125-140; Dio Chrysost. Or. xi. p.
  338-342.

  Trôilus is only once named in the Iliad (xxiv. 253); he was
  mentioned also in the Cypria; but his youth, beauty, and untimely
  end made him an object of great interest with the subsequent
  poets. Sophoklês had a tragedy called _Trôilus_ (Welcker,
  Griechisch. Tragöd. i. p. 124); Τὸν ἀνδρόπαιδα δεσπότην ἀπώλεσα,
  one of the Fragm. Even earlier than Sophoklês, his beauty was
  celebrated by the tragedian Phrynichus (Athenæ, xiii. p. 564;
  Virgil, Æneid, i. 474; Lycophrôn, 307).

  [712] Argument. Cypr. p. 11, Düntz. Καὶ μετὰ ταῦτα Ἀχιλλεὺς
  Ἑλένην ἐπιθυμεῖ θεάσασθαι, καὶ συνήγαγον αὐτοὺς εἰς τὸ αὐτὸ
  Ἀφροδίτη καὶ Θέτις. A scene which would have been highly
  interesting in the hands of Homer.

At this period of the war the Grecian army was deprived of Palamêdês,
one of its ablest chiefs. Odysseus had not forgiven the artifice
by which Palamêdês had detected his simulated insanity, nor was he
without jealousy of a rival clever and cunning in a degree equal, if
not superior, to himself; one who had enriched the Greeks with the
invention of letters, of dice for amusement, of night-watches, as
well as with other useful suggestions. According to the old Cyprian
epic, Palamêdês was drowned while fishing, by the hands of Odysseus
and Diomêdês.[713] Neither in the Iliad nor the Odyssey does the name
of Palamêdês occur: the lofty position which Odysseus occupies in both
those poems—noticed with some degree of displeasure even by Pindar,
who described Palamêdês as the wiser man of the two—is sufficient to
explain the omission.[714] But in the more advanced period of the
Greek mind, when intellectual superiority came to acquire a higher
place in the public esteem as compared with military prowess, the
character of Palamêdês, combined with his unhappy fate, rendered him
one of the most interesting personages in the Trojan legend. Æschylus,
Sophoklês and Euripidês each consecrated to him a special tragedy; but
the mode of his death as described in the old epic was not suitable
to Athenian ideas, and accordingly he was represented as having been
falsely accused of treason by Odysseus, who caused gold to be buried
in his tent, and persuaded Agamemnôn and the Grecian chiefs that
Palamêdês had received it from the Trojans.[715] He thus forfeited his
life, a victim to the calumny of Odysseus and to the delusion of the
leading Greeks. In the last speech made by the philosopher Socratês to
his Athenian judges, he alludes with solemnity and fellow-feeling to
the unjust condemnation of Palamêdês, as analogous to that which he
himself was about to suffer, and his companions seem to have dwelt with
satisfaction on the comparison. Palamêdês passed for an instance of
the slanderous enmity and misfortune which so often wait upon superior
genius.[716]

  [713] Argum. Cypr. 1. 1.; Pausan. x. 31. The concluding portion
  of the Cypria seems to have passed under the title of Παλαμηδεία
  (see Fragm. 16 and 18. p. 15, Düntz.; Welcker, Der Episch. Cycl.
  p. 459; Eustath. ad Hom. Odyss. i. 107).

  The allusion of Quintus Smyrnæus (v. 197) seems rather to point
  to the story in the Cypria, which Strabo (viii. p. 368) appears
  not to have read.

  [714] Pindar, Nem. vii. 21; Aristidês, Orat. 46. p. 260.

  [715] See the Fragments of the three tragedians,
  Παλαμήδης—Aristeidês, Or. xlvi. p. 260; Philostrat. Heroic. x.;
  Hygin. fab. 95-105. Discourses for and against Palamêdês, one
  by Alkidamas, and one under the name of Gorgias, are printed in
  Reiske’s Orr. Græc. t. viii. pp. 64, 102; Virgil, Æneid, ii. 82,
  with the ample commentary of Servius—Polyæn. Proœ. p. 6.

  Welcker (Griechisch. Tragöd. v. i. p. 130, vol. ii. p. 500)
  has evolved with ingenuity the remaining fragments of the lost
  tragedies.

  According to Diktys, Odysseus and Diomêdês prevail upon Palamêdês
  to be let down into a deep well, and then cast stones upon him
  (ii. 15).

  Xenophôn (De Venatione, c. 1) evidently recognizes the story
  in the Cypria, that Odysseus and Diomêdês caused the death of
  Palamêdês; but he _cannot_ believe that two such exemplary men
  were really guilty of so iniquitous an act—κακοὶ δὲ ἔπραξαν τὸ
  ἔργον.

  One of the eminences near Napoli still bears the name of
  _Palamidhi_.

  [716] Plato, Apolog. Socr. c. 32; Xenoph. Apol. Socr. 26; Memor.
  iv. 2, 33; Liban. pro Socr. p. 242, ed. Morell.; Lucian, Dial.
  Mort 20.

In these expeditions the Grecian army consumed nine years, during which
the subdued Trojans dared not give battle without their walls for fear
of Achilles. Ten years was the fixed epical duration of the siege of
Troy, just as five years was the duration of the siege of Kamikus by
the Krêtan armament which came to avenge the death of Minôs:[717] ten
years of preparation, ten years of siege, and ten years of wandering
for Odysseus, were periods suited to the rough chronological dashes of
the ancient epic, and suggesting no doubts nor difficulties with the
original hearers. But it was otherwise when the same events came to be
contemplated by the historicizing Greeks, who could not be satisfied
without either finding or inventing satisfactory bonds of coherence
between the separate events. Thucydidês tells us that the Greeks
were less numerous than the poets have represented, and that being
moreover very poor, they were unable to procure adequate and constant
provisions: hence they were compelled to disperse their army, and to
employ a part of it in cultivating the Chersonese,—a part in marauding
expeditions over the neighborhood. Could the whole army have been
employed against Troy at once (he says), the siege would have been much
more speedily and easily concluded.[718] If the great historian could
permit himself thus to amend the legend in so many points, we might
have imagined that the simpler course would have been to include the
duration of the siege among the list of poetical exaggerations, and to
affirm that the real siege had lasted only one year instead of ten.
But it seems that the ten years’ duration was so capital a feature in
the ancient tale, that no critic ventured to meddle with it.

  [717] Herodot. vii. 170. Ten years is a proper mythical period
  for a great war to last: the war between the Olympic gods and the
  Titan gods lasts ten years (Hesiod, Theogon. 636). Compare δεκάτῳ
  ἐνιαυτῷ (Hom. Odyss. xvi. 17).

  [718] Thucyd. i. 11.

A period of comparative intermission however was now at hand for the
Trojans. The gods brought about the memorable fit of anger of Achilles,
under the influence of which he refused to put on his armor, and kept
his Myrmidons in camp. According to the Cypria, this was the behest of
Zeus, who had compassion on the Trojans: according to the Iliad, Apollo
was the originating cause,[719] from anxiety to avenge the injury which
his priest Chrysês had endured from Agamemnôn. For a considerable
time, the combats of the Greeks against Troy were conducted without
their best warrior, and severe indeed was the humiliation which they
underwent in consequence. How the remaining Grecian chiefs vainly
strove to make amends for his absence—how Hectôr and the Trojans
defeated and drove them to their ships—how the actual blaze of the
destroying flame, applied by Hectôr to the ship of Protesilaus, roused
up the anxious and sympathizing Patroclus, and extorted a reluctant
consent from Achilles, to allow his friend and his followers to go
forth and avert the last extremity of ruin—how Achilles, when Patroclus
had been killed by Hectôr, forgetting his anger in grief for the death
of his friend, reëntered the fight, drove the Trojans within their
walls with immense slaughter, and satiated his revenge both upon the
living and the dead Hectôr—all these events have been chronicled,
together with those divine dispensations on which most of them are made
to depend, in the immortal verse of the Iliad.

  [719] Homer, Iliad, i. 21.

Homer breaks off with the burial of Hectôr, whose body has just been
ransomed by the disconsolate Priam; while the lost poem of Arktinus,
entitled the Æthiopis, so far as we can judge from the argument still
remaining of it, handled only the subsequent events of the siege. The
poem of Quintus Smyrnæus, composed about the fourth century of the
Christian æra, seems in its first books to coincide with the Æthiopis,
in the subsequent books partly with the Ilias Minor of Leschês.[720]

  [720] Tychsen, Commentat. de Quinto Smyrnæo, § iii. c. 5-7. The
  Ἰλίου Πέρσις was treated both by Arktinus and by Leschês: with
  the latter it formed a part of the Ilias Minor.

The Trojans, dismayed by the death of Hectôr, were again animated
with hope by the appearance of the warlike and beautiful queen of the
Amazons, Penthesileia, daughter of Arês, hitherto invincible in the
field, who came to their assistance from Thrace at the head of a band
of her countrywomen. She again led the besieged without the walls
to encounter the Greeks in the open field; and under her auspices
the latter were at first driven back, until she too was slain by the
invincible arm of Achilles. The victor, on taking off the helmet of
his fair enemy as she lay on the ground, was profoundly affected and
captivated by her charms, for which he was scornfully taunted by
Thersitês: exasperated by this rash insult, he killed Thersitês on
the spot with a blow of his fist. A violent dispute among the Grecian
chiefs was the result, for Diomêdês, the kinsman of Thersitês, warmly
resented the proceeding; and Achilles was obliged to go to Lesbus,
where he was purified from the act of homicide by Odysseus.[721]

  [721] Argument of the Æthiopis, p. 16, Düntzer; Quint. Smyrn.
  lib. i.; Diktys Cret. iv. 2-3.

  In the Philoktêtês, of Sophoklês, Thersitês survives Achilles
  (Soph. Phil. 358-445).

Next arrived Memnôn, son of Tithônus and Eôs, the most stately of
living men, with a powerful band of black Æthiopians, to the assistance
of Troy. Sallying forth against the Greeks, he made great havoc among
them: the brave and popular Antilochus perished by his hand, a victim
to filial devotion in defence of Nestôr.[722] Achilles at length
attacked him, and for a long time the combat was doubtful between
them: the prowess of Achilles and the supplication of Thetis with Zeus
finally prevailed; whilst Eôs obtained for her vanquished son the
consoling gift of immortality. His tomb, however,[723] was shown near
the Propontis, within a few miles of the mouth of the river Æsêpus,
and was visited annually by the birds called Memnonides, who swept it
and bedewed it with water from the stream. So the traveller Pausanias
was told, even in the second century after the Christian æra, by the
Hellespontine Greeks.

  [722] Odyss. xi. 522. Κεῖνον δὴ κάλλιστον ἴδον, μετὰ Μέμνονα
  δῖον: see also Odyss. iv. 187; Pindar, Pyth. vi. 31. Æschylus
  (ap. Strabo. xv. p. 728) conceives Memnôn as a Persian starting
  from Susa.

  Ktêsias gave in his history full details respecting the
  expedition of Memnôn, sent by the king of Assyria to the relief
  of his dependent, Priam of Troy; all this was said to be recorded
  in the royal archives. The Egyptians affirmed that Memnôn had
  come from Egypt (Diodôr. ii. 22; compare iv. 77): the two stories
  are blended together in Pausanias, x. 31, 2. The Phrygians
  pointed out the road along which he had marched.

  [723] Argum. Æth. _ut sup._; Quint. Smyrn. ii. 396-550; Pausan.
  x. 31, 1. Pindar, in praising Achilles, dwells much on his
  triumphs over Hectôr, Têlephus, Memnôn, and Cycnus, but never
  notices Penthesileia (Olymp. ii. 90; Nem. iii. 60; vi. 52. Isthm.
  v. 43).

  Æschylus, in the Ψυχοστασία, introduced Thetis and Eôs, each in
  an attitude of supplication for her son, and Zeus weighing in his
  golden scales the souls of Achilles and Memnôn (Schol. Ven. ad
  Iliad, viii. 70: Pollux, iv. 130; Plutarch, De Audiend. Poet. p.
  17). In the combat between Achilles and Memnôn, represented on
  the chest of Kypselus at Olympia, Thetis and Eôs were given each
  as aiding her son (Pausan. v. 19, 1).

But the fate of Achilles himself was now at hand. After routing the
Trojans and chasing them into the town, he was slain near the Skæan
gate by an arrow from the quiver of Paris, directed under the unerring
auspices of Apollo.[724] The greatest efforts were made by the Trojans
to possess themselves of the body, which was however rescued and borne
off to the Grecian camp by the valor of Ajax and Odysseus. Bitter
was the grief of Thetis for the loss of her son: she came into the
camp with the Muses and the Nêreids to mourn over him; and when a
magnificent funeral-pile had been prepared by the Greeks to burn him
with every mark of honor, she stole away the body and conveyed it to
a renewed and immortal life in the island of Leukê in the Euxine Sea.
According to some accounts he was there blest with the nuptials and
company of Helen.[725]

  [724] Iliad, xxii. 360; Sophokl. Philokt. 334; Virgil, Æneid, vi.
  56.

  [725] Argum. Æthiop. _ut sup._; Quint. Smyrn. 151-583; Homer,
  Odyss. v. 310; Ovid, Metam. xiii. 284; Eurip. Androm. 1262;
  Pausan. iii. 19, 13. According to Diktys (iv. 11), Paris and
  Deiphobus entrap Achilles by the promise of an interview with
  Polyxena and kill him.

  A minute and curious description of the island Leukê, or Ἀχιλλέως
  νῆσος, is given in Arrian (Periplus, Pont. Euxin. p. 21; ap.
  Geogr. Min. t. 1).

  The heroic or divine empire of Achilles in Scythia was recognized
  by Alkæus the poet (Alkæi Fragm. Schneidew. Fr. 46), Ἀχιλλεῦ, ὃς
  γᾶς Σκυθικᾶς μέδεις. Eustathius (ad Dionys. Periêgêt. 307) gives
  the story of his having followed Iphigeneia thither: compare
  Antonin. Liberal. 27.

  Ibykus represented Achilles as having espoused Mêdea in the
  Elysian Field (Idyk. Fragm. 18. Schneidewin). Simonidês followed
  this story (ap. Schol. Apoll. Rhod. iv. 815).

Thetis celebrated splendid funeral games in honor of her son, and
offered the unrivalled panoply, which Hêphæstos had forged and wrought
for him, as a prize to the most distinguished warrior in the Grecian
army. Odysseus and Ajax became rivals for the distinction, when
Athênê, together with some Trojan prisoners, who were asked from which
of the two their country had sustained greatest injury, decided in
favor of the former. The gallant Ajax lost his senses with grief and
humiliation: in a fit of phrenzy he slew some sheep, mistaking them for
the men who had wronged him, and then fell upon his own sword.[726]

  [726] Argument of Æthiopis and Ilias Minor, and Fragm. 2 of the
  latter, pp. 17, 18, Düntz.; Quint. Smyrn. v. 120-482; Hom. Odyss.
  xi. 550; Pindar, Nem. vii. 26. The Ajax of Sophoklês, and the
  contending speeches between Ajax and Ulysses in the beginning of
  the thirteenth book of Ovid’s Metamorphoses, are too well known
  to need special reference.

  The suicide of Ajax seems to have been described in detail in
  the Æthiopis: compare Pindar. Isthm. iii. 51, and the Scholia
  _ad loc._, which show the attention paid by Pindar to the minute
  circumstances of the old epic. See Fragm. 2 of the Ἰλίου Πέρσις
  of Arktinus, in Düntz. p. 22, which would seem more properly to
  belong to the Æthiopis. Diktys relates the suicide of Ajax, as a
  consequence of his unsuccessful competition with Odysseus, not
  about the arms of Achilles, but about the Palladium, after the
  taking of the city (v. 14).

  There were, however, many different accounts of the manner in
  which Ajax had died, some of which are enumerated in the argument
  to the drama of Sophoklês. Ajax is never wounded in the Iliad:
  Æschylus made him invulnerable except under the armpits (see
  Schol. ad Sophok. Ajac. 833); the Trojans pelted him with mud—εἴ
  πως βαρηθείῃ ὑπὸ τοῦ πήλου (Schol. Iliad. xiv. 404).

Odysseus now learnt from Helenus son of Priam, whom he had captured
in an ambuscade,[727] that Troy could not be taken unless both
Philoktêtês, and Neoptolemus, son of Achilles, could be prevailed
upon to join the besiegers. The former, having been stung in the
foot by a serpent, and becoming insupportable to the Greeks from the
stench of his wound, had been left at Lemnus in the commencement
of the expedition, and had spent ten years[728] in misery on that
desolate island; but he still possessed the peerless bow and arrows
of Hêraklês, which were said to be essential to the capture of Troy.
Diomêdês fetched Philoktêtês from Lemnus to the Grecian camp, where
he was healed by the skill of Machaôn,[729] and took an active part
against the Trojans—engaging in single combat with Paris, and killing
him with one of the Hêrakleian arrows. The Trojans were allowed to
carry away for burial the body of this prince, the fatal cause of
all their sufferings; but not until it had been mangled by the hand
of Menelaus.[730] Odysseus went to the island of Skyrus to invite
Neoptolemus to the army. The untried but impetuous youth gladly obeyed
the call, and received from Odysseus his father’s armor, while on the
other hand, Eurypylus, son of Têlephus, came from Mysia as auxiliary
to the Trojans and rendered to them valuable service—turning the tide
of fortune for a time against the Greeks, and killing some of their
bravest chiefs, amongst whom was numbered Peneleôs, and the unrivalled
leech Machaôn.[731] The exploits of Neoptolemus were numerous, worthy
of the glory of his race and the renown of his father. He encountered
and slew Eurypylus, together with numbers of the Mysian warriors: he
routed the Trojans and drove them within their walls, from whence they
never again emerged to give battle: nor was he less distinguished for
his good sense and persuasive diction, than for forward energy in the
field.[732]

  [727] Soph. Philokt. 604.

  [728] Soph. Philokt. 703. Ὦ μελέα ψυχὰ, Ὃς μηδ᾽ οἰνοχύτου πόματος
  Ἥσθη δεκετῆ χρόνον, etc.

  In the narrative of Diktys (ii. 47), Philoktêtês returns from
  Lemnus to Troy much earlier in the war before the death of
  Achilles, and without any assigned cause.

  [729] According to Sophoklês, Hêraklês sends Asklêpius to Troy to
  heal Philoktêtês (Soph. Philokt. 1415).

  The subject of Philoktêtês formed the subject of a tragedy both
  by Æschylus and by Euripidês (both lost) as well as by Sophoklês.

  [730] Argument. Iliad. Minor. Düntz. _l. c._ Καὶ τὸν νεκρὸν
  ὑπὸ Μενελάου καταικισθέντα ἀνελόμενοι θάπτουσιν οἱ Τρῶες. See
  Quint. Smyrn. x. 240: he differs here in many respects from the
  arguments of the old poems as given by Proclus, both as to the
  incidents and as to their order in time (Diktys, iv. 20). The
  wounded Paris flees to Œnônê, whom he had deserted in order
  to follow Helen, and entreats her to cure him by her skill in
  simples: she refuses, and permits him to die; she is afterwards
  stung with remorse, and hangs herself (Quint. Smyrn. x. 285-331;
  Apollodôr. iii. 12, 6; Conôn. Narrat. 23; see Bachet de Meziriac,
  Comment. sur les Epîtres d’Ovide, t. i. p. 456). The story of
  Œnônê is as old as Hellanikus and Kephalôn of Gergis (see Hellan.
  Fragm. 126, Didot).

  [731] To mark the way in which these legendary events pervaded
  and became embodied in the local worship, I may mention the
  received practice in the great temple of Asklêpius (father of
  Machaôn) at Pergamus, even in the time of Pausanias. Têlephus,
  father of Eurypylus, was the local hero and mythical king of
  Teuthrania, in which Pergamus was situated. In the hymns there
  sung, the poem and the invocation were addressed to Têlephus; but
  nothing was said in them about Eurypylus, nor was it permitted
  even to mention his name in the temple,—“they knew him to be
  the slayer of Machaôn:” ἄρχονται μὲν ἀπὸ Τηλέφου τῶν ὕμνων,
  προσᾴδουσι δὲ οὐδὲν ἐς τὸν Εὐρύπυλον, οὐδὲ ἀρχὴν ἐν τῷ ναῷ
  θέλουσιν ὀνομάζειν αὐτὸν, οἷα ἐπιστάμενοι φονέα ὄντα Μαχάονος
  (Pausan. iii. 26, 7).

  The combination of these qualities in other Homeric chiefs is
  noted in a subsequent chapter of his work, ch. xx. vol. ii.

  [732] Argument. Iliad. Minor. p. 17, Düntzer. Homer, Odyss. xi.
  510-520. Pausan. iii. 26, 7. Quint. Smyrn. vii. 553; viii. 201.

Troy however was still impregnable so long as the Palladium, a statue
given by Zeus himself to Dardanus, remained in the citadel; and great
care had been taken by the Trojans not only to conceal this valuable
present, but to construct other statues so like it as to mislead any
intruding robber. Nevertheless the enterprising Odysseus, having
disguised his person with miserable clothing and self-inflicted
injuries, found means to penetrate into the city and to convey the
Palladium by stealth away: Helen alone recognized him; but she was now
anxious to return to Greece, and even assisted Odysseus in concerting
means for the capture of the town.[733]

  [733] Argument. Iliad. Minor, p. 18, Düntz.; _Arktinus_ ap.
  Dionys. Hal. i. 69; Homer, Odyss. iv. 246; Quint. Smyrn. x. 354:
  Virgil, Æneid, ii. 164, and the 9th Excursus of Heyne on that
  book.

  Compare with this legend about the Palladium, the Roman legend
  respecting the Ancylia (Ovid, Fasti, III. 381).

To accomplish this object, one final stratagem was resorted to. By
the hands of Epeius of Panopeus, and at the suggestion of Athênê, a
capacious hollow wooden horse was constructed, capable of containing
one hundred men: the _élite_ of the Grecian heroes, Neoptolemus,
Odysseus, Menelaus and others, concealed themselves in the inside
of it, and the entire Grecian army sailed away to Tenedos, burning
their tents and pretending to have abandoned the siege. The Trojans,
overjoyed to find themselves free, issued from the city and
contemplated with astonishment the fabric which their enemies had left
behind: they long doubted what should be done with it; and the anxious
heroes from within heard the surrounding consultations, as well as
the voice of Helen when she pronounced their names and counterfeited
the accents of their wives.[734] Many of the Trojans were anxious to
dedicate it to the gods in the city as a token of gratitude for their
deliverance; but the more cautious spirits inculcated distrust of an
enemy’s legacy; and Laocoôn, the priest of Poseidôn, manifested his
aversion by striking the side of the horse with his spear. The sound
revealed that the horse was hollow, but the Trojans heeded not this
warning of possible fraud; and the unfortunate Laocoôn, a victim to
his own sagacity and patriotism, miserably perished before the eyes of
his countrymen, together with one of his sons,—two serpents being sent
expressly by the gods out of the sea to destroy him. By this terrific
spectacle, together with the perfidious counsels of Sinon, a traitor
whom the Greeks had left behind for the special purpose of giving false
information, the Trojans were induced to make a breach in their own
walls, and to drag the fatal fabric with triumph and exultation into
their city.[735]

  [734] Odyss. iv. 275; Virgil, Æneid, ii. 14; Heyne, Excurs. 3. ad
  Æneid. ii. Stesichorus, in his Ἰλίου Πέρσις, gave the number of
  heroes in the wooden horse as one hundred (Stesichor. Fragm. 26,
  ed. Kleine; compare Athenæ. xiii. p. 610).

  [735] Odyss. viii. 492; xi. 522. Argument of the Ἰλίου Πέρσις
  of Arktinus, p. 21. Düntz. Hydin. f. 108-135. Bacchylidês and
  Euphorion ap. Servium ad Virgil. Æneid. ii. 201.

  Both Sinon and Laocoôn came originally from the old epic poem of
  Arktinus, though Virgil may perhaps have immediately borrowed
  both them, and other matters in his second book, from a poem
  passing under the name of Pisander (see Macrob. Satur. v. 2;
  Heyne, Excurs. 1. ad Æn. ii.; Welcker, Der Episch. Kyklus, v.
  97). We cannot give credit either to Arktinus or Pisander for the
  masterly specimen of oratory which is put into the mouth of Sinon
  in the Æneid.

  In Quintus Smyrnæus (xii. 366), the Trojans torture and mutilate
  Sinon to extort from him the truth: his endurance, sustained
  by the inspiration of Hêrê, is proof against the extremity of
  suffering, and he adheres to his false tale. This is probably an
  incident of the old epic, though the delicate taste of Virgil,
  and his sympathy with the Trojans, has induced him to omit it.
  Euphorion ascribed the proceedings of Sinon to Odysseus: he also
  gave a different cause for the death of Laocoôn (Fr. 33-36. p.
  55, ed. Düntz., in the Fragments of Epic Poets after Alexander
  the Great). Sinon is ἐταῖρος Ὀδυσσέως in Pausan. x. 27, 1.

The destruction of Troy, according to the decree of the gods, was now
irrevocably sealed. While the Trojans indulged in a night of riotous
festivity, Sinon kindled the fire-signal to the Greeks at Tenedos,
loosening the bolts of the wooden horse, from out of which the enclosed
heroes descended. The city, assailed both from within and from without,
was thoroughly sacked and destroyed, with the slaughter or captivity of
the larger portion of its heroes as well as its people. The venerable
Priam perished by the hand of Neoptolemus, having in vain sought
shelter at the domestic altar of Zeus Herkeios; but his son Deiphobus,
who since the death of Paris had become the husband of Helen, defended
his house desperately against Odysseus and Menelaus, and sold his life
dearly. After he was slain, his body was fearfully mutilated by the
latter.[736]

  [736] Odyss. viii. 515; Argument of Arktinas, _ut sup._; Euripid.
  Hecub. 903; Virg. Æn. vi. 497; Quint. Smyrn. xiii. 35-229;
  Leschês ap. Pausan. x. 27, 2; Diktys, v. 12. Ibykus and Simonidês
  also represented Deiphobus as the ἀντεράστης Ἑλένης (Schol. Hom.
  Iliad. xiii. 517).

  The night-battle in the interior of Troy was described with all
  its fearful details both by Leschês and Arktinus: the Ἰλίου
  Πέρσις of the latter seems to have been a separate poem, that of
  the former constituted a portion of the Ilias Minor (see Welcker,
  Der Epische Kyklus, p. 215): the Ἰλίου Πέρσις by the lyric poets
  Sakadas and Stesichorus probably added many new incidents.
  Polygnôtus had painted a succession of the various calamitous
  scenes, drawn from the poem of Leschês, on the walls of the
  leschê at Delphi, with the name written over each figure (Pausan.
  x. 25-26).

  Hellanikus fixed the precise day of the month on which the
  capture took place (Hellan. Fr. 143-144), the twelfth day of
  Thargeliôn.

Thus was Troy utterly destroyed—the city, the altars and temples,[737]
and the population. Æneas and Antenôr were permitted to escape, with
their families, having been always more favorably regarded by the
Greeks than the remaining Trojans. According to one version of the
story, they had betrayed the city to the Greeks: a panther’s skin
had been hung over the door of Antenôr’s house as a signal for the
victorious besiegers to spare it in the general plunder.[738] In the
distribution of the principal captives, Astyanax, the infant son of
Hectôr, was cast from the top of the wall and killed, by Odysseus or
Neoptolemus: Polyxena, the daughter of Priam, was immolated on the tomb
of Achilles, in compliance with a requisition made by the shade of
the deceased hero to his countrymen;[739] while her sister Kassandra
was presented as a prize to Agamemnôn. She had sought sanctuary at
the altar of Athênê, where Ajax, the son of Oileus, making a guilty
attempt to seize her, had drawn both upon himself and upon the army the
serious wrath of the goddess, insomuch that the Greeks could hardly be
restrained from stoning him to death.[740] Andromachê and Helenus were
both given to Neoptolemus, who, according to the Ilias Minor, carried
away also Æneas as his captive.[741]

  [737] Æschyl. Agamemn. 527.—

    Βωμοὶ δ᾽ ἄϊστοι καὶ θεῶν ἱδρύματα,
    Καὶ σπέρμα πάσης ἐξαπόλλυται χθονός.

  [738] This symbol of treachery also figured in the picture of
  Polygnôtus. A different story appears in Schol. Iliad. iii. 206.

  [739] Euripid. Hecub. 38-114, and Troad. 716; Leschês ap. Pausan.
  x. 25, 9; Virgil, Æneid, iii. 322, and Servius _ad loc._

  A romantic tale is found in Diktys respecting the passion of
  Achilles for Polyxena (iii. 2).

  [740] Odyss. xi. 422. Arktinus, Argum. p. 21, Düntz. Theognis,
  1232. Pausan. i. 15, 2; x. 26, 3; 31, 1. As an expiation of
  this sin of their national hero, the Lokrians sent to Ilium
  periodically some of their maidens, to do menial service in the
  temple of Athênê (Plutarch. Ser. Numin. Vindict. p. 557, with the
  citation from Euphorion or Kallimachus, Düntzer, Epicc. Vet. p.
  118).

  [741] Leschês, Fr. 7, Düntz.; ap. Schol. Lycophr. 1263. Compare
  Schol. ad. 1232, for the respectful recollection of Andromachê,
  among the traditions of the Molossian kings, as their heroic
  mother, and Strabo, xiii. p. 594.

Helen gladly resumed her union with Menelaus: she accompanied him
back to Sparta, and lived with him there many years in comfort and
dignity,[742] passing afterwards to a happy immortality in the Elysian
fields. She was worshipped as a goddess with her brothers the Dioskuri
and her husband, having her temple, statue and altar at Therapnæ and
elsewhere, and various examples of her miraculous interventions were
cited among the Greeks.[743] The lyric poet Stesichorus had ventured
to denounce her, conjointly with her sister Klytæmnêstra, in a tone
of rude and plain-spoken severity, resembling that of Euripidês
and Lycophrôn afterwards, but strikingly opposite to the delicacy
and respect with which she is always handled by Homer, who never
admits reproaches against her except from her own lips.[744] He was
smitten with blindness, and made sensible of his impiety; but having
repented and composed a special poem formally retracting the calumny,
was permitted to recover his sight. In his poem of recantation (the
famous palinode now unfortunately lost) he pointedly contradicted the
Homeric narrative, affirming that Helen had never been to Troy at
all, and that the Trojans had carried thither nothing but her image
or _eidôlon_.[745] It is, probably, to the excited religious feelings
of Stesichorus that we owe the first idea of this glaring deviation
from the old legend, which could never have been recommended by any
considerations of poetical interest.

  [742] Such is the story of the old epic (see Odyss. iv. 260,
  and the fourth book generally; Argument of Ilias Minor, p. 20.
  Düntz.). Polygnôtus, in the paintings above alluded to, followed
  the same tale (Pausan. x. 25, 3).

  The anger of the Greeks against Helen, and the statement that
  Menelaus after the capture of Troy approached her with revengeful
  purposes, but was so mollified by her surpassing beauty as
  to cast away his uplifted sword, belongs to the age of the
  tragedians (Æschyl. Agamem. 685-1455: Eurip. Androm. 600-629;
  Helen. 75-120; Troad. 890-1057; compare also the fine lines in
  the Æneid, ii. 567-588).

  [743] See the description in Herodot. vi. 61, of the prayers
  offered to her, and of the miracle which she wrought, to remove
  the repulsive ugliness of a little Spartan girl of high family.
  Compare also Pindar, Olymp. iii. 2, and the Scholia at the
  beginning of the ode; Eurip. Helen. 1662, and Orest. 1652-1706;
  Isokrat. Encom. Helen. ii. p. 368, Auger; Dio Chrysost. Or. xi.
  p. 311. θεὸς ἐνομίσθη παρὰ τοῖς Ἕλλησι; Theodectês ap. Aristot.
  Pol. i. 2, 19. Θείων ἀπ᾽ ἀμφοῖν ἔκγονον ῥιζωμάτων.

  [744] Euripid. Troad. 982 _seq._; Lycophrôn ap. Steph. Byz. v.
  Αἰγύς; Stesichorus ap. Schol. Eurip. Orest. 239; Fragm. 9 and 10
  of the Ἰλίου Πέρσις, Schneidewin:—

    Οὕνεκα Τυνδάρεως ῥέζων ἁπᾶσι θεοῖς μιᾶς λαθετ᾽ ἠπιοδώρου
    Κύπριδος· κείνα δὲ Τυνδάρεω κούραισι χολωσαμένα
    Διγάμους τριγάμους τίθησι
    Καὶ λιπεσάνορας ...

  Further

    ... Ἑλένη ἑκοῦσ᾽ ἄπηρε, etc.

  He had probably contrasted her with other females carried away by
  force.

  Stesichorus also affirmed that Iphigeneia was the daughter
  of Helen, by Thêseus, born at Argos before her marriage with
  Menelaus and made over to Klytæmnêstra: this tale was perpetuated
  by the temple of Eileithyia at Argos, which the Argeians affirmed
  to have been erected by Helen (Pausan. ii. 22, 7). The ages
  ascribed by Hellanikus and other logographers (Hellan. Fr. 74)
  to Thêseus and Helen—he fifty years of age and she a child of
  seven—when he carried her off to Aphidnæ, can never have been the
  original form of any poetical legend: these ages were probably
  imagined in order to make the mythical chronology run smoothly;
  for Thêseus belongs to the generation before the Trojan war. But
  we ought always to recollect that Helen never grows old (τὴν γὰρ
  φάτις ἔμμεν᾽ ἀγήρω—Quint. Smyrn. x. 312), and that her chronology
  consists only with an immortal being. Servius observes (ad Æneid.
  ii. 601)—“Helenam _immortalem_ fuisse indicat tempus. Nam constat
  fratres ejus cum Argonautis fuisse. Argonautarum filii cum
  Thebanis (Thebano Eteoclis et Polynicis bello) dimicaverunt. Item
  illorum filii contra Trojam bella gesserunt. Ergo, si immortalis
  Helena non fuisset, tot sine dubio seculis durare non posset.” So
  Xenophon, after enumerating many heroes of different ages, all
  pupils of Cheirôn, says that the life of Cheirôn suffices for
  all, he being brother of Zeus (De Venatione, c. 1).

  The daughters of Tyndareus are Klytæmnêstra, Helen, and Timandra,
  all open to the charge advanced by Stesichorus: see about
  Timandra, wife of the Tegeate Echemus, the new fragment of the
  Hesiodic Catalogue, recently restored by Geel (Göttling, Pref.
  Hesiod. p. lxi.).

  It is curious to read, in Bayle’s article _Hélène_, his critical
  discussion of the adventures ascribed to her—as if they were
  genuine matter of history, more or less correctly reported.

  [745] Plato, Republic. ix. p. 587. c. 10. ὥσπερ τὸ τῆς Ἑλένης
  εἴδωλον Στησίχορός φησι περιμάχητον γενέσθαι ἐν Τροίῃ, ἀγνοίᾳ τοῦ
  ἀληθοῦς.

  Isokrat. Encom. Helen. t. ii. p. 370, Auger; Plato, Phædr. c. 44.
  p. 243-244; Max. Tyr. Diss. xi. p. 320, Davis; Conôn, Narr. 18;
  Dio Chrysost. Or. xi. p. 323. Τὸν μὲν Στησίχορον ἐν τῇ ὕστερον
  ὠδῇ λέγειν, ὡς ~τὸ παράπαν οὐδὲ πλεύσειεν ἡ Ἑλένη οὐδάμοσε~.
  Horace, Od. i. 17, Epod. xvii. 42.—

    “Infamis Helenæ Castor offensus vice,
    Fraterque magni Castoris, victi prece,
    Adempta vati reddidere lumina.”

  Pausan. iii. 19, 5. Virgil, surveying the war from the point
  of view of the Trojans, had no motive to look upon Helen with
  particular tenderness: Deiphobus imputes to her the basest
  treachery (Æneid, vi. 511. “_scelus exitiale Lacænæ_;” compare
  ii. 567).

Other versions were afterwards started, forming a sort of compromise
between Homer and Stesichorus, admitting that Helen had never really
been at Troy, without altogether denying her elopement. Such is the
story of her having been detained in Egypt during the whole term of the
siege. Paris, on his departure from Sparta, had been driven thither by
storms, and the Egyptian king Prôteus, hearing of the grievous wrong
which he had committed towards Menelaus, had sent him away from the
country with severe menaces, detaining Helen until her lawful husband
should come to seek her. When the Greeks reclaimed Helen from Troy, the
Trojans assured them solemnly, that she neither was, nor ever had been,
in the town; but the Greeks, treating this allegation as fraudulent,
prosecuted the siege until their ultimate success confirmed the
correctness of the statement, nor did Menelaus recover Helen until, on
his return from Troy, he visited Egypt.[746] Such was the story told by
the Egyptian priests to Herodotus, and it appeared satisfactory to his
historicizing mind. “For if Helen had really been at Troy (he argues)
she would certainly have been given up, even had she been mistress of
Priam himself instead of Paris: the Trojan king, with all his family
and all his subjects, would never knowingly have incurred utter and
irretrievable destruction for the purpose of retaining her: their
misfortune was, that while they did not possess, and therefore could
not restore her, they yet found it impossible to convince the Greeks
that such was the fact.” Assuming the historical character of the war
of Troy, the remark of Herodotus admits of no reply; nor can we greatly
wonder that he acquiesced in the tale of Helen’s Egyptian detention, as
a substitute for the “incredible insanity” which the genuine legend
imputes to Priam and the Trojans. Pausanias, upon the same ground
and by the same mode of reasoning, pronounces that the Trojan horse
must have been in point of fact a battering-engine, because to admit
the literal narrative would be to impute utter childishness to the
defenders of the city. And Mr. Payne Knight rejects Helen altogether as
the real cause of the Trojan war, though she may have been the pretext
of it; for he thinks that neither the Greeks nor the Trojans could have
been so mad and silly as to endure calamities of such magnitude “for
one little woman.”[747] Mr. Knight suggests various political causes as
substitutes; these might deserve consideration, either if any evidence
could be produced to countenance them, or if the subject on which they
are brought to bear could be shown to belong to the domain of history.

  [746] Herodot. ii. 120. οὐ γὰρ δὴ οὕτω γε φρενοβλαβὴς ἦν ὁ
  Πρίαμος, οὐδ᾽ οἱ ἄλλοι οἱ προσήκοντες αὐτῷ, etc. The passage is
  too long to cite, but is highly curious: not the least remarkable
  part is the religious coloring which he gives to the new version
  of the story which he is adopting,—“the Trojans, though they
  had not got Helen, yet could not persuade the Greeks that this
  was the fact; for it was the divine will that they should be
  destroyed root and branch, in order to make it plain to mankind
  that upon great crimes the gods inflict great punishments.”

  Dio Chrysostom (Or. xi. p. 333) reasons in the same way as
  Herodotus against the credibility of the received narrative. On
  the other hand, Isokratês, in extolling Helen, dwells on the
  calamities of the Trojan war as a test of the peerless value
  of the prize (Encom. Hel. p. 360, Aug.): in the view of Pindar
  (Olymp. xiii. 56), as well as in that of Hesiod (Opp. Di. 165),
  Helen is the one prize contended for.

  Euripidês, in his tragedy of Helen, recognizes the detention of
  Helen in Egypt and the presence of her εἴδωλον at Troy, but he
  follows Stesichorus in denying her elopement altogether,—Hermês
  had carried her to Egypt in a cloud (Helen. 35-45, 706): compare
  Von Hoff, De Mytho Helenæ Euripideæ, cap. 2. p. 35 (Leyden, 1843).

  [747] Pausan. i. 23, 8; Payne Knight, Prolegg. ad Homer. c. 53.
  Euphorion construed the wooden horse into a Grecian ship called
  Ἵππος, “_The Horse_” (Euphorion, Fragm. 34. ap. Düntzer, Fragm.
  Epicc. Græc. p. 55).

  See Thucyd. i. 12; vi. 2.

The return of the Grecian chiefs from Troy furnished matter to the
ancient epic hardly less copious than the siege itself, and the more
susceptible of indefinite diversity, inasmuch as those who had before
acted in concert were now dispersed and isolated. Moreover the stormy
voyages and compulsory wanderings of the heroes exactly fell in with
the common aspirations after an heroic founder, and enabled even the
most remote Hellenic settlers to connect the origin of their town with
this prominent event of their ante-historical and semi-divine world.
And an absence of ten years afforded room for the supposition of many
domestic changes in their native abode, and many family misfortunes and
misdeeds during the interval. One of these heroic “Returns,” that of
Odysseus, has been immortalized by the verse of Homer. The hero, after
a series of long-protracted suffering and expatriation, inflicted on
him by the anger of Poseidôn, at last reaches his native island, but
finds his wife beset, his youthful son insulted, and his substance
plundered, by a troop of insolent suitors; he is forced to appear as
a wretched beggar, and to endure in his own person their scornful
treatment; but finally, by the interference of Athênê coming in aid of
his own courage and stratagem, he is enabled to overwhelm his enemies,
to resume his family position, and to recover his property. The return
of several other Grecian chiefs was the subject of an epic poem by
Hagias, which is now lost, but of which a brief abstract or argument
still remains: there were in antiquity various other poems of similar
title and analogous matter.[748]

  [748] Suidas, v. Νόστος. Wüllner, De Cyclo Epico, p. 93. Also a
  poem Ἀτρειδῶν κάθοδος (Athenæ. vii. p. 281).

As usual with the ancient epic, the multiplied sufferings of this
back-voyage are traced to divine wrath, justly provoked by the sins of
the Greeks; who, in the fierce exultation of a victory purchased by so
many hardships, had neither respected nor even[749] spared the altars
of the gods in Troy; and Athênê, who had been their most zealous ally
during the siege, was so incensed by their final recklessness, more
especially by the outrage of Ajax, son of Oïleus, that she actively
harassed and embittered their return, in spite of every effort to
appease her. The chiefs began to quarrel among themselves; their
formal assembly became a scene of drunkenness; even Agamemnôn and
Menelaus lost their fraternal harmony, and each man acted on his own
separate resolution.[750] Nevertheless, according to the Odyssey,
Nestôr, Diomêdês, Neoptolemus, Idomeneus and Philoktêtês reached
home speedily and safely: Agamemnôn also arrived in Peloponnêsus, to
perish by the hand of a treacherous wife; but Menelaus was condemned
to long wanderings and to the severest privations in Egypt, Cyprus and
elsewhere, before he could set foot in his native land. The Lokrian
Ajax perished on the Gyræan rock.[751] Though exposed to a terrible
storm, he had already reached this place of safety, when he indulged
in the rash boast of having escaped in defiance of the gods: no sooner
did Poseidôn hear this language, than he struck with his trident the
rock which Ajax was grasping and precipitated both into the sea.[752]
Kalchas the soothsayer, together with Leonteus and Polypœtês, proceeded
by land from Troy to Kolophôn.[753]

  [749] Upon this the turn of fortune in Grecian affairs depends
  (Æschyl. Agamemn. 338; Odyss. iii. 130; Eurip. Troad. 69-95).

  [750] Odyss. iii. 130-161; Æschyl. Agamemn. 650-662.

  [751] Odyss. iii. 188-196; iv. 5-87. The Egyptian city of
  Kanopus, at the mouth of the Nile, was believed to have taken
  its name from the pilot of Menelaus, who had died and was buried
  there (Strabo, xvii. p. 801; Tacit. Ann. ii. 60). Μενελάϊος
  νόμος, so called after Menelaus (Dio Chrysost. xi p. 361).

  [752] Odyss. iv. 500. The epic Νόστοι of Hagias placed this
  adventure of Ajax on the rocks of Kaphareus, a southern
  promontory of Eubœa (Argum. Νόστοι, p. 23, Düntzer). Deceptive
  lights were kindled on the dangerous rocks by Nauplius, the
  father of Palamêdês, in revenge for the death of his son
  (Sophoklês, Ναύπιος Πυρκαεὺς, a lost tragedy; Hygin. f. 116;
  Senec. Agamemn. 567).

  [753] Argument. Νόστοι _ut sup._ There were monuments of Kalchas
  near Sipontum in Italy also (Strabo, vi. p. 284), as well as at
  Selgê in Pisidia (Strabo, xii. p. 570).

In respect however to these and other Grecian heroes, tales were
told different from those in the Odyssey, assigning to them a long
expatriation and a distant home. Nestôr went to Italy, where he
founded Metapontum, Pisa and Hêrakleia:[754] Philoktêtês[755] also
went to Italy, founded Petilia and Krimisa, and sent settlers to
Egesta in Sicily. Neoptolemus, under the advice of Thetis, marched
by land across Thrace, met with Odysseus, who had come by sea, at
Maroneia, and then pursued his journey to Epirus, where he became
king of the Molossians.[756] Idomeneus came to Italy, and founded
Uria in the Salentine peninsula. Diomêdês, after wandering far and
wide, went along the Italian coast into the innermost Adriatic gulf,
and finally settled in Daunia, founding the cities of Argyrippa,
Beneventum, Atria and Diomêdeia: by the favor of Athênê he became
immortal, and was worshipped as a god in many different places.[757]
The Lokrian followers of Ajax founded the Epizephyrian Lokri on
the southernmost corner of Italy,[758] besides another settlement
in Libya. I have spoken in another place of the compulsory exile of
Teukros, who, besides founding the city of Salamis in Cyprus, is said
to have established some settlements in the Iberian peninsula.[759]
Menestheus the Athenian did the like, and also founded both Elæa in
Mysia and Skylletium in Italy.[760] The Arcadian chief Agapenôr founded
Paphus in Cyprus.[761] Epeius, of Panopeus in Phôkis, the constructor
of the Trojan horse with the aid of the goddess Athênê, settled at
Lagaria near Sybaris on the coast of Italy; and the very tools which
he had employed in that remarkable fabric were shown down to a late
date in the temple of Athênê at Metapontum.[762] Temples, altars and
towns were also pointed out in Asia Minor, in Samos and in Krête, the
foundation of Agamemnôn or of his followers.[763] The inhabitants of
the Grecian town of Skionê, in the Thracian peninsula called Pallênê
or Pellênê, accounted themselves the offspring of the Pellênians from
Achæa in Peloponnêsus, who had served under Agamemnôn before Troy,
and who on their return from the siege had been driven on the spot
by a storm and there settled.[764] The Pamphylians, on the southern
coast of Asia Minor, deduced their origin from the wanderings of
Amphilochus and Kalchas after the siege of Troy: the inhabitants of the
Amphilochian Argos on the Gulf of Ambrakia revered the same Amphilochus
as their founder.[765] The Orchomenians under Ialmenus, on quitting
the conquered city, wandered or were driven to the eastern extremity
of the Euxine Sea; and the barbarous Achæans under Mount Caucasus
were supposed to have derived their first establishment from this
source.[766] Merionês with his Krêtan followers settled at Engyion in
Sicily, along with the preceding Krêtans who had remained there after
the invasion of Minôs. The Elyminians in Sicily also were composed
of Trojans and Greeks separately driven to the spot, who, forgetting
their previous differences, united in the joint settlements of Eryx
and Egesta.[767] We hear of Podaleirius both in Italy and on the coast
of Karia;[768] of Akamas son of Thêseus, at Amphipolis in Thrace, at
Soli in Cyprus, and at Synnada in Phrygia;[769] of Guneus, Prothous
and Eurypylus, in Krête as well as in Libya.[770] The obscure poem of
Lycophrôn enumerates many of these dispersed and expatriated heroes,
whose conquest of Troy was indeed a Kadmeian victory (according to
the proverbial phrase of the Greeks), wherein the sufferings of the
victor were little inferior to those of the vanquished.[771] It was
particularly among the Italian Greeks, where they were worshipped with
very special solemnity, that their presence as wanderers from Troy was
reported and believed.[772]

  [754] Strabo, v. p. 222; vi. p. 264. Vellei. Paterc. i. 1;
  Servius ad Æn. x. 179. He had built a temple to Athênê in the
  island of Keôs (Strabo, x. p. 487).

  [755] Strabo, vi. pp. 254, 272; Virgil, Æn. iii. 401, and Servius
  _ad loc._; Lycophrôn, 912.

  Both the tomb of Philoktêtês and the arrows of Hêraklês which
  he had used against Troy, were for a long time shown at Thurium
  (Justin, xx. 1).

  [756] Argument. Νόστοι, p. 23, Düntz.; Pindar, Nem. iv. 51.
  According to Pindar, however, Neoptolemus comes from Troy by sea,
  misses the island of Skyrus, and sails round to the Epeirotic
  Ephyra (Nem. vii. 37).

  [757] Pindar, Nem. x. 7, with the Scholia. Strabo, iii. p. 150;
  v. p. 214-215; vi, p. 284. Stephan. Byz. Ἀργύριππα, Διομηδεία.
  Aristotle recognizes him as buried in the Diomedean islands in
  the Adriatic (Anthol. Gr. Brunck. i. p. 178).

  The identical tripod which had been gained by Diomêdês, as victor
  in the chariot-race at the funeral games of Patroclus, was shown
  at Delphi in the time of Phanias, attested by an inscription,
  as well as the dagger which had been worn by Helikaôn, son of
  Antenôr (Athenæ. vi. p. 232).

  [758] Virgil, Æneid, iii. 399.; xi. 265; and Servius, _ibid._
  Ajax, the son of Oïleus, was worshipped there as a hero (Conôn,
  Narr. 18).

  [759] Strabo, iii. p. 257; Isokratês, Evagor. Encom. p. 192;
  Justin, xliv. 3. Ajax, the son of Teukros, established a temple
  of Zeus, and an hereditary priesthood always held by his
  descendants (who mostly bore the name of Ajax or Teukros), at
  Olbê in Kilikia (Strabo, xiv. p. 672). Teukros carried with him
  his Trojan captives to Cyprus (Athenæ. vi. p. 256).

  [760] Strabo, iii. p. 140-150; vi. p. 261; xiii. p. 622. See the
  epitaphs on Teukros and Agapenôr by Aristotle (Antholog. Gr. ed.
  Brunck. i. p. 179-180).

  [761] Strabo, xiv. p. 683; Pausan. viii. 5, 2.

  [762] Strabo, vi. p. 263; Justin, xx. 2; Aristot. Mirab. Ausc.
  c. 108. Also the epigram of the Rhodian Simmias called Πελεκύς
  (Antholog. Gr. Brunck. i. p. 210).

  [763] Vellei. Patercul. i. 1. Stephan. Byz. v. Λάμπη. Strabo,
  xiii. p. 605; xiv p. 639. Theopompus (Fragm. III, Didot)
  recounted that Agamemnôn and his followers had possessed
  themselves of the larger portion of Cyprus.

  [764] Thucydid. iv. 120.

  [765] Herodot. vii. 91; Thucyd. ii. 68. According to the old
  elegiac poet Kallinos, Kalchas himself had died at Klarus near
  Kolophôn after his march from Troy, but Mopsus, his rival in the
  prophetic function, had conducted his followers into Pamphylia
  and Kilikia (Strabo, xii. p. 570; xiv.p. 668). The oracle of
  Amphilochus at Mallus in Kilikia bore the highest character for
  exactness and truth-telling in the time of Pausanias, μαντεῖον
  ἀψευδέστατον τῶν ἐπ᾽ ἐμοῦ. (Paus. i. 34, 2). Another story
  recognized Leonteus and Polypætês as the founders of Aspendus in
  Kilikia (Eustath. ad Iliad. ii. 138).

  [766] Strabo, ix. p. 416.

  [767] Diodôr. iv. 79; Thucyd. vi. 2.

  [768] Stephan, Byz. v. Σύρνα; Lycophrôn, 1047.

  [769] Æschines, De Falsâ Legat. c. 14; Strabo, xiv. p. 683;
  Stephan. Byz. v. Σύνναδα.

  [770] Lycophrôn, 877-902, with Scholia; Apollodôr. Fragm. p.
  386, Heyne. There is also a long enumeration of these returning
  wanderers and founders of new settlements in Solinus (Polyhist.
  c. 2).

  [771] Strabo, iii. p. 150.

  [772] Aristot. Mirabil. Auscult. 79, 106, 107, 109, 111.

I pass over the numerous other tales which circulated among the
ancients, illustrating the ubiquity of the Grecian and Trojan heroes as
well as that of the Argonauts,—one of the most striking features in the
Hellenic legendary world.[773] Amongst them all, the most interesting,
individually, is Odysseus, whose romantic adventures in fabulous
places and among fabulous persons have been made familiarly known by
Homer. The goddesses Kalypso and Circê; the semi-divine mariners of
Phæacia, whose ships are endowed with consciousness and obey without
a steersman; the one-eyed Cyclôpes, the gigantic Læstrygones, and
the wind-ruler Æolus; the Sirens who ensnare by their song, as the
Lotophagi fascinate by their food—all these pictures formed integral
and interesting portions of the old epic. Homer leaves Odysseus
reëstablished in his house and family; but so marked a personage could
never be permitted to remain in the tameness of domestic life: the
epic poem called the Telegonia ascribed to him a subsequent series of
adventures. After the suitors had been buried by their relatives, he
offered sacrifice to the Nymphs, and then went to Elis to inspect his
herds of cattle there pasturing: the Eleian Polyxenus welcomed him
hospitably, and made him a present of a bowl: Odysseus then returned
to Ithaka, and fulfilled the rites and sacrifices prescribed to him by
Teiresias in his visit to the under-world. This obligation discharged,
he went to the country of the Thesprotians, and there married the
queen Kallidikê: he headed the Thesprotians in a war against the
Brygians, the latter being conducted by Arês himself, who fiercely
assailed Odysseus; but the goddess Athênê stood by him, and he was
enabled to make head against Arês until Apollo came and parted them.
Odysseus then returned to Ithaka, leaving the Thesprotian kingdom to
Polypœtês, his son by Kallidikê. Telegonus, his son by Circê, coming to
Ithaka in search of his father, ravaged the island and killed Odysseus
without knowing who he was. Bitter repentance overtook the son for
his undesigned parricide: at his prayer and by the intervention of
his mother Circê, both Penelopê and Têlemachus were made immortal:
Telegonus married Penelopê, and Têlemachus married Circê.[774]

  [773] Strabo, i. p. 48. After dwelling emphatically on the long
  voyages of Dionysus, Hêraklês, Jasôn, Odysseus, and Menelaus,
  he says, Αἰνείαν δὲ καὶ Ἀντήνορα καὶ Ἐνετοὺς, καὶ ἁπλῶς τοὺς
  ἐκ τοῦ Τρωϊκοῦ πολέμου πλανηθέντας εἰς ~πᾶσαν τὴν οἰκουμένην~,
  ἄξιον μὴ τῶν παλαιῶν ἀνθρώπων νομίσαι; Συνέβη γὰρ δὴ τοῖς τότε
  Ἕλλησιν, ὁμοίως καὶ τοῖς βαρβάροις, διὰ τὸν τῆς στρατείας
  χρόνον, ἀποβαλεῖν τά τε ἐν οἴκῳ καὶ τῇ στρατείᾳ πορισθέντα· ὥστε
  μετὰ τὴν τοῦ Ἰλίου καταστροφὴν τούς τε νικήσαντας ἐπὶ λῄστειαν
  τραπέσθαι διὰ τὰς ἀπορίας, καὶ πολλῷ μᾶλλον τοὺς ἡττηθέντας καὶ
  περιγενομένους ἐκ τοῦ πολέμου. Καὶ δὴ καὶ πόλεις ὑπὸ ~τούτων
  κτισθῆναι~ λέγονται ~κατὰ πᾶσαν τὴν ἔξω τῆς Ἑλλάδος παραλίαν~,
  ἔστι δ᾽ ὅπου καὶ τὴν μεσόγαιαν.

  [774] The Telegonia, composed by Eugammôn of Kyrênê, is lost, but
  the Argument of it has been preserved by Proclus (p. 25, Düntzer;
  Dictys, vi. 15).

  Pausanias quotes a statement from the poem called _Thesprôtis_,
  respecting a son of Odysseus and Penelopê, called Ptoliporthus,
  born after his return from Troy (viii. 12, 3). Nitzsch (Hist.
  Homer. p. 97) as well as Lobeck seem to imagine that this is the
  same poem as the Telegonia, under another title.

  Aristotle notices an oracle of Odysseus among the Eurytanes, a
  branch of the Ætôlian nation: there were also places in Epirus
  which boasted of Odysseus as their founder (Schol. ad Lycophrôn.
  800; Stephan. Byz. v. Βούνειμα; Etymolog. Mag. Ἀρκείσιος;
  Plutarch, Quæst. Gr. c. 14).

We see by this poem that Odysseus was represented as the mythical
ancestor of the Thesprotian kings, just as Neoptolemus was of the
Molossian.

It has already been mentioned that Antenôr and Æneas stand
distinguished from the other Trojans by a dissatisfaction with Priam
and a sympathy with the Greeks, which is by Sophoklês and others
construed as treacherous collusion,[775]—a suspicion indirectly
glanced, though emphatically repelled, by the Æneas of Virgil.[776]
In the old epic of Arktinus, next in age to the Iliad and Odyssey,
Æneas abandons Troy and retires to Mount Ida, in terror at the
miraculous death of Laocoôn, before the entry of the Greeks into the
town and the last night-battle: yet Leschês, in another of the ancient
epic poems, represented him as having been carried away captive by
Neoptolemus.[777] In a remarkable passage of the Iliad, Poseidôn
describes the family of Priam as having incurred the hatred of Zeus,
and predicts that Æneas and his descendants shall reign over the
Trojans: the race of Dardanus, beloved by Zeus more than all his other
sons, would thus be preserved, since Æneas belonged to it. Accordingly,
when Æneas is in imminent peril from the hands of Achilles, Poseidôn
specially interferes to rescue him, and even the implacable miso-Trojan
goddess Hêrê assents to the proceeding.[778] These passages have
been construed by various able critics to refer to a family of
philo-Hellenic or semi-Hellenic Æneadæ, known even in the time of the
early singers of the Iliad as masters of some territory in or near the
Troad, and professing to be descended from, as well as worshipping,
Æneas. In the town of Skêpsis, situated in the mountainous range of
Ida, about thirty miles eastward of Ilium, there existed two noble and
priestly families who professed to be descended, the one from Hectôr,
the other from Æneas. The Skêpsian critic Dêmêtrius (in whose time both
these families were still to be found) informs us that Skamandrius son
of Hectôr, and Ascanius son of Æneas, were the archegets or heroic
founders of his native city, which had been originally situated on
one of the highest ranges of Ida, and was subsequently transferred
by them to the less lofty spot on which it stood in his time.[779]
In Arisbê and Gentinus there seem to have been families professing
the same descent, since the same archegets were acknowledged.[780] In
Ophrynium, Hectôr had his consecrated edifice, and in Ilium both he and
Æneas were worshipped as gods:[781] and it was the remarkable statement
of the Lesbian Menekratês, that Æneas, “having been wronged by Paris
and stripped of the sacred privileges which belonged to him, avenged
himself by betraying the city, and then became one of the Greeks.”[782]

  [775] Dionys. Hal. i. 46-48; Sophokl. ap. Strab. xiii. p. 608;
  Livy, i. 1; Xenophon, Venat. i. 15.

  [776] Æn. ii. 433.

  [777] Argument of Ἰλίου Πέρσις; Fragm. 7. of Leschês, in
  Düntzer’s Collection, p. 19-21.

  Hellanikus seems to have adopted this retirement of Æneas to the
  strongest parts of Mount Ida, but to have reconciled it with
  the stories of the migration of Æneas, by saying that he only
  remained in Ida a little time, and then quitted the country
  altogether by virtue of a convention concluded with the Greeks
  (Dionys. Hal. i. 47-48). Among the infinite variety of stories
  respecting this hero, one was, that after having effected his
  settlement in Italy, he had returned to Troy and resumed the
  sceptre, bequeathing it at his death to Ascanius (Dionys. Hal. i.
  53): this was a comprehensive scheme for apparently reconciling
  _all_ the legends.

  [778] Iliad, xx. 300. Poseidôn speaks, respecting Æneas—

    Ἀλλ᾽ ἄγεθ᾽ ἡμεῖς πέρ μιν ὑπ᾽ ἐκ θανάτου ἀγάγωμεν,
    Μήπως καὶ Κρονίδης κεχολώσεται, αἴκεν Ἀχιλλεὺς
    Τόνδε κατακτείνῃ· μόριμον δέ οἱ ἐστ᾽ ἀλέασθαι,
    Ὄφρα μὴ ἄσπερμος γενεὴ καὶ ἄφαντος ὄληται
    Δαρδάνου, ὃν Κρονίδης περὶ πάντων φίλατο παίδων,
    Οἱ ἕθεν ἐξεγένοντο, γυναικῶν τε θνητάων.
    Ἤδη γὰρ Πριάμου γενεὴν ἤχθῃρε Κρονίων·
    Νῦν δὲ δὴ Αἰνείαο βίη Τρώεσσιν ἀνάξει,
    Καὶ παίδων παῖδες, τοί κεν μετόπισθε γένωνται.

  Again, v. 339, Poseidôn tells Æneas that he has nothing to dread
  from any other Greek than Achilles.

  [779] See O. Müller, on the causes of the mythe of Æneas and
  his voyage to Italy, in Classical Journal, vol. xxvi. p. 308;
  Klausen, Æneas und die Penaten, vol. i. p. 43-52.

  Dêmêtrius Skêps. ab. Strab. xiii. p. 607; Nicolaus ap. Steph.
  Byz. v. Ἀσκανία. Dêmêtrius conjectured that Skêpsis had been the
  regal seat of Æneas: there was a village called Æneia near to it
  (Strabo, xiii. p. 603).

  [780] Steph. Byz. v. Ἀρίσβη, Γεντῖνος. Ascanius is king of Ida
  after the departure of the Greeks (Conôn, Narr. 41; Mela, i. 18).
  _Ascanius portus_ between Phokæ and Kymê.

  [781] Strabo, xiii. p. 595; Lycophrôn, 1208, and Sch.;
  Athenagoras, Legat. 1. Inscription in Clarke’s Travels, vol. ii.
  p. 86, Οἱ Ἰλιεῖς τὸν πάτριον θεὸν Αἰνείαν. Lucian, Deor. Concil.
  c. 12. i. 111. p. 534, Hemst.

  [782] Menekrat. ap. Dionys. Hal. i. 48. Ἀχαιοὺς δὲ ἀνίη εἶχε
  (after the burial) καὶ ἐδόκεον τῆς στρατιῆς τὴν κεφαλὴν
  ἀπηράχθαι. Ὅμως δὲ τάφον αὐτῷ δαίσαντες, ἐπολέμεον γῇ πάσῃ,
  ἄχρις Ἴλιος ἑάλω Αἰνείεω ἐνδόντος. Αἰνείης γὰρ ἄτιτος ἐὼν ὑπὸ
  Ἀλεξάνδρου, καὶ ἀπὸ γερέων ἱερῶν ἐξειργόμενος, ἀνέτρεψε Πρίαμον,
  ἐργασάμενος δὲ ταῦτα, εἷς Ἀχαιῶν ἐγεγόνει.

  Abas, in his _Troica_, gave a narrative different from any
  other preserved: “Quidam ab Abante, qui _Troica_ scripsit,
  relatum ferunt, post discessum a Trojâ Græcorum Astyanacti ibi
  datum regnum, hunc ab Antenore expulsum sociatis sibi finitimis
  civitatibus, inter quas et Arisba fuit: Ænean hoc ægre tulisse,
  et pro Astyanacte arma cepisse ac prospere gestâ re Astyanact
  restituisse regnum” (Servius ad Virg. Æneid. ix. 264). According
  to Diktys, Antenôr remains king and Æneas goes away (Dikt. v.
  17): Antenôr brings the Palladium to the Greeks (Dikt. v. 8).
  Syncellus, on the contrary, tells us that the sons of Hectôr
  recovered Ilium by the suggestions of Helenus, expelling the
  Atenorids (Syncell. p. 322, ed. Bonn).

One tale thus among many respecting Æneas, and that too the most
ancient of all, preserved among the natives of the Troad, who
worshipped him as their heroic ancestor, was, that after the capture
of Troy he continued in the country as king of the remaining Trojans,
on friendly terms with the Greeks. But there were other tales
respecting him, alike numerous and irreconcilable: the hand of
destiny marked him as a wanderer (_fato profugus_), and his ubiquity
is not exceeded even by that of Odysseus. We hear of him at Ænus
in Thrace, in Pallênê, at Æneia in the Thermaic Gulf, in Delus, at
Orchomenos and Mantineia in Arcadia, in the islands of Kythêra and
Zakynthus, in Leukas and Ambrakia, at Buthrotum in Epirus, on the
Salentine peninsula and various other places in the southern region
of Italy; at Drepana and Segesta in Sicily, at Carthage, at Cape
Palinurus, Cumæ, Misenum, Caieta, and finally in Latium, where he lays
the first humble foundation of the mighty Rome and her empire.[783]
And the reason why his wanderings were not continued still further
was, that the oracles and the pronounced will of the gods directed
him to settle in Latium.[784] In each of these numerous places his
visit was commemorated and certified by local monuments or special
legends, particularly by temples and permanent ceremonies in honor of
his mother Aphroditê, whose worship accompanied him everywhere: there
were also many temples and many different tombs of Æneas himself.[785]
The vast ascendency acquired by Rome, the ardor with which all the
literary Romans espoused the idea of a Trojan origin, and the fact
that the Julian family recognized Æneas as their gentile primary
ancestor,—all contributed to give to the Roman version of his legend
the preponderance over every other. The various other places in which
monuments of Æneas were found came thus to be represented as places
where he had halted for a time on his way from Troy to Latium. But
though the legendary pretensions of these places were thus eclipsed
in the eyes of those who constituted the literary public, the local
belief was not extinguished: they claimed the hero as their permanent
property, and his tomb was to them a proof that he had lived and died
among them.

  [783] Dionys. Halic. A. R. i. 48-54; Heyne, Excurs. 1 ad Æneid.
  iii.; De Æneæ Erroribus, and Excurs. 1 ad Æn. v.; Conôn. Narr.
  46; Livy, xl. 4; Stephan. Byz. Αἴνεια. The inhabitants of Æneia
  in the Thermaic Gulf worshipped him with great solemnity as their
  heroic founder (Pausan. iii. 22, 4; viii. 12, 4). The tomb of
  Anchisês was shown on the confines of the Arcadian Orchomenus and
  Mantineia (compare Steph. Byz. v. Κάφυαι), under the mountain
  called Anchisia, near a temple of Aphroditê: on the discrepancies
  respecting the death of Anchisês (Heyne. Excurs. 17 ad Æn. iii.):
  Segesta in Sicily founded by Æneas (Cicero, Verr. iv. 33).

  [784] Τοῦ δὲ μηκέτι προσωτέρω τῆς Εὐρώπης πλεῦσαι τὸν Τρωϊκὸν
  στόλον, οἵ τε χρησμοὶ ἐγένοντο αἴτιοι, etc. (Dionys. Hal. i. 55).

  [785] Dionys. Hal. i. 54. Among other places, his tomb was shown
  at Berecynthia, in Phrygia (Festus, v. _Romam_, p. 224, ed.
  Müller): a curious article, which contains an assemblage of the
  most contradictory statements respecting both Æneas and Latinus.

Antenôr, who shares with Æneas the favorable sympathy of the Greeks,
is said by Pindar to have gone from Troy along with Menelaus and Helen
into the region of Kyrênê in Libya.[786] But according to the more
current narrative, he placed himself at the head of a body of Eneti or
Veneti from Paphlagonia, who had come as allies of Troy, and went by
sea into the inner part of the Adriatic Gulf, where he conquered the
neighboring barbarians and founded the town of Patavium (the modern
Padua); the Veneti in this region were said to owe their origin to his
immigration.[787] We learn further from Strabo, that Opsikellas, one
of the companions of Antenôr, had continued his wanderings even into
Ibêria, and that he had there established a settlement bearing his
name.[788]

  [786] Pindar, Pyth. v., and the citation from the Νόστοι of
  Lysimachus in the Scholia; given still more fully in the Scholia
  ad Lycophrôn. 875. There was a λόφος Ἀντηνορίδων at Kyrênê.

  [787] Livy, i. 1. Servius ad Æneid. i. 242. Strabo, i. 48; v.
  212. Ovid, Fasti, iv. 75.

  [788] Strabo, iii. p. 157.

Thus endeth the Trojan war; together with its sequel, the dispersion
of the heroes, victors as well as vanquished. The account here
given of it has been unavoidably brief and imperfect; for in a work
intended to follow consecutively the real history of the Greeks, no
greater space can be allotted even to the most splendid gem of their
legendary period. Indeed, although it would be easy to fill a large
volume with the separate incidents which have been introduced into the
“Trojan cycle,” the misfortune is that they are for the most part so
contradictory as to exclude all possibility of weaving them into one
connected narrative. We are compelled to select one out of the number,
generally without any solid ground of preference, and then to note
the variations of the rest. No one who has not studied the original
documents can imagine the extent to which this discrepancy proceeds;
it covers almost every portion and fragment of the tale.[789]

  [789] These diversities are well set forth in the useful
  Dissertation of Fuchs De Varietate Fabularum Troicarum (Cologne,
  1830).

  Of the number of romantic statements put forth respecting Helen
  and Achilles especially, some idea may be formed from the fourth,
  fifth and sixth chapters of Ptolemy Hêphæstion (apud Westermann,
  Scriptt. Mythograph. p. 188, etc.).

But though much may have been thus omitted of what the reader
might expect to find in an account of the Trojan war, its genuine
character has been studiously preserved, without either exaggeration
or abatement. The real Trojan war is that which was recounted by
Homer and the old epic poets, and continued by all the lyric and
tragic composers. For the latter, though they took great liberties
with the particular incidents, and introduced to some extent a new
moral tone, yet worked more or less faithfully on the Homeric scale:
and even Euripidês, who departed the most widely from the feeling
of the old legend, never lowered down his matter to the analogy
of contemporary life. They preserved its well-defined object, at
once righteous and romantic, the recovery of the daughter of Zeus
and sister of the Dioskuri—its mixed agencies, divine, heroic and
human—the colossal force and deeds of its chief actors—its vast
magnitude and long duration, as well as the toils which the conquerors
underwent, and the Nemesis which followed upon their success. And
these were the circumstances which, set forth in the full blaze of
epic and tragic poetry, bestowed upon the legend its powerful and
imperishable influence over the Hellenic mind. The enterprise was one
comprehending all the members of the Hellenic body, of which each
individually might be proud, and in which, nevertheless, those feelings
of jealous and narrow patriotism, so lamentably prevalent in many of
the towns, were as much as possible excluded. It supplied them with
a grand and inexhaustible object of common sympathy, common faith,
and common admiration; and when occasions arose for bringing together
a Pan-Hellenic force against the barbarians, the precedent of the
Homeric expedition was one upon which the elevated minds of Greece
could dwell with the certainty of rousing an unanimous impulse, if not
always of counterworking sinister by-motives, among their audience.
And the incidents comprised in the Trojan cycle were familiarized, not
only to the public mind but also to the public eye, by innumerable
representations both of the sculptor and the painter,—those which were
romantic and chivalrous being better adapted for this purpose, and
therefore more constantly employed, than any other.

Of such events the genuine Trojan war of the old epic was for the most
part composed. Though literally believed, reverentially cherished,
and numbered among the gigantic phænomena of the past, by the
Grecian public, it is in the eyes of modern inquiry essentially a
legend and nothing more. If we are asked whether it be not a legend
embodying portions of historical matter, and raised upon a basis of
truth,—whether there may not really have occurred at the foot of
the hill of Ilium a war purely human and political, without gods,
without heroes, without Helen, without Amazons, without Ethiopians
under the beautiful son of Eôs, without the wooden horse, without the
characteristic and expressive features of the old epical war,—like
the mutilated trunk of Deïphobus in the under-world; if we are asked
whether there was not really some such historical Trojan war as this,
our answer must be, that as the possibility of it cannot be denied,
so neither can the reality of it be affirmed. We possess nothing but
the ancient epic itself without any independent evidence: had it
been an age of records indeed, the Homeric epic in its exquisite and
unsuspecting simplicity would probably never have come into existence.
Whoever therefore ventures to dissect Homer, Arktinus and Leschês,
and to pick out certain portions as matters of fact, while he sets
aside the rest as fiction, must do so in full reliance on his own
powers of historical divination, without any means either of proving
or verifying his conclusions. Among many attempts, ancient as well as
modern, to identify real objects in this historical darkness, that of
Dio Chrysostom deserves attention for its extraordinary boldness. In
his oration addressed to the inhabitants of Ilium, and intended to
demonstrate that the Trojans were not only blameless as to the origin
of the war, but victorious in its issue—he overthrows all the leading
points of the Homeric narrative, and re-writes nearly the whole from
beginning to end: Paris is the lawful husband of Helen, Achilles is
slain by Hectôr, and the Greeks retire without taking Troy, disgraced
as well as baffled. Having shown without difficulty that the Iliad,
if it be looked at as a history, is full of gaps, incongruities and
absurdities, he proceeds to compose a more plausible narrative of his
own, which he tenders as so much authentic matter of fact. The most
important point, however, which his Oration brings to view is, the
literal and confiding belief with which the Homeric narrative was
regarded, as if it were actual history, not only by the inhabitants of
Ilium, but also by the general Grecian public.[790]

  [790] Dio Chrysost. Or. xi. p. 310-322.

The small town of Ilium, inhabited by Æolic Greeks,[791] and raised
into importance only by the legendary reverence attached to it, stood
upon an elevated ridge forming a spur from Mount Ida, rather more than
three miles from the town and promontory of Sigeium, and about twelve
stadia, or less than two miles, from the sea at its nearest point.
From Sigeium and the neighboring town of Achilleium (with its monument
and temple of Achilles), to the town of Rhœteium on a hill higher
up the Hellespont (with its monument and chapel of Ajax called the
Aianteium[792]), was a distance of sixty stadia, or seven miles and a
half in the straight course by sea: in the intermediate space was a bay
and an adjoining plain, comprehending the embouchure of the Scamander,
and extending to the base of the ridge on which Ilium stood. This plain
was the celebrated plain of Troy, in which the great Homeric battles
were believed to have taken place: the portion of the bay near to
Sigeium went by the name of the Naustathmon of the Achæans (_i. e._ the
spot where they dragged their ships ashore), and was accounted to have
been the camp of Agamemnôn and his vast army.[793]

  [791] Herodot. v. 122. Pausan. v. 8, 3: viii. 12, 4. Αἰολεὺς ἐκ
  πόλεως Τρῴαδος, the title proclaimed at the Olympic games; like
  Αἰολεὺς ἀπὸ Μουρίνας, from Myrina in the more southerly region of
  Æolis, as we find in the list of visitors at the Charitêsia, at
  Orchomenos in Bœôtia (Corp. Inscrip. Boeckh. No. 1583).

  [792] See Pausanias, i. 35, 3, for the legends current at Ilium
  respecting the vast size of the bones of Ajax in his tomb. The
  inhabitants affirmed that after the shipwreck of Odysseus, the
  arms of Achilles, which he was carrying away with him, were
  washed up by the sea against the tomb of Ajax. Pliny gives the
  distance at thirty stadia: modern travellers make it some thing
  more than Pliny, but considerably less than Strabo.

  [793] Strabo, xiii. p. 596-598. Strabo distinguishes the Ἀχαιῶν
  Ναύσταθμον, which was near to Sigeium, from the Ἀχαιῶν λιμήν,
  which was more towards the middle of the bay between Sigeium and
  Rhœteium; but we gather from his language that this distinction
  was not universally recognized. Alexander landed at the Ἀχαιῶν
  λιμήν (Arrian, i. 11).

Historical Ilium was founded, according to the questionable statement
of Strabo, during the last dynasty of the Lydian kings,[794] that is,
at some period later than 720 B. C. Until after the days of Alexander
the Great—indeed until the period of Roman preponderance—it always
remained a place of inconsiderable power and importance, as we learn
not only from the assertion of the geographer, but also from the fact
that Achilleium, Sigeium and Rhœteium were all independent of it.[795]
But inconsiderable as it might be, it was the only place which ever
bore the venerable name immortalized by Homer. Like the Homeric Ilium,
it had its temple of Athênê,[796] wherein she was worshipped as the
presiding goddess of the town: the inhabitants affirmed that Agamemnôn
had not altogether destroyed the town, but that it had been reoccupied
after his departure, and had never ceased to exist.[797] Their
acropolis was called Pergamum, and in it was shown the house of Priam
and the altar of Zeus Herkeius where that unhappy old man had been
slain: moreover there were exhibited, in the temples, panoplies which
had been worn by the Homeric heroes,[798] and doubtless many other
relics appreciated by admirers of the Iliad.

  [794] Strabo, xiii. p. 593.

  [795] Herodot. v. 95 (his account of the war between the
  Athenians and Mitylenæans about Sigeium and Achilleium); Strabo,
  xiii. p. 593. Τὴν δὲ τῶν Ἰλιέων πόλιν τὴν νῦν τέως μὲν κωμόπολιν
  εἶναί φασι, τὸ ἱερὸν ἔχουσαν τῆς Ἀθηνᾶς μικρὸν καὶ εὐτελές.
  Ἀλέξανδρον δὲ ἀναβάντα μετὰ τὴν ἐπὶ Γρανίκῳ νίκην, ἀναθήμασι τε
  κοσμῆσαι τὸ ἱερὸν καὶ προσαγορεῦσαι πόλιν, etc.

  Again, Καὶ τὸ Ἴλιον, ὃ νῦν ἐστὶ, κωμόπολίς τις ἦν ὅτε πρῶτον
  Ῥωμαῖοι τῆς Ἀσίας ἐπέβησαν.

  [796] Besides Athênê, the Inscriptions authenticate Ζεὺς Πολιεὺς
  at Ilium (Corp. Inscrip. Bœckh. No. 3599).

  [797] Strabo, xiii. p. 600. Λέγουσι δ᾽ οἱ νῦν Ἰλιεῖς καὶ τοῦτο,
  ὡς οὐδὲ τέλεως συνέβαινεν ἠφανίσθαι τὴν πόλιν κατὰ τὴν ἅλωσιν ὑπὸ
  τῶν Ἀχαιῶν, οὐδ᾽ ἐξηλείφθη οὐδέποτε.

  The situation of Ilium (or as it is commonly, but erroneously,
  termed, _New Ilium_) appears to be pretty well ascertained, about
  two miles from the sea (Rennell, On the Topography of Troy, p.
  41-71; Dr. Clarke’s Travels, vol. ii. p. 102).

  [798] Xerxês passing by Adramyttium, and leaving the range of
  Mount Ida on his left hand, ἤϊε ἐς τὴν Ἰλιάδα γῆν.... Ἀπικομένου
  δὲ τοῦ στρατοῦ ἐπὶ τὸν Σκάμανδρον ... ἐς τὸ Πριάμου Πέργαμον
  ἀνέβη, ἵμερον ἔχων θεήσασθαι. Θεησάμενος δὲ, καὶ ~πυθόμενος
  κείνων ἕκαστα~, τῇ Ἀθηναίῃ τῇ Ἰλιάδι ἔθυσε βοῦς χιλίας· χοὰς δὲ
  οἱ μάγοι τοῖσιν ἥρωσιν ἐχέαντο.... Ἅμα ἡμέρῃ δὲ ἐπορεύετο, ἐν
  ἀριστερῇ μὲν ἀπέργων Ῥοιτεῖον πόλιν καὶ Ὀφρυνεῖον καὶ Δάρδανον,
  ἥπερ δὴ Ἀβύδῳ ὅμουρος ἐστιν· ἐν δεξιῇ δὲ, Γέργιθας Τευκρούς
  (Herod. vii. 43).

  Respecting Alexander (Arrian, i. 11), Ἀνελθόντα δὲ ἐς Ἴλιον, τῇ
  Ἀθηνᾷ θῦσαι τῇ Ἰλιάδι, καὶ τὴν πανοπλίαν τὴν αὑτοῦ ἀναθεῖναι ἐς
  τὸν ναὸν, καὶ καθελεῖν ἀντὶ ταύτης τῶν ἱερῶν τινα ὅπλων ἔτι ἐκ
  τοῦ Τρωϊκοῦ ἔργου σωζόμενα· καὶ ταῦτα λέγουσιν ὅτι οἱ ὑπασπισταὶ
  ἔφερον πρὸ αὐτοῦ ἐς τὰς μάχας. Θῦσαι δὲ αὐτὸν ἐπὶ τοῦ βωμοῦ τοῦ
  Διὸς τοῦ Ἑρκείου λόγος κατέχει, μῆνιν Πριάμου παραιτούμενον τῷ
  Νεοπτολέμου γένει, ὃ δὴ ἐς αὐτὸν καθῆκε.

  The inhabitants of Ilium also showed the lyre which had belonged
  to Paris (Plutarch, Alexand. c. 15).

  Chandler, in his History of Ilium, chap. xxii. p. 89, seems to
  think that the place called by Herodotus the Pergamum of Priam
  is different from the historical Ilium. But the mention of the
  Iliean Athênê identifies them as the same.

These were testimonies which few persons in those ages were inclined
to question, when combined with the identity of name and general
locality; nor does it seem that any one did question them until the
time of Dêmêtrius of Skêpsis. Hellanikus expressly described this Ilium
as being the Ilium of Homer, for which assertion Strabo (or probably
Dêmêtrius, from whom the narrative seems to be copied) imputes to him
very gratuitously an undue partiality towards the inhabitants of the
town.[799] Herodotus relates, that Xerxês in his march into Greece
visited the place, went up to the Pergamum of Priam, inquired with
much interest into the details of the Homeric siege, made libations to
the fallen heroes, and offered to the Athênê of Ilium his magnificent
sacrifice of a thousand oxen: he probably represented and believed
himself to be attacking Greece as the avenger of the Priamid family.
The Lacedæmonian admiral Mindarus, while his fleet lay at Abydus, went
personally to Ilium to offer sacrifice to Athênê, and saw from that
elevated spot the battle fought between the squadron of Dorieus and
the Athenians, on the shore near Rhœteium.[800] During the interval
between the Peloponnesian war and the Macedonian invasion of Persia,
Ilium was always garrisoned as a strong position; but its domain was
still narrow, and did not extend even to the sea which was so near to
it.[801] Alexander, on crossing the Hellespont, sent his army from
Sestus to Abydus, under Parmenio, and sailed personally from Elæeus
in the Chersonese, after having solemnly sacrificed at the Elæuntian
shrine of Prôtesilaus, to the harbor of the Achæans between Sigeium and
Rhœteium. He then ascended to Ilium, sacrificed to the Iliean Athênê,
and consecrated in her temple his own panoply, in exchange for which
he took some of the sacred arms there suspended, which were said to
have been preserved from the time of the Trojan war. These arms were
carried before him when he went to battle by his armor-bearers. It is
a fact still more curious, and illustrative of the strong working of
the old legend on an impressible and eminently religious mind, that he
also sacrificed to Priam himself, on the very altar of Zeus Herkeius
from which the old king was believed to have been torn by Neoptolemus.
As that fierce warrior was his heroic ancestor by the maternal side, he
desired to avert from himself the anger of Priam against the Achilleid
race.[802]

  [799] Strabo, xiii. p. 602. Ἑλλάνικος δὲ χαριζόμενος τοῖς
  Ἰλιεῦσιν, οἷος ὁ ἐκείνου μῦθος, συνηγορεῖ τῷ τὴν αὐτὴν εἶναι
  πόλιν τὴν νῦν τῇ τότε. Hellanikus had written a work called
  Τρωϊκά.

  [800] Xenoph. Hellen. i. 1, 10. Skylax places Ilium twenty-five
  stadia, or about three miles, from the sea (c. 94). But I do not
  understand how he can call Skêpsis and Kebrên πόλεις ἐπὶ θαλάσσῃ.

  [801] See Xenoph. Hellen. iii. i. 16; and the description of the
  seizure of Ilium, along with Skêpsis and Kebrên, by the chief of
  mercenaries, Charidêmus, in Demosthen. cont. Aristocrat. c. 38.
  p. 671: compare Æneas Poliorcetic. c. 24, and Polyæn. iii. 14.

  [802] Arrian, _l. c._ Dikæarchus composed a separate work
  respecting this sacrifice of Alexander, περὶ τῆς ἐν Ἰλίῳ θυσίας
  (Athenæ. xiii. p. 603; Dikæarch. Fragm. p. 114, ed. Fuhr).

  Theophrastus, in noticing old and venerable trees, mentions the
  φηγοὶ (_Quercus æsculus_) on the tomb of Ilus at Ilium, without
  any doubt of the authenticity of the place (De Plant. iv. 14);
  and his contemporary, the harper Stratonikos, intimates the same
  feeling, in his jest on the visit of a bad sophist to Ilium
  during the festival of the Ilieia (Athenæ. viii. p. 351). The
  same may be said respecting the author of the tenth epistle
  ascribed to the orator Æschinês (p. 737), in which his visit of
  curiosity to Ilium is described—as well as about Apollônius of
  Tyana, or the writer who describes his life and his visit to the
  Trôad; it is evident that he did not distrust the ἀρχαιολογία
  of the Ilieans, who affirmed their town to be the real Troy
  (Philostrat. Vit. Apollôn. Tyan. iv. 11).

  The goddess Athênê of Ilium was reported to have rendered
  valuable assistance to the inhabitants of Kyzikus, when they were
  besieged by Mithridatês, commemorated by inscriptions set up in
  Ilium (Plutarch, Lucull. 10).

Alexander made to the inhabitants of Ilium many munificent promises,
which he probably would have executed, had he not been prevented
by untimely death: for the Trojan war was amongst all the Grecian
legends the most thoroughly Pan-Hellenic, and the young king of
Macedôn, besides his own sincere legendary faith, was anxious to
merge the local patriotism of the separate Greek towns in one general
Hellenic sentiment under himself as chief. One of his successors,
Antigonus,[803] founded the city of Alexandreia in the Trôad, between
Sigeium and the more southerly promontory of Lektum; compressing into
it the inhabitants of many of the neighboring Æolic towns in the region
of Ida,—Skêpsis, Kebrên, Hamaxitus, Kolônæ, and Neandria, though the
inhabitants of Skêpsis were subsequently permitted by Lysimachus to
resume their own city and autonomous government. Ilium however remained
without any special mark of favor until the arrival of the Romans in
Asia and their triumph over Antiochus (about 190 B. C.). Though it
retained its walls and its defensible position, Dêmêtrius of Skêpsis,
who visited it shortly before that event, described it as being then
in a state of neglect and poverty, many of the houses not even having
tiled roofs.[804] In this dilapidated condition, however, it was still
mythically recognized both by Antiochus and by the Roman consul
Livius, who went up thither to sacrifice to the Iliean Athênê. The
Romans, proud of their origin from Troy and Æneas, treated Ilium with
signal munificence; not only granting to it immunity from tribute,
but also adding to its domain the neighboring territories of Gergis,
Rhœteium and Sigeium—and making the Ilieans masters of the whole
coast[805] from the Peræa (or continental possessions) of Tenedos
(southward of Sigeium) to the boundaries of Dardanus, which had its own
title to legendary reverence as the special sovereignty of Æneas. The
inhabitants of Sigeium could not peaceably acquiesce in this loss of
their autonomy, and their city was destroyed by the Ilieans.

  [803] Strabo, xiii. p. 603-607.

  [804] Livy, xxxv. 43; xxxvii. 9. Polyb. v. 78-111 (passages which
  prove that Ilium was fortified and defensible about B. C. 218).
  Strabo, xiii. p. 594. Καὶ τὸ Ἴλιον δ᾽, ὃ νῦν ἐστι, κωμόπολίς τις
  ἦν, ὅτε πρῶτον Ῥωμαῖοι τῆς Ἀσίας ἐπέβησαν καὶ ἐξέβαλον Ἀντίοχον
  τὸν μέγαν ἐκ τῆς ἐντὸς τοῦ Ταύρου. Φησὶ γοῦν Δημήτριος ὁ Σκήψιος,
  μειράκιον ἐπιδημήσας εἰς τὴν πόλιν κατ᾽ ἐκείνους τοὺς καιροὺς,
  οὕτως ὠλιγωρημένην ἰδεῖν τὴν κατοικίαν, ὥστε μηδὲ κεραμωτὰς
  ἔχειν τὰς στέγας. Ἡγησιάναξ δὲ, τοὺς Γαλάτας περαιωθέντας ἐκ τῆς
  Εὐρώπης, ἀναβῆναι μὲν εἰς τὴν πόλιν δεομένους ἐρύματος, παραχρῆμα
  δ᾽ ἐκλιπεῖν διὰ τὸ ἀτείχιστον· ὕστερον δ᾽ ἐπανόρθωσιν ἔσχε
  πολλήν. Εἶτ᾽ ἐκάκωσαν αὐτὴν πάλιν οἱ μετὰ Φιμβρίου, etc.

  This is a very clear and precise statement, attested by an
  eye-witness. But it is thoroughly inconsistent with the statement
  made by Strabo in the previous chapter, a dozen lines before, as
  the text now stands; for he there informs us that Lysimachus,
  after the death of Alexander, paid great attention to Ilium,
  surrounded it with a wall of forty stadia in circumference,
  erected a temple, and aggregated to Ilium the ancient cities
  around, which were in a state of decay. We know from Livy that
  the aggregation of Gergis and Rhœteium to Ilium was effected, not
  by Lysimachus, but by the Romans (Livy, xxxviii. 37); so that the
  _first_ statement of Strabo is not only inconsistent with his
  second, but is contradicted by an independent authority.

  I cannot but think that this contradiction arises from a
  confusion of the text in Strabo’s _first_ passage, and that
  in that passage Strabo really meant to speak only of the
  improvements brought about by Lysimachus in _Alexandreia Trôas_;
  that he never meant to ascribe to Lysimachus any improvements in
  _Ilium_, but, on the contrary, to assign the remarkable attention
  paid by Lysimachus to _Alexandreia Trôas_, as the reason why
  he had neglected to fulfil the promises held out by Alexander
  to _Ilium_. The series of facts runs thus:—1. Ilium is nothing
  better than a κώμη; at the landing of Alexander; 2. Alexander
  promises great additions, but never returns from Persia to
  accomplish them; 3. Lysimachus is absorbed in Alexandreia Trôas,
  into which he aggregates several of the adjoining old towns,
  and which flourishes under his hands; 4. Hence Ilium remained a
  κώμη when the Romans entered Asia, as it had been when Alexander
  entered.

  This alteration in the text of Strabo might be effected by the
  simple transposition of the words as they now stand, and by
  omitting ὅτε καὶ, ἤδη ἐπεμελήθη, without introducing a single
  new or conjectural word, so that the passage would read thus:
  Μετὰ δὲ τὴν ἐκείνου (Alexander’s) τελευτὴν Λυσίμαχος μάλιστα τῆς
  Ἀλεξανδρείας ἐπεμελήθη, συνῳκισμένης μὲν ἤδη ὑπ᾽ Ἀντιγόνου, καὶ
  προσηγορευμένης Ἀντιγόνιας, μεταβαλούσης δὲ τοὔνομα· (ἔδοξε γὰρ
  εὐσεβὲς εἶναι τοὺς Ἀλεξάνδρον διαδεξαμένους ἐκείνου πρότερον
  κτίζειν ἐπωνύμους πόλεις, εἶθ᾽ ἑαυτῶν) καὶ νέων κατεσκεύασε καὶ
  τεῖχος περιεβάλετο ὅσον 40 σταδίων· συνῴκισε δὲ εἰς αὐτὴν τὰς
  κύκλῳ πόλεις ἀρχαίας, ἤδη κεκακωμένας. Καὶ δὴ καὶ συνέμεινε
  ... πόλεων. If this reading be adopted, the words beginning
  that which stands in Tzschucke’s edition as sect. 27, and which
  immediately follow the last word πόλεων, will read quite suitably
  and coherently,—Καὶ τὸ Ἴλιον δ᾽, ὃ νῦν ἐστὶ, κωμόπολίς τις ἦν,
  ὅτε πρῶτον Ῥωμαῖοι τῆς Ἀσίας ἐπέβησαν, etc., whereas with the
  present reading of the passage they show a contradiction, and the
  whole passage is entirely confused.

  [805] Livy, xxxviii. 39; Strabo, xiii. p. 600. Κατέσκαπται δὲ καὶ
  τὸ Σίγειον ὑπὸ τῶν Ἰλιέων διὰ τὴν ἀπείθειαν· ὑπ᾽ ἐκείνοις γὰρ ἦν
  ὕστερον ἡ παραλία πᾶσα ἡ μέχρι Δαρδάνου, καὶ νῦν ὑπ᾽ ἐκείνοις
  ἔστι.

The dignity and power of Ilium being thus prodigiously enhanced, we
cannot doubt that the inhabitants assumed to themselves exaggerated
importance as the recognized parents of all-conquering Rome. Partly,
we may naturally suppose, from the jealousies thus aroused on the
part of their neighbors at Skêpsis and Alexandreia Trôas—partly from
the pronounced tendency of the age (in which Kratês at Pergamus and
Aristarchus at Alexandria divided between them the palm of literary
celebrity) towards criticism and illustration of the old poets—a
blow was now aimed at the mythical legitimacy of Ilium. Dêmêtrius of
Skêpsis, one of the most laborious of the Homeric critics, had composed
thirty books of comment upon the Catalogue in the Iliad: Hestiæa, an
authoress of Alexandreia Trôas, had written on the same subject: both
of them, well-acquainted with the locality, remarked that the vast
battles described in the Iliad could not be packed into the narrow
space between Ilium and the Naustathmon of the Greeks; the more so,
as that space, too small even as it then stood, had been considerably
enlarged since the date of the Iliad by deposits at the mouth of the
Skamander.[806] They found no difficulty in pointing out topographical
incongruities and impossibilities as to the incidents in the Iliad,
which they professed to remove by the startling theory that the Homeric
Ilium had not occupied the site of the city so called. There was a
village, called the village of the Ilieans, situated rather less than
four miles from the city in the direction of Mount Ida, and further
removed from the sea; here, they affirmed the “holy Troy” had stood.

  [806] Strabo, xiii. 599. Παρατίθησι δὲ ὁ Δημήτριος καὶ τὴν
  Ἀλεξανδρίνην Ἑστίαιαν μάρτυρα, τὴν συγγράψασαν περὶ τῆς Ὁμήρου
  Ἰλιάδος, πυνθανομένην, εἰ περὶ τὴν νῦν πόλιν ὁ πόλεμος συνέστη,
  καὶ τὸ Τρωϊκὸν πεδίον ποῦ ἔστιν, ὃ μέταξυ τῆς πόλεως καὶ τῆς
  θαλάσσης ὁ ποιητὴς φράζει· τὸ μὲν γὰρ πρὸ τῆς νῦν πόλεως
  ὁρώμενον, πρόχωμα εἶναι τῶν ποταμῶν, ὕστερον γεγονός.

  The words ποῦ ἔστιν are introduced conjecturally by Grosskurd,
  the excellent German translator of Strabo, but they seem to me
  necessary to make the sense complete.

  Hesitæa is cited more than once in the Homeric Scholia (Schol.
  Venet. ad Iliad, iii. 64; Enstath. ad Iliad, ii. 538).

No positive proof was produced to sustain the conclusion, for Strabo
expressly states that not a vestige of the ancient city remained at
the Village of the Ilieans:[807] but the fundamental supposition was
backed by a second accessory supposition, to explain how it happened
that all such vestiges had disappeared. Nevertheless Strabo adopts the
unsupported hypothesis of Dêmêtrius as if it were an authenticated
fact—distinguishing pointedly between Old and New Ilium, and even
censuring Hellanikus for having maintained the received local faith.
But I cannot find that Dêmêtrius and Hestiæa have been followed in
this respect by any other writer of ancient times excepting Strabo.
Ilium still continued to be talked of and treated by every one as the
genuine Homeric Troy: the cruel jests of the Roman rebel Fimbria, when
he sacked the town and massacred the inhabitants—the compensation made
by Sylla, and the pronounced favor of Julius Cæsar and Augustus,—all
prove this continued recognition of identity.[808] Arrian, though a
native of Nicomedia, holding a high appointment in Asia Minor, and
remarkable for the exactness of his topographical notices, describes
the visit of Alexander to Ilium, without any suspicion that the place
with all its relics was a mere counterfeit: Aristidês, Dio Chrysostom,
Pausanias, Appian, and Plutarch hold the same language.[809] But modern
writers seem for the most part to have taken up the supposition from
Strabo as implicitly as he took it from Dêmêtrius. They call Ilium by
the disrespectful appellation of _New_ Ilium—while the traveller in the
Trôad looks for _Old_ Ilium as if it were the unquestionable spot where
Priam had lived and moved; the name is even formally enrolled on the
best maps recently prepared of the ancient Trôad.[810]

  [807] Strabo, xiii. p. 599. Οὐδὲν δ᾽ ἴχνος σώζεται τῆς ἀρχαίας
  πόλεως—εἰκότως· ἅτε γὰρ ἐκπεπορθημένων τῶν κύκλῳ πόλεων, οὐ
  τελέως δὲ κατεσπασμένων, οἱ λίθοι πάντες εἰς τὴν ἐκείνων ἀνάληψιν
  μετηνέχθησαν.

  [808] Appian, Mithridat. c. 53; Strabo, xiii. p. 594; Plutarch,
  Sertorius, c. 1; Velleius Paterc. ii. 23.

  The inscriptions attest Panathenaic games celebrated at Ilium
  in honor of Athênê by the Ilieans conjointly with various other
  neighboring cities (see Corp. Inscr. Boeckh. No. 3601-3602, with
  Boeckh’s observations). The valuable inscription No. 3595 attests
  the liberality of Antiochus Soter towards the Iliean Athênê as
  early as 278 B. C.

  [809] Arrian, i. 11; Appian _ut sup._; also Aristidês, Or. 43,
  Rhodiaca, p. 820 (Dindorf p. 369). The curious Oratio xi. of Dio
  Chrysostom, in which he writes his new version of the Trojan war,
  is addressed to the inhabitants of Ilium.

  [810] The controversy, now half a century old, respecting Troy
  and the Trojan war—between Bryant and his various opponents,
  Morritt, Gilbert Wakefield, the British Critic, etc., seems
  now nearly forgotten, and I cannot think that the pamphlets on
  either side would be considered as displaying much ability, if
  published at the present day. The discussion was first raised by
  the publication of Le Chevalier’s account of the plain of Troy,
  in which the author professed to have discovered the true site of
  Old Ilium (the supposed Homeric Troy), about twelve miles from
  the sea near Bounarbashi. Upon this account Bryant published some
  animadversions, followed up by a second treatise, in which he
  denied the historical reality of the Trojan war, and advanced the
  hypothesis that the tale was of Egyptian origin (Dissertation on
  the War of Troy, and the Expedition of the Grecians as described
  by Homer, showing that no such Expedition was ever undertaken,
  and that no such city of Phrygia existed, by Jacob Bryant;
  seemingly 1797, though there is no date in the title-page:
  Morritt’s reply was published in 1798). A reply from Mr. Bryant
  and a rejoinder from Mr. Morritt, as well as a pamphlet from G.
  Wakefield, appeared in 1799 and 1800, besides an Expostulation by
  the former addressed to the British Critic.

  Bryant, having dwelt both on the incredibilities and the
  inconsistencies of the Trojan war, as it is recounted in
  Grecian legend generally, nevertheless admitted that Homer had
  a groundwork for his story, and maintained that that groundwork
  was Egyptian. Homer (he thinks) was an Ithacan, descended
  from a family originally emigrant from Egypt: the war of Troy
  was originally an Egyptian war, which explains how Memnôn the
  Ethiopian came to take part in it: “upon this history, which
  was originally Egyptian, Homer founded the scheme of his two
  principal poems, adapting things to Greece and Phrygia by an
  ingenious transposition:” he derived information from priests
  of Memphis or Thêbes (Bryant, pp. 102, 108, 126). The Ἥρως
  Αἰγύπτιος, mentioned in the second book of the Odyssey (15), is
  the Egyptian hero, who affords, in his view, an evidence that the
  population of that island was in part derived from Egypt. No one
  since Mr. Bryant, I apprehend, has ever construed the passage in
  the same sense.

  Bryant’s Egyptian hypothesis is of no value; but the negative
  portion of his argument, summing up the particulars of the Trojan
  legend, and contending against its historical credibility, is
  not so easily put aside. Few persons will share in the zealous
  conviction by which Morritt tries to make it appear that the
  1100 ships, the ten years of war, the large confederacy of
  princes from all parts of Greece, etc., have nothing but what
  is consonant with historical probability; difficulties being
  occasionally eliminated by the plea of our ignorance of the time
  and of the subject (Morritt, p. 7-21). Gilbert Wakefield, who
  maintains the historical reality of the siege with the utmost
  intensity, and even compares Bryant to Tom Paine (W. p. 17), is
  still more displeased with those who propound doubts, and tells
  us that “grave disputation in the midst of such darkness and
  uncertainty is a conflict with chimæras” (W. p. 14).

  The most plausible line of argument taken by Morritt and
  Wakefield is, where they enforce the positions taken by Strabo
  and so many other authors, ancient as well as modern, that a
  superstructure of fiction is to be distinguished from a basis
  of truth, and that the latter is to be maintained while the
  former is rejected (Morritt, p. 5; Wake. p. 7-8). To this Bryant
  replies, that “if we leave out every absurdity, we can make
  anything plausible; that a fable may be made consistent, and we
  have many romances that are very regular in the assortment of
  characters and circumstances: this may be seen in plays, memoirs,
  and novels. But this regularity and correspondence alone will not
  ascertain the truth” (Expostulation, pp. 8, 12, 13). “That there
  are a great many other fables besides that of Troy, regular and
  consistent among themselves, believed and chronologized by the
  Greeks, and even looked up to by them in a religious view (p.
  13), which yet no one now thinks of admitting as history.”

  Morritt, having urged the universal belief of antiquity as
  evidence that the Trojan war was historically real, is met
  by Bryant, who reminds him that the same persons believed in
  centaurs, satyrs, nymphs, augury, aruspicy; Homer maintaining
  that horses could speak, etc. To which Morritt replies, “What has
  religious belief to do with historical facts? Is not the evidence
  on which our faith rests in matters of religion totally different
  in all its parts from that on which we ground our belief in
  history?” (Addit. Remarks, p. 47).

  The separation between the grounds of religious and historical
  belief is by no means so complete as Mr. Morritt supposes, even
  in regard to modern times; and when we apply his position to the
  ancient Greeks, it will be found completely the reverse of the
  truth. The contemporaries of Herodotus and Thucydidês conceived
  their early history in the most intimate conjunction with their
  religion.

Strabo has here converted into geographical matter of fact an
hypothesis purely gratuitous, with a view of saving the accuracy of
the Homeric topography; though in all probability the locality of the
pretended Old Ilium would have been found open to difficulties not
less serious than those which it was introduced to obviate.[811] It
may be true that Dêmêtrius and he were justified in their negative
argument, so as to show that the battles described in the Iliad could
not possibly have taken place if the city of Priam had stood on the
hill inhabited by the Ilieans. But the legendary faith subsisted
before, and continued without abatement afterwards, notwithstanding
such topographical impossibilities. Hellanikus, Herodotus, Mindarus,
the guides of Xerxês, and Alexander, had not been shocked by them: the
case of the latter is the strongest of all, because he had received the
best education of his time under Aristotle—he was a passionate admirer
and constant reader of the Iliad—he was moreover personally familiar
with the movements of armies, and lived at a time when maps, which
began with Anaximander, the disciple of Thalês, were at least known to
all who sought instruction. Now if, notwithstanding such advantages,
Alexander fully believed in the identity of Ilium, unconscious of
these many and glaring topographical difficulties, much less would
Homer himself, or the Homeric auditors, be likely to pay attention to
them, at a period, five centuries earlier, of comparative rudeness and
ignorance, when prose records as well as geographical maps were totally
unknown.[812] The inspired poet might describe, and his hearers would
listen with delight to the tale, how Hectôr, pursued by Achilles, ran
thrice round the city of Troy, while the trembling Trojans were all
huddled into the city, not one daring to come out even at this last
extremity of their beloved prince—and while the Grecian army looked on,
restraining unwillingly their uplifted spears at the nod of Achilles,
in order that Hectôr might perish by no other hand than his; nor were
they, while absorbed by this impressive recital, disposed to measure
distances or calculate topographical possibilities with reference to
the site of the real Ilium.[813] The mistake consists in applying to
Homer and to the Homeric siege of Troy, criticisms which would be
perfectly just if brought to bear on the Athenian siege of Syracuse, as
described by Thucydidês;[814] in the Peloponnesian war[815]—but which
are not more applicable to the epic narrative than they would be to the
exploits of Amadis or Orlando.

  [811] For example, adopting his own line of argument (not to
  mention those battles in which the pursuit and the flight reaches
  from the city to the ships and back again), it might have been
  urged to him, that by supposing the Homeric Troy to be four miles
  farther off from the sea, he aggravated the difficulty of rolling
  the Trojan horse into the town: it was already sufficiently hard
  to propel this vast wooden animal full of heroes from the Greek
  Naustathmon to the town of Ilium.

  The Trojan horse, with its accompaniments Sinon and Laocoôn, is
  one of the capital and indispensable events in the epic: Homer,
  Arktinus, Leschês, Virgil, and Quintus Smyrnæus, all dwell upon
  it emphatically as the proximate cause of the capture.

  The difficulties and inconsistencies of the movements ascribed to
  Greeks and Trojans in the Iliad, when applied to real topography,
  are well set forth in Spohn, _De Agro Trojano_, Leipsic, 1814;
  and Mr. Maclaren has shown (Dissertation on the Topography of the
  Trojan War, Edinburgh, 1822) that these difficulties are nowise
  obviated by removing Ilium a few miles further from the sea.

  [812] Major Rennell argues differently from the visit of
  Alexander, employing it to confute the hypothesis of Chevalier,
  who had placed the Homeric Troy at Bounarbashi, the site supposed
  to have been indicated by Dêmêtrius and Strabo:—

  “Alexander is said to have been a passionate admirer of the
  Iliad, and he had an opportunity of deciding on the spot how far
  the topography was consistent with the narrative. Had he been
  shown the site of Bounarbashi for that of Troy, he would probably
  have questioned the fidelity either of the historical part of the
  poem or his guides. It is not within credibility, that a person
  of so correct a judgment as Alexander could have admired a poem,
  which contained a long history of military details, and other
  transactions that could not physically have had an existence.
  What pleasure could he receive, in contemplating as subjects
  of history, events which could not have happened? Yet he did
  admire the poem, and _therefore must have found the topography
  consistent_: that is, Bounarbashi, surely, was not shown to him
  for Troy.” (Rennell, Observations on the Plain of Troy, p. 128).

  Major Rennell here supposes in Alexander a spirit of
  topographical criticism quite foreign to his real character. We
  have no reason to believe that the site of Bounarbashi was shown
  to Alexander as the Homeric Troy, or that any site was shown
  to him _except Ilium_, or what Strabo calls New Ilium. Still
  less reason have we to believe that any scepticism crossed his
  mind, or that his deep-seated faith required to be confirmed by
  measurement of distances.

  [813] Strabo, xiii. p. 599. Οὐδ᾽ ἡ τοῦ Ἕκτορος δὲ περιδρομὴ ἡ
  περὶ τὴν πόλιν ἔχει τι εὔλογον· οὐ γάρ ἐστι περίδρομος ἡ νῦν, διὰ
  τὴν συνεχῆ ῥάχιν· ἡ δὲ παλαιὰ ἔχει περιδρομήν.

  [814] Mannert (Geographie der Griechen und Römer, th. 6. heft 3.
  b. 8. cap. 8) is confused in his account of Old and New Ilium: he
  represents that Alexander raised up a new spot to the dignity of
  having been the Homeric Ilium, which is not the fact: Alexander
  adhered to the received local belief. Indeed, as far as our
  evidence goes, no one but Dêmêtrius, Hestiæa, and Strabo appears
  ever to have departed from it.

  [815] There can hardly be a more singular example of this same
  confusion, than to find elaborate military criticisms from the
  Emperor Napoleon, upon the description of the taking of Troy in
  the second book of the Æneid. He shows that gross faults are
  committed in it, when looked at from the point of view of a
  general (see an interesting article by Mr. G. C. Lewis, in the
  Classical Museum, vol. i. p. 205, “Napoleon on the Capture of
  Troy”).

  Having cited this criticism from the highest authority on the
  art of war, we may find a suitable parallel in the works of
  distinguished publicists. The attack of Odysseus on the Ciconians
  (described in Homer, Odyss. ix. 39-61) is cited both by Grotius
  (De Jure Bell. et Pac. iii. 3, 10) and by Vattel (Droit des Gens,
  iii. 202) as a case in point in international law. Odysseus is
  considered to have sinned against the rules of international law
  by attacking them as allies of the Trojans, without a formal
  declaration of war.

There is every reason for presuming that the Ilium visited by Xerxês
and Alexander was really the “holy Ilium” present to the mind of
Homer; and if so, it must have been inhabited, either by Greeks or by
some anterior population, at a period earlier than that which Strabo
assigns. History recognizes neither Troy the city, nor Trojans, as
actually existing; but the extensive region called Trôas, or the
Trôad (more properly Trôïas), is known both to Herodotus and to
Thucydidês: it seems to include the territory westward of an imaginary
line drawn from the north-east corner of the Adramyttian gulf to the
Propontis at Parium, since both Antandrus, Kolônæ, and the district
immediately round Ilium, are regarded as belonging to the Trôad.[816]
Herodotus further notices the Teukrians of Gergis[817] (a township
conterminous with Ilium, and lying to the eastward of the road
from Ilium to Abydus), considering them as the remnant of a larger
Teukrian population which once resided in the country, and which
had in very early times undertaken a vast migration from Asia into
Europe.[818] To that Teukrian population he thinks that the Homeric
Trojans belonged:[819] and by later writers, especially by Virgil and
the other Romans, the names Teukrians and Trojans are employed as
equivalents. As the name Trojans is not mentioned in any contemporary
historical monument, so the name _Teukrians_ never once occurs in the
old epic. It appears to have been first noticed by the elegiac poet
Kallinus, about 660 B. C., who connected it by an alleged immigration
of Teukrians from Krête into the region round about Ida. Others again
denied this, asserting that the primitive ancestor, Teukrus, had come
into the country from Attica,[820] or that he was of indigenous origin,
born from Skamander and the nymph Idæa—all various manifestations of
that eager thirst after an eponymous hero which never deserted the
Greeks. Gergithians occur in more than one spot in Æolis, even so far
southward as the neighborhood of Kymê:[821] the name has no place
in Homer, but he mentions Gorgythion and Kebriones as illegitimate
sons of Priam, thus giving a sort of epical recognition both to
Gergis and Kebrên. As Herodotus calls the old epical Trojans by the
name Teukrians, so the Attic Tragedians call them Phrygians; though
the Homeric hymn to Aphroditê represents Phrygians and Trojans as
completely distinct, specially noting the diversity of language;[822]
and in the Iliad the Phrygians are simply numbered among the allies
of Troy from the far Ascania, without indication of any more intimate
relationship.[823] Nor do the tales which connect Dardanus with
Samothrace and Arcadia find countenance in the Homeric poems, wherein
Dardanus is the son of Zeus, having no root anywhere except in
Dardania.[824] The mysterious solemnities of Samothrace, afterwards so
highly venerated throughout the Grecian world, date from a period much
later than Homer; and the religious affinities of that island as well
as of Krête with the territories of Phrygia and Æolis, were certain,
according to the established tendency of the Grecian mind, to beget
stories of a common genealogy.

  [816] Compare Herodot. v. 24-122; Thucyd. i. 131. The Ἰλιὰς γῆ is
  a part of the Trôad.

  [817] Herodot. vii. 43.

  [818] Herodot. v. 122. εἷλε μὲν Αἰολέας πάντας, ὅσοι τὴν Ἰλιάδα
  γῆν νέμονται, εἷλε δὲ Γέργιθας, τοὺς ἀπολειφθέντας τῶν ἀρχαίων
  Τεύκρων.

  For the migration of the Teukrians and Mysians into Europe, see
  Herodot. vii. 20; the Pæonians, on the Strymôn, called themselves
  their descendants.

  [819] Herodot. ii. 118; v. 13.

  [820] Strabo, xiii. p. 604; Apollodôr. iii. 12, 4.

  Kephalôn of Gergis called Teukrus a Krêtan (Stephan. Byz. v.
  Ἀρίσβη).

  [821] Clearchus ap. Athæne. vi. p. 256; Strabo, xiii. p. 589-616.

  [822] Homer, Hymn. in Vener. 116.

  [823] Iliad, ii. 863. Asius, the brother of Hecabê, lives in
  Phrygia on the banks of the Sangarius (Iliad, xvi. 717).

  [824] See Hellanik. Fragm. 129, 130. ed. Didot; and Kephalôn
  Gergithius ap. Steph. Byz. v. Ἀρισβή.

To pass from this legendary world,—an aggregate of streams distinct and
heterogeneous, which do not willingly come into confluence, and cannot
be forced to intermix,—into the clearer vision afforded by Herodotus,
we learn from him that in the year 500 B. C. the whole coast-region
from Dardanus southward to the promontory of Lektum (including the
town of Ilium), and from Lektum eastward to Adramyttium, had been
Æolized, or was occupied by Æolic Greeks—likewise the inland towns
of Skêpsis[825] and Krebên. So that if we draw a line northward from
Adramyttium to Kyzikus on the Propontis, throughout the whole territory
westward from that line, to the Hellespont and the Ægean Sea, all the
considerable towns would be Hellenic, with the exception of Gergis and
the Teukrian population around it,—all the towns worthy of note were
either Ionic or Æolic. A century earlier, the Teukrian population would
have embraced a wider range—perhaps Skêpsis and Krebên, the latter
of which places was colonized by Greeks from Kymê:[826] a century
afterwards, during the satrapy of Pharnabazus, it appears that Gergis
had become Hellenized as well as the rest. The four towns, Ilium,
Gergis, Kebrên and Skêpsis, all in lofty and strong positions, were
distinguished each by a solemn worship and temple of Athênê, and by the
recognition of that goddess as their special patroness.[827]

  [825] Skêpsis received some colonists from the Ionic Miletus
  (Anaximenês apud Strabo, xiv. p. 635); but the coins of the place
  prove that its dialect was Æolic. See Klausen, Æneas und die
  Penaten, tom. i. note 180.

  Arisbê also, near Abydus, seems to have been settled from
  Mitylênê (Eustath. ad Iliad. xii. 97).

  The extraordinary fertility and rich black mould of the plain
  around Ilium is noticed by modern travellers (see Franklin,
  Remarks and Observations on the Plain of Troy, London, 1800, p.
  44): it is also easily worked: “a couple of buffaloes or oxen
  were sufficient to draw the plough, whereas near Constantinople
  it takes twelve or fourteen.”

  [826] Ephôrus ap. Harpocrat. v. Κεβρῆνα.

  [827] Xenoph. Hellen. i. 1, 10; iii. 1, 10-15.

  One of the great motives of Dio in setting aside the Homeric
  narrative of the Trojan war, is to vindicate Athênê from the
  charge of having unjustly destroyed her own city of Ilium (Orat.
  xi. p. 310: μάλιστα διὰ τὴν Ἀθηνᾶν ὅπως μὴ δοκῇ ἀδίκως διαφθεῖραι
  τὴν ἑαυτῆς πόλιν).

The author of the Iliad conceived the whole of this region as occupied
by people not Greek,—Trojans, Dardanians, Lykians, Lelegians,
Pelasgians, and Kilikians. He recognizes a temple and worship of Athênê
in Ilium, though the goddess is bitterly hostile to the Trojans: and
Arktinus described the Palladium as the capital protection of the city.
But perhaps the most remarkable feature of identity between the Homeric
and the historical Æolis, is, the solemn and diffused worship of the
Sminthian Apollo. Chrysê, Killa and Tenedos, and more than one place
called Sminthium, maintain the surname and invoke the protection of
that god during later times, just as they are emphatically described to
do by Homer.[828]

  [828] Strabo, x. p. 473; xiii. p. 604-605. Polemon. Fragm. 31. p.
  63, ed. Preller.

  Polemon was a native of Ilium, and had written a periegesis of
  the place (about 200 B. C., therefore earlier than Dêmêtrius of
  Skêpsis): he may have witnessed the improvement in its position
  effected by the Romans. He noticed the identical stone upon which
  Palamêdês had taught the Greeks to play at dice.

  The Sminthian Apollo appears inscribed on the coins of
  Alexandreia Trôas; and the temple of the god was memorable even
  down to the time of the emperor Julian (Ammian. Marcellin. xxii.
  8). Compare Menander (the Rhetor) περὶ Ἐπιδεικτικῶν, iv. 14; apud
  Walz. Collect. Rhetor. t. ix. p. 304; also περὶ Σμινθιακῶν, iv.
  17.

  Σμίνθος, both in the Krêtan and the Æolic dialect, meant a
  _field-mouse_: the region seems to have been greatly plagued by
  these little animals.

  Polemon could not have accepted the theory of Dêmêtrius, that
  Ilium was not the genuine Troy: his Periegesis, describing the
  localities and relics of Ilium, implied the legitimacy of the
  place as a matter of course.

When it is said that the Post-Homeric Greeks gradually Hellenized
this entire region, we are not to understand that the whole previous
population either retired or was destroyed. The Greeks settled in the
leading and considerable towns, which enabled them both to protect one
another and to gratify their predominant tastes. Partly by force—but
greatly also by that superior activity, and power of assimilating
foreign ways of thought to their own, which distinguished them from
the beginning—they invested all the public features and management of
the town with an Hellenic air, distributed all about it their gods,
their heroes and their legends, and rendered their language the medium
of public administration, religious songs and addresses to the gods,
and generally for communications wherein any number of persons were
concerned. But two remarks are here to be made: first, in doing this
they could not avoid taking to themselves more or less of that which
belonged to the parties with whom they fraternized, so that the result
was not pure Hellenism; next, that even this was done only in the
towns, without being fully extended to the territorial domain around,
or to those smaller townships which stood to the town in a dependent
relation. The Æolic and Ionic Greeks borrowed from the Asiatics whom
they had Hellenized, musical instruments and new laws of rhythm and
melody, which they knew how to turn to account: they further adopted
more or less of those violent and maddening religious rites, manifested
occasionally in self-inflicted suffering and mutilation, which were
indigenous in Asia Minor in the worship of the Great Mother. The
religion of the Greeks in the region of Ida as well as at Kyzikus was
more orgiastic than the native worship of Greece Proper, just as that
of Lampsacus, Priapus and Parium was more licentious. From the Teukrian
region of Gergis, and from the Gergithes near Kymê, sprang the original
Sibylline prophecies, and the legendary Sibyll who plays so important
a part in the tale of Æneas: the mythe of the Sibyll, whose prophecies
are supposed to be heard in the hollow blast bursting out from obscure
caverns and apertures in the rocks,[829] was indigenous among the
Gergithian Teukrians, and passed from the Kymæans in Æolis, along with
the other circumstances of the tale of Æneas, to their brethren the
inhabitants of Cumæ in Italy. The date of the Gergithian Sibyll, or
rather of the circulation of her supposed prophecies, is placed during
the reign of Crœsus, a period when Gergis was thoroughly Teukrian.
Her prophecies, though embodied in Greek verses, had their root in a
Teukrian soil and feelings; and the promises of future empire which
they so liberally make to the fugitive hero escaping from the flames of
Troy into Italy, become interesting from the remarkable way in which
they were realized by Rome.[830]

  [829] Virgil, Æneid, vi. 42:—

    Excisum Euboicæ latus ingens rupis in antrum,
    Quo lati ducunt aditus centum, ostia centum;
    Unde ruunt totidem voces, responsa Sibyllæ.

  [830] Pausanias, x. 12, 8; Lactantius, i. 6, 12; Steph. Byz. v.
  Μέρμησσος; Schol. Plat. Phædr. p. 315, Bekker.

  The date of this Gergithian Sibyll, or of the prophecies passing
  under her name, is stated by Hêrakleidês of Pontus, and there
  seems no reason for calling it in question.

  Klausen (Æneas und die Penaten, book ii. p. 205) has worked out
  copiously the circulation and legendary import of the Sibylline
  prophecies.

At what time Ilium and Dardanus became Æolized we have no information.
We find the Mitylenæans in possession of Sigeium in the time of the
poet Alkæus, about 600 B. C.; and the Athenians during the reign of
Peisistratus, having wrested it from them and trying to maintain their
possession, vindicate the proceeding by saying that they had as much
right to it as the Mitylenæans, “for the latter had no more claim to
it than any of the other Greeks who had aided Menelaus in avenging
the abduction of Helen.”[831] This is a very remarkable incident,
as attesting the celebrity of the legend of Troy, and the value of
a mythical title in international disputes—yet seemingly implying
that the establishment of the Mitylenæans on that spot must have been
sufficiently recent. The country near the junction of the Hellespont
and the Propontis is represented as originally held[832] by Bebrykian
Thracians, while Abydus was first occupied by Milesian colonists in
the reign and by the permission of the Lydian king Gygês[833]—to whom
the whole Trôad and the neighboring territory belonged, and upon
whom therefore the Teukrians of Ida must have been dependent. This
must have been about 700 B. C., a period considerably earlier than
the Mitylenæan occupation of Sigeium. Lampsacus and Pæsus, on the
neighboring shores of the Propontis, were also Milesian colonies,
though we do not know their date: Parium was jointly settled from
Miletus, Erythræ and Parus.

  [831] Herodot. v. 94. Σίγειον ... τὸ εἷλε Πεισίστρατος αἰχμῇ
  παρὰ Μιτυληναίων ... Ἀθηναῖοι, ἀποδεικνύντες λόγῳ οὐδὲν μᾶλλον
  Αἰολεῦσι μετεὸν τῆς Ἰλιάδος χώρης, ἢ οὐ καὶ σφι καὶ τοῖσι
  ἄλλοισι, ὅσοι Ἑλλήνων συνεξεπρήξαντο Μενέλεῳ τὰς Ἑλένης ἁρπαγάς.
  In Æschylus (Eumenid. 402) the goddess Athênê claims the land
  about the Skamander, as having been presented to the sons of
  Thêseus by the general vote of the Grecian chiefs:—

    Ἀπὸ Σκαμάνδρου γῆν καταφθατουμένη,
    Ἣν δὴ τ᾽ Ἀχαιῶν ἄκτορές τε καὶ πρόμοι
    Τῶν αἰχμαλώτων χρημάτων λάχος μέγα,
    Ἔνειμαν αὐτόπρεμνον εἰς τὸ πᾶν ἐμοὶ,
    Ἐξαιρετὸν δώρημα Θησέως τόκοις.

  In the days of Peisistratus, it seems Athens was not bold enough
  or powerful enough to advance this vast pretension.

  [832] Charôn of Lampsacus ap. Schol. Apollôn. Rhod. ii. 2;
  Bernhardy ad Dionys. Periêgêt. 805. p. 747.

  [833] Such at least is the statement of Strabo (xii. p. 590);
  though such an extent of Lydian role at that time seems not easy
  to reconcile with the proceedings of the subsequent Lydian kings.




CHAPTER XVI.

GRECIAN MYTHES, AS UNDERSTOOD, FELT AND INTERPRETED BY THE GREEKS
THEMSELVES.


The preceding sections have been intended to exhibit a sketch of that
narrative matter, so abundant, so characteristic and so interesting,
out of which early Grecian history and chronology have been extracted.
Raised originally by hands unseen and from data unassignable, it
existed first in the shape of floating talk among the people, from
whence a large portion of it passed into the song of the poets, who
multiplied, transformed and adorned it in a thousand various ways.

These mythes or current stories, the spontaneous and earliest growth of
the Grecian mind, constituted at the same time the entire intellectual
stock of the age to which they belonged. They are the common root of
all those different ramifications into which the mental activity of
the Greeks subsequently diverged; containing, as it were, the preface
and germ of the positive history and philosophy, the dogmatic theology
and the professed romance, which we shall hereafter trace each in
its separate development. They furnished aliment to the curiosity,
and solution to the vague doubts and aspirations of the age; they
explained the origin of those customs and standing peculiarities
with which men were familiar; they impressed moral lessons, awakened
patriotic sympathies, and exhibited in detail the shadowy, but anxious
presentiments of the vulgar as to the agency of the gods: moreover they
satisfied that craving for adventure and appetite for the marvellous,
which has in modern times become the province of fiction proper.

It is difficult, we may say impossible, for a man of mature age to
carry back his mind to his conceptions such as they stood when he was a
child, growing naturally out of his imagination and feelings, working
upon a scanty stock of materials, and borrowing from authorities whom
he blindly followed but imperfectly apprehended. A similar difficulty
occurs when we attempt to place ourselves in the historical and
quasi-philosophical point of view which the ancient mythes present to
us. We can follow perfectly the imagination and feeling which dictated
these tales, and we can admire and sympathize with them as animated,
sublime, and affecting poetry; but we are too much accustomed to matter
of fact and philosophy of a positive kind, to be able to conceive a
time when these beautiful fancies were construed literally and accepted
as serious reality.

Nevertheless it is obvious that Grecian mythes cannot be either
understood or appreciated except with reference to the system of
conceptions and belief of the ages in which they arose. We must
suppose a public not reading and writing, but seeing, hearing and
telling—destitute of all records, and careless as well as ignorant of
positive history with its indispensable tests, yet at the same time
curious and full of eagerness for new or impressive incidents—strangers
even to the rudiments of positive philosophy and to the idea of
invariable sequences of nature either in the physical or moral world,
yet requiring some connecting theory to interpret and regularize the
phænomena before them. Such a theory was supplied by the spontaneous
inspirations of an early fancy, which supposed the habitual agency
of beings intelligent and voluntary like themselves, but superior in
extent of power, and different in peculiarity of attributes. In the
geographical ideas of the Homeric period, the earth was flat and round,
with the deep and gentle ocean-stream flowing around and returning into
itself: chronology, or means of measuring past time, there existed
none; but both unobserved regions might be described, the forgotten
past unfolded, and the unknown future predicted—through particular men
specially inspired by the gods, or endowed by them with that peculiar
vision which detected and interpreted passing signs and omens.

If even the rudiments of scientific geography and physics, now so
universally diffused and so invaluable as a security against error and
delusion, were wanting in this early stage of society, their place was
abundantly supplied by vivacity of imagination and by personifying
sympathy. The unbounded tendency of the Homeric Greeks to multiply
fictitious persons, and to construe the phænomena which interested them
into manifestations of design, is above all things here to be noticed,
because the form of personal narrative, universal in their mythes,
is one of its many manifestations. Their polytheism (comprising some
elements of an original fetichism, in which particular objects had
themselves been supposed to be endued with life, volition, and design)
recognized agencies of unseen beings identified and confounded with the
different localities and departments of the physical world. Of such
beings there were numerous varieties, and many gradations both in power
and attributes; there were differences of age, sex and local residence,
relations both conjugal and filial between them, and tendencies
sympathetic as well as repugnant. The gods formed a sort of political
community of their own, which had its hierarchy, its distribution of
ranks and duties, its contentions for power and occasional revolutions,
its public meetings in the agora of Olympus, and its multitudinous
banquets or festivals.[834] The great Olympic gods were in fact only
the most exalted amongst an aggregate of quasi-human or ultra-human
personages,—dæmons, heroes, nymphs, eponymous (or name-giving) genii,
identified with each river, mountain,[835] cape, town, village, or
known circumscription of territory,—besides horses, bulls, and dogs,
of immortal breed and peculiar attributes, and monsters of strange
lineaments and combinations, “Gorgons and Harpies and Chimæras dire.”
As there were in every _gens_ or family special gentile deities and
foregone ancestors who watched over its members, forming in each the
characteristic symbol and recognized guarantee of their union, so there
seem to have been in each guild or trade peculiar beings whose vocation
it was to coöperate or to impede in various stages of the business.[836]

  [834] Homer, Iliad, i. 603; xx. 7. Hesiod, Theogon. 802.

  [835] We read in the Iliad that Asteropæus was grandson of the
  beautiful river Axius, and Achilles, after having slain him,
  admits the dignity of this parentage, but boasts that his own
  descent from Zeus was much greater, since even the great river
  Achelôus and Oceanus himself is inferior to Zeus (xxi. 157-191).
  Skamander fights with Achilles, calling his brother Simoïs to his
  aid (213-308). Tyrô, the daughter of Salmôneus, falls in love
  with Enipeus, the most beautiful of rivers (Odyss. xi. 237).
  Achelôus appears as a suitor of Deianira (Sophokl. Trach. 9).

  There cannot be a better illustration of this feeling than what
  is told of the New Zealanders at the present time. The chief
  Heu-Heu appeals to his ancestor, the great mountain Tonga Riro:
  “I am the Heu-Heu, and rule over you all, just as my ancestor
  Tonga Riro, the mountain of snow, stands above all this land.”
  (E. J. Wakefield, Adventures in New Zealand, vol. i. ch. 17.
  p. 465). Heu-Heu refused permission to any one to ascend the
  mountain, on the ground that it was his _tipuna_ or ancestor: “he
  constantly identified himself with the mountain and called it
  his sacred ancestor” (vol. ii. c. 4. p. 113). The mountains in
  New Zealand are accounted by the natives masculine and feminine:
  Tonga Riro, and Taranaki, two male mountains, quarrelled about
  the affections of a small volcanic female mountain in the
  neighborhood (_ibid._ ii. c. 4. p. 97).

  The religious imagination of the Hindoos also (as described by
  Colonel Sleeman in his excellent work, Rambles and Recollections
  of an Indian Official), affords a remarkable parallel to that of
  the early Greeks. Colonel Sleeman says,—

  “I asked some of the Hindoos about us why they called the river
  Mother Nerbudda, if she was really never married. Her Majesty
  (said they with great respect) would really never consent to
  be married after the indignity she suffered from her affianced
  bridegroom the Sohun: and we call her _mother_ because she
  blesses us all, and we are anxious to accost her by the name
  which we consider to be the most respectful and endearing.

  “Any Englishman can easily conceive a poet in his highest
  calenture of the brain, addressing the Ocean as a steed that
  knows his rider, and patting the crested billow as his flowing
  mane. But he must come to India to understand how every
  individual of a _whole community of many millions can address
  a fine river as a living being—a sovereign princess who hears
  and understands all they say, and exercises a kind of local
  superintendence over their affairs_, without a single temple in
  which her image is worshipped, or a single priest to profit by
  the delusion. As in the case of the Ganges, _it is the river
  itself to whom they address themselves, and not to any deity
  residing in it, or presiding over it_—the stream itself is the
  deity which fills their imaginations, and receives their homage”
  (Rambles and Recollections of an Indian Official, ch. iii. p.
  20). Compare also the remarks in the same work on the sanctity
  of _Mother Nerbudda_ (chapter xxvii. p. 261); also of the holy
  personality of the earth. “The land is considered as the MOTHER
  of the prince or chief who holds it, the great parent from
  whom he derives all that maintains him, his family, and his
  establishments. If well-treated, she yields this in abundance
  to her son; but if he presumes to look upon her with the eye of
  _desire_, she ceases to be fruitful; or the Deity sends down
  hail or blight to destroy all that she yields. The measuring the
  surface of the fields, and the frequently inspecting the crops
  by the chief himself or his immediate agents, were considered by
  the people in this light—either it should not be done at all,
  or the duty should be delegated to inferior agents, whose close
  inspection of the _great parent_ could not be so displeasing to
  the Deity” (Ch. xxvii. p. 248).

  See also about the gods who are believed to reside in trees—the
  Peepultree, the cotton-tree, etc. (ch. ix. p. 112), and the
  description of the annual marriage celebrated between the sacred
  pebble, or pebble-god, Saligram, and the sacred shrub Toolsea,
  celebrated at great expense and with a numerous procession (chap.
  xix. p. 158; xxiii. p. 185).

  [836] See the song to the potters, in the Homeric Epigrams (14):—

    Εἰ μὲν δώσετε μίσθον, ἀείσω, ὦ κεραμῆες·
    Δεῦρ᾽ ἄγ᾽ Ἀθηναίη, καὶ ὑπείρεχε χεῖρα καμίνου.
    Εὖ δὲ μελανθεῖεν κότυλοι, καὶ πάντα κάναστρα
    Φρυχθῆναί τε καλῶς, καὶ τιμῆς ὦνον ἀρέσθαι.
    ... Ἦν δ᾽ ἐπ᾽ ἀναιδείην τρεφθέντες ψευδῆ ἄρῃσθε,
    Συγκαλέω δὴ ᾽πειτα καμίνῳ δηλητῆρας·
    Σύντριβ᾽ ὅμως, Σμάραγόν τε, καὶ Ἄσβετον, ἠδὲ Σαβάκτην,
    Ὠμόδαμόν θ᾽, ὃς τῇδε τέχνῃ κακὰ πολλὰ πορίζει, etc.

  A certain kindred between men and serpents (συγγένειάν τινα πρὸς
  τοὺς ὄφεις) was recognized in the peculiar gens of the ὀφιογενεῖς
  near Parion, who possessed the gift of healing by their touches
  the bite of the serpent: the original hero of this gens was said
  to have been transformed from a serpent into a man (Strabo, xiii.
  p. 588).

The extensive and multiform personifications, here faintly sketched,
pervaded in every direction the mental system of the Greeks, and
were identified intimately both with their conception and with their
description of phenomena, present as well as past. That which to
us is interesting as the mere creation of an exuberant fancy, was
to the Greek genuine and venerated reality. Both the earth and the
solid heaven (Gæa and Uranos) were both conceived and spoken of by
him as endowed with appetite, feeling, sex, and most of the various
attributes of humanity. Instead of a sun such as we now see, subject
to astronomical laws, and forming the centre of a system the changes
of which we can ascertain and foreknow, he saw the great god Hêlios,
mounting his chariot in the morning in the east, reaching at mid-day
the height of the solid heaven, and arriving in the evening at the
western horizon, with horses fatigued and desirous of repose. Hêlios,
having favorite spots wherein his beautiful cattle grazed, took
pleasure in contemplating them during the course of his journey, and
was sorely displeased if any man slew or injured them: he had moreover
sons and daughters on earth, and as his all-seeing eye penetrated
everywhere, he was sometimes in a situation to reveal secrets even to
the gods themselves—while on other occasions he was constrained to turn
aside in order to avoid contemplating scenes of abomination.[837] To us
these now appear puerile, though pleasing fancies, but to an Homeric
Greek they seemed perfectly natural and plausible. In his view, the
description of the sun, as given in a modern astronomical treatise,
would have appeared not merely absurd, but repulsive and impious.
Even in later times, when the positive spirit of inquiry had made
considerable progress, Anaxagoras and other astronomers incurred the
charge of blasphemy for dispersonifying Hêlios, and trying to assign
invariable laws to the solar phænomena.[838] Personifying fiction was
in this way blended by the Homeric Greeks with their conception of
the physical phænomena before them, not simply in the way of poetical
ornament, but as a genuine portion of their every-day belief.

  [837] Odyss. ii. 388; viii. 270; xii. 4, 128, 416; xxiii. 362.
  Iliad, xiv. 344. The Homeric Hymn to Dêmêtêr expresses it neatly
  (63)—

    Ἡέλιον δ᾽ ἵκοντο, θεῶν σκόπον ἠδὲ καὶ ἀνδρῶν.

  Also the remarkable story of Euênius of Apollônia, his neglect
  of the sacred cattle of Hêlios, and the awful consequences of it
  (Herodot. ix. 93: compare Theocr. Idyll, xxv. 130).

  I know no passage in which this conception of the heavenly bodies
  as Persons is more strikingly set forth than in the words of
  the German chief Boiocalus, pleading the cause of himself and
  his tribe the Ansibarii before the Roman legate Avitus. This
  tribe, expelled by other tribes from its native possessions,
  had sat down upon some of that wide extent of lands on the
  Lower Rhine which the Roman government reserved for the use of
  its soldiers, but which remained desert, because the soldiers
  had neither the means nor the inclination to occupy them. The
  old chief, pleading his cause before Avitus, who had issued
  an order to him to evacuate the lands, first dwelt upon his
  fidelity of fifty years to the Roman cause, and next touched
  upon the enormity of retaining so large an area in a state of
  waste (Tacit. Ann. xiii. 55): “Quotam partem campi jacere, in
  quam pecora et armenta militum aliquando transmitterentur?
  Servarent sane receptos gregibus, inter hominum famam: modo ne
  vastitatem et solitudinem mallent, quam amicos populos Chamavorum
  quondam ea arva, mox Tubantum, et post Usipiorum fuisse. Sicuti
  cœlum Diis, ita terras generi mortalium datas: quæque vacuæ,
  eas publicas esse. _Solem_ deinde respiciens, et _cœtera sidera
  vocans, quasi coram_ interrogabat—_vellentne contueri inane
  solum? potius mare superfunderent adversus terrarum ereptores_.
  Commotus his Avitus,” etc. The legate refused the request, but
  privately offered to Boiocalus lands for himself apart from the
  tribe, which that chief indignantly spurned. He tried to maintain
  himself in the lands, but was expelled by the Roman arms, and
  forced to seek a home among the other German tribes, all of whom
  refused it. After much wandering and privation, the whole tribe
  of the Ansibarii was annihilated: its warriors were all slain,
  its women and children sold as slaves.

  I notice this afflicting sequel, in order to show that the
  brave old chief was pleading before Avitus a matter of life and
  death both to himself and his tribe, and that the occasion was
  one least of all suited for a mere rhetorical prosopopœia. His
  appeal is one sincere and heartfelt to the personal feelings and
  sympathies of Hêlios.

  Tacitus, in reporting the speech, accompanies it with the
  gloss “quasi coram,” to mark that the speaker here passes into
  a different order of ideas from that to which himself or his
  readers were accustomed. If Boiocalus could have heard, and
  reported to his tribe, an astronomical lecture, he would have
  introduced some explanation, in order to facilitate to his tribe
  the comprehension of Hêlios under a point of view so new to them.
  While Tacitus finds it necessary to illustrate by a comment the
  _personification of the sun_, Boiocalus would have had some
  trouble to make his tribe comprehend the _re-ification of the god
  Hêlios_.

  [838] Physical astronomy was both new and accounted impious
  in the time of the Peloponnesian war: see Plutarch, in his
  reference to that eclipse which proved so fatal to the Athenian
  army at Syracuse, in consequence of the religious feelings of
  Nikias: οὐ γὰρ ἠνείχοντο τοὺς φυσικοὺς καὶ μετεωρολέσχας τότε
  καλουμένους ὡς, εἰς αἰτίας ἀλόγους καὶ δυνάμεις ἀπρονοήτους καὶ
  κατηναγκασμένα πάθη διατρίβοντας τὸ θεῖον (Plutarch, Nikias, c.
  23, and Periklês, c. 32; Diodôr. xii. 39; Dêmêtr. Phaler. ap.
  Diogen. Laërt, ix. 9, 1).

  “You strange man, Melêtus,” said Socratês, on his trial, to his
  accuser, “are you seriously affirming that I do not think Hêlios
  and Selênê to be gods, as the rest of mankind think?” “Certainly
  not, gentlemen of the Dikastery (_this is the reply of Melêtus_),
  Socratês says that the sun is a stone, and the moon earth.” “Why,
  my dear Melêtus, you think you are preferring an accusation
  against Anaxagoras! You account these Dikasts so contemptibly
  ignorant, as not to know that the books of Anaxagoras are
  full of such doctrines! Is it from me that the youth acquire
  such teaching, when they may buy the books for a drachma in
  the theatre, and may thus laugh me to scorn if I pretended to
  announce such views as my own—_not to mention their extreme
  absurdity_?” (ἄλλως τε καὶ οὕτως ἄτοπα ὄντα, Plato, Apolog.
  Socrat. c. 14. p. 26).

  The divinity of Hêlios and Selênê is emphatically set forth by
  Plato, Legg. x. p. 886-889. He permits physical astronomy only
  under great restrictions and to a limited extent. Compare Xenoph.
  Memor. iv. 7, 7; Diogen. Laërt. ii. 8; Plutarch, De Stoicor.
  Repugnant. c. 40. p. 1053; and Schaubach ad Anaxagoræ Fragmenta,
  p. 6.

It was in this early state of the Grecian mind, stimulating so forcibly
the imagination and the feelings, and acting through them upon the
belief, that the great body of the mythes grew up and obtained
circulation. They were, from first to last, personal narratives and
adventures; and the persons who predominated as subjects of them
were the gods, the heroes, the nymphs, etc., whose names were known
and reverenced, and in whom every one felt interested. To every god
and every hero it was consistent with Grecian ideas to ascribe great
diversity of human motive and attribute: each indeed has his own
peculiar type of character, more or less strictly defined; but in all
there was a wide foundation for animated narrative and for romantic
incident. The gods and heroes of the land and the tribe belonged, in
the conception of a Greek, alike to the present and to the past: he
worshipped in their groves and at their festivals; he invoked their
protection, and believed in their superintending guardianship, even
in his own day: but their more special, intimate, and sympathizing
agency was cast back into the unrecorded past.[839] To give suitable
utterance to this general sentiment,—to furnish body and movement and
detail to these divine and heroic pre-existences, which were conceived
only in shadowy outline,—to lighten up the dreams of what the past must
have been,[840] in the minds of those who knew not what it really had
been—such was the spontaneous aim and inspiration of productive genius
in the community, and such were the purposes which the Grecian mythes
preëminently accomplished.

  [839] Hesiod, Catalog. Fragm. 76. p. 48, ed. Düntzer:—

    Ξυναὶ γὰρ τότε δαῖτες ἔσαν ξυνοί τε θόωκει,
    Ἀθανάτοις τε θοῖσι καταθνήτοις τ᾽ ἀνθρώποις.

  Both the Theogonia and the Works and Days bear testimony to
  the same general feeling. Even the heroes of Homer suppose a
  preceding age, the inmates of which were in nearer contact with
  the gods than they themselves (Odyss. viii. 223; Iliad, v. 304;
  xii. 382). Compare Catullus, Carm. 64; Epithalam. Peleôs et
  Thetidos, v. 382-408.

  Menander the Rhetor (following generally the steps of Dionys.
  Hal. Art Rhetor. cap. 1-8) suggests to his fellow-citizens at
  Alexandria Trôas, proper and complimentary forms to invite a
  great man to visit their festival of the Sminthia:—ὥσπερ γὰρ
  Ἀπόλλωνα πολλάκις ἐδέχετο ἡ πόλις τοῖς Σμινθίοις, ~ἥνικα ἐξῆν
  θεοὺς προφανῶς ἐπιδημεῖν τοῖς ἀνθρώποις~, οὕτω καὶ σὲ ἡ πόλις
  νῦν προσδέχεται (περὶ Ἐπιδεικτικ. s. iv. c. 14. ap. Walz. Coll.
  Rhetor, t. ix. p. 304). Menander seems to have been a native
  of Alexandria Trôas, though Suidas calls him a Laodicean (see
  Walz. Præf. ad t. ix. p. xv.-xx.; and περὶ Σμινθιακῶν, sect. iv.
  c. 17). The festival of the Sminthia lasted down to his time,
  embracing the whole duration of paganism from Homer downwards.

  [840] P. A. Müller observes justly, in his _Saga-Bibliothek_, in
  reference to the Icelandic mythes, “In dem Mythischen wird das
  Leben der Vorzeit dargestellt, wie es wirklich dem kindlichen
  Verstande, der jugendlichen Einbildungskraft, und dem vollen
  Herzen, erscheint.”

  (Lange’s Untersuchungen über die Nordische und Deutsche
  Heldensage, translated from P. A. Müller, Introd. p. 1.)

The love of antiquities, which Tacitus notices as so prevalent among
the Greeks of his day,[841] was one of the earliest, the most durable,
and the most widely diffused of the national propensities. But the
antiquities of every state were divine and heroic, reproducing the
lineaments, but disregarding the measure and limits, of ordinary
humanity. The gods formed the starting-point, beyond which no man
thought of looking, though some gods were more ancient than others:
their progeny, the heroes, many of them sprung from human mothers,
constitute an intermediate link between god and man. The ancient epic
usually recognizes the presence of a multitude of nameless men, but
they are introduced chiefly for the purpose of filling the scene,
and of executing the orders, celebrating the valor, and bringing
out the personality, of a few divine or heroic characters.[842] It
was the glory of bards and storytellers to be able to satisfy those
religious and patriotic predispositions of the public, which caused the
primary demand for their tales, and which were of a nature eminently
inviting and expansive. For Grecian religion was many-sided and many
colored; it comprised a great multiplicity of persons, together with
much diversity in the types of character; it divinized every vein
and attribute of humanity, the lofty as well as the mean—the tender
as well as the warlike—the self-devoting and adventurous as well as
the laughter-loving and sensual. We shall hereafter reach a time when
philosophers protested against such identification of the gods with
the more vulgar appetites and enjoyments, believing that nothing
except the spiritual attributes of man could properly be transferred
to superhuman beings, and drawing their predicates respecting the gods
exclusively from what was awful, majestic and terror-striking in human
affairs. Such restrictions on the religious fancy were continually
on the increase, and the mystic and didactic stamp which marked the
last century of paganism in the days of Julian and Libanius, contrasts
forcibly with the concrete and vivacious forms, full of vigorous
impulse and alive to all the capricious gusts of the human temperament,
which people the Homeric Olympus.[843] At present, however, we have
only to consider the early, or Homeric and Hesiodic paganism, and its
operation in the genesis of the mythical narratives. We cannot doubt
that it supplied the most powerful stimulus, and the only one which the
times admitted, to the creative faculty of the people; as well from the
sociability, the gradations, and the mutual action and reaction of its
gods and heroes, as from the amplitude, the variety, and the purely
human cast, of its fundamental types.

  [841] Titus visited the temple of the Paphian Venus in Cyprus,
  “spectatâ opulentiâ donisque regum, quæque _alia lætum_
  antiquitatibus Græcorum genus _incertæ vetustati adfingit_, de
  navigatione primum consuluit” (Tacit. Hist. ii. 4-5).

  [842] Aristotel. Problem. xix. 48. Οἱ δὲ ἡγεμόνες τῶν ἀρχαίων
  μόνοι ἦσαν ἥρωες· οἱ δὲ λαοὶ ἄνθρωποι. Istros followed this
  opinion also: but the more common view seems to have considered
  all who combated at Troy as heroes (see Schol. Iliad, ii. 110;
  xv. 231), and so Hesiod treats them (Opp. Di. 158).

  In reference to the Trojan war, Aristotle says—καθάπερ ἐν τοῖς
  ~Ἡρωϊκοῖς~ περὶ Πριάμου μυθεύεται (Ethic. Nicom. i. 9; compare
  vii. 1).

  [843] Generation by a god is treated in the old poems as an act
  entirely human and physical (ἐμιγη—παρελέξατο); and this was the
  common opinion in the days of Plato (Plato, Apolog. Socrat. c.
  15. p. 15); the hero Astrabakus is father of the Lacedæmonian
  king Demaratus (Herod. vi. 66). [Herodotus does not believe the
  story told him at Babylon respecting Belus (i. 182).] Euripidês
  sometimes expresses disapprobation of the idea (Ion. 350), but
  Plato passed among a large portion of his admirers for the
  actual son of Apollo, and his reputed father Aristo on marrying
  was admonished in a dream to respect the person of his wife
  Periktionê, then pregnant by Apollo, until after the birth of the
  child Plato (Plutarch, Quæst. Sympos. p. 717. viii. 1; Diogen.
  Laërt. iii. 2; Origen, cont. Cels. i. p. 29). Plutarch (in Life
  of Numa, c. 4; compare Life of Thêseus, 2) discusses the subject,
  and is inclined to disallow everything beyond mental sympathy
  and tenderness in a god: Pausanias deals timidly with it, and
  is not always consistent with himself; while the later rhetors
  spiritualize it altogether. Meander, περὶ Ἐπιδεικτικῶν, (towards
  the end of the third century B. C.) prescribes rules for praising
  a king: you are to praise him for the gens to which he belongs:
  perhaps you may be able to make out that he really is the son
  of some god; for many who seem to be from men, are really _sent
  down by God_ and are _emanations from the Supreme Potency_—πολλοὶ
  τὸ μὲν δοκεῖν ἐξ ἀνθρώπων εἰσὶ, τῇ δ᾽ ἀληθείᾳ παρὰ τοῦ θεοῦ
  καταπέμπονται καὶ εἰσιν ἀπόῤῥοιαι ὄντως τοῦ κρείττονος· καὶ γὰρ
  Ἡρακλῆς ἐνομίζετο μὲν Ἀμφιτρύωνος, τῇ δὲ ἀληθείᾳ ἦν Διός. Οὕτω
  καὶ βασιλεὺς ὁ ἡμέτερος το μὲν δοκεῖν ἐξ ἀνθρώπων, τῇ δὲ ἀληθείᾳ
  τὴν καταβολὴν οὐράνοθεν ἔχει, etc. (Menander ap. Walz. Collect.
  Rhetor. t. ix. c. i. p. 218). Again—περὶ Σμινθιακῶν—Ζεὺς γένεσιν
  παιδῶν ~δημιουργεῖν~ ἐνενόησε—Ἀπόλλων τὴν Ἀσκληπιοῦ γένεσιν
  ~ἐδημιούργησε~, p. 322-327; compare Hermogenês, about the story
  of Apollo and Daphnê, Progymnasm. c. 4; and Julian. Orat. vii. p.
  220.

  The contrast of the pagan phraseology of this age (Menander had
  himself composed a hymn of invocation to Apollo—περὶ Ἐγκωμίων,
  c. 3. t. ix. p. 136, Walz.) with that of Homer is very worthy
  of notice. In the Hesiodic Catalogue of Women much was said
  respecting the marriages and amours of the gods, so as to furnish
  many suggestions, like the love-songs of Sapphô, to the composers
  of Epithalamic Odes (Menand. _ib._ sect. iv. c. 6. p. 268).

  Menander gives a specimen of a prose hymn fit to be addressed to
  the Sminthian Apollo (p. 320); the spiritual character of which
  hymn forms the most pointed contrast with the Homeric hymn to the
  same god.

  We may remark an analogous case in which the Homeric hymn to
  Apollo is modified by Plutarch. To provide for the establishment
  of his temple at Delphi, Apollo was described as having himself,
  in the shape of a dolphin, swam before a Krêtan vessel and guided
  it to Krissa, where he directed the terrified crew to open the
  Delphian temple. But Plutarch says that this old statement was
  not correct: the god had not himself appeared in the shape of
  a dolphin—he had sent a dolphin expressly to guide the vessel
  (Plutarch. de Solertiâ Animal. p. 983). See also a contrast
  between the Homeric Zeus, and the genuine Zeus, (ἀληθινὸς)
  brought out in Plutarch, Defect. Oracul. c 30. p. 426.

  Illicit amours seem in these later times to be ascribed to
  the δαίμονες: see the singular controversy started among the
  fictitious pleadings of the ancient rhetors—Νόμου ὄντος,
  παρθένους καὶ καθαρὰς εἶναι τὰς ἱερείας, ἱερεία τις εὑρέθη
  ἀτόκιον φέρουσα, καὶ κρίνεται.... Ἀλλ᾽ ἐρεῖ, φασὶ, διὰ τὰς τῶν
  δαιμόνων ἐπιφοιτήσεις καὶ ἐπιβουλὰς περιτεθεῖσθαι. Καὶ πῶς οὐκ
  ἀνόητον κομιδῆ τὸ τοιοῦτον; ἔδει γὰρ πρὸς τὸ μὴ ἀφαιρεθῆναι τὴν
  παρθενίαν φορεῖν τι ἀποτρόπαιον, οὐ μὴν πρὸς τὸ τεκεῖν (Anonymi
  Scholia ad Hermogen. Στάσεις, ap. Walz. Coll. Rh. t. vii. p. 162).

  Apsinês of Gadara, a sophist of the time of Diocletian, pretended
  to be a son of Pan (see Suidas, v. Ἀψίνης). The anecdote
  respecting the rivers Skamander and Mæander, in the tenth epistle
  ascribed to the orator Æschines (p. 737), is curious, but we do
  not know the date of that epistle.

Though we may thus explain the mythopœic fertility of the Greeks, I am
far from pretending that we can render any sufficient account of the
supreme beauty of their chief epic and artistical productions. There is
something in the first-rate productions of individual genius which lies
beyond the compass of philosophical theory: the special breath of the
Muse (to speak the language of ancient Greece) must be present in order
to give them being. Even among her votaries, many are called, but few
are chosen; and the peculiarities of those few remain as yet her own
secret.

We shall not however forget that Grecian language was also an
indispensable requisite to the growth and beauty of Grecian mythes—its
richness, its flexibility and capacity of new combinations, its vocalic
abundance and metrical pronunciation: and many even among its proper
names, by their analogy to words really significant, gave direct
occasion to explanatory or illustrative stories. Etymological mythes
are found in sensible proportion among the whole number.

To understand properly then the Grecian mythes, we must try to
identify ourselves with the state of mind of the original mythopœic
age; a process not very easy, since it requires us to adopt a string
of poetical fancies not simply as realities, but as the governing
realities of the mental system;[844] yet a process which would only
reproduce something analogous to our own childhood. The age was one
destitute both of recorded history and of positive science, but full
of imagination and sentiment and religious impressibility; from these
sources sprung that multitude of supposed persons around whom all
combinations of sensible phænomena were grouped, and towards whom
curiosity, sympathies, and reverence were earnestly directed. The
adventures of such persons were the only aliment suited at once both
to the appetites and to the comprehension of an early Greek; and the
mythes which detailed them, while powerfully interesting his emotions,
furnished to him at the same time a quasi-history and quasi-philosophy:
they filled up the vacuum of the unrecorded past, and explained many of
the puzzling incognita of the present.[845] Nor need we wonder that the
same plausibility which captivated his imagination and his feelings
was sufficient to engender spontaneous belief; or rather, that no
question as to truth or falsehood of the narrative suggested itself to
his mind. His faith is ready, literal and uninquiring, apart from all
thought of discriminating fact from fiction, or of detecting hidden and
symbolized meaning; it is enough that what he hears be intrinsically
plausible and seductive, and that there be no special cause to provoke
doubt. And if indeed there were, the poet overrules such doubts by the
holy and all-sufficient authority of the Muse, whose omniscience is the
warrant for his recital, as her inspiration is the cause of his success.

  [844] The mental analogy between the early stages of human
  civilization and the childhood of the individual is forcibly
  and frequently set forth in the works of Vico. That eminently
  original thinker dwells upon the poetical and religious
  susceptibilities as the first to develop themselves in the human
  mind, and as furnishing not merely connecting threads for the
  explanation of sensible phænomena, but also aliment for the
  hopes and fears, and means of socializing influence to men of
  genius, at a time when reason was yet asleep. He points out
  the _personifying instinct_ (“istinto d’ animazione”) as the
  spontaneous philosophy of man, “to make himself the rule of the
  universe,” and to suppose everywhere a quasi-human agency as
  the determining cause. He remarks that in an age of fancy and
  feeling, the conceptions and language of poetry coincide with
  those of reality and common life, instead of standing apart as a
  separate vein. These views are repeated frequently (and with some
  variations of opinion as he grew older) in his Latin work _De
  Uno Universi Juris Principio_, as well as in the two successive
  _rédactions_ of his great Italian work, _Scienza Nuova_ (it
  must be added that Vico as an expositor is prolix, and does not
  do justice to his own powers of original thought): I select
  the following from the second edition of the latter treatise,
  published by himself in 1744, _Della Metafisica Poetica_ (see
  vol. v. p. 189 of Ferrari’s edition of his Works, Milan, 1836):
  “Adunque la sapienza poetica, che fu la prima sapienza della
  Gentilità, dovette incominciare da una Metafisica, non _ragionata
  ed astratta_, qual è questa or degli addottrinati, ma _sentita
  ed immaginata_, quale dovett’ essere di tai primi uomini,
  siccome quelli ch’ erano di niun raziocinio, e tutti robusti
  sensi e vigorosissime fantasie, come è stato nelle degnità (the
  _Axioms_) stabilito. Questa fu la loro propria poesia, la qual
  in essi fu una facultà loro connaturale, perchè erano di tali
  sensi e di si fatte fantasie naturalmente forniti, nata da
  _ignoranza di cagioni_—la qual fu loro madre di maraviglia di
  tutte le cose, che quelli ignoranti di tutte le cose fortemente
  ammiravano. Tal poesia incominciò in essi divina: perchè nello
  stesso tempo ch’essi immaginavano le cagioni delle cose, che
  sentivano ed ammiravano, essere Dei, come ora il confermiamo con
  gli Americani, i quali tutte le cose che superano la loro picciol
  capacità, dicono esser Dei ... nello stesso tempo, diciamo, alle
  cose ammirate davano l’essere di sostanze dalla propria lor idea:
  ch’è appunto la natura dei fanciulli, che osserviamo prendere tra
  mani cose inanimate, e transtullarsi e favellarvi, come fussero
  quelle persone vive. In cotal guisa i primi uomini delle nazioni
  gentili, come fanciulli del nascente gener umano, dalla lor
  idea creavan essi le cose ... per la loro robusta ignoranza, il
  facevano in forza d’una corpolentissima fantasia, e perch’ era
  corpolentissima, il facevano con una maravigliosa sublimità, tal
  e tanta, che perturbava all’eccesso essi medesimi, che fingendo
  le si creavano.... Di questa natura di cose umane restò eterna
  proprietà spiegata con nobil espressione da Tacito, che vanamente
  gli uomini spaventati _fingunt simul creduntque_.”

  After describing the condition of rude men, terrified with
  thunder and other vast atmospheric phænomena, Vico proceeds
  (_ib._ p. 172)—“In tal caso la natura della mente umana porta
  ch’ella attribuisca all’effetto la sua natura: e la natura loro
  era in tale stato d’uomini tutti robuste forze di corpo, che
  urlando, brontolando, spiegavano le loro violentissime passioni,
  si finsero il cielo esser un gran corpo animato, che per tal
  aspetto chiamavano Giove, che col fischio dei fulmini e col
  fragore dei tuoni volesse lor dire qualche cosa.... E si fanno
  di tutta la natura un vasto corpo animato, che senta passioni ed
  affetti.”

  Now the contrast with modern habits of thought:—

  “Ma siccome _ora_ per la natura delle nostre umane menti troppo
  ritirata dai sensi nel medesimo volgo—con le tante astrazioni,
  di quante sono piene le lingue—con tanti vocaboli astratti—e
  di troppo assottigliata con l’arti dello scrivere, e quasi
  spiritualezzata con la practica dei numeri—_ci e naturalmente
  niegato di poter formare_ la vasta imagine di cotal donna che
  dicono Natura simpatetica, che mentre con la bocca dicono, non
  hanno nulla in lor mente, perocchè la lor mente è dentro il
  falso, che è nulla; nè sono soccorsi dalla fantasia a poterne
  formare una falsa vastissima imagine. Così _ora ci è naturalmente
  niegato di poter entrare nella vasta immaginativa di quei primi
  uomini_, le menti dei quali di nulla erano assottigliate, di
  nulla astratte, di nulla spiritualezzate.... Onde dicemmo
  sopra ch’_ora appena intender si può, affatto immaginar non sì
  può_, come pensassero i primi uomini che fondarono la umanità
  gentilesca.”

  In this citation (already almost too long for a note) I have
  omitted several sentences not essential to the general meaning.
  It places these early divine fables and theological poets (so
  Vico calls them) in their true point of view, and assigns to them
  their proper place in the ascending movement of human society:
  it refers the mythes to an early religious and poetical age, in
  which feeling and fancy composed the whole fund of the human
  mind, over and above the powers of sense: the great mental change
  which has since taken place has robbed us of the power, not
  merely of believing them as they were originally believed, but
  even of conceiving completely that which their first inventors
  intended to express.

  The views here given from this distinguished Italian (the
  precursor of F. A. Wolf in regard to the Homeric poems, as well
  as of Niebuhr in regard to the Roman history) appear to me no
  less correct than profound; and the obvious inference from them
  is, that attempts to _explain_ (as it is commonly called) the
  mythes (_i. e._ to translate them into some physical, moral or
  historical statements, suitable to our order of thought) are,
  even as guesses, essentially unpromising. Nevertheless Vico,
  inconsistently with his own general view, bestows great labor and
  ingenuity in attempting to discover internal meaning symbolized
  under many of the mythes; and even lays down the position, “che
  i primi uomini della Gentilità essendo stati semplicissimi,
  quanto i fanciulli, i quali per natura son veritieri: le prime
  favole non poterono finger nulla di falso: per lo che dovettero
  necessariamente essere _vere narrazioni_.” (See vol. v. p. 194;
  compare also p. 99, Axiom xvi.) If this position be meant simply
  to exclude the idea of designed imposture, it may for the most
  part be admitted; but Vico evidently intends something more. He
  thinks that there lies hid under the fables a basis of matter of
  fact—not literal but symbolized—which he draws out and exhibits
  under the form of a civil history of the divine and heroic times:
  a confusion of doctrine the more remarkable, since he distinctly
  tells us (in perfect conformity with the long passage above
  transcribed from him) that the special matter of these early
  mythes is “impossibility accredited as truth,”—“che la di lei
  propria materia è _l’impossibile credibile_” (p. 176, and still
  more fully in the first _rédaction_ of the _Scienza Nuova_, b.
  iii. c. 4; vol. iv. p. 187 of his Works).

  When we read the _Canones Mythologici_ of Vico (De Constantia
  Philologiæ, Pars Posterior, c. xxx.; vol. iii. p. 363), and
  his explanation of the legends of the Olympic gods, Hercules,
  Thêseus, Kadmus, etc., we see clearly that the meaning which he
  professes to bring out is one previously put in by himself.

  There are some just remarks to the same purpose in Karl Ritter’s
  _Vorhalle Europäischer Volkergeschichten_, Abschn. ii. p. 150
  _seq._ (Berlin, 1820). He too points out how much the faith of
  the old world (der Glaube der Vorwelt) has become foreign to our
  minds, since the recent advances of “Politik und Kritik,” and how
  impossible it is for us to elicit history from their conceptions
  by our analysis, in cases where they have not distinctly laid it
  out for us. The great length of this note prevents me from citing
  the passage: and he seems to me also (like Vico) to pursue his
  own particular investigations in forgetfulness of the principle
  laid down by himself.

  [845] O. Müller, in his _Prolegomena zu einer wissenschaftlichen
  Mythologie_ (cap. iv. p. 108), has pointed out the mistake of
  supposing that there existed originally some nucleus of pure
  reality as the starting-point of the mythes, and that upon this
  nucleus fiction was superinduced afterwards: he maintains that
  the real and the ideal were blended together in the primitive
  conception of the mythes. Respecting the general state of mind
  out of which the mythes grew, see especially pages 78 and 110 of
  that work, which is everywhere full of instruction on the subject
  of the Grecian mythes, and is eminently suggestive, even where
  the positions of the author are not completely made out.

  The short _Heldensage der Griechen_ by Nitzsch (Kiel, 1842, t.
  v.) contains more of just and original thought on the subject
  of the Grecian mythes than any work with which I am acquainted.
  I embrace completely the subjective point of view in which he
  regards them; and although I have profited much from reading
  his short tract, I may mention that before I ever saw it, I had
  enforced the same reasonings on the subject in an article in the
  Westminster Review, May 1843, on the _Heroen-Geschichten_ of
  Niebuhr.

  Jacob Grimm, in the preface to his _Deutsche Mythologie_ (p. 1,
  1st edit. Gött. 1835), pointedly insists on the distinction
  between “_Sage_” and history, as well as upon the fact that the
  former has its chief root in religious belief “Legend and history
  (he says) are powers each by itself, adjoining indeed on the
  confines, but having each its own separate and exclusive ground;”
  also p. xxvii. of the same introduction.

  A view substantially similar is adopted by William Grimm, the
  other of the two distinguished brothers whose labors have so much
  elucidated Teutonic philology and antiquities. He examines the
  extent to which either historical matter of fact or historical
  names can be traced in the _Deutsche Heldensage_; and he comes to
  the conclusion that the former is next to nothing, the latter not
  considerable. He draws particular attention to the fact, that the
  audience for whom these poems were intended had not learned to
  distinguish history from poetry (W. Grimm, _Deutsche Heldensage_,
  pp. 8, 337, 342, 345, 399, Gött. 1829).

The state of mind, and the relation of speaker to hearers, thus
depicted, stand clearly marked in the terms and tenor of the ancient
epic, if we only put a plain meaning upon what we read. The poet—like
the prophet, whom he so much resembles—sings under heavenly guidance,
inspired by the goddess to whom he has prayed for her assisting
impulse: she puts the word into his mouth and the incidents into his
mind: he is a privileged man, chosen as her organ and speaking from
her revelations.[846] As the Muse grants the gift of song to whom she
will, so she sometimes in her anger snatches it away, and the most
consummate human genius is then left silent and helpless.[847] It is
true that these expressions, of the Muse inspiring and the poet singing
a tale of past times, have passed from the ancient epic to compositions
produced under very different circumstances, and have now degenerated
into unmeaning forms of speech; but they gained currency originally in
their genuine and literal acceptation. If poets had from the beginning
written or recited, the predicate of singing would never have been
ascribed to them; nor would it have ever become customary to employ the
name of the Muse as a die to be stamped on licensed fiction, unless the
practice had begun when her agency was invoked and hailed in perfect
good faith. Belief, the fruit of deliberate inquiry and a rational
scrutiny of evidence, is in such an age unknown: the simple faith of
the time slides in unconsciously, when the imagination and feeling are
exalted; and inspired authority is at once understood, easily admitted,
and implicitly confided in.

  [846] Hesiod, Theogon. 32.—

    ... ἐνέπνευσαν δέ (the Muses) μοι αὐδὴν
    Θείην, ὡς κλείοιμι τά τ᾽ ἐσσόμενα, πρό τ᾽ ἐόντα,
    Καί με κέλονθ᾽ ὑμνεῖν μακάρων γένος αἰὲν ἐόντων, etc.

  Odyss. xxii. 347; viii. 63, 73, 481, 489. Δημόδοκ᾽ ... ἢ σέ γε
  Μοῦσ᾽ ἐδίδαξε, Διὸς παῖς, ἢ σέγ᾽ Ἀπόλλων: that is, Demodocus has
  either been inspired as a poet by the Muse, or as a prophet by
  Apollo: for the Homeric Apollo is not the god of song. Kalchas
  the prophet receives his inspiration from Apollo, who confers
  upon him the same knowledge both of past and future as the Muses
  give to Hesiod (Iliad, i. 69):—

    Κάλχας Θεστορίδης, οἰωνοπόλων ὄχ᾽ ἄριστος
    Ὃς ᾔδη τά τ᾽ ἐόντα, τά τ᾽ ἐσσόμενα, πρό τ᾽ ἐόντα
    Ἣν διὰ μαντοσύνην, τὴν οἱ πόρε Φοῖβος Ἀπόλλων.

  Also Iliad, ii. 485.

  Both the μάντις and the ἀοιδὸς are standing, recognized
  professions (Odyss. xvii. 383), like the physician and the
  carpenter, δημιόεργοι.

  [847] Iliad, ii. 599.

The word mythe (μῦθος, _fabula_, _story_), in its original meaning,
signified simply a statement or current narrative, without any
connotative implication either of truth or falsehood. Subsequently the
meaning of the word (in Latin and English as well as in Greek) changed,
and came to carry with it the idea of an old personal narrative,
always uncertified, sometimes untrue or avowedly fictitious.[848]
And this change was the result of a silent alteration in the mental
state of the society,—of a transition on the part of the superior
minds (and more or less on the part of all) to a stricter and more
elevated canon of credibility, in consequence of familiarity with
recorded history, and its essential tests, affirmative as well as
negative. Among the original hearers of the mythes, all such tests were
unknown; they had not yet learned the lesson of critical disbelief;
the mythe passed unquestioned from the mere fact of its currency, and
from its harmony with existing sentiments and preconceptions. The
very circumstances which contributed to rob it of literal belief in
after-time, strengthened its hold upon the mind of the Homeric man. He
looked for wonders and unusual combinations in the past; he expected
to hear of gods, heroes and men, moving and operating together upon
earth; he pictured to himself the fore-time as a theatre in which the
gods interfered directly, obviously and frequently, for the protection
of their favorites and the punishment of their foes. The rational
conception, then only dawning in his mind, of a systematic course of
nature was absorbed by this fervent and lively faith. And if he could
have been supplied with as perfect and philosophical a history of his
own real past time, as we are now enabled to furnish with regard to
the last century of England or France, faithfully recording all the
successive events, and accounting for them by known positive laws,
but introducing no special interventions of Zeus and Apollo—such a
history would have appeared to him not merely unholy and unimpressive,
but destitute of all plausibility or title to credence. It would
have provoked in him the same feeling of incredulous aversion as a
description of the sun (to repeat the previous illustration) in a
modern book on scientific astronomy.

  [848] In this later sense it stands pointedly opposed to ἱστορία,
  _history_, which seems originally to have designated matter
  of fact, present and seen by the describer, or the result of
  his personal inquiries (see Herodot. i. 1; Verrius Flacc. ap.
  Aul. Gell. v. 18; Eusebius, Hist. Eccles. iii. 12; and the
  observations of Dr. Jortin, Remarks on Ecclesiastical History,
  vol. i. p. 59).

  The original use of the word λόγος was the same as that of
  μῦθος—a current tale, true or false, as the case might be;
  and the term designating a person much conversant with the
  old legends (λόγιος) is derived from it (Herod. i. 1; ii. 3).
  Hekatæus and Herodotus both use λόγος in this sense. Herodotus
  calls both Æsop and Hekatæus λογοποιοί (ii. 134-143).

  Aristotle (Metaphys. i. p. 8, ed. Brandis) seems to use μῦθος
  in this sense, where he says—διὸ καὶ φιλόμυθος ὁ φιλόσοφος
  πώς ἐστιν· ὁ γὰρ μῦθος συγκεῖται ἐκ θαυμασίων, etc. In the
  same treatise (xi. p. 254), he uses it to signify fabulous
  amplification and transformation of a doctrine true in the main.

To us these mythes are interesting fictions; to the Homeric and
Hesiodic audience they were “rerum divinarum et humanarum scientia,”—an
aggregate of religious, physical and historical revelations, rendered
more captivating, but not less true and real, by the bright coloring
and fantastic shapes in which they were presented. Throughout the
whole of “mythe-bearing Hellas”[849] they formed the staple of the
uninstructed Greek mind, upon which history and philosophy were by
so slow degrees superinduced; and they continued to be the aliment of
ordinary thought and conversation, even after history and philosophy
had partially supplanted the mythical faith among the leading men, and
disturbed it more or less in the ideas of all. The men, the women,
and the children of the remote dêmes and villages of Greece, to whom
Thucydidês, Hippocratês, Aristotle, or Hipparchus were unknown, still
continued to dwell upon the local fables which formed their religious
and patriotic antiquity. And Pausanias, even in his time, heard
everywhere divine or heroic legends yet alive, precisely of the type
of the old epic; he found the conceptions of religious and mythical
faith, coëxistent with those of positive science, and contending
against them at more or less of odds, according to the temper of the
individual. Now it is the remarkable characteristic of the Homeric age,
that no such coëxistence or contention had yet begun. The religious and
mythical point of view covers, for the most part, all the phænomena
of nature; while the conception of invariable sequence exists only
in the background, itself personified under the name of the Mœræ,
or Fates, and produced generally as an exception to the omnipotence
of Zeus for all ordinary purposes. Voluntary agents, visible and
invisible, impel and govern everything. Moreover this point of view
is universal throughout the community,—adopted with equal fervor, and
carried out with equal consistency, by the loftiest minds and by the
lowest. The great man of that day is he who, penetrated like others
with the general faith, and never once imagining any other system of
nature than the agency of these voluntary Beings, can clothe them in
suitable circumstances and details, and exhibit in living body and
action those types which his hearers dimly prefigure. Such men were
the authors of the Iliad and the Odyssey; embodying in themselves the
whole measure of intellectual excellence which their age was capable of
feeling: to us, the first of poets—but to their own public, religious
teachers, historians, and philosophers besides—inasmuch as all that
then represented history and philosophy was derived from those epical
effusions and from others homogeneous with them. Herodotus recognizes
Homer and Hesiod as the main authors of Grecian belief respecting the
names and generations, the attributes and agency, the forms and the
worship of the gods.[850]

  [849] M. Ampère, in his _Histoire Littéraire de la France_ (ch.
  viii. v. i. p. 310) distinguishes the Saga (which corresponds
  as nearly as possible with the Greek μῦθος, λόγος, ἐπιχώριος
  λόγος), as a special product of the intellect, not capable of
  being correctly designated either as history, or as fiction, or
  as philosophy:—

  “Il est un pays, la Scandinavie, où la tradition racontée s’est
  développée plus complètement qu’ailleurs, où ses produits ont
  été plus soigneusement recueillis et mieux conservés: dans ce
  pays, ils ont reçu un nom particulier, dont l’équivalent exact ne
  se trouve pas hors des langues Germaniques: c’est le mot _Saga,
  Sage, ce qu’on dit, ce qu’on raconte_,—la tradition orale. Si
  l’on prend ce mot non dans une acception restreinte, mais dans
  le sens général où le prenait Niebuhr quand il l’appliquoit, par
  exemple, aux traditions populaires qui ont pu fournir à Tite Live
  une portion de son histoire, la Saga doit être comptée parmi
  les produits spontanés de l’imagination humaine. La Saga a son
  existence propre comme la poësie, comme l’histoire, comme le
  roman. Elle n’est pas la poësie, parcequ’elle n’est pas chantée,
  mais parlée; elle n’est pas l’histoire, parcequ’elle est denuée
  de critique; elle n’est pas le roman, parcequ’elle est sincère,
  parcequ’elle a foi à ce qu’elle raconte. Elle n’invente pas, mais
  répète: elle peut se tromper, mais elle ne ment jamais. Ce récit
  souvent merveilleux, que personne ne fabrique sciemment, et que
  tout le monde altère et falsifie sans le vouloir, qui se perpétue
  à la manière des chants primitifs et populaires,—ce récit, quand
  il se rapporte non à un héros, mais à un saint, s’appelle une
  légende.”

  [850] Herodot. ii. 53.

History, philosophy, etc., properly so called and conforming to our
ideas (of which the subsequent Greeks were the first creators), never
belonged to more than a comparatively small number of thinking men,
though their influence indirectly affected more or less the whole
national mind. But when positive science and criticism, and the idea of
an invariable sequence of events, came to supplant in the more vigorous
intellects the old mythical creed of omnipresent personification,
an inevitable scission was produced between the instructed few and
the remaining community. The opposition between the scientific and
the religious point of view was not slow in manifesting itself: in
general language, indeed, both might seem to stand together, but in
every particular case the admission of one involved the rejection of
the other. According to the theory which then became predominant,
the course of nature was held to move invariably on, by powers and
attributes of its own, unless the gods chose to interfere and reverse
it; but they had the power of interfering as often and to as great an
extent as they thought fit. Here the question was at once opened,
respecting a great variety of particular phænomena, whether they were
to be regarded as natural or miraculous. No constant or discernible
test could be suggested to discriminate the two: every man was called
upon to settle the doubt for himself, and each settled it according to
the extent of his knowledge, the force of his logic, the state of his
health, his hopes, his fears, and many other considerations affecting
his separate conclusion. In a question thus perpetually arising, and
full of practical consequences, instructed minds, like Periklês,
Thucydidês, and Euripidês, tended more and more to the scientific
point of view,[851] in cases where the general public were constantly
gravitating towards the religious.

  [851] See Plutarch, Perikl. capp. 5, 32, 38; Cicero, De Republ.
  i. 15-16, ed. Maii.

  The phytologist Theophrastus, in his valuable collection of facts
  respecting vegetable organization, is often under the necessity
  of opposing his scientific interpretation of curious incidents in
  the vegetable world to the religious interpretation of them which
  he found current. Anomalous phænomena in the growth or decay of
  trees were construed as signs from the gods, and submitted to a
  prophet for explanation (see Histor. Plantar. ii. 3, iv. 16; v.
  3).

  We may remark, however, that the old faith had still a certain
  hold over his mind. In commenting on the story of the willow-tree
  at Philippi, and the venerable old plane-tree at Antandros
  (more than sixty feet high, and requiring four men to grasp it
  round in the girth), having been blown down by a high wind, and
  afterwards spontaneously resuming their erect posture, he offers
  some explanations how such a phænomenon might have happened, but
  he admits, at the end, that there may be something extra-natural
  in the case, Ἀλλὰ ταῦτα μὲν ἴσως ἔξω φυσικῆς αἰτίας ἔστιν, etc.
  (De Caus. Plant. v. 4): see a similar miracle in reference to the
  cedar-tree of Vespasian (Tacit. Hist. ii. 78).

  Euripidês, in his lost tragedy called Μελανίππη Σοφὴ, placed in
  the month of Melanippê a formal discussion and confutation of the
  whole doctrine of τέρατα, or supernatural indications (Dionys.
  Halicar. Ars Rhetoric. p. 300-356, Reisk). Compare the Fables of
  Phædrus, iii. 3; Plutarch, Sept. Sap. Conviv. ch. 3. p. 149; and
  the curious philosophical explanation by which the learned men of
  Alexandria tranquillized the alarms of the vulgar, on occasion of
  the serpent said to have been seen entwined round the head of the
  crucified Kleomenês (Plutarch, Kleomen. c. 39).

  It is one part of the duty of an able physician, according to
  the Hippocratic treatise called Prognosticon (c. 1. t. ii. p.
  112, ed. Littré), when he visits his patient, to examine whether
  there is anything divine in the malady, ἅμα δὲ καὶ εἴ τι θεῖον
  ἔνεστιν ἐν τῇσι νούσοισι: this, however, does not agree with the
  memorable doctrine laid down in the treatise, De Aëre, Locis et
  Aquis (c. 22. p. 78, ed. Littré), and cited hereafter, in this
  chapter. Nor does Galen seem to have regarded it as harmonizing
  with the general views of Hippocratês. In the excellent
  Prolegomena of M. Littré to his edition of Hippocratês (t. i.
  p. 76) will be found an inedited scholium, wherein the opinion
  of Baccheius and other physicians is given, that the affections
  of the plague were to be looked upon as divine, inasmuch as the
  disease came from God; and also the opinion of Xenophôn, the
  friend of Praxagoras, that the “genus of days of crisis” in fever
  was divine; “For (said Xenophôn) just as the Dioskuri, being
  gods, appear to the mariner in the storm and bring him salvation,
  so also do the days of crisis, when they arrive, in fever.”
  Galen, in commenting upon this doctrine of Xenophôn, says that
  the author “has expressed his own individual feeling, but has
  no way set forth the opinion of Hippocratês:” Ὁ δὲ τῶν κρισίμων
  γένος ἡμερῶν εἰπὼν εἶναι θεῖον, ἑαυτοῦ τι πάθος ὡμολόγησεν· οὐ
  μὴν Ἱπποκράτους γε τὴν γνώμην ἔδειξεν (Galen, Opp. t. v. p. 120,
  ed. Basil).

  The comparison of the Dioskuri appealed to by Xenophôn is a
  precise reproduction of their function as described in the
  Homeric Hymn (Hymn xxxiii. 10): his personification of the “days
  of crisis” introduces the old religious agency to fill up a gap
  in his medical science.

  I annex an illustration from the Hindoo vein of thought:—“It is
  a rule with the Hindoos to bury, and not to burn, the bodies of
  those who die of the small pox: for (say they) the small pox is
  not only caused by the goddess Davey, but is, in fact, _Davey
  herself_; and to burn the body of a person affected with this
  disease, is, in reality, neither more nor less than _to burn the
  goddess_.” (Sleeman, Rambles and Recollections, etc., vol. i. ch.
  xxv. p. 221.)

The age immediately prior to this unsettled condition of thought is
the really mythopœic age; in which the creative faculties of the
society know no other employment, and the mass of the society no
other mental demand. The perfect expression of such a period, in
its full peculiarity and grandeur, is to be found in the Iliad and
Odyssey,—poems of which we cannot determine the exact date, but which
seem both to have existed prior to the first Olympiad, 776 B. C., our
earliest trustworthy mark of Grecian time. For some time after that
event, the mythopœic tendencies continued in vigor (Arktinus, Leschês,
Eumêlus, and seemingly most of the Hesiodic poems, fall within or
shortly after the first century of recorded Olympiads); but from and
after this first century, we may trace the operation of causes which
gradually enfeebled and narrowed them, altering the point of view from
which the mythes were looked at. What these causes were, it will be
necessary briefly to intimate.

The foremost and most general of all is, the expansive force of Grecian
intellect itself,—a quality in which this remarkable people stand
distinguished from all their neighbors and contemporaries. Most, if
not all nations have had mythes, but no nation except the Greeks have
imparted to them immortal charm and universal interest; and the same
mental capacities, which raised the great men of the poetic age to this
exalted level, also pushed forward their successors to outgrow the
early faith in which the mythes had been generated and accredited.

One great mark, as well as means, of such intellectual expansion,
was the habit of attending to, recording, and combining, positive
and present facts, both domestic and foreign. In the genuine Grecian
epic, the theme was an unknown and aoristic past; but even as early
as the Works and Days of Hesiod, the present begins to figure: the
man who tills the earth appears in his own solitary nakedness, apart
from gods and heroes—bound indeed by serious obligations to the gods,
but contending against many difficulties which are not to be removed
by simple reliance on their help. The poet denounces his age in the
strongest terms as miserable, degraded and profligate, and looks back
with reverential envy to the extinct heroic races who fought at Troy
and Thêbes. Yet bad as the present time is, the Muse condescends to
look at it along with him, and to prescribe rules for human life—with
the assurance that if a man be industrious, frugal, provident, just
and friendly in his dealings, the gods will recompense him with
affluence and security. Nor does the Muse disdain, while holding out
such promise, to cast herself into the most homely details of present
existence and to give advice thoroughly practical and calculating. Men
whose minds were full of the heroes of Homer, called Hesiod in contempt
the poet of the Helots; and the contrast between the two is certainly
a remarkable proof of the tendency of Greek poetry towards the present
and the positive.

Other manifestations of the same tendency become visible in the age of
Archilochus (B. C. 680-660). In an age when metrical composition and
the living voice are the only means whereby the productive minds of a
community make themselves felt, the invention of a new metre, new forms
of song and recitation, or diversified accompaniments, constitute an
epoch. The iambic, elegiac, choric, and lyric poetry, from Archilochus
downwards, all indicate purposes in the poet, and impressibilities
of the hearers, very different from those of the ancient epic. In
all of them the personal feeling of the poet and the specialties of
present time and place, are brought prominently forward, while in the
Homeric hexameter the poet is a mere nameless organ of the historical
Muse—the hearers are content to learn, believe, and feel, the incidents
of a foregone world, and the tale is hardly less suitable to one
time and place than to another. The iambic metre (we are told) was
first suggested to Archilochus by the bitterness of his own private
antipathies; and the mortal wounds inflicted by his lampoons, upon the
individuals against whom they were directed, still remain attested,
though the verses themselves have perished. It was the metre (according
to the well-known judgment of Aristotle) most nearly approaching to
common speech, and well suited both to the coarse vein of sentiment,
and to the smart and emphatic diction of its inventor.[852] Simonidês
of Amorgus, the younger contemporary of Archilochus, employed the same
metre, with less bitterness, but with an anti-heroic tendency not
less decided. His remaining fragments present a mixture of teaching
and sarcasm, having a distinct bearing upon actual life,[853] and
carrying out the spirit which partially appears in the Hesiodic Works
and Days. Of Alkæus and Sapphô, though unfortunately we are compelled
to speak of them upon hearsay only, we know enough to satisfy us that
their own personal sentiments and sufferings, their relations private
or public with the contemporary world, constituted the soul of those
short effusions which gave them so much celebrity:[854] and in the
few remains of the elegiac poets preserved to us—Kallinus, Mimnermus,
Tyrtæus—the impulse of some present motive or circumstance is no
less conspicuous. The same may also be said of Solôn, Theognis and
Phokylidês, who preach, encourage, censure, or complain, but do not
recount—and in whom a profound ethical sensibility, unknown to the
Homeric poems, manifests itself: the form of poetry (to use the words
of Solôn himself) is made the substitute for the public speaking of the
agora.[855]

  [852] Horat. de Art. Poet. 79:—

    “Archilochum proprio rabies armavit Iambo,” etc.

  Compare Epist. i. 19, 23, and Epod. vi. 12; Aristot. Rhetor. iii.
  8, 7, and Poetic. c. 4—also Synesius de Somniis—ὥσπερ Ἀλκαῖος καὶ
  Ἀρχίλοχος, οἳ δεδαπανήκασι τὴν εὐστομίαν εἰς τὸν οἰκεῖον βίον
  ἑκάτερος (Alcæi Fragment. Halle, 1810, p. 205). Quintilian speaks
  in striking language of the power of expression manifested by
  Archilochus (x. 1, 60).

  [853] Simonidês of Amorgus touches briefly, but in a tone of
  contempt upon the Trojan war—~γυναικὸς οὕνεκ᾽~ ἀμφιδηριωμένους
  (Simonid. Fragm. 8. p. 36. v. 118); he seems to think it absurd
  that so destructive a struggle should have taken place “_pro unâ
  mulierculâ_,” to use the phrase of Mr. Payne Knight.

  [854] See Quintilian, x. 1, 63. Horat. Od. i. 32; ii. 13.
  Aristot. Polit. iii. 10, 4. Dionys. Halic. observes (Vett.
  Scriptt. Censur. v. p. 421) respecting Alkæus—πολλαχοῦ γοῦν τὸ
  μέτρον εἴ τις περιέλοι, ῥητορικὴν ἂν εὕροι πολιτείαν; and Strabo
  (xiii. p. 617), τὰ στασιωτικὰ καλούμενα τοῦ Ἀλκαίου ποιήματα.

  There was a large dash of sarcasm and homely banter aimed at
  neighbors and contemporaries in the poetry of Sapphô, apart from
  her impassioned love-songs—ἄλλως σκώπτει τὸν ἄγροικον νύμφιον καὶ
  τὸν θυρωρὸν τὸν ἐν τοῖς γάμοις, εὐτελέστατα καὶ ἐν πέζοις ὀνόμασι
  μᾶλλον ἢ ἐν ποιητικοῖς. Ὥστε αὐτῆς μᾶλλόν ἐστι τὰ ποιήματα ταῦτα
  διαλέγεσθαι ἢ ἄδειν· οὐδ᾽ ἂν ἅρμοσαι πρὸς τὸν χόρον ἢ πρὸς τὴν
  λύραν, εἰ μή τις εἴη χόρος διαλεκτικός (Dêmêtr. Phaler, De
  Interpret. c. 167).

  Compare also Herodot. ii. 135, who mentions the satirical talent
  of Sapphô, employed against her brother for an extravagance about
  the courtezan Rhodôpis.

  [855] Solôn, Fragm. iv. 1, ed. Schneidewin:—

    Αὐτὸς κήρυξ ἦλθον ἀφ᾽ ἱμερτῆς Σαλαμῖνος
      Κόσμον ἐπέων ᾠδὴν ἀντ᾽ ἀγορῆς θέμενος, etc.

  See _Brandis_, Handbuch der Griechischen Philosophie, sect.
  xxiv.-xxv. Plato states that Solôn, in his old age, engaged in
  the composition of an epic poem, which he left unfinished, on the
  subject of the supposed island of Atlantis and Attica (Plato,
  Timæus, p. 21, and Kritias, p. 113). Plutarch, Solôn, c. 31.

Doubtless all these poets made abundant use of the ancient mythes, but
it was by turning them to present account, in the way of illustration,
or flattery, or contrast,—a tendency which we may usually detect even
in the compositions of Pindar, in spite of the lofty and heroic strain
which they breathe throughout. That narrative or legendary poetry still
continued to be composed during the seventh and sixth centuries before
the Christian æra is not to be questioned; but it exhibited the old
epical character without the old epical genius; both the inspiration
of the composer and the sympathies of the audience had become more
deeply enlisted in the world before them, and disposed to fasten on
incidents of their own actual experience. From Solôn and Theognis
we pass to the abandonment of all metrical restrictions and to the
introduction of prose writing,—a fact, the importance of which it is
needless to dwell upon,—marking as well the increased familiarity with
written records, as the commencement of a separate branch of literature
for the intellect, apart from the imagination and emotions wherein the
old legends had their exclusive root.

Egypt was first unreservedly opened to the Greeks during the reign of
Psammetichus, about B. C. 660; gradually it became much frequented by
them for military or commercial purposes, or for simple curiosity, and
enlarged the range of their thoughts and observations, while it also
imparted to them that vein of mysticism, which overgrew the primitive
simplicity of the Homeric religion, and of which I have spoken in a
former chapter. They found in it a long-established civilization,
colossal wonders of architecture, and a certain knowledge of astronomy
and geometry, elementary indeed, but in advance of their own. Moreover
it was a portion of their present world, and it contributed to form in
them an interest for noting and describing the actual realities before
them. A sensible progress is made in the Greek mind during the two
centuries from B. C. 700 to B. C. 500, in the record and arrangement
of historical facts: an _historical sense_ arises in the superior
intellects, and some idea of evidence as a discriminating test between
fact and fiction. And this progressive tendency was further stimulated
by increased communication and by more settled and peaceful social
relations between the various members of the Hellenic world, to which
may be added material improvements, purchased at the expense of a
period of turbulence and revolution, in the internal administration
of each separate state. The Olympic, Pythian, Nemean, and Isthmian
games became frequented by visitors from the most distant parts of
Greece: the great periodical festival in the island of Dêlos brought
together the citizens of every Ionic community, with their wives and
children, and an ample display of wealth and ornaments.[856] Numerous
and flourishing colonies were founded in Sicily, the south of Italy,
the coasts of Epirus and of the Euxine Sea: the Phokæans explored the
whole of the Adriatic, established Massalia, and penetrated even as
far as the south of Ibêria, with which they carried on a lucrative
commerce.[857] The geographical ideas of the Greeks were thus both
expanded and rectified: the first preparation of a map, by Anaximander
the disciple of Thalês, is an epoch in the history of science. We may
note the ridicule bestowed by Herodotus both upon the supposed people
called Hyperboreans and upon the idea of a circumfluous ocean-stream,
as demonstrating the progress of the age in this department of
inquiry.[858] And even earlier than Herodotus, Xanthus had noticed the
occurrence of fossil marine productions in the interior of Asia Minor,
which led him to reflections on the changes of the earth’s surface with
respect to land and water.[859]

  [856] Homer, Hymn. ad Apollin. 155; Thucydid. iii. 104.

  [857] Herodot. i. 163.

  [858] Herodot. iv. 36. γελῶ δὲ ὁρέων Γῆς περιόδους γράψαντας
  πολλοὺς ἤδη, καὶ οὐδένα νόον ἔχοντας ἐξηγησάμενον· οἳ Ὠκέανόν τε
  ῥέοντα γράφουσι πέριξ τὴν γῆν, ἐοῦσαν κυκλοτερέα ὡς ἀπὸ τόρνου,
  etc., a remark probably directed against Hekatæus.

  Respecting the map of Anaximander, Strabo, i. p. 7; Diogen.
  Laërt. ii. 1; Agathemer ap. Geograph. Minor. i. 1. πρῶτος
  ἐτόλμησε τὴν οἰκουμένην ἐν πίνακι γράψαι.

  Aristagoras of Milêtus, who visited Sparta to solicit aid for the
  revolted Ionians against Darius, brought with him a brazen tablet
  or map, by means of which he exhibited the relative position of
  places in the Persian empire (Herodot. v. 49).

  [859] Xanthus ap. Strabo. i. p. 50; xii. p. 579. Compare Creuzer,
  Fragmenta Xanthi, p. 162.

If then we look down the three centuries and a half which elapsed
between the commencement of the Olympic æra and the age of
Herodotus and Thucydidês, we shall discern a striking advance in
the Greeks,—ethical, social and intellectual. Positive history and
chronology has not only been created, but in the case of Thucydidês,
the qualities necessary to the historiographer, in their application to
recent events, have been developed with a degree of perfection never
since surpassed. Men’s minds have assumed a gentler as well as a juster
cast; and acts come to be criticized with reference to their bearing
on the internal happiness of a well-regulated community, as well as
upon the standing harmony of fraternal states. While Thucydidês
treats the habitual and licensed piracy, so coolly alluded to in the
Homeric poems, as an obsolete enormity, many of the acts described in
the old heroic and Theogonic legends were found not less repugnant to
this improved tone of feeling. The battles of the gods with the Giants
and Titans,—the castration of Uranus by his son Kronus,—the cruelty,
deceit and licentiousness, often supposed both in the gods and heroes,
provoked strong disapprobation. And the language of the philosopher
Xenophanês, who composed both elegiac and iambic poems for the express
purpose of denouncing such tales, is as vehement and unsparing as that
of the Christian writers, who, eight centuries afterwards, attacked the
whole scheme of paganism.[860]

  [860] Xenophan. ap. Sext. Empiric. adv. Mathemat. ix. 193. Fragm.
  1. Poet. Græc. ed. Schneidewin. Diogen. Laërt. ix. 18.

Nor was it alone as an ethical and social critic that Xenophanês
stood distinguished. He was one of a great and eminent triad—Thalês
and Pythagoras being the others—who, in the sixth century before the
Christian æra, first opened up those veins of speculative philosophy
which occupied afterwards so large a portion of Grecian intellectual
energy. Of the material differences between the three I do not here
speak; I regard them only in reference to the Homeric and Hesiodic
philosophy which preceded them, and from which all three deviated by a
step, perhaps the most remarkable in all the history of philosophy. In
the scheme of ideas common to Homer and to the Hesiodic Theogony (as
has been already stated), we find nature distributed into a variety of
personal agencies, administered according to the free-will of different
Beings more or less analogous to man—each of these Beings having his
own character, attributes and powers, his own sources of pain and
pleasure, and his own especial sympathies or antipathies with human
individuals; each being determined to act or forbear, to grant favor
or inflict injury in his own department of phænomena, according as
men, or perhaps other Beings analogous to himself, might conciliate or
offend him. The Gods, properly so called, (those who bore a proper name
and received some public or family worship,) were the most commanding
and capital members amidst this vast network of agents visible and
invisible, spread over the universe.[861] The whole view of nature
was purely religious and subjective, the spontaneous suggestion of
the early mind. It proceeded from the instinctive tendencies of the
feelings and imagination to transport, to the world without, the
familiar type of free-will and conscious personal action: above all, it
took deep hold of the emotions, from the widely extended sympathy which
it so perpetually called forth between man and nature.[862]

  [861] Hesiod, Opp. Di. 122; Homer, Hymn. ad Vener. 260.

  [862] A defence of the primitive faith, on this ground, is found
  in Plutarch, Quæstion. Sympos. vii. 4, 4, p. 703.

The first attempt to disenthral the philosophic intellect from this
all-personifying religious faith, and to constitute a method of
interpreting nature distinct from the spontaneous inspirations of
untaught minds, is to be found in Thalês, Xenophanês and Pythagoras, in
the sixth century before the Christian æra. It is in them that we first
find the idea of Person tacitly set aside or limited, and an impersonal
Nature conceived as the object of study. The divine husband and wife,
Oceanus and Têthys, parents of many gods and of the Oceanic nymphs,
together with the avenging goddess Styx, are translated into the
material substance _water_, or, as we ought rather to say, the Fluid:
and Thalês set himself to prove that water was the primitive element,
out of which all the different natural substances had been formed.[863]
He, as well as Xenophanês and Pythagoras, started the problem of
physical philosophy, with its objective character and invariable laws,
to be discoverable by a proper and methodical application of the human
intellect. The Greek word Φύσις, denoting _nature_, and its derivatives
_physics_ and _physiology_, unknown in that large sense to Homer or
Hesiod, as well as the word _Kosmos_, to denote the mundane system,
first appears with these philosophers.[864] The elemental analysis
of Thalês—the one unchangeable cosmic substance, varying only in
appearance, but not in reality, as suggested by Xenophanês,—and the
geometrical and arithmetical combinations of Pythagoras,—all these were
different ways of approaching the explanation of physical phænomena,
and each gave rise to a distinct school or succession of philosophers.
But they all agreed in departing from the primitive method, and in
recognizing determinate properties, invariable sequences, and objective
truth, in nature—either independent of willing or designing agents, or
serving to these latter at once as an indispensable subject-matter and
as a limiting condition. Xenophanês disclaimed openly all knowledge
respecting the gods, and pronounced that no man could have any means of
ascertaining when he was right and when he was wrong, in affirmations
respecting them:[865] while Pythagoras represents in part the
scientific tendencies of his age, in part also the spirit of mysticism
and of special fraternities for religious and ascetic observance, which
became diffused throughout Greece in the sixth century before the
Christian æra. This was another point which placed him in antipathy
with the simple, unconscious and demonstrative faith of the old poets,
as well as with the current legends.

  [863] Aristotel. Metaphys. i. 3.

  [864] Plutarch, Placit. Philos. ii. 1; also Stobæus, Eclog.
  Physic. i. 22, where the difference between the Homeric
  expressions and those of the subsequent philosophers is seen.
  Damm, Lexic. Homeric. v. Φύσις; Alexander von Humboldt, _Kosmos_,
  p. 76, the note 9 on page 62 of that admirable work.

  The title of the treatises of the early philosophers (Melissus,
  Dêmokritus, Parmenidês, Empedoclês, Alkmæôn, etc.) was frequently
  Περὶ Φύσεως (Galen. Opp. tom. i. p. 56, ed. Basil).

  [865] Xenophan. ap. Sext. Empiric. vii. 50; viii. 326.—

    Καὶ τὸ μὲν οὖν σαφὲς οὔτις ἀνὴρ ἴδεν, οὔτε τίς ἐστιν
    Εἰδὼς ἀμφὶ θεῶν τε καὶ ἅσσα λέγω περὶ πάντων·
    Εἰ γὰρ καὶ τὰ μάλιστα τύχοι τετελεσμένον εἰπῶν,
    Αὐτος ὅμως οὐκ οἶδε, δόκος δ᾽ ἐπὶ πᾶσι τέτυκται.

  Compare Aristotel. De Xenophane, Zenone, et Georgiâ, capp. 1-2.

If these distinguished men, when they ceased to follow the primitive
instinct of tracing the phænomena of nature to personal and designing
agents, passed over, not at once to induction and observation, but to a
misemployment of abstract words, substituting metaphysical _eideôla_ in
the place of polytheism, and to an exaggerated application of certain
narrow physical theories—we must remember that nothing else could be
expected from the scanty stock of facts then accessible, and that the
most profound study of the human mind points out such transition as an
inevitable law of intellectual progress.[866] At present, we have to
compare them only with that state of the Greek mind[867] which they
partially superseded, and with which they were in decided opposition.
The rudiments of physical science were conceived and developed among
superior men; but the religious feeling of the mass was averse to them;
and the aversion, though gradually mitigated, never wholly died away.
Some of the philosophers were not backward in charging others with
irreligion, while the multitude seems to have felt the same sentiment
more or less towards all—or towards that postulate of constant
sequences, with determinate conditions of occurrence, which scientific
study implies, and which they could not reconcile with their belief
in the agency of the gods, to whom they were constantly praying for
special succor and blessings.

  [866] See the treatise of M. Auguste Comte (_Cours de Philosophie
  Positive_), and his doctrine of the three successive stages of
  the human mind in reference to scientific study—the theological,
  the metaphysical, and the positive;—a doctrine laid down
  generally in his first lecture (vol. i. p. 4-12), and largely
  applied and illustrated throughout his instructive work. It is
  also re-stated and elucidated by Mr. John Stuart Mill, in his
  System of Logic, Ratiocinative and Inductive, vol. ii. p. 610.

  [867] “Human wisdom (ἀνθρωπίνη σοφία), as contrasted with
  the primitive theology (οἱ ἀρχαῖοι καὶ διατρίβοντες περὶ τὰς
  θεολογίας),” to take the words of Aristotle (Meteorolog. ii. 1.
  pp. 41-42, ed. Tauchnitz).

The discrepancy between the scientific and the religious point of
view was dealt with differently by different philosophers. Thus
Socratês openly admitted it, and assigned to each a distinct and
independent province. He distributed phænomena into two classes: one,
wherein the connection of antecedent and consequent was invariable
and ascertainable by human study, and therefore future results
accessible to a well-instructed foresight; the other, and those, too,
the most comprehensive and important, which the gods had reserved
for themselves and their own unconditional agency, wherein there was
no invariable or ascertainable sequence, and where the result could
only be foreknown by some omen, prophecy, or other special inspired
communication from themselves. Each of these classes was essentially
distinct, and required to be looked at and dealt with in a manner
radically incompatible with the other. Socratês held it wrong to
apply the scientific interpretation to the latter, or the theological
interpretation to the former. Physics and astronomy, in his opinion,
belonged to the divine class of phænomena, in which human research was
insane, fruitless, and impious.[868]

  [868] Xenoph. Memor. i. 1, 6-9. Τὰ μὲν ἀναγκαῖα (Σωκράτης)
  συνεβούλευε καὶ πράττειν, ὡς ἐνόμιζεν ἄριστ᾽ ἂν πραχθῆναι·
  περὶ δὲ τῶν ἀδήλων ὅπως ἀποβήσοιτο, μαντευσομένους ἔπεμπεν, εἰ
  ποιητέα. Καὶ τοὺς μέλλοντας οἴκους τε καὶ πόλεις καλῶς οἰκήσειν
  μαντικῆς ἔφη προσδεῖσθαι· τεκτονικὸν μὲν γὰρ ἢ χαλκευτικὸν ἢ
  γεωργικὸν ἢ ἀνθρώπων ἀρχικὸν, ἢ τῶν τοιούτων ἔργων ἐξεταστικὸν,
  ἢ λογιστικὸν, ἢ οἰκονομικὸν, ἢ στρατηγικὸν γενέσθαι, πάντα τὰ
  τοιαῦτα, μαθήματα καὶ ἀνθρώπου γνώμῃ αἱρετέα, ἐνόμιζεν εἶναι· τὰ
  δὲ μέγιστα τῶν ἐν τούτοις ἔφη ~τοὺς θεοὺς ἑαυτοῖς καταλείπεσθαι~,
  ὧν οὐδὲν δῆλον εἶναι τοῖς ἀνθρώποις.... Τοὺς δὲ μηδὲν τῶν
  τοιούτων οἰομένους εἶναι δαιμόνιον, ἀλλὰ πάντα τῆς ἀνθρωπίνης
  γνώμης, δαιμονᾷν ἔφη· δαιμονᾷν δὲ καὶ τοὺς μαντευομένους ἃ τοῖς
  ἀνθρώποις ἔδωκαν οἱ θεοὶ μαθοῦσι διακρίνειν.... Ἔφη δὲ δεῖν, ἃ
  μὲν μαθόντας ποιεῖν ἔδωκαν οἱ θεοὶ, μανθάνειν· ἃ δὲ μὴ δῆλα τοῖς
  ἀνθρώποις ἔστι, πειρᾶσθαι διὰ μαντικῆς παρὰ τῶν θεῶν πυνθάνεσθαι·
  τοὺς θεοὺς γὰρ, οἷς ἂν ὦσιν ἵλεῳ, σημαίνειν. Compare also
  Memorab. iv. 7. 7; and Cyropæd. i. 6, 3, 23-46.

  Physical and astronomical phænomena are classified by Socratês
  among the divine class, interdicted to human study (Memor. i.
  1,13): τὰ θεῖα or δαιμόνια as supposed to τἀνθρώπεια. Plato
  (Phileb. c. 16; Legg. x. p. 886-889; xii. p. 967) held the sun
  and stars to be gods, each animated with its special soul: he
  allowed astronomical investigation to the extent necessary
  for avoiding blasphemy respecting these beings—μέχρι τοῦ μὴ
  βλασφημεῖν περὶ αὐτά (vii. 821).

On the other hand, Hippocratês, the contemporary of Socratês,
denied the discrepancy, and merged into one those two classes of
phænomena,—the divine and the scientifically determinable,—which the
latter had put asunder. Hippocratês treated all phænomena as at once
both divine and scientifically determinable. In discussing certain
peculiar bodily disorders found among the Scythians, he observes, “The
Scythians themselves ascribe the cause of this to God, and reverence
and bow down to such sufferers, each man fearing that he may suffer
the like; and I myself think too that these affections, as well as all
others, are divine: no one among them is either more divine or more
human than another, but all are on the same footing, and all divine;
nevertheless each of them has its own physical conditions, and not one
occurs without such physical conditions.”[869]

  [869] Hippocratês, De Aëre, Locis et Aquis, c. 22 (p. 78, ed.
  Littré, sect. 106 ed. Petersen): Ἔτι τε πρὸς τουτέοισι εὐνούχιαι
  γίγνονται οἱ πλεῖστοι ἐν Σκύθῃσι, καὶ γυναικηΐα ἐργάζονται καὶ
  ὡς αἱ γυναῖκες διαλέγονταί τε ὁμοίως· καλεῦνταί τε οἱ τοιοῦτοι
  ἀνανδριεῖς. Οἱ μὲν οὖν ἐπιχώριοι τὴν αἰτίην προστιθέασι θεῷ καὶ
  σέβονται τουτέους τοὺς ἀνθρώπους καὶ προσκυνέουσι, δεδοικότες
  περὶ ἑωϋτέων ἕκαστοι. Ἐμοὶ δὲ καὶ αὐτέῳ δοκέει ταῦτα τὰ πάθεα
  θεῖα εἶναι, καὶ τἄλλα πάντα, καὶ οὐδὲν ἕτερον ἑτέρου θειότερον
  οὐδὲ ἀνθρωπινώτερον, ἀλλὰ πάντα θεῖα· ἕκαστον δὲ ἔχει φύσιν τῶν
  τοιουτέων, καὶ οὐδὲν ἄνευ φύσιος γίγνεται. Καὶ τοῦτο τὸ πάθος, ὥς
  μοι δοκέει γίγνεσθαι, φράσω, etc.

  Again, sect. 112. Ἀλλὰ γὰρ, ὥσπερ καὶ πρότερον ἔλεξα, θεῖα μὲν
  καὶ ταῦτά ἐστι ὁμοίως τοῖσι ἄλλοισι, γίγνεται δὲ κατὰ φύσιν
  ἕκαστα.

  Compare the remarkable treatise of Hippocratês, De Morbo Sacro,
  capp. 1 and 18, vol. vi. p. 352-394, ed. Littré. See this opinion
  of Hippocratês illustrated by the doctrines of some physical
  philosophers stated in Aristotle, Physic. ii. 8. ὥσπερ ὕει ὁ
  Ζεὺς, οὐχ ὅπως τὸν σῖτον αὐξήσῃ, ἀλλ᾽ ἐξ ἀνάγκης, etc. Some
  valuable observations on the method of Hippocratês are also found
  in Plato, Phædr. p. 270.

A third distinguished philosopher of the same day, Anaxagoras,
allegorizing Zeus and the other personal gods, proclaimed the doctrine
of one common pervading Mind, as having first established order and
system in the mundane aggregate, which had once been in a state of
chaos—and as still manifesting its uninterrupted agency for wise and
good purposes. This general doctrine obtained much admiration from
Plato and Aristotle; but they at the same time remarked with surprise,
that Anaxagoras never made any use at all of his own general doctrine
for the explanation of the phænomena of nature,—that he looked for
nothing but physical causes and connecting laws,[870]—so that in fact
the spirit of his particular researches was not materially different
from those of Demokritus or Leukippus, whatever might be the difference
in their general theories. His investigations in meteorology and
astronomy, treating the heavenly bodies as subjects for calculation,
have been already noticed as offensive, not only to the general public
of Greece, but even to Socratês himself among them: he was tried at
Athens, and seems to have escaped condemnation only by voluntary
exile.[871]

  [870] See the graphic picture in Plato, Phædon. p. 97-98 (cap.
  46-47): compare Plato, Legg. xii. p. 967; Aristotel. Metaphysic.
  i. p. 13-14 (ed. Brandis); Plutarch, Defect. Oracul. p. 435.

  Simplicius, Commentar. in Aristotel. Physic. p. 38. καὶ ὅπερ
  δὲ ὁ ἐν Φαίδωνι Σωκράτης ἐγκαλεῖ τῷ Ἀναξαγόρᾳ, τὸ ἐν ταῖς τῶν
  κατὰ μέρος αἰτιολογίαις μὴ τῷ νῷ κεχρῆσθαι, ἀλλὰ ταῖς ὑλικαῖς
  ἀποδόσεσιν, οἰκεῖον ἦν τῇ φυσιολογίᾳ. Anaxagoras thought that the
  superior intelligence of men, as compared with other animals,
  arose from his possession of hands (Aristot. de Part. Animal. iv.
  10. p. 687, ed. Bekk.).

  [871] Xenophôn, Memorab. iv. 7. Socratês said, καὶ παραφρονῆσαι
  τὸν ταῦτα μεριμνῶντα οὐδὲν ἧττον ἢ Ἀναξαγόρας παρεφρόνησεν, ὁ
  μέγιστον φρονήσας ἐπὶ τῷ τὰς τῶν θεῶν μηχανὰς ἐξηγεῖσθαι, etc.
  Compare Schaubach, Anaxagoræ Fragment. p. 50-141; Plutarch,
  Nikias, 23, and Periklês, 6-32; Diogen. Laërt. ii. 10-14.

  The Ionic philosophy, from which Anaxagoras receded more in
  language than in spirit, seems to have been the least popular
  of all the schools, though some of the commentators treat it as
  conformable to vulgar opinion, because it confined itself for the
  most part to phænomenal explanations, and did not recognize the
  _noumena_ of Plato, or the τὸ ἓν νοητὸν of Parmenidês,—“qualis
  fuit Ionicorum, quæ tum dominabatur, ratio, vulgari opinione
  et communi sensu comprobata” (Karsten, Parmenidis Fragment,
  De Parmenidis Philosophiâ, p. 154). This is a mistake: the
  Ionic philosophers, who constantly searched for and insisted
  upon physical laws, came more directly into conflict with the
  sentiment of the multitude than the Eleatic school.

  The larger atmospheric phænomena were connected in the most
  intimate manner with Grecian religious feeling and uneasiness
  (see Demokritus ap. Sect. Empiric. ix. sect. 19-24. p. 552-554,
  Fabric.): the attempts of Anaxagoras and Demokritus to explain
  them were more displeasing to the public than the Platonic
  speculations (Demokritus ap. Aristot. Meteorol. ii. 7; Stobæus,
  Eclog. Physic. p. 594: compare Mullach, Democriti Fragmenta, lib.
  iv. p. 394).

The three eminent men just named, all essentially different from
each other, may be taken as illustrations of the philosophical mind
of Greece during the last half of the fifth century B. C. Scientific
pursuits had acquired a powerful hold, and adjusted themselves in
various ways with the prevalent religious feelings of the age. Both
Hippocratês and Anaxagoras modified their ideas of the divine agency
so as to suit their thirst for scientific research. According to the
former, the gods were the really efficient agents in the production
of all phænomena,—the mean and indifferent not less than the terrific
or tutelary. Being thus alike connected with all phænomena, they were
specially associated with none—and the proper task of the inquirer
was, to find out those rules and conditions by which (he assumed)
their agency was always determined, and according to which it might
be foretold. And this led naturally to the proceeding which Plato and
Aristotle remark in Anaxagoras,—that the all-governing and Infinite
Mind, having been announced in sublime language at the beginning of
his treatise, was afterward left out of sight, and never applied to
the explanation of particular phænomena, being as much consistent with
one modification of nature as with another. Now such a view of the
divine agency could never be reconciled with the religious feelings
of the ordinary Grecian believer, even as they stood in the time
of Anaxagoras; still less could it have been reconciled with those
of the Homeric man, more than three centuries earlier. By him Zeus
and Athênê were conceived as definite Persons, objects of special
reverence, hopes, and fears, and animated with peculiar feelings,
sometimes of favor, sometimes of wrath, towards himself or his family
or country. They were propitiated by his prayers, and prevailed upon
to lend him succor in danger—but offended and disposed to bring evil
upon him if he omitted to render thanks or sacrifice. This sense of
individual communion with, and dependence upon them was the essence of
his faith; and with that faith, the all-pervading Mind proclaimed by
Anaxagoras—which had no more concern with one man or one phænomenon
than with another,—could never be brought into harmony. Nor could the
believer, while he prayed with sincerity for special blessings or
protection from the gods, acquiesce in the doctrine of Hippocratês,
that their agency was governed by constant laws and physical conditions.

That radical discord between the mental impulses of science and
religion, which manifests itself so decisively during the most
cultivated ages of Greece, and which harassed more or less so many
of the philosophers, produced its most afflicting result in the
condemnation of Socratês by the Athenians. According to the remarkable
passage recently cited from Xenophôn, it will appear that Socratês
agreed with his countrymen in denouncing physical speculations as
impious,—that he recognized the religious process of discovery as a
peculiar branch, coördinate with the scientific,—and that he laid down
a theory, of which the basis was, the confessed divergence of these
two processes from the beginning—thereby seemingly satisfying the
exigencies of religious hopes and fears on the one hand, and those of
reason, in her ardor for ascertaining the invariable laws of phænomena,
on the other. We may remark that the theory of this religious and
extra-scientific process of discovery was at that time sufficiently
complete; for Socratês could point out, that those anomalous phænomena
which the gods had reserved for themselves, and into which science was
forbidden to pry, were yet accessible to the seekings of the pious man,
through oracles, omens, and other exceptional means of communication
which divine benevolence vouchsafed to keep open. Considering thus
to how great an extent Socratês was identified in feeling with
the religious public of Athens, and considering moreover that his
performance of open religious duties was assiduous—we might wonder, as
Xenophôn does wonder,[872] how it could have happened that the Athenian
dikasts mistook him at the end of his life for an irreligious man.
But we see, by the defence which Xenophôn as well as Plato gives for
him, that the Athenian public really considered him, in spite of his
own disclaimer, as homogeneous with Anaxagoras and the other physical
inquirers, because he had applied similar scientific reasonings
to moral and social phænomena. They looked upon him with the same
displeasure as he himself felt towards the physical philosophers, and
we cannot but admit that in this respect they were more unfortunately
consistent than he was. It is true that the mode of defence adopted
by Socratês contributed much to the verdict found against him, and
that he was further weighed down by private offence given to powerful
individuals and professions; but all these separate antipathies found
their best account in swelling the cry against him as an over-curious
sceptic, and an impious innovator.

  [872] Xenophôn, Memorab. i. 1.

Now the scission thus produced between the superior minds and the
multitude, in consequence of the development of science and the
scientific point of view, is a fact of great moment in the history
of Greek progress, and forms an important contrast between the age
of Homer and Hesiod and that of Thucydidês; though in point of fact
even the multitude, during this later age, were partially modified by
those very scientific views which they regarded with disfavor. And
we must keep in view the primitive religious faith, once universal
and unobstructed, but subsequently disturbed by the intrusions of
science; we must follow the great change, as well in respect to
enlarged intelligence as to refinement of social and ethical feeling,
among the Greeks, from the Hesiodic times downward, in order to render
some account of the altered manner in which the ancient mythes came
to be dealt with. These mythes, the spontaneous growth of a creative
and personifying interpretation of nature, had struck root in Grecian
associations at a time when the national faith required no support
from what we call evidence. They were _now_ submitted, not simply to a
feeling, imagining, and believing public, but also to special classes
of instructed men,—philosophers, historians, ethical teachers, and
critics,—and to a public partially modified by their ideas[873] as well
as improved by a wider practical experience. They were not intended
for such an audience; they had ceased to be in complete harmony even
with the lower strata of intellect and sentiment,—much more so with
the higher. But they were the cherished inheritance of a past time;
they were interwoven in a thousand ways with the religious faith,
the patriotic retrospect, and the national worship, of every Grecian
community; the general type of the mythe was the ancient, familiar,
and universal form of Grecian thought, which even the most cultivated
men had imbibed in their childhood from the poets,[874] and by which
they were to a certain degree unconsciously enslaved. Taken as a whole
the mythes had acquired prescriptive and ineffaceable possession: to
attack, call in question, or repudiate them, was a task painful even to
undertake, and far beyond the power of any one to accomplish.

  [873] It is curious to see that some of the most recondite
  doctrines of the Pythagorean philosophy were actually brought
  before the general Syracusan public in the comedies of
  Epicharmus: “In comœdiis suis personas sæpe ita colloqui fecit,
  ut sententias Pythagoricas et in universum sublimia vitæ præcepta
  immisceret” (Grysar, De Doriensium Comœdiâ, p. 111, Col. 1828).
  The fragments preserved in Diogen. Laërt. (iii. 9-17) present
  both criticisms upon the Hesiodic doctrine of a primæval chaos,
  and an exposition of the archetypal and immutable ideas (as
  opposed to the fluctuating phænomena of sense) which Plato
  afterwards adopted and systematized.

  Epicharmus seems to have combined with this abstruse philosophy
  a strong vein of comic shrewdness and some turn to scepticism
  (Cicero, Epistol. ad Attic. i. 19): “ut crebro mihi vafer ille
  Siculus Epicharmus insusurret cantilenam suam.” Clemens Alex.
  Strom. v. p. 258. Νᾶφε καὶ μέμνασ᾽ ἀπιστεῖν· ἄρθρα ταῦτα τῶν
  φρενῶν. Ζῶμεν ἀριθμῷ καὶ λογισμῷ· ταῦτα γὰρ σώζει βροτοὺς. Also
  his contemptuous ridicule of the prophetesses of his time who
  cheated foolish women out of their money, pretending to universal
  knowledge, καὶ πάντα γιγνώσκοντι τῷ τηνᾶν λόγῳ (ap Polluc. ix.
  81). See, about Epicharmus. O. Müller, Dorians, iv. 7, 4.

  These dramas seem to have been exhibited at Syracuse between
  480-460 B. C., anterior even to Chionidês and Magnês at Athens
  (Aristot. Poet. c. 3): he says ~πολλῷ~ πρότερος, which can hardly
  be literally exact. The critics of the Horatian age looked upon
  Epicharmus as the prototype of Plautus (Hor. Epistol. ii. 1. 58).

  [874] The third book of the republic of Plato is particularly
  striking in reference to the use of the poets in education:
  see also his treatise De Legg. vii. p. 810-811. Some teachers
  made their pupils learn whole poets by heart (ὅλους ποιητὰς
  ἐκμανθάνων), others preferred extracts and selections.

For these reasons the anti-mythic vein of criticism was of no effect
as a destroying force, but nevertheless its dissolving decomposing
and transforming influence was very considerable. To accommodate the
ancient mythes to an improved tone of sentiment and a newly created
canon of credibility, was a function which even the wisest Greeks
did not disdain, and which occupied no small proportion of the whole
intellectual activity of the nation.

The mythes were looked at from a point of view completely foreign to
the reverential curiosity and literal imaginative faith of the Homeric
man; they were broken up and recast in order to force them into new
moulds such as their authors had never conceived. We may distinguish
four distinct classes of minds, in the literary age now under
examination, as having taken them in hand—the poets, the logographers,
the philosophers, and the historians.

With the poets and logographers, the mythical persons are real
predecessors, and the mythical world an antecedent fact; but it is
divine and heroic reality, not human; the present is only half-brother
of the past (to borrow[875] an illustration from Pindar in his allusion
to gods and men), remotely and generically, but not closely and
specifically, analogous to it. As a general habit, the old feelings
and the old unconscious faith, apart from all proof or evidence,
still remain in their minds; but recent feelings have grown up which
compel them to omit, to alter, sometimes even to reject and condemn,
particular narratives.

  [875] Pindar, Nem. vi. 1. Compare Simonidês, Fragm. 1 (Gaisford).

Pindar repudiates some stories and transforms others, because they
are inconsistent with his conceptions of the gods. Thus he formally
protests against the tale that Pelops had been killed and served up at
table by his father, for the immortal gods to eat; he shrinks from the
idea of imputing to them so horrid an appetite; he pronounces the tale
to have been originally fabricated by a slanderous neighbor. Nor can
he bring himself to recount the quarrels between different gods.[876]
The amours of Zeus and Apollo are no way displeasing to him; but he
occasionally suppresses some of the simple details of the old mythe,
as deficient in dignity: thus, according to the Hesiodic narrative,
Apollo was informed by a raven of the infidelity of the nymph Korônis:
but the mention of the raven did not appear to Pindar consistent
with the majesty of the god, and he therefore wraps up the mode of
detection in vague and mysterious language.[877] He feels considerable
repugnance to the character of Odysseus, and intimates more than once
that Homer has unduly exalted him, by force of poetical artifice.
With the character of the Æakid Ajax, on the other hand, he has the
deepest sympathy, as well as with his untimely and inglorious death,
occasioned by the undeserved preference of a less worthy rival.[878]
He appeals for his authority usually to the Muse, but sometimes to
“ancient sayings of men,” accompanied with a general allusion to
story-tellers and bards,—admitting, however, that these stories
present great discrepancy, and sometimes that they are false.[879]
Yet the marvellous and the supernatural afford no ground whatever for
rejecting a story: Pindar makes an express declaration to this effect
in reference to the romantic adventures of Perseus and the Gorgon’s
head.[880] He treats even those mythical characters, which conflict
the most palpably with positive experience, as connected by a real
genealogical thread with the world before him. Not merely the heroes
of Troy and Thêbes, and the demigod seamen of Jasôn and the ship Argô,
but also the Centaur Cheirôn, the hundred-headed Typhôs, the giant
Alkyoneus, Antæus, Bellerophôn and Pegasus, the Chimæra, the Amazons
and the Hyperboreans—all appear painted on the same canvas, and touched
with the same colors, as the men of the recent and recorded past,
Phalaris and Krœsus; only they are thrown back to a greater distance
in the perspective.[881] The heroic ancestors of those great Æginetan,
Thessalian, Thêban, Argeian, etc. families, whose present members the
poet celebrates for their agonistic victories, sympathize with the
exploits and second the efforts of their descendants: the inestimable
value of a privileged breed and of the stamp of nature is powerfully
contrasted with the impotence of unassisted teaching and practice.[882]
The power and skill of the Argeian Theæus and his relatives as
wrestlers, are ascribed partly to the fact that their ancestors
Pamphaês in aforetime had hospitably entertained the Tyndarids Kastôr
and Pollux.[883] Perhaps however the strongest proof of the sincerity
of Pindar’s mythical faith is afforded when he notices a guilty
incident with shame and repugnance, but with an unwilling confession of
its truth, as in the case of the fratricide committed on Phôkus by his
brothers Pêleus and Telamôn.[884]

  [876] Pindar, Olymp. i. 30-55; ix. 32-45.

  [877] Pyth. iii. 25. See the allusions to Semelê, Alkmêna, and
  Danaê, Pyth. iii. 98; Nem. x. 10. Compare also _supra_, chap. ix.
  p. 245.

  [878] Pindar, Nem. vii. 20-30; viii. 23-31. Isthm. iii. 50-60.

  It seems to be sympathy for Ajax, in odes addressed to noble
  Æginetan victors, which induces him thus to depreciate Odysseus;
  for he eulogizes Sisyphus, specially on account of his cunning
  and resources (Olymp. xiii. 50) in the ode addressed to Xenophôn
  the Corinthian.

  [879] Olymp. i. 28; Nem. viii. 20; Pyth. i. 93; Olymp. vii. 55;
  Nem. vi. 43. φάντι δ᾽ ἀνθρώπων παλαιαὶ ῥήσιες, etc.

  [880] Pyth. x. 49. Compare Pyth. xii. 11-22.

  [881] Pyth. i. 17; iii. 4-7; iv. 12; viii. 16. Nem. iv. 27-32;
  v. 89. Isthm. v. 31; vi. 44-48. Olymp. iii. 17; viii. 63; xiii.
  61-87.

  [882] Nem. iii. 39; v. 40. συγγενὴς εὐδοξία—πότμος συγγενής; v.
  8. Olymp. ix. 103. Pindar seems to introduce φύᾳ in cases where
  Homer would have mentioned the divine assistance.

  [883] Nem. x. 37-51. Compare the family legend of the Athenian
  Dêmocrates, in Plato, Lysis, p. 295.

  [884] Nem. v. 12-16.

Æschylus and Sophoklês exhibit the same spontaneous and uninquiring
faith as Pindar in the legendary antiquities of Greece, taken as a
whole; but they allow themselves greater license as to the details. It
was indispensable to the success of their compositions that they should
recast and group anew the legendary events, preserving the names and
general understood relation of those characters whom they introduced.
The demand for novelty of combination increased with the multiplication
of tragic spectacles at Athens: moreover the feelings of the Athenians,
ethical as well as political, had become too critical to tolerate the
literal reproduction of many among the ancient stories.

Both of them exalted rather than lowered the dignity of the mythical
world, as something divine and heroic rather than human. The
Promêtheus of Æschylus is a far more exalted conception than his
keen-witted namesake in Hesiod, and the more homely details of
the ancient Thêbaïs and Œdipodia were in like manner modified by
Sophoklês.[885] The religious agencies of the old epic are constantly
kept prominent, and the paternal curse,—the wrath of deceased persons
against those from whom they have sustained wrong,—the judgments of
the Erinnys against guilty or foredoomed persons, sometimes inflicted
directly, sometimes brought about through dementation of the sufferer
himself (like the Homeric Atê),—are frequent in their tragedies.[886]

  [885] See above, chap. xiv. p. 368. on the Legend of the Siege of
  Thêbes.

  [886] The curse of Œdipus is the determining force in the
  Sept. ad Thêb., Ἀρά τ᾽, Ἐριννὺς πατρὸς ἡ μεγασθενής (v. 70);
  it reappears several times in the course of the drama, with
  particular solemnity in the mouth of Eteoklês (695-709, 725, 785,
  etc.); he yields to it as an irresistible force, as carrying the
  family to ruin:—

    Ἐπεὶ τὸ πρᾶγμα κάρτ᾽ ἐπισπέρχει θεὸς,
    Ἴτω κατ᾽ οὖρον, κῦμα Κωκυτοῦ λαχὸν,
    Φοίβῳ στυγηθὲν πᾶν τὸ Λαΐου γένος.
       ·   ·   ·   ·   ·
    Φίλου γὰρ ἐχθρά μοι πατρὸς τέλει᾽ ἄρα
    Ξηροῖς ἀκλαύστοις ὄμμασιν προσιζάνει, etc.

  So again at the opening of the Agamemnôn, the μνάμων μῆνις
  τεκνόποινος (v. 155) and the sacrifice of Iphigeneia are dwelt
  upon as leaving behind them an avenging doom upon Agamemnôn,
  though he took precautions for gagging her mouth during the
  sacrifice and thus preventing her from giving utterance to
  imprecations—Φθόγγον ἀραῖον οἴκοις, Βίᾳ χαλινῶν τ᾽ ἀναύδῳ μένει
  (κατασχεῖν), v. 346. The Erinnys awaits Agamemnôn even at the
  moment of his victorious consummation at Troy (467; compare
  762-990, 1336-1433): she is most to be dreaded after great
  good fortune: she enforces the curse which ancestral crimes
  have brought upon the house of Atreus—πρώταρχος ἄτη—παλαιαὶ
  ἁμαρτίαι δόμων (1185-1197, Choëph. 692)—the curse imprecated by
  the outraged Thyestês (1601). In the Choëphoræ, Apollo menaces
  Orestês with the wrath of his deceased father, and all the
  direful visitations of the Erinnyes, unless he undertakes to
  revenge the murder (271-296). Αἶσα and Ἐριννὺς bring on blood
  for blood (647). But the moment that Orestês, placed between
  these conflicting obligations (925), has achieved it, he becomes
  himself the victim of the Erinnyes, who drive him mad even at
  the end of the Choëphoræ (ἕως δ᾽ ἔτ᾽ ἔμφρων εἰμὶ, 1026), and who
  make their appearance bodily, and pursue him throughout the third
  drama of this fearful trilogy. The Eidôlon of Klytæmnêstra impels
  them to vengeance (Eumenid. 96) and even spurs them on when they
  appear to relax. Apollo conveys Orestês to Athens, whither the
  Erinnyes pursue him, and prosecute him before the judgment-seat
  of the goddess Athênê, to whom they submit the award; Apollo
  appearing as his defender. The debate between “the daughters of
  Night” and the god, accusing and defending, is eminently curious
  (576-730): the Erinnyes are deeply mortified at the humiliation
  put upon them when Orestês is acquitted, but Athênê at length
  reconciles them, and a covenant is made whereby they become
  protectresses of Attica, accepting of a permanent abode and
  solemn worship (1006): Orestês returns to Argos, and promises
  that even in his tomb he will watch that none of his descendants
  shall ever injure the land of Attica (770). The solemn trial and
  acquittal of Orestês formed the consecrating legend of the Hill
  and Judicature of Areiopagus.

  This is the only complete trilogy of Æschylus which we possess,
  and the avenging Erinnyes (416) are the movers throughout the
  whole—unseen in the first two dramas, visible and appalling in
  the third. And the appearance of Cassandra under the actual
  prophetic fever in the first, contributes still farther to impart
  to it a coloring different from common humanity.

  The general view of the movement of the Oresteia given in Welcker
  (Æschyl. Trilogie, p. 445) appears to me more conformable to
  Hellenic ideas than that of Klausen (Theologumena Æschyli, pp.
  157-169), whose valuable collection and comparison of passages
  is too much affected, both here and elsewhere, by the desire to
  bring the agencies of the Greek mythical world into harmony with
  what a religious mind of the present day would approve. Moreover,
  he sinks the personality of Athênê too much in the supreme
  authority of Zeus (p. 158-168).

Æschylus in two of his remaining pieces brings forward the gods as
the chief personages, and far from sharing the objection of Pindar
to dwell upon dissensions of the gods, he introduces Promêtheus and
Zeus in the one, Apollo and the Eumenidês in the other, in marked
opposition. The dialogue, first superinduced by him upon the primitive
Chorus, gradually became the most important portion of the drama, and
is more elaborated in Sophoklês than in Æschylus. Even in Sophoklês,
however, it still generally retains its ideal majesty as contrasted
with the rhetorical and forensic tone which afterwards crept in; it
grows out of the piece, and addresses itself to the emotions more than
to the reason of the audience. Nevertheless, the effect of Athenian
political discussion and democratical feeling is visible in both these
dramatists. The idea of rights and legitimate privileges as opposed to
usurping force, is applied by Æschylus even to the society of the gods:
the Eumenidês accuse Apollo of having, with the insolence of youthful
ambition, “ridden down” their old prerogatives[887]—while the Titan
Promêtheus, the champion of suffering humanity against the unfriendly
dispositions of Zeus, ventures to depict the latter as a recent usurper
reigning only by his superior strength, exalted by one successful
revolution, and destined at some future time to be overthrown by
another,—a fate which cannot be averted except through warnings
communicable only by Promêtheus himself.[888]

  [887] Eumenidês, 150.—

    Ἰὼ παῖ Διὸς, ἐπίκλοπος πέλει,
    Νέος δὲ γραίας δαίμονας καθιππάσω, etc.

  The same metaphor again, v. 731. Æschylus seems to delight in
  contrasting the young and the old gods: compare 70-162, 882.

  The Erinnyes tell Apollo that he assumes functions which do not
  belong to him, and will thus desecrate those which do belong to
  him (715-754):—

    Ἀλλ᾽ αἱματηρὰ πράγματ᾽, ~οὐ λαχὼν~, σέβεις,
    Μαντεῖα δ᾽ οὐκ ἔθ᾽ ἁγνὰ μαντεύσει μένων.

  The refusal of the king Pelasgos, in the Supplices, to undertake
  what he feels to be the sacred duty of protecting the suppliant
  Danaïdes, without first submitting the matter to his people
  and obtaining their expressed consent, and the fear which he
  expresses of their blame (κατ᾽ ἀρχὰς γὰρ φιλαίτιος λέως), are
  more forcibly set forth than an old epic poet would probably have
  thought necessary (see Supplices, 369, 397, 485, 519). The solemn
  wish to exclude both anarchy and despotism from Athens bears
  still more the mark of political feeling of the time—μήτ᾽ ἄναρχον
  μήτε δεσποτουμένον (Eumenid. 527-696).

  [888] Promêtheus, 35, 151, 170, 309, 524, 910, 940, 956.

It is commonly understood that Æschylus disapproved of the march of
democracy at Athens during his later years, and that the Eumenidês
is intended as an indirect manifestation in favor of the senate of
Areiopagus. Without inquiring at present whether such a special
purpose can be distinctly made out, we may plainly see that the poet
introduces, into the relations of the gods with each other, a feeling
of political justice, arising out of the times in which he lived and
the debates of which he was a witness. But though Æschylus incurred
reproaches of impiety from Plato, and seemingly also from the Athenian
public, for particular speeches and incidents in his tragedies,[889]
and though he does not adhere to the received vein of religious
tradition with the same strictness as Sophoklês—yet the ascendency
and interference of the gods is never out of sight, and the solemnity
with which they are represented, set off by a bold, figurative, and
elliptical style of expression (often but imperfectly intelligible to
modern readers), reaches its maximum in his tragedies. As he throws
round the gods a kind of airy grandeur, so neither do his men or
heroes appear like tenants of the common earth: the mythical world from
which he borrows his characters is peopled only with “the immediate
seed of the gods, in close contact with Zeus, in whom the divine blood
has not yet had time to degenerate:”[890] his individuals are taken,
not from the iron race whom Hesiod acknowledges with shame as his
contemporaries, but from the extinct heroic race which had fought at
Troy and Thêbes. It is to them that his conceptions aspire, and he is
even chargeable with frequent straining, beyond the limits of poetical
taste, to realize his picture. If he does not consistently succeed in
it, the reason is because consistency in such a matter is unattainable,
since, after all, the analogies of common humanity, the only materials
which the most creative imagination has to work upon, obtrude
themselves involuntarily, and the lineaments of the man are thus seen
even under a dress which promises superhuman proportions.

  [889] Plato, Republ. ii. 381-383; compare Æschyl. Fragment. 159,
  ed. Dindorf. He was charged also with having divulged in some of
  his plays secret matters of the mysteries of Dêmêtêr, but is said
  to have excused himself by alleging ignorance: he was not aware
  that what he had said was comprised in the mysteries (Aristot.
  Ethic. Nicom. iii. 2; Clemens Alex. Strom. ii. p. 387); the story
  is different again in Ælian, V. H. v. 19.

  How little can be made out distinctly respecting this last
  accusation may be seen in Lobeck, Aglaopham. p. 81.

  Cicero (Tusc. Dis. ii. 10) calls Æschylus “almost a Pythagorean:”
  upon what the epithet is founded we do not know.

  There is no evidence to prove to us that the Promêtheus Vinctus
  was considered as impious by the public before whom it was
  represented; but its obvious meaning has been so regarded by
  modern critics, who resort to many different explanations of it,
  in order to prove that when properly construed it is not impious.
  But if we wish to ascertain what Æschylus really meant, we ought
  not to consult the religious ideas of modern times; we have no
  test except what we know of the poet’s own time and that which
  had preceded him. The explanations given by the ablest critics
  seem generally to exhibit a predetermination to bring out Zeus
  as a just, wise, merciful, and all-powerful Being; and all, in
  one way or another, distort the figures, alter the perspective,
  and give far-fetched interpretations of the meaning, of this
  striking drama, which conveys an impression directly contrary
  (see Welcker, Trilogie, Æsch. p. 90-117, with the explanation
  of Dissen there given; Klausen, Theologum. Æsch. p. 140-154;
  Schömann, in his recent translation of the play, and the
  criticism on that Translation in the Wiener Jahrbücher, vol. cix.
  1845, p. 245, by F. Ritter). On the other hand, Schutz (Excurs.
  ad Prom. Vinct. p. 149) thinks that Æschylus wished by means of
  this drama to enforce upon his countrymen the hatred of a despot.
  Though I do not agree in this interpretation, it appears to me
  less wide of the truth than the forcible methods employed by
  others to bring the poet into harmony with their own religious
  ideas.

  Without presuming to determine whether Æschylus proposed to
  himself any special purpose, if we look at the Æschylean
  Promêtheus in reference only to ancient ideas, it will be found
  to borrow both its characters and all its main circumstances
  from the legend in the Hesiodic Theogony. Zeus acquires his
  supremacy only by overthrowing Kronos and the Titans; the Titan
  god Promêtheus is the pronounced champion of helpless man, and
  negotiates with Zeus on their behalf: Zeus wishes to withhold
  from them the most essential blessings, which Promêtheus employs
  deceit and theft to procure for them, and ultimately with
  success; undergoing, however, severe punishment for so doing from
  the superior force of Zeus. These are the main features of the
  Æschylean Promêtheus, and they are all derived from the legend
  as it stands in the Theogony. As for the human race, they are
  depicted as abject and helpless in an extreme degree, in Æschylus
  even more than in Hesiod: they appear as a race of aboriginal
  savages, having the god Promêtheus for their protector.

  Æschylus has worked up the old legend, homely and unimpressive as
  we read it in Hesiod, into a sublime ideal. We are not to forget
  that Promêtheus is not a man, but a god,—the equal of Zeus in
  race, though his inferior in power, and belonging to a family of
  gods who were once superior to Zeus: he has moreover deserted his
  own kindred, and lent all his aid and superior sagacity to Zeus,
  whereby chiefly the latter was able to acquire supremacy (this
  _last_ circumstance is an addition by Æschylus himself to the
  Hesiodic legend). In spite of such essential service, Zeus had
  doomed him to cruel punishment, for no other reason than because
  he conferred upon helpless man the prime means of continuance and
  improvement, thus thwarting the intention of Zeus to extinguish
  the race.

  Now Zeus, though superior to all the other gods and exercising
  general control, was never considered, either in Grecian legend
  or in Grecian religious belief, to be superior in so immeasurable
  a degree as to supersede all free action and sentiment on the
  part of gods less powerful. There were many old legends of
  dissension among the gods, and several of disobedience against
  Zeus: when a poet chose to dramatize one of these, he might
  so turn his composition as to sympathize either with Zeus or
  with the inferior god, without in either case shocking the
  general religious feeling of the country. And if there ever
  was an instance in which preference of the inferior god would
  be admissible, it is that of Promêtheus, whose proceedings are
  such as to call forth the maximum of human sympathy,—superior
  intelligence pitted against superior force, and resolutely
  encountering foreknown suffering, for the sole purpose of
  rendering inestimable and gratuitous service to mortals.

  Of the Promêtheus Solutus, which formed a sequel to the
  Promêtheus Vinctus (the entire trilogy is not certainly known),
  the fragments preserved are very scanty, and the guesses of
  critics as to its plot have little base to proceed upon. They
  contend that, in one way or other, the apparent objections which
  the Promêth. Vinctus presents against the justice of Zeus were
  in the Promêth. Solutus removed. Hermann, in his _Dissertatio
  de Æschyli Prometheo Soluto_ (Opuscula, vol. iv. p. 256), calls
  this position in question: I transcribe from his Dissertation one
  passage, because it contains an important remark in reference
  to the manner in which the Greek poets handled their religious
  legends: “while they recounted and believed many enormities
  respecting individual gods, they always described the Godhead in
  the abstract as holy and faultless.” ...

  “Immo illud admirari oportet, quod quum de singulis Diis
  indignissima quæque crederent, tamen ubi sine certo nomine Deum
  dicebant, immunem ab omni vitio, summâque sanctitate præditum
  intelligebant. Illam igitur Jovis sævitiam ut excusent defensores
  Trilogiæ, et jure punitum volunt Prometheum—et in sequente fabulâ
  reconciliato Jove, restitutam arbitrantur divinam justitiam. Quo
  invento, vereor ne non optime dignitati consuluerint supremi
  Deorum, quem decuerat potius non sævire omnino, quam placari eâ
  lege, ut alius Promethei vice lueret.”

  [890] Æschyl. Fragment. 146, Dindorf; ap. Plato. Repub. iii. p.
  391; compare Strabo, xii. p. 580.—

    ... οἱ θεῶν ἀγχίσποροι
    Οἱ Ζηνὸς ἐγγὺς, οἷς ἐν Ἰδαίῳ πάγῳ
    Διὸς πατρῴου βωμός ἐστ᾽ ἐν αἰθέρι,
    Κοὔπω σφιν ἐξίτηλον αἷμα δαιμόνων.

  There is one real exception to this statement—the Persæ—which
  is founded upon an event of recent occurrence; and one apparent
  exception—the Promêtheus Vinctus. But in that drama no individual
  mortal is made to appear; we can hardly consider Iô as an
  ἐφήμερος (253).

Sophoklês, the most illustrious ornament of Grecian tragedy, dwells
upon the same heroic characters, and maintains their grandeur, on the
whole, with little abatement, combining with it a far better dramatic
structure, and a wider appeal to human sympathies. Even in Sophoklês,
however, we find indications that an altered ethical feeling and a
more predominant sense of artistic perfection are allowed to modify
the harsher religious agencies of the old epic; occasional misplaced
effusions[891] of rhetoric, as well as of didactic prolixity, may
also be detected. It is Æschylus, not Sophoklês, who forms the marked
antithesis to Euripidês; it is Æschylus, not Sophoklês, to whom
Aristophanês awards the prize of tragedy, as the poet who assigns most
perfectly to the heroes of the past those weighty words, imposing
equipments, simplicity of great deeds with little talk, and masculine
energy superior to the corruptions of Aphroditê, which beseem the
comrades of Agamemnôn and Adrastus.[892]

  [891] For the characteristics of Æschylus see Aristophan. Ran.
  755, _ad fin. passim_. The competition between Æschylus and
  Euripidês turns upon γνῶμαι ἀγαθαὶ, 1497; the weight and majesty
  of the words, 1362; πρῶτος τῶν Ἑλλήνων πυργώσας ῥήματα σεμνά,
  1001, 921, 930 (“sublimis et gravis et grandiloquus sæpe usque ad
  vitium,” Quintil. x. 1); the imposing appearance of his heroes,
  such as Memnôn and Cycnus, 961; their reserve in speech, 908; his
  dramas “full of Arês” and his lion-hearted chiefs, inspiring the
  auditors with fearless spirit in defence of their country,—1014,
  1019, 1040; his contempt of feminine tenderness, 1042.—

     ÆSCH. Οὐδ᾽ οἶδ᾽ οὐδεὶς ἥντιν᾽ ἐρῶσαν πώποτ᾽ ἐποίησα γυναῖκα.
    EURIP. Μὰ Δί᾽, οὐδὲ γὰρ ἦν τῆς Ἀφροδίτης οὐδέν σοι.
     ÆSCH.                                             μηδέ γ᾽ ἐπείη·
           Ἀλλ᾽ ἐπὶ σοί τοι καὶ τοῖς σοῖσιν πολλὴ πολλοῦ ᾽πικάθοιτο.

  To the same general purpose Nubes (1347-1356), composed so many
  years earlier. The weight and majesty of the Æschylean heroes
  (βάρος, τὸ μεγαλοπρεπὲς) is dwelt upon in the life of Æschylus,
  and Sophoklês is said to have derided it—Ὥσπερ γὰρ ὁ Σοφοκλῆς
  ἔλεγε, τὸν Αἰσχύλου διαπεπαιχὼς ~ὄγκον~, etc. (Plutarch, De
  Profect. in Virt. Sent. c. 7), unless we are to understand this
  as a mistake of Plutarch quoting Sophoklês instead of Euripidês,
  as he speaks in the Frogs of Aristophanês, which is the opinion
  both of Lessing in his Life of Sophoklês and of Welcker (Æschyl.
  Trilogie, p. 525).

  [892] See above, Chapters xiv. and xv.

  Æschylus seems to have been a greater innovator as to the matter
  of the mythes than either Sophoklês or Euripidês (Dionys. Halic.
  Judic. de Vett. Script. p. 422, Reisk.). For the close adherence
  of Sophoklês to the Homeric epic, see Athena, vii. p. 277;
  Diogen. Laërt. iv. 20; Suidas, v. Πολέμων. Æschylus puts into
  the mouth of the Eumenidês a serious argument derived from the
  behavior of Zeus in chaining his father Kronos (Eumen. 640).

How deeply this feeling, of the heroic character of the mythical
world, possessed the Athenian mind, may be judged by the bitter
criticisms made on Euripidês, whose compositions were pervaded, partly
by ideas of physical philosophy learnt under Anaxagoras, partly by
the altered tone of education and the wide diffusion of practical
eloquence, forensic as well as political, at Athens.[893] While
Aristophanês assails Euripidês as the representative of this “young
Athens,” with the utmost keenness of sarcasm,—other critics also
concur in designating him as having vulgarized the mythical heroes,
and transformed them into mere characters of common life,—loquacious,
subtle, and savoring of the market-place.[894] In some of his plays,
sceptical expressions and sentiments were introduced, derived from his
philosophical studies, sometimes confounding two or three distinct gods
into one, sometimes translating the personal Zeus into a substantial
Æthêr with determinate attributes. He put into the mouths of some of
his unprincipled dramatic characters, apologetic speeches which were
denounced as ostentatious sophistry, and as setting out a triumphant
case for the criminal.[895] His thoughts, his words, and the rhythm
of his choric songs, were all accused of being deficient in dignity
and elevation. The mean attire and miserable attitude in which he
exhibited Œneus, Têlephus, Thyestês, Inô, and other heroic characters,
were unmercifully derided,[896] though it seems that their position and
circumstances had always been painfully melancholy; but the effeminate
pathos which Euripidês brought so nakedly into the foreground, was
accounted unworthy of the majesty of a legendary hero. And he incurred
still greater obloquy on another point, on which he is allowed even
by his enemies to have only reproduced in substance the preëxisting
tales,—the illicit and fatal passion depicted in several of his female
characters, such as Phædra and Sthenobœa. His opponents admitted that
these stories were true, but contended that they ought to be kept back
and not produced upon the stage,—a proof both of the continued mythical
faith and of the more sensitive ethical criticism of his age.[897] The
marriage of the six daughters to the six sons of Æolus is of Homeric
origin, and stands now, though briefly stated, in the Odyssey: but the
incestuous passion of Macareus and Canacê, embodied by Euripidês[898]
in the lost tragedy called _Æolus_, drew upon him severe censure.
Moreover, he often disconnected the horrors of the old legends with
those religious agencies by which they had been originally forced on,
prefacing them by motives of a more refined character, which carried
no sense of awful compulsion: thus the considerations by which the
Euripidean Alkmæôn was reduced to the necessity of killing his mother
appeared to Aristotle ridiculous.[899] After the time of this great
poet, his successors seem to have followed him in breathing into their
characters the spirit of common life, but the names and plot were still
borrowed from the stricken mythical families of Tantalus, Kadmus, etc.:
and the heroic exaltation of all the individual personages introduced,
as contrasted with the purely human character of the Chorus, is still
numbered by Aristotle among the essential points of the theory of
tragedy.[900]

  [893] See Valckenaer, Diatribe in Euripid. Fragm. capp. 5 and 6.

  The fourth and fifth lectures among the _Dramatische Vorlesungen_
  of August Wilhelm Schlegel depict both justly and eloquently the
  difference between Æschylus, Sophoklês and Euripidês, especially
  on this point of the gradual sinking of the mythical colossus
  into an ordinary man; about Euripidês especially in lecture 5,
  vol. i. p. 206, ed. Heidelberg 1809.

  [894] Aristot. Poetic. c. 46. Οἷον καὶ Σοφοκλῆς ἔφη, αὐτὸς μὲν
  οἵους δεῖ ποιεῖν, Εὐριπίδης δὲ, οἷοί εἰσι.

  The Ranæ and Acharneis of Aristophanês exhibit fully the
  reproaches urged against Euripidês: the language put into
  the mouth of Euripidês in the former play (vv. 935-977)
  illustrates specially the point here laid down. Plutarch (De
  Gloriâ Atheniens. c. 5) contrasts ἡ Εὐριπίδου σοφία καὶ ἡ
  Σοφοκλεοῦς λογιότης. Sophoklês either adhered to the old mythes
  or introduced alterations into them in a spirit comformable to
  their original character, while Euripidês refined upon them. The
  comment of Dêmêtrius Phalereus connects τὸ λόγιον expressly with
  the maintenance of the dignity of the tales. Ἄρξομαι δὲ ἀπὸ τοῦ
  μεγαλοπρεποῦς, ὅπερ νῦν ~λόγιον~ ὀνομάζουσιν (c. 38).

  [895] Aristophan. Ran. 770, 887, 1066.

  Euripidês says to Æschylus, in regard to the language employed by
  both of them,—

              Ἦν οὖν σὺ λέγῃς Λυκαβήττους
    Καὶ Παρνάσσων ἡμῖν μεγέθη, τοῦτ᾽ ἐστὶ τὸ χρηστὰ διδάσκειν,
    Ὃν χρὴ φράζειν ἀνθρωπείως;

  Æschylus replies,—

              Ἀλλ᾽, ὦ κακόδαιμον, ἀνάγκη
    Μεγάλων γνωμῶν καὶ διανοιῶν ἴσα καὶ τὰ ῥήματα τίκτειν.
    Κἄλλως εἰκὸς τοὺς ~ἡμιθέους~ τοῖς ῥήμασι μείζοσι χρῆσθαι·
    Καὶ γὰρ τοῖς ἱματίοις ἡμῶν χρῶνται πολὺ σεμνοτέροισι.
    Ἃ ᾽μοῦ χρηστῶς καταδείξαντος διελυμήνω συ.
    EURIP.                                     Τί δράσας;
     ÆSCH. Πρῶτον μὲν τοὺς βασιλεύοντας ῥάκι᾽ ἀμπίσχων, ἵν᾽ ἐλεινοὶ
           Τοῖς ἀνθρώποις φαίνοιντ᾽ εἶναι.

  For the character of the language and measures of Euripidês,
  as represented by Æschylus, see also v. 1297, and Pac. 527.
  Philosophical discussion was introduced by Euripidês (Dionys.
  Hal. Ars Rhetor. viii. 10-ix. 11) about the Melanippê, where
  the doctrine of prodigies (τέρας) appears to have been argued.
  Quintilian (x. 1) remarks that to young beginners in judicial
  pleading, the study of Euripidês was much more specially
  profitable than that of Sophoklês: compare Dio Chrysostom, Orat.
  xviii. vol. i. p. 477, Reisk.

  In Euripidês the heroes themselves sometimes delivered
  moralizing discourses:—εἰσάγων τὸν Βελλεροφόντην γνωμολογοῦντα
  (Welcker, Griechisch. Tragöd. Eurip. Stheneb. p. 782). Compare
  the fragments of his Bellerophôn (15-25, Matthiæ), and of his
  Chrysippus (7, _ib._). A striking story is found in Seneca,
  Epistol. 115; and Plutarch, de Audiend. Poetis, c. 4. t. i. p.
  70, Wytt.

  [896] Aristophan. Ran. 840.—

                      ὦ στωμυλιοσυλλεκτάδη
    Καὶ πτωχοποιὲ καὶ ῥακιοσυῤῥαπτάδη·

  See also Aristophan. Acharn. 385-422. For an unfavorable
  criticism upon such proceeding, see Aristot. Poet. 27.

  [897] Aristophan. Ran. 1050.—

  EURIP. Πότερον δ᾽ ~οὐκ ὄντα λόγον~ τοῦτον περὶ τῆς Φαίδρας ξυνέθηκα;

   ÆSCH. Μὰ Δί᾽, ἀλλ᾽ ὄντ᾽· ἀλλ᾽ ἀποκρύπτειν χρὴ τὸ πονηρὸν τόν γε ποιητὴν,
           Καὶ μὴ παράγειν μηδὲ διδάσκειν.

  In the Hercules Furens, Euripidês puts in relief and even
  exaggerates the worst elements of the ancient mythes: the
  implacable hatred of Hêrê towards Hêraklês is pushed so far as to
  deprive him of his reason (by sending down Iris and the unwilling
  Λύσσα), and thus intentionally to drive him to slay his wife and
  children with his own hands.

  [898] Aristoph. Ran. 849, 1041, 1080; Thesmophor. 547; Nubes,
  1354. Grauert, De Mediâ Græcorum Comœdiâ in Rheinisch. Museum,
  2nd Jahrs. 1 Heft, p. 51. It suited the plan of the drama of
  Æolus, as composed by Euripidês, to place in the mouth of
  Macareus a formal recommendation of incestuous marriages:
  probably this contributed much to offend the Athenian public. See
  Dionys. Hal. Rhetor. ix. p. 355.

  About the liberty of intermarriage among relatives, indicated in
  Homer, parents and children being alone excepted, see Terpstra,
  Antiquitas Homerica, cap. xiii. p. 104.

  Ovid, whose poetical tendencies led him chiefly to copy
  Euripidês, observes (Trist. ii. 1, 380)—

    “Omne genus scripti gravitate Tragœdia vincit,
      Hæc quoque materiam semper amoris habet.
    Nam quid in Hippolyto nisi cæcæ flamma novercæ?
      Nobilis est Canace fratris amore sui.”

  This is the reverse of the truth in regard to Æschylus and
  Sophoklês, and only very partially true in respect to Euripidês.

  [899] Aristot. Ethic. Nicom. iii. 1, 8. καὶ γὰρ τὸν Εὐριπίδου
  Ἀλκμαίωνα γελοῖα φαίνεται τὰ ἀναγκάσαντα μητροκτονῆσαι (In the
  lost tragedy called Ἀλκμαίων ὁ διὰ Ψωφῖδος).

  [900] Aristot. Poetic. 26-27. And in his Problemata also, in
  giving the reason why the Hypo-Dorian and Hypo-Phrygian musical
  modes were never assigned to the Chorus, he says—

  Ταῦτα δὲ ἄμφω χόρῳ μὲν ἀναρμοστὰ, τοῖς δὲ ἀπὸ σκηνῆς οἰκειότερα.
  Ἐκεῖνοι μὲν γὰρ ἡρώων μίμηται· οἱ δὲ ἡγεμόνες τῶν ἀρχαίων μόνοι
  ἦσαν ἥρωες, οἱ δὲ λαοὶ ἄνθρωποι, ὧν ἐστὶν ὁ χόρος. Διὸ καὶ
  ἁρμόζει αὐτῷ τὸ γοερὸν καὶ ἡσύχιον ἦθος καὶ μέλος· ἀνθρωπικὰ γάρ.

The tendency then of Athenian tragedy—powerfully manifested in
Æschylus, and never wholly lost—was to uphold an unquestioning faith
and a reverential estimate of the general mythical world and its
personages, but to treat the particular narratives rather as matter
for the emotions than as recitals of actual fact. The logographers
worked along with them to the first of these two ends, but not
to the second. Their grand object was, to cast the mythes into a
continuous readable series, and they were in consequence compelled to
make selection between inconsistent or contradictory narratives; to
reject some narratives as false, and to receive others as true. But
their preference was determined more by their sentiments as to what
was appropriate, than by any pretended historical test. Pherekydês,
Akusilaus and Hellanikus[901] did not seek to banish miraculous or
fantastic incidents from the mythical world; they regarded it as
peopled with loftier beings, and expected to find in it phænomena
not paralleled in their own degenerate days. They reproduced the
fables as they found them in the poets, rejecting little except the
discrepancies, and producing ultimately what they believed to be
not only a continuous but an exact and trustworthy history of the
past—wherein they carry indeed their precision to such a length, that
Hellanikus gives the year, and even the day of the capture of Troy.[902]

  [901] See Müller, Prolegom. zu einer wissenschaftlichen
  Mythologie, c. iii. p. 93.

  [902] Hellanic. Fragment. 143, ed. Didot.

Hekatæus of Milêtus (500 B. C.), anterior to Pherekydês and Hellanikus,
is the earliest writer in whom we can detect any disposition to
disallow the prerogative and specialty of the mythes, and to soften
down their characteristic prodigies, some of which however still find
favor in his eyes, as in the case of the speaking ram who carried
Phryxus over the Hellespont. He pronounced the Grecian fables to be
“many and ridiculous;” whether from their discrepancies or from their
intrinsic improbabilities we do not know: and we owe to him the first
attempt to force them within the limits of historical credibility; as
where he transforms the three-headed Cerberus, the dog of Hadês, into
a serpent inhabiting a cavern on Cape Tænarus—and Geryôn of Erytheia
into a king of Epirus rich in herds of oxen.[903] Hekatæus traced the
genealogy of himself and the gens to which he belonged through a line
of fifteen progenitors up to an initial god,[904]—the clearest proof
both of his profound faith in the reality of the mythical world, and
of his religious attachment to it as the point of junction between the
human and the divine personality.

  [903] Hekatæi Fragm. ed. Didot. 332, 346, 349; Schol. Apollôn.
  Rhod. 1. 256; Athenæ. ii. p. 133; Skylax, c. 26.

  Perhaps Hekatæus was induced to look for Erytheia in Epirus by
  the brick-red color of the earth there in many places, noticed
  by Pouqueville and other travellers (Voyage dans la Grèce,
  vol. ii. 248: see Klausen, Æneas und die Penaten, vol. i. p.
  222). Ἑκαταῖος ὁ Μιλήσιος—λόγον εὖρεν εἰκότα, Pausan. iii. 25,
  4. He seems to have written expressly concerning the fabulous
  Hyperboreans, and to have upheld the common faith against doubts
  which had begun to rise in his time: the derisory notice of
  Hyperboreans in Herodotus is probably directed against Hekatæus,
  iv. 36; Schol. Apollôn. Rhod. ii. 675; Diodôr. ii. 47.

  It is maintained by Mr. Clinton (Fast. Hell. ii. p. 480) and
  others (see not. ad Fragment. Hecatæi, p. 30, ed. Didot), that
  the work on the Hyperboreans was written by Hekatæus of Abdêra, a
  literary Greek of the age of Ptolemy Philadelphus—not by Hekatæus
  of Milêtus. I do not concur in this opinion. I think it much more
  probable that the earlier Hekatæus was the author spoken of.

  The distinguished position held by Hekatæus at Milêtus is marked
  not only by the notice which Herodotus takes of his opinions on
  public matters, but also by his negotiation with the Persian
  satrap Artaphernes on behalf of his countrymen (Diodôr. Excerpt.
  xlvii. p. 41, ed. Dindorf).

  [904] Herodot. ii. 143.

We have next to consider the historians, especially Herodotus and
Thucydidês. Like Hekatæus, Thucydidês belonged to a gens which traced
its descent from Ajax, and through Ajax to Æakus and Zeus.[905]
Herodotus modestly implies that he himself had no such privilege to
boast of.[906] Their curiosity respecting the past had no other
materials to work upon except the mythes; but these they found already
cast by the logographers into a continuous series, and presented as an
aggregate of antecedent history, chronologically deduced from the times
of the gods. In common with the body of the Greeks, both Herodotus
and Thucydidês had imbibed that complete and unsuspecting belief in
the general reality of mythical antiquity, which was interwoven with
the religion and the patriotism, and all the public demonstrations of
the Hellenic world. To acquaint themselves with the genuine details
of this foretime, was an inquiry highly interesting to them: but the
increased positive tendencies of their age, as well as their own habits
of personal investigation, had created in them an _historical sense_
in regard to the past as well as to the present. Having acquired a
habit of appreciating the intrinsic tests of historical credibility
and probability, they found the particular narratives of the poets and
logographers, inadmissible as a whole even in the eyes of Hekatæus,
still more at variance with their stricter canons of criticism. And we
thus observe in them the constant struggle, as well as the resulting
compromise, between these two opposite tendencies; on one hand a firm
belief in the reality of the mythical world, on the other hand an
inability to accept the details which their only witnesses, the poets
and logographers, told them respecting it.

  [905] Marcellin. Vit. Thucyd. init.

  [906] Herodot. ii. 143.

Each of them however performed the process in his own way. Herodotus is
a man of deep and anxious religious feeling; he often recognizes the
special judgments of the gods as determining historical events: his
piety is also partly tinged with that mystical vein which the last two
centuries had gradually infused into the religion of the Greeks—for
he is apprehensive of giving offence to the gods by reciting publicly
what he has heard respecting them; he frequently stops short in his
narrative and intimates that there is a sacred legend, but that he will
not tell it: in other cases, where he feels compelled to speak out, he
entreats forgiveness for doing so from the gods and heroes. Sometimes
he will not even mention the name of a god, though he generally
thinks himself authorized to do so, the names being matter of public
notoriety.[907] Such pious reserve, which the open-hearted Herodotus
avowedly proclaims as chaining up his tongue, affords a striking
contrast with the plain-spoken and unsuspecting tone of the ancient
epic, as well as of the popular legends, wherein the gods and their
proceedings were the familiar and interesting subjects of common talk
as well as of common sympathy, without ceasing to inspire both fear and
reverence.

  [907] Herodot. ii. 3, 51, 61, 65, 170. He alludes briefly (c.
  51) to an ἱρὸς λόγος which was communicated in the Samothracian
  mysteries, but he does not mention what it was: also about the
  Thesmophoria, or τελετὴ of Dêmêtêr (c. 171).

  Καὶ περὶ μὲν τούτων τοσαῦτα ἡμῖν εἰποῦσι, καὶ παρὰ τῶν θεῶν καὶ
  ἡρώων εὐμένεια εἴη (c. 45).

  Compare similar scruples on the part of Pausanias (viii. 25 and
  37).

  The passage of Herodotus (ii. 3) is equivocal, and has been
  understood in more ways than one (see Lobeck, Aglaopham. p. 1287).

  The aversion of Dionysius of Halikarnassus to reveal the divine
  secrets is not less powerful (see A. R. i. 67, 68), and Pausanias
  _passim_.

Herodotus expressly distinguishes, in the comparison of Polykratês with
Minôs, the human race to which the former belonged, from the divine or
heroic race which comprised the latter.[908] But he has a firm belief
in the authentic personality and parentage of all the names in the
mythes, divine, heroic and human, as well as in the trustworthiness of
their chronology computed by generations. He counts back 1600 years
from his own day to that of Semelê, mother of Dionysus; 900 years to
Hêraklês, and 800 years to Penelopê, the Trojan war being a little
earlier in date.[909] Indeed even the longest of these periods must
have seemed to him comparatively short, seeing that he apparently
accepts the prodigious series of years which the Egyptians professed to
draw from a recorded chronology—17,000 years from their god Hêraklês,
and 15,000 years from their god Osiris or Dionysus, down to their king
Amasis[910] (550 B. C.) So much was his imagination familiarized with
these long chronological computations barren of events, that he treats
Homer and Hesiod as “men of yesterday,” though separated from his own
age by an interval which he reckons as four hundred years.[911]

  [908] Herod. iii. 122.

  [909] Herod. ii. 145.

  [910] Herodot. ii. 43-145. Καὶ ταῦτα Αἰγύπτιοι ἀτρεκέως φασὶ
  ἐπίστασθαι, ἀεί τε λογιζόμενοι καὶ ἀεὶ ἀπογραφόμενοι τὰ ἔτεα.

  [911] Herodot, ii. 53. μέχρι οὗ πρωήν τε καὶ χθὲς, ὡς εἰπεῖν
  λόγῳ. Ἡσίοδον γὰρ καὶ Ὅμηρον ἡλικίην τετρακοσίοισι ἔτεσι δοκέω
  μευ πρεσβυτέρους γενέσθαι, καὶ οὐ πλέοσι.

Herodotus had been profoundly impressed with what he saw and heard
in Egypt. The wonderful monuments, the evident antiquity, and the
peculiar civilization of that country, acquired such preponderance in
his mind over his own native legends, that he is disposed to trace
even the oldest religious names or institutions of Greece to Egyptian
or Phœnician original, setting aside in favor of this hypothesis the
Grecian legends of Dionysus and Pan.[912] The oldest Grecian mythical
genealogies are thus made ultimately to lose themselves in Egyptian
or Phœnician antiquity, and in the full extent of these genealogies
Herodotus firmly believes. It does not seem that any doubt had
ever crossed his mind as to the real personality of those who were
named or described in the popular mythes: all of them have once had
reality, either as men, as heroes, or as gods. The eponyms of cities,
dêmes and tribes, are all comprehended in this affirmative category;
the supposition of fictitious personages being apparently never
entertained. Deukaliôn, Hellên, Dôrus,[913]—Iôn, with his four sons,
the eponyms of the old Athenian tribes,[914]—the autochthonous Titakus
and Dekelus,[915]—Danaus, Lynkeus, Perseus, Amphitryôn, Alkmêna, and
Hêraklês,[916]—Talthybius, the heroic progenitor of the privileged
heraldic gens at Sparta,—the Tyndarids and Helena,[917]—Agamemnôn,
Menelaus, and Orestes,[918]—Nestôr and his son Peisistratus,—Asôpus,
Thêbê, and Ægina,—Inachus and Iô, Æêtês and Mêdea,[919]—Melanippus,
Adrastus, and Amphiaräus, as well as Jasôn and the Argô,[920]—all
these are occupants of the real past time, and predecessors of himself
and his contemporaries. In the veins of the Lacedæmonian kings flowed
the blood both of Kadmus and of Danaus, their splendid pedigree being
traceable to both of these great mythical names: Herodotus carries the
lineage up through Hêraklês first to Perseus and Danaê, then through
Danaê to Akrisios and the Egyptian Danaus; but he drops the paternal
lineage when he comes to Perseus (inasmuch as Perseus is the son of
Zeus by Danaê, without any reputed human father, such as Amphitryôn
was to Hêraklês), and then follow the higher members of the series
through Danaê alone.[921] He also pursues the same regal genealogy,
through the mother of Eurysthenês and Proclês, up to Polynikês, Œdipus,
Laius, Labdakus, Polydôrus and Kadmus; and he assigns various ancient
inscriptions which he saw in the temple of the Ismenian Apollo at
Thêbes, to the ages of Laius and Œdipus.[922] Moreover, the sieges of
Thêbes and Troy,—the Argonautic expedition,—the invasion of Attica by
the Amazons,—the protection of the Herakleids, and the defeat and death
of Eurystheus, by the Athenians,[923]—the death of Mêkisteus and Tydeus
before Thêbes by the hands of Melanippus, and the touching calamities
of Adrastus and Amphiaräus connected with the same enterprise,—the
sailing of Kastôr and Pollux in the Argô,[924]—the abductions of Iô,
Eurôpa, Mêdea and Helena,—the emigration of Kadmus in quest of Eurôpa,
and his coming to Bœôtia, as well as the attack of the Greeks upon Troy
to recover Helen,[925]—all these events seem to him portions of past
history, not less unquestionably certain, though more clouded over by
distance and misrepresentation, than the battles of Salamis and Mykalê.

  [912] Herodot. ii. 146.

  [913] Herod. i. 56.

  [914] Herod. v. 66.

  [915] Herod. ix. 73.

  [916] Herod. ii, 43-44, 91-98, 171-182 (the Egyptians admitted
  the truth of the Greek legend, that Perseus had come to Libya to
  fetch the Gorgon’s head).

  [917] Herod. ii. 113-120; iv. 145; vii. 134.

  [918] Herod. i. 67-68; ii. 113. vii. 159.

  [919] Herod. i. 1, 2, 4; v. 81, 65.

  [920] Herod. i. 52; iv. 145; v. 67; vii. 193.

  [921] Herod. vi. 52-53.

  [922] Herod. iv. 147; v. 59-61.

  [923] Herod. v. 61; ix. 27-28.

  [924] Herod. i. 52; iv. 145; v. 67.

  [925] Herod. i. 1-4; ii. 49, 113: iv. 147; v. 94.

But though Herodotus is thus easy of faith in regard both to the
persons and to the general facts of Grecian mythes, yet when he comes
to discuss particular facts taken separately, we find him applying
to them stricter tests of historical credibility, and often disposed
to reject as well the miraculous as the extravagant. Thus even with
respect to Hêraklês, he censures the levity of the Greeks in ascribing
to him absurd and incredible exploits; he tries their assertion by
the philosophical standard of nature, or of determinate powers and
conditions governing the course of events. “How is it consonant to
nature (he asks), that Hêraklês, being, as he was, according to the
statement of the Greeks, _a man_, should kill many thousand persons?
I pray that indulgence may be shown to me both by gods and heroes for
saying so much as this.” The religious feelings of Herodotus here
told him that he was trenching upon the utmost limits of admissible
scepticism.[926]

  [926] Herod. ii. 45. Λέγουσι δὲ πολλὰ καὶ ἄλλα ἀνεπισκέπτως
  οἱ Ἕλληνες· εὐήθης δὲ αὐτέων καὶ ὅδε ὁ μῦθός ἐστι, τὸν περὶ
  τοῦ Ἡρακλέος λέγουσι.... Ἔτι δὲ ἕνα ἐόντα τὸν Ἡρακλέα, καὶ ἔτι
  ἄνθρωπον ὡς δή φασι, ~κῶς φύσιν ἔχει~ πολλὰς μυριάδας φονεῦσαι;
  Καὶ περὶ μὲν τούτων τοσαῦτα ἡμῖν εἰποῦσι, καὶ παρὰ τῶν θεῶν καὶ
  παρὰ τῶν ἡρώων εὐμένεια εἴη.

  We may also notice the manner in which the historian criticizes
  the stratagem whereby Peisistratus established himself as despot
  at Athens—by dressing up the stately Athenian woman Phyê in the
  costume of the goddess Athênê, and passing off her injunctions
  as the commands of the goddess; the Athenians accepted her with
  unsuspecting faith, and received Peisistratus at her command.
  Herodotus treats the whole affair as a piece of extravagant
  silliness, πρᾶγμα εὐηθέστατον μακρῷ (i. 60).

Another striking instance of the disposition of Herodotus to
rationalize the miraculous narratives of the current mythes, is to be
found in his account of the oracle of Dôdôna and its alleged Egyptian
origin. Here, if in any case, a miracle was not only in full keeping,
but apparently indispensable to satisfy the exigences of the religious
sentiment; anything less than a miracle would have appeared tame and
unimpressive to the visitors of so revered a spot, much more to the
residents themselves. Accordingly, Herodotus heard, both from the three
priestesses and from the Dodonæans generally, that two black doves had
started at the same time from Thêbes in Egypt: one of them went to
Libya, where it directed the Libyans to establish the oracle of Zeus
Ammon; the other came to the grove of Dôdôna, and perched on one of the
venerable oaks, proclaiming with a human voice that an oracle of Zeus
must be founded on that very spot. The injunction of the speaking dove
was respectfully obeyed.[927]

  [927] Herod. ii. 55. Δωδωναίων δὲ αἱ ἱρηΐαι ... ἔλεγον ταῦτα,
  συνωμολόγεον δέ σφι καὶ οἱ ἄλλοι Δωδωναῖοι οἱ περὶ τὸ ἱρόν.

  The miracle sometimes takes another form; the oak at Dôdôna was
  itself once endued with speech (Dionys. Hal. Ars. Rhetoric. i. 6;
  Strabo).

Such was the tale related and believed at Dôdôna. But Herodotus had
also heard, from the priests at Thêbes in Egypt, a different tale,
ascribing the origin of all the prophetic establishments, in Greece as
well as in Libya, to two sacerdotal women, who had been carried away
from Thêbes by some Phœnician merchants and sold, the one in Greece,
the other in Libya. The Theban priests boldly assured Herodotus that
much pains had been taken to discover what had become of these women so
exported, and that the fact of their having been taken to Greece and
Libya had been accordingly verified.[928]

  [928] Herod. ii. 54.

The historian of Halicarnassus cannot for a moment think of admitting
the miracle which harmonized so well with the feelings of the
priestesses and the Dodonæans.[929] “How (he asks) could a dove
speak with human voice?” But the narrative of the priests at Thêbes,
though its prodigious improbability hardly requires to be stated, yet
involved no positive departure from the laws of nature and possibility,
and therefore Herodotus makes no difficulty in accepting it. The
curious circumstance is, that he turns the native Dodonæan legend
into a figurative representation, or rather a misrepresentation, of
the supposed true story told by the Theban priests. According to his
interpretation, the woman who came from Thêbes to Dôdôna was called
a dove, and affirmed to utter sounds like a bird, because she was
non-Hellenic and spoke a foreign tongue: when she learned to speak the
language of the country, it was then said that the dove spoke with a
human voice. And the dove was moreover called black, because of the
woman’s Egyptian color.

  [929] Herod. ii. 57. Ἐπεὶ τέῳ ἂν τρόπῳ πελειάς γε ἀνθρωπηΐῃ φωνῇ
  φθέγξαιτο;

  According to one statement, the word Πελειὰς in the Thessalian
  dialect meant both a dove and a prophetess (Scriptor. Rer.
  Mythicarum, ed. Bode, i. 96). Had there been any truth in this,
  Herodotus could hardly have failed to notice it, inasmuch as it
  would exactly have helped him out of the difficulty which he felt.

That Herodotus should thus bluntly reject a miracle, recounted to him
by the prophetic women themselves as the prime circumstance in the
_origines_ of this holy place, is a proof of the hold which habits
of dealing with historical evidence had acquired over his mind; and
the awkwardness of his explanatory mediation between the dove and the
woman, marks not less his anxiety, while discarding the legend, to let
it softly down into a story quasi-historical and not intrinsically
incredible.

We may observe another example of the unconscious tendency of
Herodotus to eliminate from the mythes the idea of special aid from
the gods, in his remarks upon Melampus. He designates Melampus “as a
clever man, who had acquired for himself the art of prophecy;” and
had procured through Kadmus much information about the religious
rites and customs of Egypt, many of which he introduced into
Greece[930]—especially the name, the sacrifices, and the phallic
processions of Dionysus: he adds, “that Melampus himself did not
accurately comprehend or bring out the whole doctrine, but wise men
who came after him made the necessary additions.”[931] Though the
name of Melampus is here maintained, the character described[932] is
something in the vein of Pythagoras—totally different from the great
seer and leech of the old epic mythes—the founder of the gifted family
of the Amythaonids, and the grandfather of Amphiaräus.[933] But that
which is most of all at variance with the genuine legendary spirit, is
the opinion expressed by Herodotus (and delivered with some emphasis
as _his own_), that Melampus “was a clever man, who had acquired for
himself prophetic powers.” Such a supposition would have appeared
inadmissible to Homer or Hesiod, or indeed to Solôn, in the preceding
century, in whose view even inferior arts come from the gods, while
Zeus or Apollo bestows the power of prophesying.[934] The intimation
of such an opinion by Herodotus, himself a thoroughly pious man, marks
the sensibly diminished omnipresence of the gods, and the increasing
tendency to look for the explanation of phenomena among more visible
and determinate agencies.

  [930] Herod. ii. 49. Ἐγὼ μὲν νύν φημι Μελάμποδα γενόμενον ἄνδρα
  σοφὸν, μαντικήν τε ἑωυτῷ συστῆσαι, καὶ πυθόμενον ἀπ᾽ Αἰγύπτου,
  ἄλλα τε πολλὰ ἐσηγήσασθαι Ἕλλησι, καὶ τὰ περὶ τὸν Διόνυσον, ὀλίγα
  αὐτῶν παραλλάξαντα.

  [931] Herod. ii. 49. Ἀτρεκέως μὲν οὐ πάντα συλλαβὼν τὸν λόγον
  ἔφῃνε· (Melampus) ἀλλ᾽ οἱ ἐπιγενόμενοι τούτῳ ~σοφισταὶ~ μεζόνως
  ἐξέφῃναν.

  [932] Compare Herod. iv. 95; ii. 81. Ἑλλήνων οὐ τῷ ἀσθενεστάτῳ
  ~σοφιστῇ~ Πυθαγόρᾳ.

  [933] Homer, Odyss. xi. 290; xv. 225. Apollodôr. i. 9, 11-12.
  Hesiod, Eoiai, Fragm. 55, ed. Düntzer (p. 43)—

    Ἀλκὴν μὲν γὰρ ἔδωκεν Ὀλύμπιος Αἰακίδησι,
    ~Νοῦν δ᾽ Ἀμυθαονίδαις~, πλοῦτον δ᾽ ἔπορ᾽ Ἀτρείδησι.

  also Frag. 34 (p. 38), and Frag. 65 (p. 45); Schol. Apoll. Rhod.
  i. 118.

  Herodotus notices the celebrated mythical narrative of Melampus
  healing the deranged Argive women (ix. 34); according to the
  original legend, the daughters of Prœtus. In the Hesiodic Eoiai
  (Fr. 16, Düntz.; Apollod. ii. 2) the distemper of the Prœtid
  females was ascribed to their having repudiated the rites and
  worship of Dionysus (Akusilaus, indeed, assigned a different
  cause), which shows that the old fable recognized a connection
  between Melampus and these rites.

  [934] Homer, Iliad, i. 72-87; xv. 412. Odyss. xv. 245-252; iv.
  233. Some times the gods inspired prophecy for the special
  occasion, without conferring upon the party the permanent gift
  and _status_ of a prophet (compare Odyss. i. 202; xvii. 383).
  Solôn, Fragm. xi. 48-53, Schneidewin:—

    Ἄλλον μάντιν ἔθηκεν ἄναξ ἑκάεργος Ἀπολλὼν,
      Ἔγνω δ᾽ ἀνδρὶ κακὸν τηλόθεν ἐρχόμενον,
    Ὧι συνομαρτήσωσι θεοὶ....

  Herodotus himself reproduces the old belief in the special gift
  of prophetic power by Zeus and Apollo, in the story of Euenius of
  Apollônia (ix. 94).

  See the fine ode of Pindar, describing the birth and inspiration
  of Jamus, eponymous father of the great prophetic family in Elis
  called the Jamids (Herodot. ix. 33), Pindar, Olymp. vi. 40-75.
  About Teiresias, Sophoc. Œd. Tyr. 283-410. Neither Nestôr nor
  Odysseus possesses the gift of prophecy.

We may make a similar remark on the dictum of the historian respecting
the narrow defile of Tempê, forming the embouchure of the Pêneus and
the efflux of all the waters from the Thessalian basin. The Thessalians
alleged that this whole basin of Thessaly had once been a lake,
but that Poseidôn had split the chain of mountains and opened the
efflux;[935] upon which primitive belief, thoroughly conformable to
the genius of Homer and Hesiod, Herodotus comments as follows: “The
Thessalian statement is reasonable. For whoever thinks that Poseidôn
shakes the earth, and that the rifts of an earthquake are the work of
that god, will, on seeing the defile in question, say that Poseidôn
has caused it. For the rift of the mountains is, as appeared to me
(when I saw it), the work of an earthquake.” Herodotus admits the
reference to Poseidôn, when pointed out to him, but it stands only in
the background: what is present to his mind is the phænomenon of the
earthquake, not as a special act, but as part of a system of habitual
operations.[936]

  [935] More than one tale is found elsewhere, similar to this,
  about the defile of Tempê:—

  “A tradition exists that this part of the country was once a
  lake, and that Solomon commanded two deeves, or genii, named
  Ard and Beel, to turn off the water into the Caspian, which
  they effected by cutting a passage through the mountains; and a
  city, erected in the newly-formed plain, was named after them
  Ard-u-beel.” (Sketches on the Shores of the Caspian, by W. R.
  Holmes.)

  Also about the plain of Santa Fe di Bogota, in South America,
  that it was once under water, until Bochica cleft the mountains
  and opened a channel of egress (Humboldt, Vues des Cordillères,
  p. 87-88); and about the plateau of Kashmir (Humboldt, Asie
  Centrale, vol. i. p. 102), drained in a like miraculous manner
  by the saint Kâsyapa. The manner in which conjectures, derived
  from local configuration or peculiarities, are often made to
  assume the form of _traditions_, is well remarked by the same
  illustrious traveller: “Ce qui se présente comme une tradition,
  n’est souvent que le reflet de l’impression que laisse l’aspect
  des lieux. Des bancs de coquilles à demi-fossiles, répandues
  dans les isthmes ou sur des plateaux, font naître même chez les
  hommes les moins avancés dans la culture intellectuelle, l’idée
  de grandes inondations, d’anciennes communications entre des
  bassins limitrophes. Des opinions, que l’on pourroit appeler
  systématiques, se trouvent dans les forêts de l’Orénoque comme
  dans les îles de la Mer du Sud. Dans l’une et dans l’autre de
  ces contrées, elles ont pris la forme des traditions.” (A. von
  Humboldt, Asie Centrale, vol. ii. p. 147.) Compare a similar
  remark in the same work and volume, p. 286-294.

  [936] Herodot. vii. 129. (Poseidôn was worshipped as Πετραῖος
  in Thessaly, in commemoration of this geological interference:
  Schol. Pindar. Pyth. iv. 245.) Τὸ δὲ παλαιὸν λέγεται, οὐκ ἐόντος
  κω τοῦ αὐλῶνος καὶ διεκρόου τούτου, τοὺς ποτάμους τούτους ...
  ῥέοντας ποιεῖν τὴν Θεσσαλίην πᾶσαν πέλαγος. Αὐτοὶ μέν νυν
  Θέσσαλοι λέγουσι Ποσειδέωνα ποιῆσαι τὸν αὐλῶνα, δι᾽ οὗ ῥέει ὁ
  Πηνειὸς, οἰκότα λέγοντες. Ὅστις γὰρ νομίζει Ποσειδέωνα τὴν γῆν
  σείειν, καὶ τὰ διεστεῶτα ὑπὸ σεισμοῦ τοῦ θεοῦ τούτου ἔργα εἶναι,
  καὶ ἂν ἐκεῖνο ἰδὼν φαίη Ποσειδέωνα ποιῆσαι. Ἐστὶ γὰρ σεισμοῦ
  ἔργον, ὡς ἐμοὶ ἐφαίνετο εἶναι, ἡ διάστασις τῶν οὐρέων. In another
  case (viii. 129), Herodotus believes that Poseidôn produced a
  preternaturally high tide, in order to punish the Persians, who
  had insulted his temple near Potidæa: here was a special motive
  for the god to exert his power.

  This remark of Herodotus illustrates the hostile ridicule cast
  by Aristophanês (in the Nubes) upon Socratês, on the score of
  alleged impiety, because he belonged to a school of philosophers
  (though in point of fact he discountenanced that line of
  study) who introduced physical laws and forces in place of the
  personal agency of the gods. The old man Strepsiades inquires
  from Socratês, _Who rains? Who thunders?_ To which Socratês
  replies, “_Not Zeus_, but the Nephelæ, _i. e. the clouds_: you
  never saw rain without clouds.” Strepsiades then proceeds to
  inquire—“But who is it that compels the clouds to move onward?
  is it not Zeus?” Socratês—“Not at all; it is æthereal rotation.”
  Strepsiades—“Rotation? that had escaped me: Zeus then no longer
  exists, and Rotation reigns in his place.”

  STREPS. Ὁ δ᾽ ἀναγκάζων ἐστὶ τίς αὐτὰς (Νεφέλας), οὐχ ὁ Ζεὺς,
            ὥστε φέρεσθαι;
  SOCRAT. Ἥκιστ᾽, ἀλλ᾽ αἰθέριος δῖνος.
  STREPS.                              Δῖνος; τουτί μ᾽ ἐλελήθει—
          Ὁ Ζεὺς οὐκ ὢν, ἀλλ᾽ ἀντ᾽ αὐτοῦ Δῖνος νυνὶ βασιλεύων.

  To the same effect v. 1454, Δῖνος βασιλεύει τὸν Δί᾽
  ἐξεληλακώς—“Rotation has driven out Zeus, and reigns in his
  place.”

  If Aristophanês had had as strong a wish to turn the public
  antipathies against Herodotus as against Socratês and Euripidês,
  the explanation here given would have afforded him a plausible
  show of truth for doing so; and it is highly probable that the
  Thessalians would have been sufficiently displeased with the view
  of Herodotus to sympathize in the poet’s attack upon him. The
  point would have been made (waiving metrical considerations)—

    ~Σεισμὸς~ βασιλεύει, τὸν ~Ποσειδῶν᾽~ ἐξεληλακώς.

  The comment of Herodotus upon the Thessalian view seems almost as
  if it were intended to guard against this very inference.

  Other accounts ascribed the cutting of the defile of Tempê to
  Hêraklês (Diodôr. iv. 18).

  Respecting the ancient Grecian faith, which recognized the
  displeasure of Poseidôn as the cause of earthquakes, see Xenoph.
  Hellen. iii. 3, 2; Thucydid. i. 127; Strabo, xii. p. 579; Diodôr.
  xv. 48-49. It ceased to give universal satisfaction even so early
  as the time of Thalês and Anaximenês (see Aristot. Meteorolog.
  ii. 7-8; Plutarch, Placit. Philos. iii. 15; Seneca, Natural.
  Quæst. vi. 6-23); and that philosopher, as well as Anaxagoras,
  Democritus and others, suggested different physical explanations
  of the fact. Notwithstanding a dissentient minority, however,
  the old doctrine still continued to be generally received: and
  Diodôrus, in describing the terrible earthquake in 373 B. C.,
  by which Helikê and Bura were destroyed, while he notices those
  philosophers (probably Kallisthenês, Senec. Nat. Quæst. vi. 23)
  who substituted physical causes and laws in place of the divine
  agency, rejects their views, and ranks himself with the religious
  public, who traced this formidable phænomenon to the wrath of
  Poseidôn (xv. 48-49).

  The Romans recognized many different gods as producers of
  earthquakes; an unfortunate creed, since it exposed them to the
  danger of addressing their prayers to the wrong god: “Unde in
  ritualibus et pontificiis observatur, obtemperantibus sacerdotiis
  caute, ne alio Deo pro alio nominato, cum quis eorum terram
  concutiat, piacula committantur.” (Ammian. Marcell. xvii. 7.)

Herodotus adopts the Egyptian version of the legend of Troy, founded on
that capital variation which seems to have originated with Stesichorus,
and according to which Helen never left Sparta at all—her _eidôlon_
had been taken to Troy in her place. Upon this basis a new story had
been framed, midway between Homer and Stesichorus, representing Paris
to have really carried off Helen from Sparta, but to have been driven
by storms to Egypt, where she remained during the whole siege of
Troy, having been detained by Prôteus, the king of the country, until
Menelaus came to reclaim her after his triumph. The Egyptian priests,
with their usual boldness of assertion, professed to have heard the
whole story from Menelaus himself—the Greeks had beseiged Troy, in
the full persuasion that Helen and the stolen treasures were within
the walls, nor would they ever believe the repeated denials of the
Trojans as to the fact of her presence. In intimating his preference
for the Egyptian narrative, Herodotus betrays at once his perfect and
unsuspecting confidence that he is dealing with genuine matter of
history, and his entire distrust of the epic poets, even including
Homer, upon whose authority that supposed history rested. His reason
for rejecting the Homeric version is that it teems with historical
improbabilities. If Helen had been really in Troy (he says), Priam and
the Trojans would never have been so insane as to retain her to their
own utter ruin: but it was the divine judgment which drove them into
the miserable alternative of neither being able to surrender Helen,
nor to satisfy the Greeks of the real fact that they had never had
possession of her—in order that mankind might plainly read, in the
utter destruction of Troy, the great punishments with which the gods
visit great misdeeds. Homer (Herodotus thinks) had heard this story,
but designedly departed from it, because it was not so suitable a
subject for epic poetry.[937]

  [937] Herod. ii. 116. δοκέει δέ μοι καὶ Ὅμηρος τὸν λόγον τοῦτον
  πυθέσθαι· ἀλλ᾽ οὐ γὰρ ὁμοίως εὐπρεπὴς ἐς τὴν ἐποποιΐην ἦν τῷ
  ἑτέρῳ τῷ περ ἐχρήσατο· ἑς ὃ μετῆκε αὐτὸν, δηλώσας ὡς καὶ τοῦτον
  ἐπισταῖτο τὸν λόγον.

  Herodotus then produces a passage from the Iliad, with a view to
  prove that Homer knew of the voyage of Paris and Helen to Egypt;
  but the passage proves nothing at all to the point.

  Again (c. 120), his slender confidence in the epic poets breaks
  out—εἰ χρή τι τοῖσι ἐποποιοῖσι χρεώμενον λέγειν.

  It is remarkable that Herodotus is disposed to identify Helen
  with the ξείνη Ἀφροδίτη whose temple he saw at Memphis (c. 112).

Enough has been said to show how wide is the difference between
Herodotus and the logographers with their literal transcript of the
ancient legends. Though he agrees with them in admitting the full
series of persons and generations, he tries the circumstances narrated
by a new standard. Scruples have arisen in his mind respecting
violations of the laws of nature: the poets are unworthy of trust, and
their narratives must be brought into conformity with historical and
ethical conditions, before they can be admitted as truth. To accomplish
this conformity, Herodotus is willing to mutilate the old legend in
one of its most vital points: he sacrifices the personal presence of
Helena in Troy, which ran through every one of the ancient epic poems
belonging to the Trojan cycle, and is indeed, under the gods, the great
and present moving force throughout.

Thucydidês places himself generally in the same point of view
as Herodotus with regard to mythical antiquity, yet with some
considerable differences. Though manifesting no belief in present
miracles or prodigies,[938] he seems to accept without reserve the
preexistent reality of all the persons mentioned in the mythes, and
of the long series of generations extending back through so many
supposed centuries: in this category, too, are included the eponymous
personages, Hellen, Kekrops, Eumolpus, Pandiôn, Amphilochus the son
of Amphiaräus, and Akarnan. But on the other hand, we find no trace
of that distinction between a human and an heroic ante-human race,
which Herodotus still admitted,—nor any respect for Egyptian legends.
Thucydidês, regarding the personages of the mythes as men of the same
breed and stature with his own contemporaries, not only tests the acts
imputed to them by the same limits of credibility, but presumes in them
the same political views and feelings as he was accustomed to trace in
the proceedings of Peisistratus or Periklês. He treats the Trojan war
as a great political enterprise, undertaken by all Greece; brought into
combination through the imposing power of Agamemnôn, not (according to
the legendary narrative) through the influence of the oath exacted by
Tyndareus. Then he explains how the predecessors of Agamemnôn arrived
at so vast a dominion—beginning with Pelops, who came over (as he
says) from Asia with great wealth among the poor Peloponnêsians, and
by means of this wealth so aggrandized himself, though a foreigner, as
to become the eponym of the peninsula. Next followed his son Atreus,
who acquired after the death of Eurystheus the dominion of Mykênæ,
which had before been possessed by the descendants of Perseus: here
the old legendary tale, which described Atreus as having been banished
by his father Pelops in consequence of the murder of his elder brother
Chrysippus, is invested with a political bearing, as explaining the
reason why Atreus retired to Mykênæ. Another legendary tale—the defeat
and death of Eurystheus by the fugitive Herakleids in Attica, so
celebrated in Attic tragedy as having given occasion to the generous
protecting intervention of Athens—is also introduced as furnishing the
cause why Atreus succeeded to the deceased Eurystheus: “for Atreus, the
maternal uncle of Eurystheus, had been entrusted by the latter with
his government during the expedition into Attica, and had effectually
courted the people, who were moreover in great fear of being attacked
by the Herakleids.” Thus the Pelopids acquired the supremacy in
Peloponnêsus, and Agamemnôn was enabled to get together his 1200 ships
and 100,000 men for the expedition against Troy. Considering that
contingents were furnished from every portion of Greece, Thucydidês
regards this as a small number, treating the Homeric catalogue as an
authentic muster-roll, perhaps rather exaggerated than otherwise. He
then proceeds to tell us why the armament was not larger: many more
men could have been furnished, but there was not sufficient money to
purchase provisions for their subsistence; hence they were compelled,
after landing and gaining a victory, to fortify their camp, to divide
their army, and to send away one portion for the purpose of cultivating
the Chersonese, and another portion to sack the adjacent towns. This
was the grand reason why the siege lasted so long as ten years. For
if it had been possible to keep the whole army together, and to act
with an undivided force, Troy would have been taken both earlier and at
smaller cost.[939]

  [938] “Ut conquirere fabulosa (says Tacitus, Hist. ii. 50, a
  worthy parallel of Thucydidês) et fictis oblectare legentium
  animos, procul gravitate cœpti operis crediderim, ita vulgatis
  traditisque demere fidem non ausim. Die, quo Bebriaci
  certabatur, avem inusitatâ specie, apud Regium Lepidum celebri
  vico consedisse, incolæ memorant; nec deinde cœtu hominum aut
  circumvolitantium alitum, territam pulsamque, donec Otho se ipse
  interficeret: tum ablatam ex oculis: et tempora reputantibus,
  initium finemque miraculi cum Othonis exitu competisse.”
  Suetonius (Vesp. 5) recounts a different miracle, in which three
  eagles appear.

  This passage of Tacitus occurs immediately after his magnificent
  description of the suicide of the emperor Otho, a deed which he
  contemplates with the most fervent admiration. His feelings were
  evidently so wrought up that he was content to relax the canons
  of historical credibility.

  [939] Thucyd. i. 9-12.

Such is the general sketch of the war of Troy, as given by Thucydidês.
So different is it from the genuine epical narrative, that we seem
hardly to be reading a description of the same event; still less
should we imagine that the event was known, to him as well as to us,
only through the epic poets themselves. The men, the numbers, and the
duration of the siege, do indeed remain the same; but the cast and
juncture of events, the determining forces, and the characteristic
features, are altogether heterogeneous. But, like Herodotus, and
still more than Herodotus, Thucydidês was under the pressure of two
conflicting impulses—he shared the general faith in the mythical
antiquity, but at the same time he could not believe in any facts which
contradicted the laws of historical credibility or probability. He was
thus under the necessity of torturing the matter of the old mythes
into conformity with the subjective exigencies of his own mind: he
left out, altered, recombined, and supplied new connecting principles
and supposed purposes, until the story became such as no one could
have any positive reason for calling in question: though it lost the
impressive mixture of religion, romance, and individual adventure,
which constituted its original charm, it acquired a smoothness and
plausibility, and a poetical _ensemble_, which the critics were
satisfied to accept as historical truth. And historical truth it would
doubtless have been, if any independent evidence could have been found
to sustain it. Had Thucydidês been able to produce such new testimony,
we should have been pleased to satisfy ourselves that the war of Troy,
as he recounted it, was the real event; of which the war of Troy, as
sung by the epic poets, was a misreported, exaggerated, and ornamented
recital. But in this case the poets are the only real witnesses, and
the narrative of Thucydidês is a mere extract and distillation from
their incredibilities.

A few other instances may be mentioned to illustrate the views of
Thucydidês respecting various mythical incidents. 1. He treats the
residence of the Homeric Phæakians at Corkyra as an undisputed fact,
and employs it partly to explain the efficiency of the Korkyrean
navy in times preceding the Peloponnesian war.[940] 2. He notices,
with equal confidence, the story of Têreus and Proknê, daughter of
Pandiôn, and the murder of the child Itys by Proknê his mother, and
Philomêla; and he produces this ancient mythe with especial reference
to the alliance between the Athenians and Têrês, king of the Odrysian
Thracians, during the time of the Peloponnesian war, intimating that
the Odrysian Têrês was neither of the same family nor of the same
country as Têreus the husband of Proknê.[941] The conduct of Pandiôn,
in giving his daughter Proknê in marriage to Têreus, is in his view
dictated by political motives and interests. 3. He mentions the
Strait of Messina as the place through which Odysseus is said to have
sailed.[942] 4. The Cyclôpes and the Læstrygones (he says) were the
most ancient reported inhabitants of Sicily; but he cannot tell to what
race they belonged, nor whence they came.[943] 5. Italy derived its
name from Italus, king of the Sikels. 6. Eryx and Egesto in Sicily were
founded by fugitive Trojans after the capture of Troy; also Skionê,
in the Thracian peninsula of Pallênê, by Greeks from the Achæan town
of Pellênê, stopping thither in their return from the siege of Troy:
the Amphilochian Argos in the Gulf of Ambrakia was in like manner
founded by Amphilochus son of Amphiaräus, in his return from the
same enterprise. The remorse and mental derangement of the matricidal
Alkmæôn, son of Amphiaräus, is also mentioned by Thucydidês,[944] as
well as the settlement of his son Akarnan in the country called after
him Akarnania.[945]

  [940] Thucyd. i. 25.

  [941] Thucyd. ii. 29. Καὶ τὸ ἔργον τὸ περὶ τὸν Ἴτυν αἱ γυναῖκες
  ἐν τῇ γῇ ταύτῃ ἔπραξαν· πολλοῖς δὲ καὶ τῶν ποιητῶν ἐν ἀηδόνος
  μνήμῃ Δαυλιὰς ἡ ὄρνις ἐπωνόμασται. Εἰκὸς τε καὶ τὸ κῆδος Πανδίονα
  ξυνάψασθαι τῆς θυγατρὸς διὰ τοσούτου, ἐπ᾽ ὠφελείᾳ τῇ πρὸς
  ἀλλήλους, μᾶλλον ἢ διὰ πολλῶν ἡμερῶν ἐς Ὀδρύσας ὁδοῦ. The first
  of these sentences would lead us to infer, if it came from any
  other pen than that of Thucydidês, that the writer believed the
  metamorphosis of Philomêla into a nightingale: see above, ch. xi.
  p. 270.

  The observation respecting the convenience of neighborhood for
  the marriage is remarkable, and shows how completely Thucydidês
  regarded the event as historical. What would he have said
  respecting the marriage of Oreithyia, daughter of Erechtheus,
  with Boreas, and the prodigious distance which she is reported
  to have been carried by her husband? Ὑπέρ τε πόντον πάντ᾽, ἐπ᾽
  ἔσχατα χθονὸς, etc. (Sophoklês ap. Strabo. vii. p. 295.)

  From the way in which Thucydidês introduces the mention of this
  event, we see that he intended to correct the misapprehension
  of his countrymen, who having just made an alliance with the
  Odrysian _Têrês_, were led by that circumstance to think of the
  old mythical _Têreus_, and to regard him as the ancestor of
  _Têrês_.

  [942] Thucyd. iv. 24.

  [943] Thucyd. vi. 2.

  [944] Thucyd. ii. 68-102; iv. 120; vi. 2. Antiochus of Syracuse,
  the contemporary of Thucydidês, also mentioned Italus as the
  eponymous king of Italy: he farther named Sikelus, who came to
  Morgos, son of Italus, after having been banished from Rome. He
  talks about Italus, just as Thucydidês talks about Thêseus, as
  a wise and powerful king, who first acquired a great dominion
  (Dionys. H. A. R. i. 12, 35, 73). Aristotle also mentioned Italus
  in the same general terms (Polit. vii. 9, 2).

  [945] We may here notice some particulars respecting Isokratês.
  He manifests entire confidence in the authenticity of the
  mythical genealogies and chronology; but while he treats the
  mythical personages as historically real, he regards them at the
  same time not as human, but as half-gods, superior to humanity.
  About Helena, Thêseus, Sarpêdôn, Cycnus, Memnôn, Achilles, etc.,
  see Encom. Helen. Or. x. pp. 282, 292, 295. Bek. Helena was
  worshipped in his time as a goddess at Therapnæ (_ib._ p. 295).
  He recites the settlements of Danaus, Kadmus, and Pelops in
  Greece, as undoubted historical facts (p. 297). In his discourse
  called _Busiris_, he accuses Polykratês, the sophist, of a gross
  anachronism, in having placed Busiris subsequent in point of date
  to Orpheus and Æolus (Or. xi. p. 301, Bek.), and he adds that the
  tale of Busiris having been slain by Hêraklês was chronologically
  impossible (p. 309). Of the long Athenian genealogy from Kekrops
  to Thêseus, he speaks with perfect historical confidence
  (Panathenaic. p. 349, Bek.); not less so of the adventures
  of Hêraklês and his mythical contemporaries, which he places
  in the mouth of Archidamus as a justification of the Spartan
  title to Messenia (Or. vi. _Archidamus_, p. 156, Bek.; compare
  Or. v. _Philippus_, pp. 114, 138), φάσιν, οἷς περὶ τῶν παλαιῶν
  πιστεύομεν, etc. He condemns the poets in strong language for
  the wicked and dissolute tales which they circulated respecting
  the gods: many of them (he says) had been punished for such
  blasphemies by blindness, poverty, exile, and other misfortunes
  (Or. xi. p. 309, Bek.).

  In general, it may be said that Isokratês applies no principles
  of historical criticism to the mythes; he rejects such as appear
  to him discreditable or unworthy, and believes the rest.

Such are the special allusions made by this illustrious author in
the course of his history to mythical events. From the tenor of his
language we may see that he accounted all that could be known about
them to be uncertain and unsatisfactory; but he has it much at heart
to show, that even the greatest were inferior in magnitude and
importance to the Peloponnesian war.[946] In this respect his opinion
seems to have been at variance with that which was popular among his
contemporaries.

  [946] Thucyd. i. 21-22.

  The first two volumes of this history have been noticed in an
  able article of the Quarterly Review, for October, 1846; as well
  as in the Heidelberger Jahrbücher der Literatur (1846. No. 41.
  pp. 641-655), by Professor Kortüm.

  While expressing, on several points, approbation of my work, by
  which I feel much flattered—both my English and my German critic
  take partial objection to the views respecting Grecian legend.
  While the Quarterly Reviewer contends that the mythopœic faculty
  of the human mind, though essentially loose and untrustworthy,
  is never creative, but requires some basis of fact to work
  upon—Kortüm thinks that I have not done justice to Thucydidês,
  as regards his way of dealing with legend; that I do not allow
  sufficient weight to the authority of an historian so circumspect
  and so cold-blooded (den kalt-blüthigsten und besonnensten
  Historiker des Alterthums, p. 653) as a satisfactory voucher for
  the early facts of Grecian history in his preface (Herr G. fehlt
  also, wenn er das anerkannt kritische Proœmium als Gewährsmann
  verschmäht, p. 654).

  No man feels more powerfully than I do the merits of Thucydidês
  as an historian, or the value of the example which he set in
  multiplying critical inquiries respecting matters recent and
  verifiable. But the ablest judge or advocate, in investigating
  specific facts, can proceed no further than he finds witnesses
  having the means of knowledge, and willing more or less to tell
  truth. In reference to facts prior to 776 B. C., Thucydidês had
  nothing before him except the legendary poets, whose credibility
  is not at all enhanced by the circumstance that he accepted them
  as witnesses, applying himself only to cut down and modify their
  allegations. His credibility in regard to the specific facts of
  these early times depends altogether upon theirs. Now we in our
  day are in a better position for appreciating their credibility
  than he was in his, since the foundations of historical evidence
  are so much more fully understood, and good or bad materials
  for history are open to comparison in such large extent and
  variety. Instead of wondering that he shared the general faith
  in such delusive guides—we ought rather to give him credit for
  the reserve with which he qualified that faith, and for the sound
  idea of historical possibility to which he held fast as the limit
  of his confidence. But it is impossible to consider Thucydidês
  as a _satisfactory guarantee_ (Gewährsmann) for matters of fact
  which he derives only from such sources.

  Professor Kortüm considers that I am inconsistent with myself
  in refusing to discriminate particular matters of historical
  fact among the legends—and yet in accepting these legends (in
  my chap. xx.) as giving a faithful mirror of the general state
  of early Grecian society (p. 653). It appears to me that this
  is no inconsistency, but a real and important distinction.
  Whether Hêraklês, Agamemnôn, Odysseus, etc. were real persons,
  and performed all, or a part, of the possible actions ascribed
  to them—I profess myself unable to determine. But even assuming
  both the persons and their exploits to be fictions, these very
  fictions will have been conceived and put together in conformity
  to the general social phænomena among which the describer and
  his hearers lived—and will thus serve as illustrations of the
  manners then prevalent. In fact, the real value of the Preface
  of Thucydidês, upon which Professor Kortüm bestows such just
  praise, consists, not in the particular facts which he brings out
  by altering the legends, but in the rational general views which
  he sets forth respecting early Grecian society, and respecting
  the steps as well as the causes whereby it attained its actual
  position as he saw it.

  Professor Kortüm also affirms that the mythes contain “real
  matter of fact along with mere conceptions:” which affirmation is
  the same as that of the Quarterly Reviewer, when he says that the
  mythopœic faculty is not creative. Taking the mythes in the mass,
  I doubt not that this is true, nor have I anywhere denied it.
  Taking them one by one, I neither affirm nor deny it. My position
  is, that, whether there be matter of fact or not, we have no
  test whereby it can be singled out, identified, and severed from
  the accompanying fiction. And it lies upon those, who proclaim
  the practicability of such severance, to exhibit some means of
  verification better than any which has been yet pointed out. If
  Thucydidês has failed in doing this, it is certain that none of
  the many authors who have made the same attempt after him have
  been more successful.

  It cannot surely be denied that the mythopœic faculty is
  _creative_, when we have before us so many divine legends, not
  merely in Greece, but in other countries also. To suppose that
  these religious legends are mere exaggerations, etc. of some
  basis of actual fact—that the gods of polytheism were merely
  divinized men, with qualities distorted or feigned—would be to
  embrace in substance the theory of Euêmerus.

To touch a little upon the later historians by whom these mythes were
handled, we find that Anaximenês of Lampsacus composed a consecutive
history of events, beginning from the Theogony down to the battle
of Mantineia.[947] But Ephorus professed to omit all the mythical
narratives which are referred to times anterior to the return of the
Herakleids, (such restriction would of course have banished the siege
of Troy,) and even reproved those who introduced mythes into historical
writing; adding, that everywhere truth was the object to be aimed
at.[948] Yet in practice he seems often to have departed from his own
rule.[949] Theopompus, on the other hand, openly proclaimed that he
could narrate fables in his history better than Herodotus, or Ktesias,
or Hellanikus.[950] The fragments which remain to us, exhibit some
proof that this promise was performed as to quantity;[951] though as
to his style of narration, the judgment of Dionysius is unfavorable.
Xenophôn ennobled his favorite amusement of the chase by numerous
examples chosen from the heroic world, tracing their portraits with all
the simplicity of an undiminished faith. Kallisthenês, like Ephorus,
professed to omit all mythes which referred to a time anterior to the
return of the Herakleids; yet we know that he devoted a separate book
or portion of his history to the Trojan war.[952] Philistus introduced
some mythes in the earlier portions of his Sicilian history; but Timæus
was distinguished above all others for the copious and indiscriminate
way in which he collected and repeated such legends.[953] Some of
these writers employed their ingenuity in transforming the mythical
circumstances into plausible matter of history: Ephorus, in particular,
converted the serpent Pythô, slain by Apollo, into a tyrannical
king.[954]

  [947] Diodôr. xv. 89. He was a contemporary of Alexander the
  Great.

  [948] Diodôr. iv. 1. Strabo, ix. p. 422, ἐπιτιμήσας τοῖς
  φιλομυθοῦσιν ἐν τῇ τῆς ἱστωρίας γραφῇ.

  [949] Ephorus recounted the principal adventures of Hêraklês
  (Fragm. 8, 9, ed. Marx.), the tales of Kadmus and Harmonia
  (Fragm. 12), the banishment of Ætôlus from Elis (Fragm. 15;
  Strabo, viii. p. 357); he drew inferences from the chronology of
  the Trojan and Theban wars (Fragm. 28); he related the coming
  of Dædalus to the Sikan king Kokalus, and the expedition of the
  Amazons (Fragm. 99-103).

  He was particularly copious in his information about κτίσεις,
  ἀποικίαι and συγγενείαι (Polyb. ix. 1).

  [950] Strabo, i. p. 74.

  [951] Dionys. Halic. De Vett. Scriptt. Judic. p. 428, Reisk;
  Ælian, V. H. iii. 18, Θεόπομπος ... δεινὸς μυθόλογος.

  Theopompus affirmed, that the bodies of those who went into the
  forbidden precinct (τὸ ἄβατον) of Zeus, in Arcadia, gave no
  shadow (Polyb. xvi. 12). He recounted the story of Midas and
  Silênus (Fragm. 74, 75, 76, ed. Wichers); he said a good deal
  about the heroes of Troy; and he seems to have assigned the
  misfortunes of the Νόστοι to an historical cause—the rottenness
  of the Grecian ships, from the length of the siege, while the
  genuine epic ascribes it to the anger of Athênê (Fragm. 112,
  113, 114; Schol. Homer. Iliad, ii. 135); he narrated an alleged
  expulsion of Kinyras from Cyprus by Agamemnôn (Fragm. 111);
  he gave the genealogy of the Macedonian queen Olympias up to
  Achilles and Æakus (Fragm. 232).

  [952] Cicero, Epist. ad Familiar. v. 12; Xenophôn de Venation. c.
  1.

  [953] Philistus, Fragm. 1 (Göller), Dædalus, and Kokalus; about
  Liber and Juno (Fragm. 57); about the migration of the Sikels
  into Sicily, eighty years after the Trojan war (ap. Dionys. Hal.
  i. 3).

  Timæus (Fragm. 50, 51, 52, 53, Göller) related many fables
  respecting Jasôn, Mêdea, and the Argonauts generally. The
  miscarriage of the Athenian armament under Nikias, before
  Syracuse, is imputed to the anger of Hêraklês against the
  Athenians because they came to assist the Egestans, descendants
  of Troy (Plutarch, Nikias, 1),—a naked reproduction of genuine
  epical agencies by an historian; also about Diomêdês and the
  Daunians; Phaëthôn and the river Eridanus; the combats of the
  Gigantes in the Phlegræan plains (Fragm. 97, 99, 102).

  [954] Strabo, ix. p. 422.

But the author who pushed this transmutation of legend into history
to the greatest length, was the Messenian Euêmerus, contemporary of
Kassander of Macedôn. He melted down in this way the divine persons
and legends, as well as the heroic—representing both gods and heroes
as having been mere earth-born men, though superior to the ordinary
level in respect of force and capacity, and deified or heroified
after death as a recompense for services or striking exploits. In
the course of a voyage into the Indian sea, undertaken by command of
Kassander, Euêmerus professed to have discovered a fabulous country
called Panchaia, in which was a temple of the Triphylian Zeus: he there
described a golden column, with an inscription purporting to have been
put up by Zeus himself, and detailing his exploits while on earth.[955]
Some eminent men, among whom may be numbered Polybius, followed the
views of Euêmerus, and the Roman poet Ennius[956] translated his
Historia Sacra; but on the whole he never acquired favor, and the
unblushing inventions which he put into circulation were of themselves
sufficient to disgrace both the author and his opinions. The doctrine
that all the gods had once existed as mere men offended the religious
pagans, and drew upon Euêmerus the imputation of atheism; but, on the
other hand, it came to be warmly espoused by several of the Christian
assailants of paganism,—by Minucius Felix, Lactantius, and St.
Augustin, who found the ground ready prepared for them in their efforts
to strip Zeus and the other pagan gods of the attributes of deity. They
believed not only in the main theory, but also in the copious details
of Euêmerus; and the same man whom Strabo casts aside as almost a
proverb for mendacity, was extolled by them as an excellent specimen
of careful historical inquiry.[957]

  [955] Compare Diodôr. v. 44-46; and Lactantius, De Falsâ Relig.
  i. 11.

  [956] Cicero, De Naturâ Deor. i. 42; Varro, De Re Rust. i. 48.

  [957] Strabo, ii. p. 102. Οὐ πολὺ οὖν λείπεται ταῦτα τῶν Πύθεω
  καὶ Εὐημέρου καὶ Ἀντιφάνους ψευσμάτων; compare also i. p. 47, and
  ii. p. 104.

  St. Augustin, on the contrary, tells us (Civitat. Dei, vi. 7),
  “Quid de ipse Jove senserunt, qui nutricem ejus in Capitolio
  posuerunt? Nonne attestati sunt omnes Euemero, qui non fabulosâ
  garrulitate, sed _historicâ diligentiâ_, homines fuisse
  mortalesque conscripsit?” And Minucius Felix (Octav. 20-21),
  “Euemerus exequitur Deorum natales: patrias, sepulcra dinumerat,
  et per provincias monstrat, Dictæi Jovis, et Apollinis Delphici,
  et Phariæ Isidis, et Cereris Eleusiniæ.” Compare Augustin, Civit.
  Dei, xviii. 8-14; and Clemens Alexand. Cohort. ad Gent. pp.
  15-18, Sylb.

  Lactantius (De Falsâ Relig. c. 13, 14, 16) gives copious
  citations from Ennius’s translation of the Historia Sacra of
  Euêmerus.

  Εὐήμερος, ὁ ἐπικληθεὶς ἄθεος, Sextus Empiricus, adv. Physicos,
  ix. § 17-51. Compare Cicero, De Nat. Deor. i. 42; Plutarch, De
  Iside et Osiride, c. 23. tom. ii. p. 475, ed. Wytt.

  Nitzsch assumes (Helden Sage der Griechen, sect. 7. p. 84) that
  the voyage of Euêmerus to Panchaia was intended only as an
  amusing romance, and that Strabo, Polybius, Eratosthenês and
  Plutarch were mistaken in construing it as a serious recital.
  Böttiger, in his Kunst-Mythologie der Griechen (Absch. ii. s. 6.
  p. 190), takes the same view. But not the least reason is given
  for adopting this opinion, and it seems to me far-fetched and
  improbable; Lobeck (Aglaopham. p. 989), though Nitzsch alludes to
  him as holding it, manifests no such tendency, as far as I can
  observe.

But though the pagan world repudiated that “lowering tone of
explanation,” which effaced the superhuman personality of Zeus and the
great gods of Olympus, the mythical persons and narratives generally
came to be surveyed more and more from the point of view of history,
and subjected to such alterations as might make them look more like
plausible matter of fact. Polybius, Strabo, Diodôrus, and Pausanias,
cast the mythes into historical statements—with more or less of
transformation, as the case may require, assuming always that there
is a basis of truth, which may be discovered by removing poetical
exaggerations and allowing for mistakes. Strabo, in particular, lays
down that principle broadly and unequivocally in his remarks upon
Homer. To give pure fiction, without any foundation of fact, was in his
judgment utterly unworthy of so great a genius; and he comments with
considerable acrimony on the geographer Eratosthenês, who maintains the
opposite opinion. Again, Polybius tells us that the Homeric Æolus, the
dispenser of the winds by appointment from Zeus, was in reality a man
eminently skilled in navigation, and exact in predicting the weather;
that the Cyclôpes and Læstrygones were wild and savage real men in
Sicily; and that Scylla and Charybdis were a figurative representation
of dangers arising from pirates in the Strait of Messina. Strabo speaks
of the amazing expeditions of Dionysus and Hêraklês, and of the long
wanderings of Jasôn, Menelaus, and Odysseus, in the same category
with the extended commercial range of the Phœnician merchant-ships:
he explains the report of Thêseus and Peirithöus having descended to
Hadês, by their dangerous earthly pilgrimages,—and the invocation
of the Dioskuri as the protectors of the imperiled mariner, by the
celebrity which they had acquired as real men and navigators.

Diodôrus gave at considerable length versions of the current fables
respecting the most illustrious names in the Grecian mythical world,
compiled confusedly out of distinct and incongruous authors. Sometimes
the mythe is reproduced in its primitive simplicity, but for the most
part it is partially, and sometimes wholly, historicized. Amidst this
jumble of dissentient authorities we can trace little of a systematic
view, except the general conviction that there was at the bottom of
the mythes a real chronological sequence of persons, and real matter
of fact, historical or ultra-historical. Nevertheless, there are
some few occasions on which Diodôrus brings us back a step nearer
to the point of view of the old logographers. For, in reference to
Hêraklês, he protests against the scheme of cutting down the mythes
to the level of present reality, and contends that a special standard
of ultra-historical credibility ought to be constituted, so as to
include the mythe in its native dimensions, and do fitting honor to
the grand, beneficent, and superhuman personality of Hêraklês and
other heroes or demi-gods. To apply to such persons the common measure
of humanity (he says), and to cavil at the glorious picture which
grateful man has drawn of them, is at once ungracious and irrational.
All nice criticism into the truth of the legendary narratives is
out of place: we show our reverence to the god by acquiescing in
the incredibilities of his history, and we must be content with the
best guesses which we can make, amidst the inextricable confusion
and numberless discrepancies which they present.[958] Yet though
Diodôrus here exhibits a preponderance of the religious sentiment over
the purely historical point of view, and thus reminds us of a period
earlier than Thucydidês—he in another place inserts a series of stories
which seem to be derived from Euêmerus, and in which Uranus, Kronus,
and Zeus appear reduced to the character of human kings celebrated
for their exploits and benefactions.[959] Many of the authors, whom
Diodôrus copies, have so entangled together Grecian, Asiatic, Egyptian,
and Libyan fables, that it becomes impossible to ascertain how much of
this heterogeneous mass can be considered as at all connected with the
genuine Hellenic mind.

  [958] Diodôr. iv. 1-8. Ἔνιοι γὰρ τῶν ἀναγινωσκόντων, οὐ
  δικαίᾳ χρώμενοι κρίσει, τἀκριβὲς ἐπιζητοῦσιν ἐν ταῖς ἀρχαίαις
  μυθολογίαις, ἐπίσης τοῖς πραττομένοις ἐν τῷ καθ᾽ ἡμᾶς χρόνῳ, καὶ
  τὰ δισταζόμενα τῶν ἔργων διὰ τὸ μέγεθος, ἐκ τοῦ καθ᾽ αὑτοὺς βίου
  τεκμαιρόμενοι, τὴν Ἡρακλέους δύναμιν ἐκ τῆς ἀσθενείας τῶν νῦν
  ἀνθρώπων θεωροῦσιν, ὥστε διὰ τὴν ὑπερβολὴν τοῦ μεγέθους τῶν ἔργων
  ἀπιστεῖσθαι τὴν γραφήν. Καθόλου γὰρ ἐν ταῖς ἀρχαίαις μυθολογίαις
  οὐκ ἐκ παντὸς τρόπου ~πικρῶς τὴν ἀλήθειαν ἐξεταστέον~. Καὶ γὰρ ἐν
  τοῖς θεάτροις ~πεπεισμένοι μήτε Κενταύρους~ διφυεῖς ἐξ ἑτερογενῶν
  σωμάτων ὑπάρξαι, μήτε Γηρυόνην τρισώματον, ~ὅμως προσδεχόμεθα τὰς
  τοιαύτας μυθολογίας~, καὶ ~ταῖς ἐπισημασίαις συναύξομεν τὴν τοῦ
  θεοῦ τιμήν~. Καὶ γὰρ ἄτοπον, Ἡρακλέα μὲν ἔτι κατ᾽ ἀνθρώπους ὄντα
  τοῖς ἰδίοις πόνοις ἐξημερῶσαι τὴν οἰκουμένην, τοὺς δ᾽ ἀνθρώπους,
  ἐπιλαθομένους τῆς κοινῆς εὐεργεσίας, ~συκοφαντεῖν~ τὸν ἐπὶ τοῖς
  καλλίστοις ἔργοις ἔπαινον, etc.

  This is a remarkable passage: first, inasmuch as it sets forth
  the total inapplicability of analogies drawn from the historical
  past as narratives about Hêraklês; next, inasmuch as it suspends
  the employment of critical and scientific tests, and invokes an
  acquiescence interwoven and identified with the feelings, as the
  proper mode of evincing pious reverence for the god Hêraklês.
  It aims at reproducing exactly that state of mind to which the
  mythes were addressed, and with which alone they could ever be in
  thorough harmony.

  [959] Diodôr. iii. 45-60; v. 44-46.

Pausanias is far more strictly Hellenic in his view of the Grecian
mythes than Diodôrus: his sincere piety makes him inclined to faith
generally with regard to the mythical narratives, but subject
nevertheless to the frequent necessity of historicizing or allegorizing
them. His belief in the general reality of the mythical history and
chronology is complete, in spite of the many discrepancies which he
finds in it, and which he is unable to reconcile.

Another author who seems to have conceived clearly, and applied
consistently, the semi-historical theory of the Grecian mythes, is
Palæphatus, of whose work what appears to be a short abstract has been
preserved.[960] In the short preface of this treatise “concerning
Incredible Tales,” he remarks, that some men, from want of instruction,
believe all the current narratives; while others, more searching and
cautious, disbelieve them altogether. Each of these extremes he is
anxious to avoid. On the one hand, he thinks that no narrative could
ever have acquired credence unless it had been founded in truth;
on the other, it is impossible for him to accept so much of the
existing narratives as conflicts with the analogies of present natural
phænomena. If such things ever had been, they would still continue to
be—but they never have so occurred; and the extra-analogical features
of the stories are to be ascribed to the license of the poets.
Palæphatus wishes to adopt a middle course, neither accepting all nor
rejecting all: accordingly, he had taken great pains to separate the
true from the false in many of the narratives; he had visited the
localities wherein they had taken place, and made careful inquiries
from old men and others.[961] The results of his researches are
presented in a new version of fifty legends, among the most celebrated
and the most fabulous, comprising the Centaurs, Pasiphaê, Aktæôn,
Kadmus and the Sparti, the Sphinx, Cycnus, Dædalus, the Trojan horse,
Æolus, Scylla, Geryôn, Bellerophôn, etc.

  [960] The work of Palæphatus, probably this original, is alluded
  to in the _Ciris_ of Virgil (88):—

    “Docta Palæphatiâ testatur voce papyrus.”

  The date of Palæphatus is unknown—indeed this passage of the
  _Ciris_ seems the only ground that there is for inference
  respecting it. That which we now possess is probably an extract
  from a larger work—made by another person at some later time: see
  Vossius de Historicis Græcis, p. 478, ed. Westermann.

  [961] Palæphat. init. ap. Script. Mythogr. ed. Westermann, p.
  268. Τῶν ἀνθρώπων οἱ μὲν πείθονται πᾶσι τοῖς λεγομένοις, ὡς
  ἀνομίλητοι σοφίας καὶ ἐπιστήμης—οἱ δὲ πυκνότεροι τὴν φύσιν καὶ
  πολυπράγμονες ἀπιστοῦσι τὸ παράπαν μηδὲν γενέσθαι τούτων. Ἐμοὶ δὲ
  δοκεῖ γενέσθαι πάντα τὰ λεγόμενα· ... γενόμενα δέ τινα οἱ ποιηταὶ
  καὶ λογογράφοι παρέτρεψαν εἰς τὸ ἀπιστότερον καὶ θαυμασιώτερον
  τοῦ θαυμάζειν ἕνεκα τοὺς ἀνθρώπους. Ἐγὼ δὲ γινώσκω, ὅτι οὐ
  δύναται τὰ τοιαῦτα εἶναι οἷα καὶ λέγεται· τοῦτο δὲ καὶ διείληφα,
  ὅτι εἰ μὴ ἐγένετο, οὐκ ἄν ἐλέγετο.

  The main assumption of the semi-historical theory is here shortly
  and clearly stated.

  One of the early Christian writers, Minucius Felix, is astonished
  at the easy belief of his pagan forefathers in miracles. If ever
  such things had been done in former times (he affirms), they
  would continue to be done now; as they cannot be done now, we may
  be sure that they never were _really_ done formerly (Minucius
  Felix, Octav. c. 20): “Majoribus enim nostris tam facilis in
  mendaciis fides fuit, ut temerè crediderint etiam alia monstruosa
  mira miracula, Scyllam multiplicem, Chimæram multiformem, Hydram,
  et Centauros. Quid illas aniles fabulas—de hominibus aves, et
  feras homines, et de hominibus arbores atque flores? _Quæ, si
  essent facta, fierent; quia fieri non possunt, ideo nec facta
  sunt._”

It must be confessed that Palæphatus has performed his promise of
transforming the “incredibilia” into narratives in themselves plausible
and unobjectionable, and that in doing so he always follows some
thread of analogy, real or verbal. The Centaurs (he tells us) were a
body of young men from the village of Nephelê in Thessaly, who first
trained and mounted horses for the purpose of repelling a herd of bulls
belonging to Ixiôn king of the Lapithæ, which had run wild and done
great damage: they pursued these wild bulls on horseback, and pierced
them with their spears, thus acquiring both the name of _Prickers_
(κέντορες) and the imputed attribute of joint body with the horse.
Aktæôn was an Arcadian, who neglected the cultivation of his land for
the pleasures of hunting, and was thus eaten up by the expense of
his hounds. The dragon whom Kadmus killed at Thêbes, was in reality
Drako, king of Thêbes; and the dragon’s teeth which he was said to
have sown, and from whence sprung a crop of armed men, were in point
of fact elephants’ teeth, which Kadmus as a rich Phœnician had brought
over with him: the sons of Drako sold these elephants’ teeth and
employed the proceeds to levy troops against Kadmus. Dædalus, instead
of flying across the sea on wings, had escaped from Krête in a swift
sailing-boat under a violent storm: Kottus, Briareus, and Gygês were
not persons with one hundred hands, but inhabitants of the village of
Hekatoncheiria in Upper Macedonia, who warred with the inhabitants of
Mount Olympus against the Titans: Scylla, whom Odysseus so narrowly
escaped, was a fastsailing piratical vessel, as was also Pegasus, the
alleged winged horse of Bellerophôn.[962]

  [962] Palæphat. Narrat. 1, 3, 6, 13, 20, 21, 29. Two short
  treatises on the same subject as this of Palæphatus, are printed
  along with it, both in the collection of Gale and of Westermann;
  the one, _Heracliti de Incredibilibus_, the other _Anonymi de
  Incredibilibus_. They both profess to interpret some of the
  extraordinary or miraculous mythes, and proceed in a track not
  unlike that of Palæphatus. Scylla was a beautiful courtezan,
  surrounded with abominable parasites: she ensnared and ruined
  the companions of Odysseus, though he himself was prudent enough
  to escape her (Heraclit. c. 2. p. 313, West.) Atlas was a great
  astronomer: Pasiphaê fell in love with a youth named Taurus; the
  monster called the Chimæra was in reality a ferocious queen, who
  had two brothers called Leo and Drako; the ram which carried
  Phryxus and Hellê across the Ægean was a boatman named Krias
  (Heraclit. c. 2, 6, 15, 24).

  A great number of similar explanations are scattered throughout
  the Scholia on Homer and the Commentary of Eustathius, without
  specification of their authors.

  Theôn considers such resolution of fable into plausible history
  as a proof of surpassing ingenuity (Progymnasmata, cap. 6, ap.
  Walz. Coll. Rhett. Græc. i. p. 219). Others among the Rhetors,
  too, exercised their talents sometimes in vindicating, sometimes
  in controverting, the probability of the ancient mythes. See the
  Progymnasmata of Nicolaus—Κατασκευὴ ὅτι εἰκότα τὰ κατὰ Νιόβην,
  Ἀνασκευὴ ὅτι οὐκ εἰκότα τὰ κατὰ Νιόβην (ap. Walz. Coll. Rhetor.
  i. p. 284-318), where there are many specimens of this fanciful
  mode of handling.

  Plutarch, however, in one of his treatises, accepts Minotaurs,
  Sphinxes, Centaurs, etc. as realities; he treats them as products
  of the monstrous, incestuous, and ungovernable lusts of man,
  which he contrasts with the simple and moderate passions of
  animals (Plutarch, Gryllus, p. 990).

By such ingenious conjectures, Palæphatus eliminates all the incredible
circumstances, and leaves to us a string of tales perfectly credible
and commonplace, which we should readily believe, provided a very
moderate amount of testimony could be produced in their favor. If his
treatment not only disenchants the original mythes, but even effaces
their generic and essential character, we ought to remember that this
is not more than what is done by Thucydidês in his sketch of the
Trojan war. Palæphatus handles the mythes consistently, according to
the semi-historical theory, and his results exhibit the maximum which
that theory can ever present. By aid of conjecture, we get out of the
impossible, and arrive at matters intrinsically plausible, but totally
uncertified; beyond this point we cannot penetrate, without the light
of extrinsic evidence, since there is no intrinsic mark to distinguish
truth from plausible fiction.[963]

  [963] The learned Mr. Jacob Bryant regards the explanations of
  Palæphatus as if they were founded upon real fact. He admits,
  for example, the city Nephelê alleged by that author in his
  exposition of the fable of the Centaurs. Moreover, he speaks with
  much commendation of Palæphatus generally: “He (Palæphatus) wrote
  early, and seems to have been a serious and sensible person; one
  who saw the absurdity of the fables upon which the theology of
  his country was founded.” (Ancient Mythology, vol. i. p. 411-435.)

  So also Sir Thomas Brown (Enquiry into Vulgar Errors, Book I.
  chap. vi. p. 221, ed. 1835) alludes to Palæphatus as having
  incontestably pointed out the real basis of the fables. “And
  surely the fabulous inclination of those days was greater than
  any since; which swarmed so with fables, and from such slender
  grounds took hints for fictions, poisoning the world ever
  after: wherein how far they succeeded, may be exemplified from
  Palæphatus, in his Book of Fabulous Narrations.”

It remains that we should notice the manner in which the ancient
mythes were received and dealt with by the philosophers. The earliest
expression which we hear, on the part of philosophy, is the severe
censure bestowed upon them on ethical grounds by Xenophanês of
Kolophôn, and seemingly by some others of his contemporaries.[964]
It was apparently in reply to such charges, which did not admit of
being directly rebutted, that Theagenês of Rhêgium (about 520 B. C.)
first started the idea of a double meaning in the Homeric and Hesiodic
narratives,—an interior sense, different from that which the words
in their obvious meaning bore, yet to a certain extent analogous,
and discoverable by sagacious divination. Upon this principle, he
allegorized especially the battle of the gods in the Iliad.[965] In
the succeeding century, Anaxagoras and Metrodôrus carried out the
allegorical explanation more comprehensively and systematically;
the former representing the mythical personages as mere mental
conceptions, invested with name and gender, and illustrative of ethical
precepts,—the latter connecting them with physical principles and
phænomena. Metrodôrus resolved not only the persons of Zeus, Hêrê,
and Athênê, but also those of Agamemnôn, Achilles, and Hectôr, into
various elemental combinations and physical agencies, and treated
the adventures ascribed to them as natural facts concealed under the
veil of allegory.[966] Empedoklês, Prodikus, Antisthenês, Parmenidês,
Hêrakleidês of Pontus, and in a later age, Chrysippus, and the Stoic
philosophers generally,[967] followed more or less the same principle
of treating the popular gods as allegorical personages; while the
expositors of Homer (such as Stesimbrotus, Glaukôn, and others, even
down to the Alexandrine age), though none of them proceeded to the same
extreme length as Metrodôrus, employed allegory amongst other media
of explanation for the purpose of solving difficulties, or eluding
reproaches against the poet.

  [964] Xenophan. ap. Sext. Empir. adv. Mathemat. ix. 193. He also
  disapproved of the rites, accompanied by mourning and wailing,
  with which the Eleatês worshipped Leukothea: he told them, εἰ
  μὲν θεὸν ὑπολαμβάνουσι, μὴ θρηνεῖν· εἰ δὲ ἄνθρωπον, μὴ θύειν
  (Aristotel. Rhet. ii. 23).

  Xenophanês pronounced the battles of the Titans, Gigantes, and
  Centaurs to be “fictions of our predecessors,” πλάσματα τῶν
  προτέρων (Xenophan. Fragm. 1. p. 42, ed. Schneidewin).

  See a curious comparison of the Grecian and Roman theology in
  Dionys. Halicarn. Ant. Rom. ii. 20.

  [965] Schol. Iliad. xx. 67: Tatian. adv. Græc. c. 48. Hêrakleitus
  indignantly repelled the impudent atheists who found fault
  with the divine mythes of the Iliad, ignorant of their true
  allegorical meaning: ἡ τῶν ἐπιφυομένων τῷ Ὁμήρῳ τόλμα τοὺς Ἥρας
  δεσμοὺς αἰτιᾶται, καὶ νομίζουσιν ὕλην τινα δαψιλῆ τῆς ἀθέου πρὸς
  Ὅμηρον ἔχειν μανίας ταῦτα—Ἦ οὐ μέμνῃ ὅτι τ᾽ ἐκρέμω ὑψόθεν, etc.
  λέληθε δ᾽ αὐτοὺς ὅτι τούτοις τοῖς ἔπεσιν ἐκτεθεολόγηται ἡ τοῦ
  παντὸς γένεσις, καὶ τὰ συνεχῶς ᾀδόμενα τέσσαρα στοιχεῖα τούτων
  τῶν στίχων ἐστὶ τάξις (Schol. ad Hom. Iliad. xv. 18).

  [966] Diogen. Laërt. ii. 11; Tatian. adv. Græc. c. 37; Hesychius,
  v. Ἀγαμέμνονα. See the ethical turn given to the stories of
  Circê, the Sirens, and Scylla, in Xenoph. Memorab. i. 3, 7;
  ii. 6, 11-31. Syncellus, Chronic. p. 149. Ἑρμηνεύουσι δὲ οἱ
  Ἀναξαγόρειοι τοὺς μυθώδεις θεοὺς, νοῦν μὲν τὸν Δία, τὴν δὲ Ἀθηνᾶν
  τέχνην, etc.

  Uschold and other modern German authors seem to have adopted
  in its full extent the principle of interpretation proposed by
  Metrodorus—treating Odysseus and Penelopê as personifications of
  the Sun and Moon, etc. See Helbig, Die Sittlichen Zustände des
  Griechischen Helden Alters, Einleitung, p. xxix. (Leipzig, 1839.)

  Corrections of the Homeric text were also resorted to, in order
  to escape the necessity of imputing falsehood to Zeus (Aristotel.
  De Sophist. Elench. c. 4).

  [967] Sextus Empiric. ix. 18; Diogen. viii. 76; Plutarch, De
  Placit. Philosoph. i. 3-6; De Poesi Homericâ, 92-126; De Stoicor.
  Repugn. p. 1050, Menander, De Encomiis, c. 5.

  Cicero, De Nat. Deor. i. 14, 15, 16, 41; ii. 24-25. “Physica
  ratio non inelegans inclusa in impias fabulas.”

  In the _Bacchæ_ of Euripidês, Pentheus is made to deride the tale
  of the motherless infant Dionysus having been sewn into the thigh
  of Zeus. Teiresias, while reproving him for his impiety, explains
  the story away in a sort of allegory: the μηρὸς Διὸς (he says)
  was a mistaken statement in place of the αἰθὴρ χθόνα ἐγκυλούμενος
  (Bacch. 235-290).

  Lucretius (iii. 995-1036) allegorizes the conspicuous sufferers
  in Hadês,—Tantalus, Sisyphus, Tityus, and the Danaïds, as well
  as the ministers of penal infliction, Cerberus and the Furies.
  The first four are emblematic descriptions of various defective
  or vicious characters in human nature,—the deisidæmonic, the
  ambitious, the amorous, or the insatiate and querulous man; the
  last two represent the mental terrors of the wicked.

In the days of Plato and Zenophôn, this allegorizing interpretation
was one of the received methods of softening down the obnoxious
mythes—though Plato himself treated it as an insufficient defence,
seeing that the bulk of youthful hearers could not see through the
allegory, but embraced the story literally as it was set forth.[968]
Pausanias tells us, that when he first began to write his work,
he treated many of the Greek legends as silly and undeserving of
serious attention; but as he proceeded, he gradually arrived at the
full conviction, that the ancient sages had designedly spoken in
enigmatical language, and that there was valuable truth wrapped up in
their narratives: it was the duty of a pious man, therefore, to study
and interpret, but not to reject, stories current and accredited
respecting the gods.[969] And others,—arguing from the analogy of the
religious mysteries, which could not be divulged without impiety to any
except such as had been specially admitted and initiated,—maintained
that it would be a profanation to reveal directly to the vulgar, the
genuine scheme of nature and the divine administration: the ancient
poets and philosophers had taken the only proper course, of talking
to the many in types and parables, and reserving the naked truth for
privileged and qualified intelligences.[970] The allegorical mode of
explaining the ancient fables[971] became more and more popular in
the third and fourth centuries after the Christian æra, especially
among the new Platonic philosophers; being both congenial to their
orientalized turn of thought, and useful as a shield against the
attacks of the Christians.

  [968] Οἱ νῦν περὶ Ὅμηρον δεινοί—so Plato calls these interpreters
  (Kratylus, p. 407); see also Xenoph. Sympos. iii. 6; Plato,
  Ion. p. 530; Plutarch, De Audiend. Poet. p. 19. ὑπόνοια was the
  original word, afterwards succeeded by ἀλληγορία.

  Ἥρας δὲ δεσμοὺς καὶ Ἡφαίστου ῥίψεις ὑπὸ πατρὸς, μέλλοντος τῇ
  μητρὶ τυπτομένῃ ἀμυνεῖν, καὶ θεομαχίας ὅσας Ὅμηρος πεποίηκεν, οὐ
  παραδεκτέον εἰς τὴν πόλιν, ~οὔτ᾽ ἐν ὑπονοίαις πεποιημένας, οὔτ᾽
  ἄνευ ὑπονοιῶν~. Ὁ γὰρ νέος οὐχ οἷός τε κρίνειν ὅ,τι τε ὑπόνοια
  καὶ ὃ μὴ, ἀλλ᾽ ἃ ἂν τηλικοῦτος ὢν λάβῃ ἐν ταῖς δόξαις, δυσέκνιπτά
  τε καὶ ἀμετάστατα φιλεῖ γίγνεσθαι (Plato, Republ. ii. 17. p. 378).

  The idea of an interior sense and concealed purpose in the
  ancient poets occurs several times in Plato (Theætet. c. 93. p.
  180): παρὰ μὲν τῶν ἀρχαίων, μετὰ ποιήσεως ἐπικρυπτομένων τοὺς
  πολλοὺς, etc.; also Protagor. c. 20. p. 316.

  “Modo Stoicum Homerum faciunt,—modo Epicureum,—modo
  Peripateticum,—modo Academicum. Apparet nihil horum esse in illo,
  quia omnia sunt.” (Seneca, Ep. 88.) Compare Plutarch, De Defectu
  Oracul. c. 11-12. t. ii. p. 702, Wytt., and Julian, Orat. vii. p.
  216.

  [969] Pausan. viii. 8, 2. To the same purpose (Strabo, x. p.
  474), allegory is admitted to a certain extent in the fables
  by Dionys. Halic. Ant. Rom. ii. 20. The fragment of the lost
  treatise of Plutarch, on the Platæan festival of the Dædala, is
  very instructive respecting Grecian allegory (Fragm. ix. t. 5. p.
  754-763, ed. Wyt.; ap. Euseb. Præpar. Evang. iii. 1).

  [970] This doctrine is set forth in Macrobius (i. 2). He
  distinguishes between _fabula_ and _fabulosa narratio_:
  the former is fiction pure, intended either to amuse or to
  instruct—the latter is founded upon truth, either respecting
  human or respecting divine agency. The gods did not like to
  be publicly talked of (according to his view) except under
  the respectful veil of a fable (the same feeling as that of
  Herodotus, which led him to refrain from inserting the ἱεροὶ
  λόγοι in his history). The supreme god, the τἀγαθὸν, the πρῶτον
  αἴτιον, could not be talked of in fables: but the other gods,
  the aërial or æthereal powers and the soul, might be, and
  ought to be, talked of in that manner alone. Only superior
  intellects ought to be admitted to a knowledge of the secret
  reality. “De Diis cæteris, et de animâ, non frustra se, nec ut
  oblectent, ad fabulosa convertunt; sed quia sciunt _inimicam
  esse naturæ apertam nudamque expositionem sui_: quæ sicut
  vulgaribus sensibus hominum intellectum sui, vario rerum tegmine
  operimentoque, subtraxit; ita à prudentibus arcana sua voluit
  per fabulosa tractari.... Adeo semper ita se et sciri et coli
  numina maluerunt, qualiter in vulgus antiquitus fabulata est....
  Secundum hæc Pythagoras ipse atque Empedocles, Parmenides quoque
  et Heraclides, de Diis fabulati sunt: nec secus Timæus.” Compare
  also Maximus Tyrius, Dissert. x. and xxxii. Arnobius exposes the
  allegorical interpretation as mere evasion, and holds the Pagans
  to literal historical fact (Adv. Gentes, v. p. 185, ed. Elm.).

  Respecting the allegorical interpretation applied to the Greek
  fables, Böttiger (Die Kunst—Mythologie der Griechen, Abschn. ii.
  p. 176); Nitzsch (Heldensage der Griech. sect. 6. p. 78); Lobeck
  (Aglaopham. p. 133-155).

  [971] According to the anonymous writer ap. Westermann (Script.
  Myth. p. 328), every personal or denominated god may be construed
  in three different ways: either πραγματικῶς (historically, as
  having been a king or a man)—or ψυχικῶς, in which theory Hêrê
  signifies the soul; Athênê, _prudence_; Aphroditê, _desire_;
  Zeus, _mind_, etc.—or στοιχειακῶς, in which system Apollo
  signifies the _sun_; Poseidôn, the _sea_; Hêrê, the upper stratum
  of the air, or _æther_; Athênê, the lower or denser stratum;
  Zeus, the upper hemisphere; Kronus, the lower, etc. This writer
  thinks that all the three principles of construction may be
  resorted to, each on its proper occasion, and that neither of
  them excludes the others. It will be seen that the first is pure
  Euêmerism; the two latter are modes of allegory.

  The allegorical construction of the gods and of the divine
  mythes is copiously applied in the treatises, both of Phurnutus
  and Sallustius, in Gale’s collection of mythological writers.
  Sallustius treats the mythes as of divine origin, and the chief
  poets as inspired (θεόληπτοι): the gods were propitious to those
  who recounted worthy and creditable mythes respecting them, and
  Sallustius prays that they will accept with favor his own remarks
  (cap. 3 and 4. pp. 245-251, Gale). He distributes mythes into
  five classes; theological, physical, spiritual, material, and
  mixed. He defends the practice of speaking of the gods under
  the veil of allegory, much in the same way as Macrobius (in the
  preceding note): he finds, moreover, a good excuse even for
  those mythes which imputed to the gods theft, adultery, outrages
  towards a father, and other enormities: such tales (he says) were
  eminently suitable, since the mind _must at once see_ that the
  facts as told are _not_ to be taken as being themselves the real
  truth, but simply as a veil, disguising some interior truth (p.
  247).

  Besides the Life of Homer ascribed to Plutarch (see Gale, p.
  325-332), Hêraclidês (_not_ Hêraclidês of Pontus) carries out the
  process of allegorizing the Homeric mythes most earnestly and
  most systematically. The application of the allegorizing theory
  is, in his view, the only way of rescuing Homer from the charge
  of scandalous impiety—πάντῃ γὰρ ἠσέβησεν, εἰ μηδὲν ἠλληγόρησεν
  (Hêrac. _in init._ p. 407, Gale). He proves at length, that the
  destructive arrows of Apollo, in the first book of the Iliad,
  mean nothing at the bottom except a contagious plague, caused
  by the heat of the summer sun in marshy ground (pp. 416-424).
  Athênê, who darts down from Olympus at the moment when Achilles
  is about to draw his sword on Agamemnôn, and seizes him by the
  hair, is a personification of repentant prudence (p. 435). The
  conspiracy against Zeus, which Homer (Iliad, i. 400) relates to
  have been formed by the Olympic gods, and defeated by the timely
  aid of Thetis and Briareus—the chains and suspension imposed upon
  Hêrê—the casting of Hêphæstos by Zeus out of Olympus, and his
  fall in Lêmnus—the destruction of the Grecian wall by Poseidôn,
  after the departure of the Greeks—the amorous scene between Zeus
  and Hêrê on Mount Gargarus—the distribution of the universe
  between Zeus, Poseidôn, and Hadês—all these he resolves into
  peculiar manifestations and conflicts of the elemental substances
  in nature. To the much-decried battle of the gods, he gives a
  turn partly physical, partly ethical (p. 481). In like manner,
  he transforms and vindicates the adventures of the gods in the
  Odyssey: the wanderings of Odysseus, together with the Lotophagi,
  the Cyclôps, Circê, the Sirens, Æolus, Scylla, etc., he resolves
  into a series of temptations, imposed as a trial upon a man of
  wisdom and virtue, and emblematic of human life (p. 496). The
  story of Arês, Aphroditê, and Hêphæstos, in the eighth book of
  the Odyssey, seems to perplex him more than any other: he offers
  two explanations, neither of which seems satisfactory even to
  himself (p. 494).

  An anonymous writer in the collection of Westermann (pp. 329-344)
  has discussed the wanderings of Odysseus upon the same ethical
  scheme of interpretation as Hêraclidês: he entitles his treatise
  “A short essay on the Wanderings of Odysseus in Homer, worked
  out in conjunction with ethical reflections, and rectifying what
  is rotten in the story, as well as may be, for the benefit of
  readers.” (τὸ μύθου σαθρὸν θεραπεύουσα.) The author resolves the
  adventures of Odysseus into narratives emblematic of different
  situations and trials of human life. Scylla and Charybdis, for
  example (c. 8. p. 338), represent, the one, the infirmities and
  temptations arising out of the body, the other, those springing
  from the mind, between which man is called upon to steer.
  The adventure of Odysseus with Æolus, shows how little good
  a virtuous man does himself by seeking, in case of distress,
  aid from conjurors and evil enchanters; the assistance of such
  allies, however it may at first promise well, ultimately deceives
  the person who accepts it, and renders him worse off than he was
  before (c. 3. p. 332). By such illustrations does the author
  sustain his general position, that there is a great body of
  valuable ethical teaching wrapped up in the poetry of Homer.

  Proclus is full of similar allegorization, both of Homer and
  Hesiod: the third Excursus of Heyne ad Iliad. xxiii. (vol. viii.
  p. 563), De Allegoriâ Homericâ, contains a valuable summary of
  the general subject.

  The treatise De Astrologiâ, printed among the works of Lucian,
  contains specimens of astrological explanations applied to many
  of the Grecian μῦθοι, which the author as a pious man cannot
  accept in their literal meaning. “How does it consist with
  holiness (he asks) to believe that Æneas was son of Aphroditê,
  Minôs of Zeus, or Askalaphus of Mars? No; these were men born
  under the favorable influences of the planets Venus, Jupiter, and
  Mars.” He considers the principle of astrological explanation
  peculiarly fit to be applied to the mythes of Homer and Hesiod
  (Lucian, De Astrologiâ, c. 21-22).

It was from the same strong necessity, of accommodating the old mythes
to a new standard both of belief and of appreciation, that both the
historical and the allegorical schemes of transforming them arose;
the literal narrative being decomposed for the purpose of arriving at
a base either of particular matter of fact, or of general physical
or moral truth. Instructed men were commonly disposed to historicize
only the heroic legends, and to allegorize more or less of the divine
legends: the attempt of Euêmerus to historicize the latter was for
the most part denounced as irreligious, while that of Metrodôrus to
allegorize the former met with no success. In allegorizing, moreover,
even the divine legends, it was usual to apply the scheme of allegory
only to the inferior gods, though some of the great Stoic philosophers
carried it farther, and allegorized all the separate personal gods,
leaving only an all-pervading cosmic Mind,[972] essential as a
co-efficient along with Matter, yet not separable from Matter. But
many pious pagans seem to have perceived that allegory pushed to this
extent was fatal to all living religious faith,[973] inasmuch as it
divested the gods of their character of Persons, sympathizing with
mankind and modifiable in their dispositions according to the conduct
and prayers of the believer: and hence they permitted themselves to
employ allegorical interpretation only to some of the obnoxious legends
connected with the superior gods, leaving the personality of the latter
unimpeached.

  [972] See Ritter, Geschichte der Philosophie, 2nd edit. part 3.
  book 11. chap. 4. p. 592; Varro ap. Augustin. Civitat. Dei, vi.
  5, ix. 6; Cicero, Nat. Deor. ii. 24-28.

  Chrysippus admitted the most important distinction between Zeus
  and the other gods (Plutarch. de Stoicor. Repugnant. p. 1052.)

  [973] Plutarch. de Isid. et Osirid. c. 66. p. 377; c. 70. p.
  379. Compare on this subject O. Müller, Prolegom. Mythol. p. 59
  _seq._, and Eckermann, Lehrbuch der Religionsgeschichte, vol. i.
  sect. ii. p. 46.

One novelty, however, introduced seemingly by the philosopher
Empedoklês and afterwards expanded by others, deserves notice,
inasmuch as it modified considerably the old religious creed by
drawing a pointed contrast between gods and dæmons,—a distinction
hardly at all manifested in Homer, but recognized in the Works and
Days of Hesiod.[974] Empedoklês widened the gap between the two,
and founded upon it important consequences. The gods were good,
immortal, and powerful agents, having free-will and intelligence,
but without appetite, passion, or infirmity: the dæmons were of
a mixed nature between gods and men, ministers and interpreters
from the former to the latter, but invested also with an agency
and dispositions of their own. They were very long-lived, but not
immortal, and subject to the passions and propensities of men, so
that there were among them beneficent and maleficent dæmons with
every shade of intermediate difference.[975] It had been the mistake
(according to these philosophers) of the old mythes to ascribe to
the gods proceedings really belonging to the dæmons, who were always
the immediate communicants with mortal nature, inspiring prophetic
power to the priestesses of the oracles, sending dreams and omens,
and perpetually interfering either for good or for evil. The wicked
and violent dæmons, having committed many enormities, had thus
sometimes incurred punishment from the gods: besides which, their bad
dispositions had imposed upon men the necessity of appeasing them by
religious ceremonies of a kind acceptable to such beings: hence, the
human sacrifices, the violent, cruel, and obscene exhibitions, the
wailings and fastings, the tearing and eating of raw flesh, which it
had become customary to practise on various consecrated occasions, and
especially in the Dionysiac solemnities. Moreover, the discreditable
actions imputed to the gods,—the terrific combats, the Typhonic and
Titanic convulsions, the rapes, abductions, flight, servitude, and
concealment,—all these were really the doings and sufferings of bad
dæmons, placed far below the sovereign agency—equable, undisturbed, and
unpolluted—of the immortal gods. The action of such dæmons upon mankind
was fitful and intermittent: they sometimes perished or changed their
local abode, so that oracles which had once been inspired became after
a time forsaken and disfranchized.[976]

  [974] Hesiod, Opp. et Di. 122: to the same effect Pythagoras and
  Thalês (Diogen. Laër. viii. 32; and Plutarch, Placit. Philos. i.
  8).

  The Hesiodic dæmons are all good: Athenagoras (Legat. Chr. p.
  8) says that Thalês admitted a distinction between good and bad
  dæmons, which seems very doubtful.

  [975] The distinction between Θεοὶ and Δαίμονες is especially set
  forth in the treatise of Plutarch, De Defectu Oraculorum, capp.
  10, 12, 13, 15, etc. He seems to suppose it traceable to the
  doctrine of Zoroaster or the Orphic mysteries, and he represents
  it as relieving the philosopher from great perplexities: for it
  was difficult to know where to draw the line in admitting or
  rejecting divine Providence: errors were committed sometimes in
  affirming God to be the cause of everything, at other times in
  supposing him to be the cause of nothing. Ἐπεὶ τὸ διορίσαι πῶς
  χρηστέον καὶ μέχρι τίνων τῇ προνοίᾳ, χαλεπὸν, οἱ μὲν οὐδενὸς
  ἁπλῶς τὸν θεὸν, οἱ δὲ ὁμοῦ τι πάντων αἴτιον ποιοῦντες, ἀστοχοῦσι
  τοῦ μετρίου καὶ πρέποντος. Εὖ μὲν οὖν λέγουσιν καὶ οἱ λέγοντες,
  ὅτι Πλάτων τὸ ταῖς γεννωμέναις ποιότησιν ὑποκείμενον στοιχεῖον
  ἐξευρὼν, ὃ νῦν ὕλην καὶ φύσιν καλοῦσιν, πολλῶν ἀπήλλαξε καὶ
  μεγάλων ἀποριῶν τοὺς φιλοσόφους· ἐμοὶ δὲ δοκοῦσι πλείονας λῦσαι
  καὶ μείζονας ἀπορίας οἱ τὸ τῶν δαιμόνων γένος ἐν μέσῳ θεῶν καὶ
  ἀνθρώπων, καὶ τρόπον τινὰ τὴν κοινωνίαν ἡμῶν σύναγον εἰς ταὐτὸ
  καὶ σύναπτον, ἐξευρόντες (c. 10). Ἡ δαιμόνων φύσις ἔχουσα καὶ
  πάθος θνητοῦ καὶ θεοῦ δύναμιν (c. 13).

  Εἰσὶ γὰρ, ὡς ἐν ἀνθρώποις, καὶ δαίμοσιν ἀρετῆς διάφοραὶ, καὶ τοῦ
  παθητικοῦ καὶ ἀλόγου τοῖς μὲν ἀσθενὲς καὶ ἀμαυρὸν ἔτι λείψανον,
  ὥσπερ περίττωμα, τοῖς δὲ πολὺ καὶ δυσκατάσβεστον ἔνεστιν, ὧν ἴχνη
  καὶ σύμβολα πολλαχοῦ θύσιαι καὶ τελεταὶ καὶ μυθολογίαι σώζουσι
  καὶ διαφυλάττουσιν ἐνδιεσπαρμένα (_ib._): compare Plutarch. de
  Isid. et Osir. 25. p. 360.

  Καὶ μὴν ὅσας ~ἔν τε μύθοις καὶ ὕμνοις λέγουσι καὶ ᾅδουσι~, τοῦτο
  μὲν ἁρπαγὰς, τοῦτο δὲ πλάνας θεῶν, κρύψεις τε καὶ φυγὰς καὶ
  λατρείας, οὐ θεῶν εἰσίν ἀλλὰ δαιμόνων παθήματα, etc. (c. 15):
  also c. 23; also De Isid. et Osir. c. 25. p. 366.

  Human sacrifices and other objectionable rites are excused, as
  necessary for the purpose of averting the anger of bad dæmons (c.
  14-15).

  Empedoklês is represented as the first author of the doctrine
  which imputed vicious and abominable dispositions to many of the
  dæmons (c. 15, 16, 17, 20), τοὺς εἰσαγομένους ὑπὸ Ἐμπεδοκλέους
  δαίμονας; expelled from heaven by the gods, θεήλατοι καὶ
  οὐρανοπετεὶς (Plutarch, De Vitand. Aër. Alien. p. 830); followed
  by Plato, Xenokratês, and Chrysippus, c. 17: compare Plato
  (Apolog. Socrat. p. 27; Politic. p. 271; Symposion, c. 28. p.
  203), though he seems to treat the δαίμονες as defective and
  mutable beings, rather than actively maleficent. Xenokratês
  represents some of them both as wicked and powerful in a high
  degree:—Ξενοκράτης καὶ τῶν ἡμερῶν τὰς ἀποφράδας, καὶ τῶν
  ἑορτῶν ὅσαι πληγάς τινας ἢ κοπετοὺς, ἢ νηστείας, ἢ δυσφημίας,
  ἢ αἰσχρολογίαν ἔχουσιν, οὔτε θεῶν τιμαῖς οὔτε δαιμόνων οἴεται
  προσήκειν χρηστῶν, ἀλλ᾽ εἶναι φύσεις ἐν τῷ περιέχοντι μεγάλας
  μὲν καὶ ἰσχυρὰς, δυστρόπους δὲ καὶ σκυθρωπὰς, αἳ ~χαίρουσι τοῖς
  τοιούτοις~, καὶ ~τυγχάνουσαι πρὸς οὐθὲν ἄλλο χεῖρον τρέπονται~
  (Plutarch, De Isid. ut Osir. c. 26. p. 361; Quæstion. Rom. p.
  283): compare Stobæus, Eclog. Phys. i. p. 62.

  [976] Plutarch, De Defect. Orac. c. 15. p. 418. Chrysippus
  admitted, among the various conceivable causes to account for
  the existence of evil, the supposition of some negligent and
  reckless dæmons, δαιμόνια φαυλὰ ἐν οἷς τῷ ὄντι γίνονται καὶ
  ἐγκλητέαι ἀμέλειαι (Plutarch, De Stoicor. Repugnant. p. 1051). A
  distinction, which I do not fully understand, between θεοὶ and
  δαίμονες, was also adopted among the Locrians at Opus: δαίμων
  with them seems to have been equivalent to ἥρως (Plutarch,
  Quæstion. Græc. c. 6. p. 292): see the note above, pp. 350-351.

This distinction between gods and dæmons appeared to save in a great
degree both the truth of the old legends and the dignity of the
gods: it obviated the necessity of pronouncing either that the gods
were unworthy, or the legends untrue. Yet although devised for the
purpose of satisfying a more scrupulous religious sensibility, it was
found inconvenient afterwards, when assailants arose against paganism
generally. For while it abandoned as indefensible a large portion of
what had once been genuine faith, it still retained the same word
_dæmons_ with an entirely altered signification. The Christian writers
in their controversies found ample warrant among the _earlier_ pagan
authors[977] for treating all the gods as dæmons—and not less ample
warrant among the _later_ pagans for denouncing the dæmons generally as
evil beings.[978]

  [977] Tatian. adv. Græcos, c. 20; Clemens Alexandrin. Admonit.
  ad Gentes, pp. 26-29, Sylb.; Minuc. Felix, Octav. c. 26. “Isti
  igitur impuri spiritus, ut ostensum a Magis, a philosophis, a
  Platone, sub statuis et imaginibus consecrati delitescunt, et
  afflatu suo quasi auctoritatem præsentis numinis consequuntur,”
  etc. This, like so many other of the aggressive arguments of
  the Christians against paganism, was taken from the pagan
  philosophers themselves.

  Lactantius, De Verâ Philosophiâ, iv. 28. “Ergo iidem sunt
  Dæmones, quos fatentur execrandos esse: iidem Dii, quibus
  supplicant. Si nobis credendum esse non putant, credant Homero;
  qui summum illum Jovem Dæmonibus aggregavit,” etc.

  [978] See above, Chapter II. p. 70, the remarks on the Hesiodic
  Theogony.

Such were the different modes in which the ancient mythes were
treated, during the literary life of Greece, by the four classes above
named—poets, logographers, historians, and philosophers.

Literal acceptance, and unconscious, uninquiring faith, such as they
had obtained from the original auditors to whom they were addressed,
they now found only among the multitude—alike retentive of traditional
feeling[979] and fearful of criticizing the proceedings of the
gods.[980] But with instructed men they became rather subjects of
respectful and curious analysis—all agreeing that the Word as tendered
to them was inadmissible, yet all equally convinced that it contained
important meaning, though hidden yet not undiscoverable. A very large
proportion of the force of Grecian intellect was engaged in searching
after this unknown base, by guesses, in which sometimes the principle
of semi-historical interpretation was assumed, sometimes that of
allegorical, without any collateral evidence in either case, and
without possibility of verification. Out of the one assumption grew a
string of allegorized phænomenal truths, out of the other a long series
of seeming historical events and chronological persons,—both elicited
from the transformed mythes and from nothing else.[981]

  [979] A destructive inundation took place at Pheneus in
  Arcadia, seemingly in the time of Plutarch: the subterranean
  outlet (βάραθρον) of the river had become blocked up, and the
  inhabitants ascribed the stoppage to the anger of Apollo, who had
  been provoked by the stealing of the Pythian tripod by Hêraklês:
  the latter had carried the tripod to Pheneus and deposited it
  there. Ἆρ᾽ οὖν οὐκ ἀτοπώτερος τούτων ὁ Ἀπόλλων, εἰ Φενεάτας
  ἀπόλλυσι τοὺς νῦν, ἐμφράξας τὸ βάραθρον, καὶ κατακλύσας τὴν χώραν
  ἅπασαν αὐτῶν, ὅτι πρὸ χιλίων ἐτῶν, ὥς φασιν, ὁ Ἡρακλῆς ἀνασπάσας
  τὸν τρίποδα τὸν μαντικὸν εἰς Φενεὸν ἀπήνεγκε; (Plutarch. de
  Serâ Numin. Vindictâ, p. 577; compare Pausan. viii. 14, 1.)
  The expression of Plutarch, that the abstraction of the tripod
  by Hêraklês had taken place 1000 years before, is that of the
  critic, who thinks it needful to historicize and chronologize the
  genuine legend; which, to an inhabitant of Pheneus, at the time
  of the inundation, was doubtless as little questioned as if the
  theft of Hêraklês had been laid in the preceding generation.

  Agathoclês of Syracuse committed depredations on the coasts
  of Ithaca and Korkyra: the excuse which he offered was, that
  Odysseus had come to Sicily and blinded Polyphêmus, and that
  on his return he had been kindly received by the Phæakians
  (Plutarch, _ib._).

  This is doubtless a jest, either made by Agathoclês, or more
  probably invented for him; but it is founded upon a popular
  belief.

  [980] “Sanctiusque et reverentius visum, de actis Deorum credere
  quam scire.” (Tacit. German. c. 34.)

  Aristidês, however, represents the Homeric theology (whether he
  would have included the Hesiodic we do not know) as believed
  quite literally among the multitude in his time, the second
  century after Christianity (Aristid. Orat. iii. p. 25). Ἀπορῶ,
  ὅπη πότε χρή με διαθέσθαι μεθ᾽ ὑμῶν, πότερα ὡς τοῖς πολλοῖς δοκεῖ
  καὶ Ὁμήρῳ δὲ συνδοκεῖ, θεῶν παθήματα συμπεισθῆναι καὶ ἡμᾶς,
  οἷον Ἀρέος δέσμα καὶ Ἀπόλλωνος θητείας καὶ Ἡφαίστου ῥίψεις εἰς
  θάλασσαν, οὕτω δὲ καὶ Ἰνοῦς ἄχη καὶ φυγάς τινας. Compare Lucian,
  Ζεὺς Τραγῶδος, c. 20, and De Luctu, c. 2; Dionys. Halicar. A. R.
  ii. p. 90, Sylb.

  Kallimachus (Hymn. ad Jov. 9) distinctly denied the statement
  of the Kretans that they possessed in Krête the tomb of Zeus,
  and treated it as an instance of Kretan mendacity; while
  Celsus did not deny it, but explained it in some figurative
  manner—αἰνιττόμενος τροπικὰς ὑπονοίας (Origen. cont. Celsum, iii.
  p. 137).

  [981] There is here a change as compared with my first edition;
  I had inserted here some remarks on the allegorical theory of
  interpretation, as compared with the semi-historical. An able
  article on my work (in the Edinburgh Review, October 1846),
  pointed out that those remarks required modification, and that
  the idea of allegory in reference to the construction of the
  mythes was altogether inadmissible.

The utmost which we accomplish by means of the semi-historical theory,
even in its most successful applications, is, that after leaving out
from the mythical narrative all that is miraculous or high-colored or
extravagant, we arrive at a series of credible incidents—incidents
which _may, perhaps_, have really occurred, and against which no
intrinsic presumption can be raised. This is exactly the character
of a well-written modern novel (as, for example, several among the
compositions of Defoe), the whole story of which is such as may well
have occurred in real life: it is plausible fiction, and nothing
beyond. To raise plausible fiction up to the superior dignity of
truth, some positive testimony or positive ground of inference must
be shown; even the highest measure of intrinsic probability is not
alone sufficient. A man who tells us that, on the day of the battle
of Platæa, rain fell on the spot of ground where the city of New York
now stands, will neither deserve nor obtain credit, because he can
have had no means of positive knowledge; though the statement is not
in the slightest degree improbable. On the other hand, statements in
themselves very improbable may well deserve belief, provided they be
supported by sufficient positive evidence; thus the canal dug by order
of Xerxês across the promontory of Mount Athos, and the sailing of
the Persian fleet through it, is a fact which I believe, because it
is well-attested—notwithstanding its remarkable improbability, which
so far misled Juvenal as to induce him to single out the narrative
as a glaring example of Grecian mendacity.[982] Again, many critics
have observed that the general tale of the Trojan war (apart from
the superhuman agencies) is not more improbable than that of the
Crusades, which every one admits to be an historical fact. But (even
if we grant this position, which is only true to a small extent), it
is not sufficient to show an analogy between the two cases in respect
to negative presumptions alone; the analogy ought to be shown to hold
between them in respect to positive certificate also. The Crusades
are a curious phænomenon in history, but we accept them, nevertheless,
as an unquestionable fact, because the antecedent improbability is
surmounted by adequate contemporary testimony. When the like testimony,
both in amount and kind, is produced to establish the historical
reality of a Trojan war, we shall not hesitate to deal with the two
events on the same footing.

  [982] Juvenal, Sat. x. 174:—

                                “Creditur olim
    Velificatus Athos, et quantum Græcia mendax
    Audet in historiâ,” etc.

In applying the semi-historical theory to Grecian mythical narrative,
it has been often forgotten that a certain strength of testimony, or
positive ground of belief, must first be tendered, before we can be
called upon to discuss the antecedent probability or improbability of
the incidents alleged. The belief of the Greeks themselves, without the
smallest aid of special or contemporary witnesses, has been tacitly
assumed as sufficient to support the case, provided only sufficient
deduction be made from the mythical narratives to remove all antecedent
improbabilities. It has been taken for granted that the faith of the
people must have rested originally upon some particular historical
event, involving the identical persons, things, and places which the
original mythes exhibit, or at least the most prominent among them.
But when we examine the pyschagogic influences predominant in the
society among whom this belief originally grew up, we shall see that
their belief is of little or no evidentiary value, and that the growth
and diffusion of it may be satisfactorily explained without supposing
any special basis of matters of fact. The popular faith, so far as
it counts for anything, testifies in favor of the entire and literal
mythes, which are now universally rejected as incredible.[983] We have
thus the very minimum of positive proof, and the maximum of negative
presumption: we may diminish the latter by conjectural omissions and
interpolations, but we cannot by any artifice increase the former: the
narrative ceases to be incredible, but it still remains uncertified,—a
mere commonplace possibility. Nor is fiction always, or essentially,
extravagant and incredible. It is often not only plausible and
coherent, but even more like truth (if a paradoxical phrase may be
allowed) than truth itself. Nor can we, in the absence of any extrinsic
test, reckon upon any intrinsic mark to discriminate the one from the
other.[984]

  [983] Colonel Sleeman observes, respecting the Hindoo historical
  mind—“History to this people is all a fairy tale.” (Rambles and
  Recollections of an Indian Official, vol. i. ch. ix. p. 70.) And
  again, “The popular poem of the Ramaen describes the abduction
  of the heroine by the monster king of Ceylon, Rawun; and her
  recovery by means of the monkey general, Hunnooman. Every word of
  this poem, the people assured me was written, if not by the hand
  of the Deity himself, at least by his inspiration, which was the
  same thing—and it must consequently be true. Ninety-nine out of
  a hundred, among the Hindoos, implicitly believe, not only every
  word of the poem, but every word of every poem that has ever been
  written in Sanscrit. If you ask a man whether he really believes
  any very egregious absurdity quoted from these books, he replies,
  with the greatest _naïveté_ in the world, Is it not written in
  the book; and how should it be there written, if not true? The
  Hindoo religion reposes upon an entire prostration of mind,—that
  continual and habitual surrender of the reasoning faculties,
  which we are accustomed to make occasionally, while engaged at
  the theatre, or in the perusal of works of fiction. We allow the
  scenes, characters, and incidents, to pass before our mind’s
  eye, and move our feelings—without stopping a moment to ask
  whether they are real or true. There is only this difference—that
  with people of education among us, even in such short intervals
  of illusion or _abandon_, any extravagance in the acting, or
  flagrant improbability in the fiction, destroys the charm, breaks
  the spell by which we have been so mysteriously bound, and
  restores us to reason and the realities of ordinary life. With
  the Hindoos, on the contrary, the greater the improbability, the
  more monstrous and preposterous the fiction—the greater is the
  charm it has over their minds; and the greater their learning
  in the Sanscrit, the more are they under the influence of this
  charm. Believing all to be written by the Deity, or under his
  inspirations, and the men and things of former days to have
  been very different from men and things of the present day, and
  the heroes of these fables to have been demigods, or people
  endowed with powers far superior to those of the ordinary men
  of their own day—the analogies of nature are never for a moment
  considered; nor do questions of probability, or possibility,
  according to those analogies, ever obtrude to dispel the charm
  with which they are so pleasingly bound. They go on through
  life reading and talking of these monstrous fictions, which
  shock the taste and understanding of other nations, without
  ever questioning the truth of one single incident, or hearing
  it questioned. There was a time, and that not far distant, when
  it was the same in England, and in every other European nation;
  and there are, I am afraid, some parts of Europe where it is
  so still. But the Hindoo faith, so far as religious questions
  are concerned, is not more capacious or absurd than that of the
  Greeks or Romans in the days of Socrates or Cicero: the only
  difference is, that among the Hindoos a greater number of the
  questions which interest mankind are brought under the head of
  religion.” (Sleeman, Rambles, etc., vol. i. ch. xxvi. p. 227:
  compare vol. ii. ch. v. p. 51; viii. p. 97.)

  [984] Lord Lyttleton, in commenting on the tales of the Irish
  bards, in his History of Henry II., has the following just
  remarks (book iv. vol. iii. p. 13, quarto): “One may reasonably
  suppose that in MSS. written since the Irish received the Roman
  letters from St. Patrick, _some_ traditional truths recorded
  before by the bards in their unwritten poems may have been
  preserved to our times. Yet these cannot be so separated from
  many fabulous stories derived from the same sources, as to obtain
  a firm credit; it not being sufficient to establish the authority
  of suspected traditions, that they can be shown not to be so
  improbable or absurd as others with which they are mixed—_since
  there may be specious as well as senseless fictions_. Nor can a
  poet or bard, who lived in the sixth or seventh century after
  Christ, if his poem is still extant, be any voucher for facts
  supposed to have happened before the incarnation; though his
  evidence (allowing for poetical license) may be received on such
  matters as come within his own time, or the remembrance of old
  men with whom he conversed. The most judicious historians pay no
  regard to the Welsh or British traditions delivered by Geoffrey
  of Monmouth, though it is not impossible but that some of these
  may be true.”

  One definition of a mythe given by Plutarch coincides exactly
  with a _specious fiction_: Ὁ μῦθος εἶναι βούλεται λόγος ψευδὴς
  ἐοικὼς ἀληθινῷ (Plutarch, Bellone an pace clariores fuerunt
  Athenienses, p. 348).

  “Der Grund-Trieb des Mythus (Creuzer justly expresses it) das
  Gedachte in ein Geschehenes umzusetzen.” (Symbolik der Alten
  Welt, sect. 43. p. 99.)

In the semi-historical theory respecting Grecian mythical narrative,
the critic unconsciously transports into the Homeric age those habits
of classification and distinction, and that standard of acceptance
or rejection, which he finds current in his own. Amongst us, the
distinction between historical fact and fiction is highly valued as
well as familiarly understood: we have a long history of the past,
deduced from a study of contemporary evidences; and we have a body
of fictitious literature, stamped with its own mark and interesting
in its own way. Speaking generally, no man could now hope to succeed
permanently in transferring any striking incident from the latter
category into the former, nor could any man deliberately attempt it
without incurring well-merited obloquy. But this _historical sense_,
now so deeply rooted in the modern mind that we find a difficulty
in conceiving any people to be without it, is the fruit of records
and inquiries, first applied to the present, and then preserved and
studied by subsequent generations; while in a society which has not
yet formed the habit of recording its present, the real facts of the
past can never be known; the difference between attested matter
of fact and plausible fiction—between truth and that which is like
truth—can neither be discerned nor sought for. Yet it is precisely upon
the supposition that this distinction is present to men’s habitual
thoughts, that the semi-historical theory of the mythes is grounded.

It is perfectly true, as has often been stated, that the Grecian epic
contains what are called traditions respecting the past—the larger
portion of it, indeed, consists of nothing else. But what are these
traditions? They are the matter of those songs and stories which have
acquired hold on the public mind; they are the creations of the poets
and storytellers themselves, each of whom finds some preëxisting, and
adds others of his own, new and previously untold, under the impulse
and authority of the inspiring Muse. Homer doubtless found many songs
and stories current with respect to the siege of Troy; he received and
transmitted some of these traditions, recast and transformed others,
and enlarged the whole mass by new creations of his own. To the
subsequent poets, such as Arktinus and Leschês, these Homeric creations
formed portions of preëxisting tradition, with which they dealt in the
same manner; so that the whole mass of traditions constituting the tale
of Troy became larger and larger with each successive contributor. To
assume a generic difference between the older and the newer strata of
tradition—to treat the former as morsels of history, and the latter
as appendages of fiction—is an hypothesis gratuitous at the least,
not to say inadmissible. For the further we travel back into the
past, the more do we recede from the clear day of positive history,
and the deeper do we plunge into the unsteady twilight and gorgeous
clouds of fancy and feeling. It was one of the agreeable dreams of the
Grecian epic, that the man who travelled far enough northward beyond
the Rhipæan mountains, would in time reach the delicious country and
genial climate of the virtuous Hyperboreans—the votaries and favorites
of Apollo, who dwelt in the extreme north beyond the chilling blasts
of Boreas. Now the hope that we may, by carrying our researches up the
stream of time, exhaust the limits of fiction, and land ultimately upon
some points of solid truth, appears to me no less illusory than this
northward journey in quest of the Hyperborean elysium.

The general disposition to adopt the semi-historical theory as to the
genesis of Grecian mythes, arises in part from reluctance in critics to
impute to the mythopœic ages extreme credulity or fraud; together with
the usual presumption, that where much is believed some portion of it
must be true. There would be some weight in these grounds of reasoning,
if the ages under discussion had been supplied with records and
accustomed to critical inquiry. But amongst a people unprovided with
the former and strangers to the latter, credulity is naturally at its
maximum, as well in the narrator himself as in his hearers: the idea
of deliberate fraud is moreover inapplicable,[985] for if the hearers
are disposed to accept what is related to them as a revelation from
the Muse, the _œstrus_ of composition is quite sufficient to impart a
similar persuasion to the poet whose mind is penetrated with it. The
belief of that day can hardly be said to stand apart by itself as an
act of reason. It becomes confounded with vivacious imagination and
earnest emotion; and in every case where these mental excitabilities
are powerfully acted upon, faith ensues unconsciously and as a matter
of course. How active and prominent such tendencies were among the
early Greeks, the extraordinary beauty and originality of their epic
poetry may teach us.

  [985] In reference to the loose statements of the Highlanders,
  Dr. Johnson observes, “He that goes into the Highlands with a
  mind naturally acquiescent, and a credulity eager for wonders,
  may perhaps come back with an opinion very different from mine;
  for the inhabitants, knowing the ignorance of all strangers in
  their language and antiquities, are perhaps not very scrupulous
  adherents to truth; yet I do not say that they deliberately speak
  studied falsehood, or have a settled purpose to deceive. They
  have acquired and considered little, and do not always feel their
  own ignorance. They are not much accustomed to be interrogated
  by others, and seem never to have thought of interrogating
  themselves; _so that if they do not know what they tell to be
  true, they likewise do not distinctly perceive it to be false_.
  Mr. Boswell was very diligent in his inquiries, and the result of
  his investigations was, that the answer to the second question
  was commonly such as nullified the answer to the first.” (Journey
  to the Western Islands, p. 272, 1st edit., 1775).

It is, besides, a presumption far too largely and indiscriminately
applied, even in our own advanced age, that where much is believed,
something must necessarily be true—that accredited fiction is always
traceable to some basis of historical truth.[986] The influence
of imagination and feeling is not confined simply to the process
of retouching, transforming, or magnifying narratives originally
founded on fact; it will often create new narratives of its own,
without any such preliminary basis. Where there is any general body
of sentiment pervading men living in society, whether it be religious
or political—love, admiration, or antipathy—all incidents tending to
illustrate that sentiment are eagerly welcomed, rapidly circulated
and (as a general rule) easily accredited. If real incidents are not
at hand, impressive fictions will be provided to satisfy the demand.
The perfect harmony of such fictions with the prevalent feeling stands
in the place of certifying testimony, and causes men to hear them not
merely with credence, but even with delight: to call them in question
and require proof, is a task which cannot be undertaken without
incurring obloquy. Of such tendencies in the human mind, abundant
evidence is furnished by the innumerable religious legends which
have acquired currency in various parts of the world, and of which
no country was more fertile than Greece—legends which derived their
origin, not from special facts misreported and exaggerated, but from
pious feelings pervading the society, and translated into narrative
by forward and imaginative minds—legends, in which not merely the
incidents, but often even the personages are unreal, yet in which the
generating sentiment is conspicuously discernible, providing its own
matter as well as its own form. Other sentiments also, as well as the
religious, provided they be fervent and widely diffused, will find
expression in current narrative, and become portions of the general
public belief—every celebrated and notorious character is the source
of a thousand fictions exemplifying his peculiarities. And if it be
true, as I think present observation may show us, that such creative
agencies are even now visible and effective, when the materials of
genuine history are copious and critically studied—much more are we
warranted in concluding that, in ages destitute of records, strangers
to historical testimony, and full of belief in divine inspiration both
as to the future and as to the past, narratives purely fictitious
will acquire ready and uninquiring credence, provided only they be
plausible and in harmony with the preconceptions of the auditors.

  [986] I considered this position more at large in an article
  in the “Westminster Review” for May, 1843, on Niebuhr’s Greek
  Legends, with which article much in the present chapter will be
  found to coincide.

The allegorical interpretation of the mythes has been by several
learned investigators, especially by Creuzer, connected with the
hypothesis of an ancient and highly instructed body of priests, having
their origin either in Egypt or in the East, and communicating to the
rude and barbarous Greeks religious, physical, and historical knowledge
under the veil of symbols. At a time (we are told) when language was
yet in its infancy, visible symbols were the most vivid means of acting
upon the minds of ignorant hearers: the next step was to pass to
symbolical language and expressions—for a plain and literal exposition,
even if understood at all, would at least have been listened to with
indifference, as not corresponding with any mental demand. In such
allegorizing way, then, the early priests set forth their doctrines
respecting God, nature, and humanity—a refined monotheism and a
theological philosophy—and to this purpose the earliest mythes were
turned. But another class of mythes, more popular and more captivating,
grew up under the hands of the poets—mythes purely epical, and
descriptive of real or supposed past events. The allegorical mythes,
being taken up by the poets, insensibly became confounded in the same
category with the purely narrative mythes—the matter symbolized was no
longer thought of, while the symbolizing words came to be construed in
their own literal meaning—and the basis of the early allegory, thus
lost among the general public, was only preserved as a secret among
various religious fraternities, composed of members allied together
by initiation in certain mystical ceremonies, and administered by
hereditary families of presiding priests. In the Orphic and Bacchic
sects, in the Eleusinian and Samothracian mysteries, was thus treasured
up the secret doctrine of the old theological and philosophical mythes,
which had once constituted the primitive legendary stock of Greece, in
the hands of the original priesthood and in ages anterior to Homer.
Persons who had gone through the preliminary ceremonies of initiation,
were permitted at length to hear, though under strict obligation of
secrecy, this ancient religious and cosmogonic doctrine, revealing
the destination of man and the certainty of posthumous rewards and
punishments—all disengaged from the corruptions of poets, as well as
from the symbols and allegories under which they still remained buried
in the eyes of the vulgar. The mysteries of Greece were thus traced up
to the earliest ages, and represented as the only faithful depository
channels of that purer theology and physics which had originally
been communicated, though under the unavoidable inconvenience of a
symbolical expression, by an enlightened priesthood coming from abroad
to the then rude barbarians of the country.[987]

  [987] For this general character of the Grecian mysteries, with
  their concealed treasure of doctrine, see _Warburton_, Divine
  Legation of Moses, book ii. sect. 4.

  _Payne Knight_, On the Symbolical Language of ancient Art and
  Mythology, sect. 6, 10, 11, 40, etc.

  _Saint Croix_, Recherches sur les Mystères du Paganisme, sect. 3,
  p. 106; sect. 4, p. 404, etc.

  _Creuzer_, Symbolik und Mythologie der Alten Völker, sect. 2,
  3, 23, 39, 42, etc. Meiners and Heeren adopt generally the same
  view, though there are many divergences of opinion between these
  different authors, on a subject essentially obscure. Warburton
  maintained that the interior doctrine communicated in the
  mysteries was the existence of one Supreme Divinity, combined
  with the Euemeristic creed, that the pagan gods had been mere men.

  See Clemens Alex. Strom. v. p. 582, Sylb.

  The view taken by Hermann of the ancient Greek mythology is
  in many points similar to that of Creuzer, though with some
  considerable difference. He thinks that it is an aggregate
  of doctrine—philosophical, theological, physical, and
  moral—expressed under a scheme of systematic personifications,
  each person being called by a name significant of the function
  personified: this doctrine was imported from the East into
  Greece, where the poets, retaining or translating the names, but
  forgetting their meaning and connection, distorted the primitive
  stories, the sense of which came to be retained only in the
  ancient mysteries. That true sense, however, (he thinks,) may be
  recovered by a careful analysis of the significant names: and
  his two dissertations (De Mythologiâ Græcorum Antiquissimâ, in
  the Opuscula, vol. ii.) exhibit a specimen of this systematic
  expansion of etymology into narrative. The dissent from Creuzer
  is set forth in their published correspondence, especially in his
  concluding “Brief an Creuzer über das Wesen und die Behandlung
  der Mythologie,” Leipzig, 1819. The following citation from his
  Latin dissertation sets forth his general doctrine:—

  Hermann, De Mythologiâ Græcorum Antiquissimâ, p. 4 (Opuscula,
  vol. ii. p. 171): “Videmus rerum divinarum humanarumque scientiam
  ex Asiâ per Lyciam migrantem in Europam: videmus fabulosos
  poëtas peregrinam doctrinam, monstruoso tumore orientis sive
  exutam, sive nondum indutam, quasi de integro Græcâ specie
  procreantes; videmus poëtas, illos, quorum omnium vera nomina
  nominibus—ab arte, quâ clarebant, petitis—obliterata sunt,
  diu in Thraciâ hærentes, raroque tandem etiam cum aliis
  Græciæ partibus commercio junctos: qualis Pamphus, non ipse
  Atheniensis, Atheniensibus hymnos Deorum fecit. Videmus denique
  retrusam paulatim in mysteriorum secretam illam sapientum
  doctrinam, vitiatam religionum perturbatione, corruptam inscitiâ
  interpretum, obscuratam levitate amœniora sectantium—adeo ut eam
  ne illi quidem intelligerent, qui hæreditariam a prioribus poësin
  colentes, quum ingenii præstantiâ omnes præstinguerent, tantâ
  illos oblivione merserunt, ut ipsi sint primi auctores omnis
  eruditionis habiti.”

  Hermann thinks, however, that by pursuing the suggestions of
  etymology, vestiges may still be discovered, and something like a
  history compiled, of Grecian belief as it stood anterior to Homer
  and Hesiod: “Est autem in hac omni ratione judicio maxime opus,
  quia non testibus res agitur, sed ad interpretandi solertiam
  omnia revocanda sunt” (p. 172). To the same general purpose
  the French work of M. Emérie David, Recherches sur le Dieu
  Jupiter—reviewed by O. Müller: see the Kleine Schriften of the
  latter, vol. ii. p. 82.

  Mr. Bryant has also employed a profusion of learning, and
  numerous etymological conjectures, to resolve the Greek mythes
  into mistakes, perversions, and mutilations, of the exploits and
  doctrines of oriental tribes long-lost and by-gone,—Amonians,
  Cuthites, Arkites, etc. “It was Noah (he thinks) who was
  represented under the different names of Thoth, Hermês, Menês,
  Osiris, Zeuth, Atlas, Phorôneus, Promêtheus, to which list a
  farther number of great extent might be added: the Νοῦς of
  Anaxagoras was in reality the patriarch Noah” (Ant. Mythol. vol.
  ii. pp. 253, 272). “The Cuthites or Amonians, descendants of
  Noah, settled in Greece from the east, celebrated for their skill
  in building and the arts” (_ib._ i. p. 502; ii. p. 187). “The
  greatest part of the Grecian theology arose from misconception
  and blunders, the stories concerning their gods and heroes were
  founded on terms misinterpreted or abused” (_ib._ i. p. 452).
  “The number of different actions ascribed to the various Grecian
  gods or heroes all relate to one people or family, and are at
  bottom one and the same history” (_ib._ ii. p. 57). “The fables
  of Promêtheus and Tityus were taken from ancient Amonian temples,
  from hieroglyphics misunderstood and badly explained” (i. p.
  426): see especially vol. ii. p. 160.

But this theory, though advocated by several learned men, has been
shown to be unsupported and erroneous. It implies a mistaken view
both of the antiquity and the purport of the mysteries, which cannot
be safely carried up even to the age of Hesiod, and which, though
imposing and venerable as religious ceremonies, included no recondite
or esoteric teaching.[988]

  [988] The Anti-Symbolik of Voss, and still more the Aglaophamus
  of Lobeck, are full of instruction on the subject of this
  supposed interior doctrine, and on the ancient mysteries
  in general: the latter treatise, especially, is not less
  distinguished for its judicious and circumspect criticism than
  for its copious learning.

  Mr. Halhed (Preface to the Gentoo Code of Laws, pp. xiii.-xiv.)
  has good observations on the vanity of all attempts to allegorize
  the Hindu mythology: he observes, with perfect truth, “The
  vulgar and illiterate have always understood the mythology of
  their country in its literal sense; and there was a time to
  every nation, when the highest rank in it was equally vulgar and
  illiterate with the lowest.... A Hindu esteems the astonishing
  miracles attributed to a Brima, or a Kishen, as facts of the
  most indubitable authenticity, and the relation of them as most
  strictly historical.”

  Compare also Gibbon’s remarks on the allegorizing tendencies of
  the later Platonists (Hist. Decl. and Fall, vol. iv. p. 71).

The doctrine, supposed to have been originally symbolized and
subsequently overclouded, in the Greek mythes, was in reality first
intruded into them by the unconscious fancies of later interpreters. It
was one of the various roads which instructed men took to escape from
the literal admission of the ancient mythes, and to arrive at some new
form of belief, more consonant with their ideas of what the attributes
and character of the gods ought to be. It was one of the ways of
constituting, by help of the mysteries, a philosophical religion apart
from the general public, and of connecting that distinction with the
earliest periods of Grecian society. Such a distinction was both avowed
and justified among the superior men of the later pagan world. Varro
and Scævola distributed theology into three distinct departments,—the
mythical or fabulous, the civil, and the physical. The first had
its place in the theatre, and was left without any interference to
the poets; the second belonged to the city of political community
as such,—it comprised the regulation of all the public worship and
religious rites, and was consigned altogether to the direction of
the magistrate; the third was the privilege of philosophers, but was
reserved altogether for private discussion in the schools, apart from
the general public.[989] As a member of the city, the philosopher
sympathized with the audience in the theatre, and took a devout share
in the established ceremonies, nor was he justified in trying what
he heard in the one or saw in the other by his own ethical standard.
But in the private assemblies of instructed or inquisitive men, he
enjoyed the fullest liberty of canvassing every received tenet, and
of broaching his own theories unreservedly, respecting the existence
and nature of the gods. By these discussions, the activity of the
philosophical mind was maintained and truth elicited; but it was such
truth as the body of the people ought not to hear, lest their faith in
their own established religious worship should be overthrown. In thus
distinguishing the civil theology from the fabulous, Varro was enabled
to cast upon the poets all the blame of the objectionable points in the
popular theology, and to avoid the necessity of pronouncing censure on
the magistrates, who (he contended) had made as good a compromise with
the settled prejudices of the public as the case permitted.

  [989] Varro, ap. Augustin. De Civ. Dei, iv. 27; vi. 5-6. “Dicis
  fabulosos Deos accommodatos esse ad theatrum, naturales ad
  mundum, civiles ad urbem.” “Varro, de religionibus loquens, multa
  esse vera dixit, quæ non modo vulgo scire non sit utile, sed
  etiam tametsi falsa sint, aliter existimare populum expediat:
  et ideo Græcos teletas et mysteria taciturnitate parietibusque
  clausisse” (_ibid._ iv. 31). See Villoison, De Triplici Theologiâ
  Commentatio, p. 8; and Lactantius, De Origin. Error. ii. 3. The
  doctrine of the Stoic Chrysippus, ap. Etymologicon Magn. v.
  Τελεταί—Χρύσιππος δέ φησι, τοὺς περὶ τῶν θείων λόγους εἰκότως
  καλεῖσθαι τελετὰς, χρῆναι γὰρ τούτους τελευταίους καὶ ἐπὶ πᾶσι
  διδάσκεσθαι, τῆς ψυχῆς ἐχούσης ἕρμα καὶ κεκρατημένης, καὶ πρὸς
  τοὺς ἀμυήτους σιωπᾷν δυναμένης· μέγα γὰρ εἶναι τὸ ἆθλον ὑπὲρ θεῶν
  ἀκοῦσαί τε ὀρθὰ, καὶ ἐγκρατεῖς γενέσθαι αὐτῶν.

  The triple division of Varro is reproduced in Plutarch,
  Amatorius, p. 763. τὰ μὲν μύθῳ, τὰ δὲ νόμῳ, τὰ δὲ λόγῳ, πίστιν ἐξ
  ἀρχῆς ἔσχηκε· τῆς δ᾽ οὖν περὶ θεῶν δόξης καὶ παντάπασιν ἡγεμόνες
  καὶ διδάσκαλοι γεγόνασιν ἡμῖν οἵ τε ποιηταὶ, καὶ οἱ νομοθέται,
  καὶ τρίτον, οἱ φιλόσοφοι.

The same conflicting sentiments which led the philosophers to decompose
the divine mythes into allegory, impelled the historians to melt down
the heroic mythes into something like continuous political history,
with a long series of chronology calculated upon the heroic pedigrees.
The one process as well as the other was interpretative guesswork,
proceeding upon unauthorized assumptions, and without any verifying
test or evidence: while it frittered away the characteristic beauty
of the mythe into something essentially anti-mythical, it sought to
arrive both at history and philosophy by impracticable roads. That the
superior men of antiquity should have striven hard to save the dignity
of legends which constituted the charm of their literature as well as
the substance of the popular religion, we cannot be at all surprised;
but it is gratifying to find Plato discussing the subject in a more
philosophical spirit. The Platonic Socratês, being asked whether he
believed the current Attic fable respecting the abduction of Oreithyia
(daughter of Erechtheus) by Boreas, replies, in substance,—“It would
not be strange if I disbelieved it, as the clever men do; I might
then show my cleverness by saying that a gust of Boreas blew her down
from the rocks above while she was at play, and that, having been
killed in this manner, she was reported to have been carried off by
Boreas. Such speculations are amusing enough, but they belong to men
ingenious and busy-minded overmuch, and not greatly to be envied, if
it be only for this reason, _that, after having set right one fable,
they are under the necessity of applying the same process to a host of
others_—Hippocentaurs, Chimæras, Gorgons, Pegasus, and numberless other
monsters and incredibilities. A man, who, disbelieving these stories,
shall try to find a probable basis for each of them, will display an
ill-placed acuteness and take upon himself an endless burden, for which
I at least have no leisure: accordingly, I forego such researches, and
believe in the current version of the stories.”[990]

  [990] Plato, Phædr. c. 7. p. 229:—

  PHÆDRUS. Εἶπέ μοι, ὦ Σώκρατες, σὺ τοῦτο τὸ μυθολόγημα πείθει
  ἀληθὲς εἶναι;

  SOCRATES. Ἀλλ᾽ εἰ ἀπιστοίην, ὥσπερ οἱ σοφοὶ, οὐκ ἂν ἄτοπος εἴην,
  εἶτα σοφιζόμενος φαίην αὐτὴν πνεῦμα Βορέου κατὰ τῶν πλησίον
  πετρῶν σὺν φαρμακείᾳ παίζουσαν ὦσαι, καὶ οὕτω δὴ τελευτήσασαν
  λεχθῆναι ὑπὸ τοῦ Βορέου ἀναρπαστὸν γεγονέναι.... Ἐγὼ δὲ, ὦ
  Φαῖδρε, ἄλλως μὲν τὰ τοιαῦτα χαρίεντα ἡγοῦμαι, λίαν δὲ δεινοῦ καὶ
  ἐπιπόνου καὶ οὐ πάνυ εὐτυχοῦς ἀνδρὸς, κατ᾽ ἄλλο μὲν οὐδὲν, ὅτι δ᾽
  αὐτῷ ἀνάγκη μετὰ τοῦτο τὸ τῶν Ἱπποκενταύρων εἶδος ἐπανορθοῦσθαι,
  καὶ αὖθις τὸ τῆς Χιμαίρας. Καὶ ἐπιῤῥεῖ δὲ ὄχλος τοιούτων Γοργόνων
  καὶ Πηγάσων, καὶ ἄλλων ἀμηχάνων πλήθη τε καὶ ἀτοπίαι τερατολόγων
  τινῶν φύσεων· αἷς εἴ τις ἀπιστῶν προσβιβᾷ κατὰ τὸ εἰκὸς ἕκαστον,
  ἅτε ἀγροίκῳ τινὶ σοφίᾳ χρώμενος, πολλῆς αὐτῷ σχολῆς δεήσει. Ἐμοὶ
  δὲ πρὸς ταῦτα οὐδαμῶς ἐστι σχολή.... Ὅθεν δὴ χαίρειν ἐάσας ταῦτα,
  πειθόμενος δὲ τῷ νομιζομένῳ περὶ αὐτῶν, ὃ νῦν δὴ ἔλεγον, σκοπῶ οὐ
  ταῦτα ἀλλ᾽ ἐμαυτὸν, etc.

These remarks of Plato are valuable, not simply because they point
out the uselessness of digging for a supposed basis of truth in the
mythes, but because they at the same time suggest the true reason
for mistrusting all such tentatives. The mythes form a class apart,
abundant as well as peculiar: to remove any individual mythe from its
own class into that of history or philosophy, by simple conjecture,
and without any collateral evidence, is of no advantage, unless you
can perform a similar process on the remainder. If the process be
trustworthy, it ought to be applied to all; and _e converso_, if it
be not applicable to all, it is not trustworthy as applied to any one
specially; always assuming no special evidence to be accessible. To
detach any individual mythe from the class to which it belongs, is to
present it in an erroneous point of view; we have no choice except to
admit them as they stand, by putting ourselves approximatively into the
frame of mind of those for whom they were destined and to whom they
appeared worthy of credit.

If Plato thus discountenances all attempts to transform the mythes by
interpretation into history or philosophy, indirectly recognizing the
generic difference between them—we find substantially the same view
pervading the elaborate precepts in his treatise on the Republic. He
there regards the mythes, not as embodying either matter-of-fact or
philosophical principle, but as portions of religious and patriotic
faith, and instruments of ethical tuition. Instead of allowing the
poets to frame them according to the impulses of their own genius,
and with a view to immediate popularity, he directs the legislator to
provide types of his own for the characters of the gods and heroes,
and to suppress all such divine and heroic legends as are not in
harmony with these preëstablished canons. In the Platonic system, the
mythes are not to be matters of history, nor yet of spontaneous or
casual fiction, but of prescribed faith: he supposes that the people
will believe, as a thing of course, what the poets circulate, and he
therefore directs that the latter shall circulate nothing which does
not tend to ennoble and improve the feelings. He conceives the mythes
as stories composed to illustrate the general sentiments of the poets
and the community, respecting the character and attributes of the gods
and heroes, or respecting the social relations, and ethical duties as
well as motives of mankind: hence the obligation upon the legislator to
prescribe beforehand the types of character which shall be illustrated,
and to restrain the poets from following out any opposing fancies.
“Let us neither believe ourselves (he exclaims), nor permit any one to
circulate, that Thêseus son of Poseidôn and Peirithöus son of Zeus, or
any other hero or son of a god, could ever have brought themselves to
commit abductions or other enormities such as are now falsely ascribed
to them. We must compel the poets to say, either that such persons were
not the sons of gods, or that they were not the perpetrators of such
misdeeds.”[991]

  [991] Plato, Repub. iii. 5. p. 391. The perfect ignorance of
  all men respecting the gods, rendered the task of fiction easy
  (Plato, Kritias, p. 107).

Most of the mythes which the youth hear and repeat (according to Plato)
are false, but some of them are true: the great and prominent mythes
which appear in Homer and Hesiod are no less fictions than the rest.
But fiction constitutes one of the indispensable instruments of mental
training as well as truth; only the legislator must take care that
the fiction so employed shall be beneficent and not mischievous.[992]
As the mischievous fictions (he says) take their rise from wrong
preconceptions respecting the character of the gods and heroes, so the
way to correct them is to enforce, by authorized compositions, the
adoption of a more correct standard.[993]

  [992] Plato, Repub. ii. 16. p. 377. Λόγων δὲ διττὸν εἶδος, τὸ
  μὲν ἀληθὲς, ψεῦδος δ᾽ ἕτερον; Ναί. Παιδευτέον δ᾽ ἐν ἀμφοτέροις,
  πρότερον δ᾽ ἐν τοῖς ψεύδεσιν· ... Οὐ μανθάνεις, ὅτι πρῶτον τοῖς
  παιδίοις μύθους λέγομεν· τοῦτο δέ που ὡς τὸ ὅλον εἰπεῖν ψεῦδος,
  ἔνι δὲ καὶ ἀληθῆ.... Πρῶτον ἡμῖν ἐπιστατητέον τοῖς μυθοποιοῖς,
  καὶ ὃν μὲν ἂν καλὸν μῦθον ποιήσωσιν, ἐγκριτέον, ὃν δ᾽ ἂν μὴ,
  ἀποκριτέον ... ὧν δὲ νῦν λέγουσι, τοὺς πολλοὺς ἐκβλητέον ... οὓς
  Ἡσίοδος καὶ Ὅμηρος ἡμῖν ἐλεγέτην, καὶ οἱ ἄλλοι ποιηταί. Οὗτοι
  γάρ που μύθους τοῖς ἀνθρώποις ψευδεῖς συντιθέντες ἔλεγόν τε καὶ
  λέγουσι. Ποίους δὴ, ἦ δ᾽ ὅς, καὶ τί αὐτῶν μεμφόμενος λέγεις;
  Ὅπερ, ἦν δ᾽ ἐγὼ, χρὴ καὶ πρῶτον καὶ μάλιστα μέμφεσθαι, ἄλλως τε
  καὶ ἐάν τις μὴ καλῶς ψεύδηται. Τί τοῦτο; Ὅταν τις εἰκάζῃ κακῶς
  τῷ λόγῳ περὶ θεῶν τε καὶ ἡρώων, οἷοί εἰσιν, ὥσπερ γραφεὺς μηδὲν
  ἐοικότα γράφων οἷς ἂν ὅμοια βουληθῇ γράψαι.

  The same train of thought, and the precepts founded upon it, are
  followed up through chaps. 17, 18, and 19; compare De Legg. xii.
  p. 941.

  Instead of recognizing the popular or dramatic theology as
  something distinct from the civil (as Varro did), Plato
  suppresses the former as a separate department and merges it in
  the latter.

  [993] Plato, Repub. ii. c. 21. p. 382. Τὸ ἐν τοῖς λόγοις ψεῦδος
  πότε καὶ τί χρήσιμον, ὥστε μὴ ἄξιον εἶναι μίσους; Ἆρ᾽ οὐ πρός
  τε τοὺς πολεμίους καὶ τῶν καλουμένων φίλων, ὅταν διὰ μανίαν ἤ
  τινα ἄνοιαν κακόν τι ἐπιχειρῶσι πράττειν, τότε ἀποτροπῆς ἕνεκα
  ὡς φάρμακον χρήσιμον γίγνεται; ~Καὶ ἐν αἷς νῦν δὴ ἐλέγομεν ταῖς
  μυθολογίαις, διὰ τὸ μὴ εἰδέναι ὅπῃ τἀληθὲς ἔχει περὶ τῶν παλαιῶν,
  ἀφομοιοῦντες τῷ ἀληθεῖ τὸ ψεῦδος~, ὅτι μάλιστα, οὕτω χρήσιμον
  ποιοῦμεν;

The comments which Plato has delivered with so much force in his
Republic, and the enactments which he deduces from them, are in the
main an expansion of that sentiment of condemnation, which he shared
with so many other philosophers, towards a large portion of the Homeric
and Hesiodic stories.[994] But the manner in which he has set forth
this opinion, unfolds to us more clearly the real character of the
mythical narratives. They are creations of the productive minds in the
community, deduced from the supposed attributes of the gods and heroes:
so Plato views them, and in such character he proposes to amend them.
The legislator would cause to be prepared a better and truer picture of
the foretime, because he would start from truer (that is to say, more
creditable) conceptions of the gods and heroes. For Plato rejects the
mythes respecting Zeus and Hêrê, or Thêseus and Peirithöus, not from
any want of evidence, but because they are unworthy of gods and heroes:
he proposes to call forth new mythes, which, though he admits them at
the outset to be fiction, he knows will soon be received as true, and
supply more valuable lessons of conduct.

  [994] The censure which Xenophanês pronounced upon the Homeric
  legends has already been noticed: Herakleitus (Diogen. Laërt. ix.
  1) and Metrodôrus, the companion and follower of Epicurus, were
  not less profuse in their invectives, ἐν γράμμασι τοσούτοις τῷ
  ποιητῇ λελοιδόρηται (Plutarch, Non posse suaviter vivi secundum
  Epicurum, p. 1086). He even advised persons not to be ashamed
  to confess their utter ignorance of Homer, to the extent of not
  knowing whether Hectôr was a Greek or a Trojan (Plut. _ib._ p.
  1094).

We may consider, then, that Plato disapproves of the attempt to
identify the old mythes either with exaggerated history or with
disguised philosophy. He shares in the current faith, without any
suspicion or criticism, as to Orpheus, Palamêdês, Dædalus, Amphiôn,
Thêseus, Achilles, Cheirôn, and other mythical personages;[995] but
what chiefly fills his mind is, the inherited sentiment of deep
reverence for these superhuman characters and for the age to which they
belonged,—a sentiment sufficiently strong to render him not only an
unbeliever in such legends as conflict with it, but also a deliberate
creator of new legends for the purpose of expanding and gratifying it.
The more we examine this sentiment, both in the mind of Plato as well
as in that of the Greeks generally, the more shall we be convinced
that it formed essentially and inseparably a portion of Hellenic
religious faith. The mythe both presupposes, and springs out of, a
settled basis, and a strong expansive force of religious, social, and
patriotic feeling, operating upon a past which is little better than
a blank as to positive knowledge. It resembles history, in so far as
its form is narrative; it resembles philosophy, in so far as it is
occasionally illustrative; but in its essence and substance, in the
mental tendencies by which it is created as well as in those by which
it is judged and upheld, it is a popularized expression of the divine
and heroic faith of the people.

  [995] Plato, Republic iii. 4-5. p. 391; De Legg. iii. 1. p. 677.

Grecian antiquity cannot be at all understood except in connection
with Grecian religion. It begins with gods and it ends with historical
men, the former being recognized not simply as gods, but as primitive
ancestors, and connected with the latter by a long mythical genealogy,
partly heroic and partly human. Now the whole value of such genealogies
arises from their being taken entire; the god or hero at the top is
in point of fact the most important member of the whole;[996] for
the length and continuity of the series arises from anxiety on the
part of historical men to join themselves by a thread of descent
with the being whom they worshipped in their gentile sacrifices.
Without the ancestorial god, the whole pedigree would have become not
only acephalous, but worthless and uninteresting. The pride of the
Herakleids, Asklêpiads, Æakids, Nêleids, Dædalids, etc. was attached
to the primitive eponymous hero and to the god from whom they sprung,
not to the line of names, generally long and barren, through which the
divine or heroic dignity gradually dwindled down into common manhood.
Indeed, the length of the genealogy (as I have before remarked) was an
evidence of the humility of the historical man, which led him to place
himself at a respectful distance from the gods or heroes; for Hekatæus
of Milêtus, who ranked himself as the fifteenth descendant of a god,
might perhaps have accounted it an overweening impiety in any living
man to claim a god for his immediate father.

  [996] For a description of similar tendencies in the Asiatic
  religions, see Mövers, Die Phönizier, ch. v. p. 153 (Bonn, 1841):
  he points out the same phænomena as in the Greek,—coalescence
  between the ideas of ancestry and worship,—confusion between
  gods and men in the past,—increasing tendency to Euêmerize (pp.
  156-157).

The whole chronology of Greece, anterior to 776 B. C., consists of
calculations founded upon these mythical genealogies, especially upon
that of the Spartan kings and their descent from Hêraklês,—thirty
years being commonly taken as the equivalent of a generation, or
about three generations to a century. This process of computation
was altogether illusory, as applying historical and chronological
conditions to a case on which they had no bearing. Though the domain
of history was seemingly enlarged, the religious element was tacitly
set aside: when the heroes and gods were chronologized, they became
insensibly approximated to the limits of humanity, and the process
indirectly gave encouragement to the theory of Euêmerus. Personages
originally legendary and poetical were erected into definite landmarks
for measuring the duration of the foretime, thus gaining in respect to
historical distinctness, but not without loss on the score of religious
association. Both Euêmerus and the subsequent Christian writers, who
denied the original and inherent divinity of the pagan gods, had a
great advantage in carrying their chronological researches strictly and
consistently upwards—for all chronology fails as soon as we suppose a
race superior to common humanity.

Moreover, it is to be remarked that the pedigree of the Spartan kings,
which Apollodôrus and Eratosthenês selected as the basis of their
estimate of time, is nowise superior in credibility and trustworthiness
to the thousand other gentile and family pedigrees with which Greece
abounded; it is rather indeed to be numbered among the most incredible
of all, seeing that Hêraklês as a progenitor is placed at the head of
perhaps more pedigrees than any other Grecian god or hero.[997] The
descent of the Spartan king Leonidas from Hêraklês rests upon no better
evidence than that of Aristotle or Hippocratês from Asklêpius,[998]—of
Evagoras or Thucydidês from Æakus,—of Socratês from Dædalus,—of the
Spartan heraldic family from Talthybius,—of the prophetic Iamid family
in Elis from Iamus,—of the root-gatherers in Pêlion from Cheirôn,—and
of Hekatæus and his gens from some god in the sixteenth ascending
line of the series. There is little exaggeration in saying, indeed,
that no permanent combination of men in Greece, religious, social, or
professional, was without a similar pedigree; all arising out of the
same exigences of the feelings and imagination, to personify as well
as to sanctify the bond of union among the members. Every one of these
_gentes_ began with a religious and ended with an historical person.
At some point or other in the upward series, entities of history were
exchanged for entities of religion; but where that point is to be
found we are unable to say, nor had the wisest of the ancient Greeks
any means of determining. Thus much, however, we know, that the series
taken as a whole, though dear and precious to the believing Greek,
possesses no value as chronological evidence to the historian.

  [997] According to that which Aristotle seems to recognize
  (Histor. Animal. vii. 6), Hêraklês was father of seventy-two
  sons, but of only one daughter—he was essentially ἀῤῥενόγονος,
  illustrating one of the physical peculiarities noticed by
  Aristotle. Euripidês, however, mentions daughters of Hêraklês in
  the plural number (Euripid. Herakleid. 45).

  [998] Hippocratês was twentieth in descent from Hêraklês,
  and nineteenth from Asklêpius (Vita Hippocr. by Soranus, ap.
  Westermann, Scriptor. Biographic. viii. 1); about Aristotle,
  see Diogen. Laërt. v. 1. Xenophôn, the physician of the emperor
  Claudius, was also an Asklêpiad (Tacit. Ann. xii. 61).

  In Rhodes, the neighboring island to Kôs, was the gens Ἁλιάδαι,
  or sons of Hêlios, specially distinguished from the Ἁλιασταὶ of
  mere associated worshippers of Hêlios, τὸ κοινὸν τῶν Ἁλιαδῶν καὶ
  τῶν Ἁλιαστῶν (see the Inscription in Boeckh’s Collection, No.
  2525, with Boeckh’s comment).

When Hekatæus visited Thêbes in Egypt, he mentioned to the Egyptian
priests, doubtless with a feeling of satisfaction and pride, the
imposing pedigree of the gens to which he belonged,—with fifteen
ancestors in ascending line, and a god as the initial progenitor. But
he found himself immeasurably overdone by the priests “who genealogized
against him.”[999] They showed to him three hundred and forty-one
wooden colossal statues, representing the succession of chief priests
in the temple in uninterrupted series from father to son, through a
space of 11,300 years. Prior to the commencement of this long period
(they said), the gods dwelling along with men, had exercised sway in
Egypt; but they repudiated altogether the idea of men begotten by gods
or of heroes.[1000]

  [999] Herodot. ii. 144. Ἑκαταίῳ δὲ γενεηλογήσαντι ἑωϋτὸν,
  καὶ ἀναδήσαντι ἐς ἑκκαιδέκατον θεὸν, ἀντεγενεηλόγησαν ἐπὶ τῇ
  ἀριθμήσει, οὐ δεκόμενοι παρ᾽ αὐτοῦ, ἀπὸ θεοῦ γένεσθαι ἄνθρωπον·
  ἀντεγενεηλόγησαν δὲ ὧδε, etc.

  [1000] Herod. ii. 143-145. Καὶ ταῦτα Αἰγύπτιοι ἀτρεκέως φασὶν
  ἐπίστασθαι, αἰεί τε λογιζόμενοι καὶ αἰεὶ ἀπογραφόμενοι τὰ ἔτεα.

But these counter-genealogies, are, in respect to trustworthiness and
evidence, on the same footing. Each represents partly the religious
faith, partly the retrospective imagination, of the persons from whom
it emanated; in each, the lower members of the series (to what extent
we cannot tell) are real, the upper members fabulous; but in each also
the series derived all its interest and all its imposing effect from
being conceived unbroken and entire. Herodotus is much perplexed by the
capital discrepancy between the Grecian and Egyptian chronologies, and
vainly employs his ingenuity in reconciling them. There is no standard
of objective evidence by which either the one or the other of them can
be tried: each has its own subjective value, in conjunction with the
faith and feelings of Egyptians and Greeks, and each presupposes in the
believer certain mental prepossessions which are not to be found beyond
its own local limits. Nor is the greater or less extent of duration at
all important, when we once pass the limits of evidence and verifiable
reality. One century of recorded time, adequately studded with
authentic and orderly events, presents a greater mass and a greater
difficulty of transition to the imagination than a hundred centuries of
barren genealogy. Herodotus, in discussing the age of Homer and Hesiod,
treats an anterior point of 400 years as if it were only yesterday;
the reign of Henry VI. is separated from us by an equal interval, and
the reader will not require to be reminded how long that interval now
appears.

The mythical age was peopled with a mingled aggregate of gods, heroes,
and men, so confounded together that it was often impossible to
distinguish to which class any individual name belonged. In regard
to the Thracian god Zalmoxis, the Hellespontic Greeks interpreted
his character and attributes according to the scheme of Euêmerism.
They affirmed that he had been a man, the slave of the philosopher
Pythagoras at Samos, and that he had by abilities and artifice
established a religious ascendency over the minds of the Thracians,
and obtained from them divine honors. Herodotus cannot bring himself
to believe this story, but he frankly avows his inability to determine
whether Zalmoxis was a god or a man,[1001] nor can he extricate
himself from a similar embarrassment in respect to Dionysus and Pan.
Amidst the confusion of the Homeric fight, the goddess Athênê confers
upon Diomêdês the miraculous favor of dispelling the mist from his
eyes, so as to enable him to discriminate gods from men; and nothing
less than a similar miracle could enable a critical reader of the
mythical narratives to draw an ascertained boundary-line between
the two.[1002] But the original hearers of the mythes felt neither
surprise nor displeasure from this confusion of the divine with the
human individual. They looked at the past with a film of faith over
their eyes—neither knowing the value, nor desiring the attainment, of
an unclouded vision. The intimate companionship, and the occasional
mistake of identity between gods and men, were in full harmony with
their reverential retrospect. And we, accordingly, see the poet Ovid
in his Fasti, when he undertakes the task of unfolding the legendary
antiquities of early Rome, re-acquiring, by the inspiration of Juno,
the power of seeing gods and men in immediate vicinity and conjunct
action, such as it existed before the development of the critical and
historical sense.[1003]

  [1001] Herod. iv. 94-96. After having related the Euemeristic
  version given by the Hellespontic Greeks, he concludes with his
  characteristic frankness and simplicity—Ἐγὼ δὲ, περὶ μὲν τούτου
  καὶ τοῦ καταγαίου οἰκήματος, οὔτε ἀπιστέω, οὔτε ὦν πιστεύω τι
  λίην, δοκέω δὲ πολλοῖσι ἔτεσι πρότερον τὸν Ζάλμοξιν τοῦτον
  γενέσθαι Πυθαγόρεω. Εἴτε δὲ ἐγένετό τις Ζάλμοξις ἄνθρωπος, εἴτ᾽
  ἐστὶ δαίμων τις Γέτησι οὗτος ἐπιχώριος, χαιρέτω. So Plutarch
  (Numa c. 19) will not undertake to determine whether Janus was a
  god or a king εἴτε δαίμων, εἴτε βασιλεὺς γενόμενος, etc.

  Herakleitus the philosopher said that men were θεοὶ θνητοὶ, and
  the gods were ἄνθρωποι ἀθάνατοι (Lucian, Vitar. Auctio. c. 13.
  vol. i. p. 303, Tauch.: compare the same author, Dialog. Mortuor.
  iii. vol. i. p. 182, ed. Tauchn).

  [1002] Iliad, v. 127:—

    Ἀχλὺν δ᾽ αὖ τοι ἀπ᾽ ὀφθαλμῶν ἕλον, ἣ πρὶν ἐπῆεν,
    Ὄφρ᾽ εὖ γιγνώσκῃς ἠμὲν θεὸν, ἠδὲ καὶ ἄνδρα.

  Of this undistinguishable confusion between gods and men,
  striking illustrations are to be found both in the third book of
  Cicero de Naturâ Deorum (16-21), and in the long disquisition of
  Strabo (x. pp. 467-474) respecting the Kabeiri, the Korybantes,
  the Dactyls of Ida; the more so, as he cites the statements of
  Pherekydês, Akusilaus, Dêmêtrius of Skêpsis, and others. Under
  the Roman empire, the lands in Greece belonging to the immortal
  gods were exempted from tribute. The Roman tax-collectors
  refused to recognize as immortal gods any persons who had once
  been men; but this rule could not be clearly applied (Cicero,
  Nat. Deor. iii. 20). See the remarks of Pausanias (ii. 26, 7)
  about Asklêpius: Galen, too, is doubtful about Asklêpius and
  Dionysus—Ἀσκληπιός γέ τοι καὶ Διόνυσος, εἶτ᾽ ἄνθρωποι πρότερον
  ἤστην, εἴτε καὶ ἀρχῆθεν θεοί (Galen in Protreptic. 9. tom. i. p.
  22, ed. Kühn). Xenophôn (De Venat. c. i.) considers Cheirôn as
  the brother of Zeus.

  The ridicule of Lucian (Deorum Concilium, t. iii. p 527-538,
  Hems.) brings out still more forcibly the confusion here
  indicated.

  [1003] Ovid, Fasti, vi. 6-20:—

    “Fas mihi præcipue vultus vidisse Deorum,
      Vel quia sum vates, vel quia sacra cano ...
       ... Ecce Deas vidi....
    Horrueram, tacitoque animum pallore fatebar:
      Cum Dea, quos fecit, sustulit ipsa metus.
    Namque ait—O vates, Romani conditor anni,
      Ause per exiguos magna referre modos;
    Jus tibi fecisti numen cœleste videndi,
      Cum placuit numeris condere festa tuis.”

To resume, in brief, what has been laid down in this and the preceding
chapters respecting the Grecian mythes:—

1. They are a special product of the imagination and feelings,
radically distinct both from history and philosophy: they cannot be
broken down and decomposed into the one, nor allegorized into the
other. There are indeed some particular and even assignable mythes,
which raise intrinsic presumption of an allegorizing tendency; and
there are doubtless some others, though not specially assignable, which
contain portions of matter of fact, or names of real persons, embodied
in them. But such matter of fact cannot be verified by any intrinsic
mark, nor we are entitled to presume its existence in any given case
unless some collateral evidence can be produced.

2. We are not warranted in applying to the mythical world the rules
either of historical credibility or chronological sequence. Its
personages are gods, heroes, and men, in constant juxtaposition and
reciprocal sympathy; men, too, of whom we know a large proportion to
be fictitious, and of whom we can never ascertain how many may have
been real. No series of such personages can serve as materials for
chronological calculation.

3. The mythes were originally produced in an age which had no records,
no philosophy, no criticism, no canon of belief, and scarcely any
tincture either of astronomy or geography—but which, on the other hand,
was full of religious faith, distinguished for quick and susceptible
imagination, seeing personal agents where we look only for objects and
connecting laws;—an age, moreover, eager for new narrative, accepting
with the unconscious impressibility of children (the question of truth
or falsehood being never formally raised) all which ran in harmony
with its pre-existing feelings, and penetrable by inspired prophets
and poets in the same proportion that it was indifferent to positive
evidence. To such hearers did the primitive poet or story-teller
address himself: it was the glory of his productive genius to provide
suitable narrative expression for the faith and emotions which he
shared in common with them, and the rich stock of Grecian mythes
attests how admirably he performed his task. As the gods and the heroes
formed the conspicuous object of national reverence, so the mythes were
partly divine, partly heroic, partly both in one.[1004] The adventures
of Achilles, Helen, and Diomêdês, of Œdipus and Adrastus, of Meleager
and Athæa, of Jasôn and the Argô, were recounted by the same tongues,
and accepted with the same unsuspecting confidence, as those of Apollo
and Artemis, of Arês and Aphroditê, of Poseidôn and Hêraklês.

  [1004] The fourth Eclogue of Virgil, under the form of a
  prophecy, gives a faithful picture of the heroic and divine past,
  to which the legends of Troy and the Argonauts belonged:—

    “Ille Deûm vitam accipiet, Divisque videbit
    Permixtos heroas,” etc.

    “Alter erit tum Tiphys et altera quæ vehat Argo
    Delectos heroas: erunt etiam altera bella,
    Atque iterum ad Trojam magnus mittetur Achilles.”

4. The time however came, when this plausibility ceased to be complete.
The Grecian mind made an important advance, socially, ethically, and
intellectually. Philosophy and history were constituted, prose writing
and chronological records became familiar; a canon of belief more
or less critical came to be tacitly recognized. Moreover, superior
men profited more largely by the stimulus, and contracted habits of
judging different from the vulgar: the god Elenchus[1005] (to use a
personification of Menander) the giver and prover of truth, descended
into their minds. Into the new intellectual medium, thus altered in its
elements, and no longer uniform in its quality, the mythes descended by
inheritance; but they were found, to a certain extent, out of harmony
even with the feelings of the people, and altogether dissonant with
those of instructed men. But the most superior Greek was still a Greek,
and cherished the common reverential sentiment towards the foretime of
his country. Though he could neither believe nor respect the mythes as
they stood, he was under an imperious mental necessity to transform
them into a state worthy of his belief and respect. Whilst the literal
mythe still continued to float among the poets and the people, critical
men interpreted, altered, decomposed, and added, until they found
something which satisfied their minds as a supposed real basis. They
manufactured some dogmas of supposed original philosophy, and a long
series of fancied history and chronology, retaining the mythical names
and generations even when they were obliged to discard or recast
the mythical events. The interpreted mythe was thus promoted into a
reality, while the literal mythe was degraded into a fiction.[1006]

  [1005] Lucian, Pseudol. c. 4. Παρακλητέος ἡμῖν τῶν Μενάνδρου
  προλόγων εἷς, ὁ Ἔλεγχος, φίλος ἀληθείᾳ καὶ παῤῥησίᾳ θεὸς, οὐχ
  ὁ ἀσημότατος τῶν ἐπὶ τὴν σκήνην ἀναβαινόντων. (See Meineke ad
  Menandr. p. 284.)

  [1006] The following passage from Dr. Ferguson’s Essay on Civil
  Society (part ii. sect. i. p. 126) bears well on the subject
  before us:—

  “If conjectures and opinions formed at a distance have not a
  sufficient authority in the history of mankind, the domestic
  antiquities of every nation must for this very reason be received
  with caution. They are, for the most part, the mere conjectures
  or the fictions of subsequent ages; and even where at first they
  contained some resemblance of truth, they still vary with the
  imagination of those by whom they were transmitted, and in every
  generation receive a different form. They are made to bear the
  stamp of the times through which they have passed in the form of
  tradition, not of the ages to which their pretended descriptions
  relate.... When traditionary fables are rehearsed by the vulgar,
  they bear the marks of a national character, and though mixed
  with absurdities, often raise the imagination and move the heart:
  when made the materials of poetry, and adorned by the skill and
  the eloquence of an ardent and superior mind, they instruct the
  understanding as well as engage the passions. It is only in the
  management of mere antiquaries, or stript of the ornaments which
  the laws of history forbid them to wear, that _they become unfit
  even to amuse the fancy or to serve any purpose whatever_.

  “It were absurd to quote the fable of the Iliad or the Odyssey,
  the legend of Hercules, Theseus, and Œdipus, as authorities
  in matters of fact relating to the history of mankind; but
  they may, with great justice, be cited to ascertain what were
  the conceptions and sentiments of the age in which they were
  composed, or to characterize the genius of that people with whose
  imaginations they were blended, and by whom they were fondly
  rehearsed and admired. In this manner, fiction may be admitted
  to vouch for the genius of nations, while history has nothing to
  offer worthy of credit.”

  To the same purpose, M. Paulin Paris (in his Lettre à M. H. de
  Monmerqué, prefixed to the Roman de Berte aux Grans Piés, Paris,
  1836), respecting the “romans” of the Middle Ages: “Pour bien
  connaître l’histoire du moyen âge, non pas celle des faits, mais
  celle des mœurs qui rendent les faits vraisemblables, il faut
  l’avoir étudiée dans les romans, et voilà pourquoi l’Histoire de
  France n’est pas encore faite.” (p. xxi.)

The habit of distinguishing the interpreted from the literal mythe
has passed from the literary men of antiquity to those of the modern
world, who have for the most part construed the divine mythes as
allegorized philosophy, and the heroic mythes as exaggerated, adorned,
and over-colored history. The early ages of Greece have thus been
peopled with quasi-historical persons and quasi-historical events,
all extracted from the mythes after making certain allowances for
poetical ornament. But we must not treat this extracted product as
if it were the original substance; we cannot properly understand it
except by viewing it in connection with the literal mythes out of
which it was obtained, in their primitive age and appropriate medium,
before the superior minds had yet outgrown the common faith in an
all-personified Nature, and learned to restrict the divine free-agency
by the supposition of invariable physical laws. It is in this point
of view that the mythes are important for any one who would correctly
appreciate the general tone of Grecian thought and feeling; for they
were the universal mental stock of the Hellenic world—common to men and
women, rich and poor, instructed and ignorant; they were in every one’s
memory and in every one’s mouth,[1007] while science and history were
confined to comparatively few. We know from Thucydidês how erroneously
and carelessly the Athenian public of his day retained the history
of Peisistratus, only one century past;[1008] but the adventures of
the gods and heroes, the numberless explanatory legends attached to
visible objects and periodical ceremonies, were the theme of general
talk, and any man unacquainted with them would have found himself
partially excluded from the sympathy of his neighbors. The theatrical
representations, exhibited to the entire city population, and listened
to with enthusiastic interest, both presupposed and perpetuated
acquaintance with the great lines of heroic fable: indeed, in later
times even the pantomimic dancers embraced in their representations
the whole field of mythical incident, and their immense success proves
at once how popular and how well known such subjects were. The names
and attributes of the heroes were incessantly alluded to in the way
of illustration, to point out a consoling, admonitory, or repressive
moral: the simple mention of any of them sufficed to call up in every
one’s mind the principal events of his life, and the poet or rhapsode
could thus calculate on touching chords not less familiar than
susceptible.[1009]

  [1007] A curious evidence of the undiminished popularity of
  the Grecian mythes to the exclusion even of recent history, is
  preserved by Vopiscus at the beginning of his Life of Aurelian.

  The præfect of the city of Rome, Junius Tiberianus, took Vopiscus
  into his carriage on the festival-day of the Hilaria; he was
  connected by the ties of relationship with Aurelian, who had
  died about a generation before—and as the carriage passed by the
  splendid Temple of the Sun, which Aurelian had consecrated, he
  asked Vopiscus, what author had written the life of that emperor?
  To which Vopiscus replied, that he had read some Greek works
  which touched upon Aurelian, but nothing in Latin. Whereat the
  venerable præfect was profoundly grieved: “Dolorem gemitûs sui
  vir sanctus per hæc verba profudit: Ergo _Thersitem, Sinonem,
  cæteraque illa prodigia vetustatis, et nos bene scimus, et
  posteri frequentabunt_: divum Aurelianum, clarissimum principem,
  severissimum Imperatorem, per quem totus Romano nomini orbis est
  restitutus, posteri nescient? Deus avertat hanc amentiam! Et
  tamen, si bene memini, ephemeridas illius viri scriptas habemus,”
  etc. (Historiæ August. Scriptt. p. 209, ed. Salmas.)

  This impressive remonstrance produced the Life of Aurelian by
  Vopiscus. The materials seem to have been ample and authentic; it
  is to be regretted that they did not fall into the hands of an
  author qualified to turn them to better account.

  [1008] Thucyd. vi. 56.

  [1009] Pausan. i. 3, 3. Λέγεται μὲν δὴ καὶ ἄλλα οὐκ ἀληθῆ παρὰ
  τοῖς πολλοῖς, οἷα ἱστορίας ἀνηκόοις οὖσι, καὶ ὅποσα ἤκουον εὐθὺς
  ἐκ παιδῶν ἔν τε χόροις καὶ τραγῳδίαις πιστὰ ἡγουμένοις, etc.
  The treatise of Lucian, De Saltatione, is a curious proof how
  much these mythes were in every one’s memory, and how large the
  range of knowledge of them was which a good dancer possessed (see
  particularly c. 76-79. t. ii. p. 308-310, Hemst).

  Antiphanês ap. Athenæ. vi. p. 223:—

                  Μακάριόν ἐστιν ἡ τραγῳδία
    ποίημα κατὰ πάντ᾽, εἴ γε πρῶτον οἱ λόγοι
    ὑπὸ τῶν θεατῶν εἰσιν ἐγνωρίσμενοι
    πρὶν καί τιν᾽ εἰπεῖν· ὡς ὑπομνῆσαι μόνον
    δεῖ τὸν ποιητήν. Οἰδίπουν γὰρ ἂν γε φῶ,
    τὰ δ᾽ ἄλλα πάντ᾽ ἴσασιν· ὁ πατὴρ Λάϊος,
    μήτηρ Ἰοκάστη, θυγατέρες, παῖδες τίνες·
    τί πείσεθ᾽ οὗτος, τί πεποίηκεν. Ἂν πάλιν
    εἴπῃ τις Ἀλκμαίωνα, καὶ τὰ παιδία
    πάντ᾽ εὐθὺς εἴρηχ᾽, ὅτι μανεὶς ἀπέκτονε
    τὴν μήτερ᾽· ἀγανακτῶν δ᾽ Ἄδραστος εὐθέως
    ἥξει, πάλιν δ᾽ ἄπεισιν, etc.

  The first pages of the eleventh Oration of Dio Chrysostom contain
  some striking passages both as to the universal acquaintance
  with the mythes, and as to their extreme popularity (Or. xi. p.
  307-312, Reisk). See also the commencement of Heraklidês, De
  Allegoriâ Homericâ (ap. Scriptt. Myth. ed. Gale, p. 408), about
  the familiarity with Homer.

  The Lydê of the poet Antimachus was composed for his own
  consolation under sorrow, by enumerating the ἡρωϊκὰς συμφοράς
  (Plutarch, Consolat. ad Apollôn. c. 9. p. 106: compare Æschines
  cont. Ktesiph. c. 48): a sepulchral inscription in Thêra, on the
  untimely death of Admêtus, a youth of the heroic gens Ægidae,
  makes a touching allusion to his ancestors Pêleus and Pherês
  (Boeckh, C. I. t. ii. p. 1087).

  A curious passage of Aristotle is preserved by Dêmêtrius
  Phalereus (Περὶ Ἑρμηνείας, c. 144),—Ὅσῳ γὰρ αὐτίτης καὶ
  μονώτης εἰμὶ, φιλομυθότερος γέγονα (compare the passage in the
  Nikomachean Ethics, i. 9, μονώτης καὶ ἄτεκνος). Stahr refers this
  to a letter of Aristotle written in his old age, the mythes being
  the consolation of his solitude (Aristotelia, i. p. 201).

  For the employment of the mythical names and incidents as
  topics of pleasing and familiar comparison, see Menander, Περὶ
  Ἐπιδεικτῖκ. § iv. capp. 9 and 11, ap. Walz. Coll. Rhett. t. ix.
  pp. 283-294. The degree in which they passed into the ordinary
  songs of women is illustrated by a touching epigram contained
  among the Chian Inscriptions published in Boeckh’s Collection
  (No. 2236):—

    Βιττὼ καὶ Φαινὶς, φίλη ἡμέρη (?), αἱ συνέριθοι,
      Αἱ πενιχραὶ, γραῖαι, τῆδ᾽ ἐκλίθημεν ὁμοῦ.
    Ἀμφότεραι Κώαι, πρῶται γένος—ὦ γλυκὺς ὄρθρος,
      Πρὸς λύχνον ᾧ μύθους ᾔδομεν ἡμιθέων.

  These two poor women were not afraid to boast of their family
  descent. They probably belonged to some noble gens which traced
  its origin to a god or a hero. About the songs of women, see also
  Agathias, i. 7. p. 29, ed. Bonn.

  In the family of the wealthy Athenian Dêmocratês was a legend,
  that his primitive ancestor (son of Zeus by the daughter of
  the Archêgetês of the dême Aixôneis, to which he belonged) had
  received Hêraklês at his table: this legend was so rife that the
  old women sung it,—ἅπερ αἱ γραῖαι ᾄδουσι (Plato, Lysis, p. 205).
  Compare also a legend of the dême Ἀναγυροῦς, mentioned in Suidas
  _ad voc._

  “Who is this virgin?” asks Orestês from Pyladês in the Iphigeneia
  in Tauris of Euripidês (662), respecting his sister Iphigeneia,
  whom he does not know as priestess of Artemis in a foreign land:—

    Τίς ἐστιν ἡ νεᾶνις; ὡς ~Ἑλληνικῶς~
    Ἀνήρεθ᾽ ἡμᾶς τούς τ᾽ ἐν Ἰλίῳ πόνους
    Νόστον τ᾽ Ἀχαιῶν τόν τ᾽ ἐν οἰωνοῖς σοφὸν
    Κάλχαντ᾽, Ἀχιλλέως τ᾽ οὔνομ᾽, etc.
    ... ἐστὶν ἡ ξένη γένος
    Ἐκεῖθεν. Ἀργεία τις, etc.

A similar effect was produced by the multiplied religious festivals
and processions, as well as by the oracles and prophecies which
circulated in every city. The annual departure of the Theôric ship
from Athens to the sacred island of Dêlos, kept alive, in the minds
of Athenians generally, the legend of Thêseus and his adventurous
enterprise in Krete;[1010] and in like manner most of the other
public rites and ceremonies were of a commemorative character,
deduced from some mythical person or incident familiarly known to
natives, and forming to strangers a portion of the curiosities of
the place.[1011] During the period of Grecian subjection under the
Romans, these curiosities, together with their works of art and
their legends, were especially clung to as a set-off against present
degradation. The Thêban citizen who found himself restrained from the
liberty enjoyed by all other Greeks, of consulting Amphiaräus as a
prophet, though the sanctuary and chapel of the hero stood in his own
city—could not be satisfied without a knowledge of the story which
explained the origin of such prohibition,[1012] and which conducted
him back to the originally hostile relations between Amphiaräus and
Thêbes. Nor can we suppose among the citizens of Sikyôn anything less
than a perfect and reverential conception of the legend of Thêbes,
when we read the account given by Herodotus of the conduct of the
despot Kleisthenês in regard to Adrastus and Melanippus.[1013] The
Trœzenian youths and maidens,[1014] who universally, when on the eve
of marriage, consecrated an offering of their hair at the Herôon of
Hippolytus, maintained a lively recollection of the legend of that
unhappy recusant whom Aphroditê had so cruelly punished. Abundant
relics preserved in many Grecian cities and temples, served both as
mementos and attestations of other legendary events; and the tombs
of the heroes counted among the most powerful stimulants of mythical
reminiscence. The sceptre of Pelops and Agamemnôn, still preserved in
the days of Pausanias at Chæroneia in Bœôtia, was the work of the god
Hêphæstos. While many other alleged productions of the same divine hand
were preserved in different cities of Greece, this is the only one
which Pausanias himself believed to be genuine: it had been carried
by Elektra, daughter of Agamemnôn to Phôkis, and received divine
honors from the citizens of Chæroneia.[1015] The spears of Mêrionês
and Odysseus were treasured up at Engyium in Sicily, that of Achilles
at Phasêlis; the sword of Memnôn adorned the temple of Asklêpius at
Nicomêdia; and Pausanias, with unsuspecting confidence, adduces the two
latter as proofs that the arms of the heroes were made of brass.[1016]
The hide of the Kalydônian boar was guarded and shown by the Tegeates
as a precious possession; the shield of Euphorbus was in like manner
suspended in the temple of Branchidæ near Milêtus, as well as in the
temple of Hêrê in Argos. Visible relics of Epeius and Philoktêtês
were not wanting, while Strabo raises his voice with indignation
against the numerous Palladia which were shown in different cities,
each pretending to be the genuine image from Troy.[1017] It would be
impossible to specify the number of chapels, sanctuaries, solemnities,
foundations of one sort or another, said to have been first commenced
by heroic or mythical personages,—by Hêraklês, Jasôn, Mêdea, Alkmæôn,
Diomêdês, Odysseus, Danaus, and his daughters,[1018] etc. Perhaps in
some of these cases particular critics might raise objections, but the
great bulk of the people entertained a firm and undoubted belief in the
current legend.

  [1010] Plato, Phædo, c. 2.

  [1011] The Philopseudes of Lucian (t. iii. p. 31, Hemst. cap.
  2, 3, 4) shows not only the pride which the general public of
  Athens and Thêbes took in their old mythes (Triptolemus, Boreas,
  and Oreithyia, the Sparti, etc.), but the way in which they
  treated every man who called the stories in question as a fool
  or as an atheist. He remarks, that if the guides who showed the
  antiquities had been restrained to tell nothing but what was
  true, they would have died of hunger; for the visiting strangers
  would not care to hear plain truth, even if they could have got
  it for nothing (μηδὲ ἀμισθὶ τῶν ξένων ἀληθὲς ἀκούειν ἐθελησάντων).

  [1012] Herodot. viii. 134.

  [1013] Herodot. v. 67.

  [1014] Euripid. Hippolyt. 1424; Pausan. ii. 32, 1; Lucian, De Deâ
  Syriâ, c. 60. vol. iv. p. 287, Tauch.

  It is curious to see in the account of Pausanias how all the
  petty peculiarities of the objects around became connected with
  explanatory details growing out of this affecting legend. Compare
  Pausan. i. 22, 2.

  [1015] Pausan. ix. 40, 6.

  [1016] Plutarch, Marcell. c. 20; Pausan. iii. 3, 6.

  [1017] Pausan. viii. 46, 1; Diogen. Laër. viii. 5; Strabo, vi. p.
  263; Appian, Bell. Mithridat. c. 77; Æschyl. Eumen. 380.

  Wachsmuth has collected the numerous citations out of Pausanias
  on this subject (Hellenische Alterthumskunde, part ii. sect. 115.
  p. 111).

  [1018] Herodot. ii. 182; Plutarch, Pyrrh. c. 32; Schol. Apoll.
  Rhod iv. 1217; Diodôr. iv. 56.

If we analyze the intellectual acquisitions of a common Grecian
townsman, from the rude communities of Arcadia or Phôkis even up to the
enlightened Athens, we shall find that, over and above the rules of
art or capacities requisite for his daily wants, it consisted chiefly
of the various mythes connected with his gens, his city, his religious
festivals, and the mysteries in which he might have chosen to initiate
himself, as well as with the works of art and the more striking
natural objects which he might see around him,—the whole set off and
decorated by some knowledge of the epic and dramatic poets. Such was
the intellectual and imaginative reach of an ordinary Greek, considered
apart from the instructed few: it was an aggregate of religion, of
social and patriotic retrospect, and of romantic fancy, blended into
one indivisible faith. And thus the subjective value of the mythes,
looking at them purely as elements of Grecian thought and feeling, will
appear indisputably great, however little there may be of objective
reality, either historical or philosophical, discoverable under them.

Nor must we omit the incalculable importance of the mythes as
stimulants to the imagination of the Grecian artist in sculpture, in
painting, in carving, and in architecture. From the divine and heroic
legends and personages were borrowed those paintings, statues, and
reliefs, which rendered the temples, porticos, and public buildings,
at Athens and elsewhere, objects of surpassing admiration; and such
visible reproduction contributed again to fix the types of the gods and
heroes familiarly and indelibly on the public mind.[1019] The figures
delineated on cups and vases, as well as on the walls of private
houses, were chiefly drawn from the same source—the mythes being the
great storehouse of artistic scenes and composition.

  [1019] Ἡμιθέων ἀρεταῖς, the subjects of the works of Polygnotus
  at Athens (Melanthius ap. Plutarch. Cimôn. c. 4): compare
  Theocrit. xv. 138.

To enlarge on the characteristic excellence of Grecian art would here
be out of place: I regard it only in so far as, having originally drawn
its materials from the mythes, it reacted upon the mythical faith and
imagination—the reaction imparting strength to the former as well
as distinctness to the latter. To one who saw constantly before him
representations of the battles of the Centaurs or the Amazons,[1020]
of the exploits performed by Perseus and Bellerophôn, of the incidents
composing the Trojan war or the Kalydônian boar-hunt—the process of
belief, even in the more fantastic of these conceptions, became easy
in proportion as the conception was familiarized. And if any person
had been slow to believe in the efficacy of the prayers of Æakus,
whereby that devout hero once obtained special relief from Zeus, at a
moment when Greece was perishing with long-continued sterility, his
doubts would probably vanish when, on visiting the Æakeion at Ægina,
there were exhibited to him the statues of the very envoys who had
come on the behalf of the distressed Greeks to solicit that Æakus
would pray for them.[1021] A Grecian temple[1022] was not simply a
place of worship, but the actual dwelling-place of a god, who was
believed to be introduced by the solemn dedicatory ceremony, and whom
the imagination of the people identified in the most intimate manner
with his statue. The presence or removal of the statue was conceived
as identical with that of the being represented,—and while the statue
was solemnly washed, dressed, and tended with all the respectful
solicitude which would have been bestowed upon a real person,[1023]
miraculous tales were often rife respecting the manifestation of real
internal feeling in the wood and the marble. At perilous or critical
moments, the statue was affirmed to have sweated, to have wept, to have
closed its eyes, or brandished the spear in its hands, in token of
sympathy or indignation.[1024] Such legends, springing up usually in
times of suffering and danger, and finding few men bold enough openly
to contradict them, ran in complete harmony with the general mythical
faith, and tended to strengthen it in all its various ramifications.
The renewed activity of the god or hero both brought to mind and
accredited the preëxisting mythes connected with his name. When Boreas,
during the invasion of Greece by Xerxês, and in compliance with the
fervent prayers of the Athenians, had sent forth a providential storm,
to the irreparable damage of the Persian armada,[1025] the sceptical
minority (alluded to by Plato), who doubted the mythe of Boreas and
Oreithyia, and his close connection thus acquired with Erechtheus,
and the Erechtheids generally, must for the time have been reduced to
absolute silence.

  [1020] The Centauromachia and the Amazonomachia are constantly
  associated together in the ancient Grecian reliefs (see
  the Expedition Scientifique de Morée, t. ii. p. 16, in the
  explanation of the temple of Apollo Epikureius at Phigaleia).

  [1021] Pausan. ii. 29, 6.

  [1022] Ernst Curtius, Die Akropolis von Athen, Berlin, 1844, p.
  18. Arnobius adv. Gentes, vi. p. 203, ed. Elmenhorst.

  [1023] See the case of the Æginêtans lending the Æakids for a
  time to the Thebans (Herodot. v. 80), who soon, however, returned
  them: likewise sending the Æakids to the battle of Salamis (viii.
  64-80). The Spartans, when they decreed that only one of their
  two kings should be out on military service, decreed at the same
  time that only one of the Tyndarids should go out with them
  (v. 75): they once lent the Tyndarids as aids to the envoys of
  Epizephyrian Locri, who prepared for them a couch on board their
  ship (Diodôr. Excerpt. xvi. p. 15, Dindorf). The Thebans grant
  their hero Melanippus to Kleisthenês of Sikyôn (v. 68). What was
  sent, must probably have been a consecrated copy of the genuine
  statue.

  Respecting the solemnities practised towards the statues, see
  Plutarch, Alkibiad. 34; Kallimach. Hymn. ad Lavacr. Palladis,
  init. with the note of Spanheim; K. O. Müller, Archæologie der
  Kunst, § 69; compare Plutarch, Quæstion. Romaic. § 61. p. 279;
  and Tacit. Mor. Germ. c. 40; Diodôr. xvii. 49.

  The manner in which the real presence of a hero was identified
  with his statue (τὸν δίκαιον δεῖ θεὸν οἴκοι μένειν σώζοντα
  ροὺς ἱδρυμένους.—Menander, Fragm. Ἡνίοχος, p. 71, Meineke),
  consecrated ground, and oracle, is nowhere more powerfully
  attested than in the Heroïca of Philostratus (capp. 2-20.
  pp. 674-692; also De Vit Apollôn. Tyan. iv. 11), respecting
  Prôtesilaus at Elæus, Ajax at the Aianteium, and Hectôr at
  Ilium: Prôtesilaus appeared exactly in the equipment of his
  statue,—χλαμύδα ἐνῆπται, ξένε, τὸν Θετταλικὸν τρόπον, ὥσπερ καὶ
  τὸ ἄγαλμα τοῦτο (p. 674). The presence and sympathy of the hero
  Lykus is essential to the satisfaction of the Athenian dikasts
  (Aristophan. Vesp. 389-820): the fragment of Lucilius, quoted by
  Lactantius, De Falsâ Religione (i. 22), is curious.—Τοῖς ἥρωσι
  τοῖς κατὰ τὴν πόλιν καὶ τὴν χώραν ~ἱδρυμένοις~ (Lycurgus cont.
  Leocrat. c. 1).

  [1024] Plutarch, Timoleon, c. 12; Strabo, vi. p. 264.
  Theophrastus treats the perspiration as a natural phænomenon in
  the statues made of cedar-wood (Histor. Plant. v. 10). Plutarch
  discusses the credibility of this sort of miracles in his Life of
  Coriolanus, c. 37-38.

  [1025] Herodot. vii. 189. Compare the gratitude of the
  Megalopolitans to Boreas for having preserved them from the
  attack of the Lacedæmonian king Agis (Pausan. viii. 27, 4.—viii.
  36, 4). When the Ten Thousand Greeks were on their retreat
  through the cold mountains of Armenia, Boreas blew in their
  faces, “parching and freezing intolerably.” One of the prophets
  recommended that a sacrifice should be offered to him, which was
  done, “and the painful effect of the wind appeared to every one
  forthwith to cease in a marked manner;” (καὶ πᾶσι δὴ περιφανῶς
  ἔδοξε λῆξαι τὸ χαλεπὸν τοῦ πνεύματος.—Xenoph. Anab. iv. 5, 3.)




CHAPTER XVII.

THE GRECIAN MYTHICAL VEIN COMPARED WITH THAT OF MODERN EUROPE.


I have already remarked that the existence of that popular narrative
talk, which the Germans express by the significant word _Sage_
or _Volks-Sage_, in a greater or less degree of perfection or
development, is a phænomenon common to almost all stages of society
and to almost all quarters of the globe. It is the natural effusion
of the unlettered, imaginative, and believing man, and its maximum
of influence belongs to an early state of the human mind; for the
multiplication of recorded facts, the diffusion of positive science,
and the formation of a critical standard of belief, tend to discredit
its dignity and to repress its easy and abundant flow. It supplies to
the poet both materials to recombine and adorn, and a basis as well
as a stimulus for further inventions of his own; and this at a time
when the poet is religious teacher, historian, and philosopher, all in
one,—not, as he becomes at a more advanced period, the mere purveyor of
avowed, though interesting, fiction.

Such popular stories, and such historical songs (meaning by historical,
simply that which is accepted as history) are found in most quarters
of the globe, and especially among the Teutonic and Celtic populations
of early Europe. The old Gothic songs were cast into a continuous
history by the historian Ablavius;[1026] and the poems of the Germans
respecting Tuisto the earth-born god, his son Mannus, and his
descendants the eponyms of the various German tribes,[1027] as they
are briefly described by Tacitus, remind us of Hesiod, or Eumêlus, or
the Homeric Hymns. Jacob Grimm, in his learned and valuable Deutsche
Mythologie, has exhibited copious evidence of the great fundamental
analogy, along with many special differences, between the German,
Scandinavian, and Grecian mythical world; and the Dissertation of
Mr. Price (prefixed to his edition of Warton’s History of English
Poetry) sustains and illustrates Grimm’s view. The same personifying
imagination—the same ever-present conception of the will, sympathies,
and antipathies of the gods as the producing causes of phænomena,
and as distinguished from a course of nature with its invariable
sequence—the same relations between gods, heroes, and men, with the
like difficulty of discriminating the one from the other in many
individual names—a similar wholesale transfer of human attributes to
the gods, with the absence of human limits and liabilities—a like
belief in Nymphs, Giants, and other beings, neither gods nor men—the
same coalescence of the religious with the patriotic feeling and
faith—these are positive features common to the early Greeks with
the early Germans: and the negative conditions of the two are not
less analogous—the absence of prose writing, positive records, and
scientific culture. The preliminary basis and encouragements for the
mythopœic faculty were thus extremely similar.

  [1026] Jornandes, De Reb. Geticis, capp. 4-6.

  [1027] Tacit. Mor. German. c. 2. “Celebrant carminibus antiquis,
  quod unum apud eos memoriæ et annalium genus est, Tuistonem Deum
  terrâ editum, et filium Mannum, originem gentis conditoresque.
  Quidam licentiâ vetustatis, plures Deo ortos, pluresque gentis
  appellationes, Marsos, Gambrivios, Suevos, Vandaliosque
  affirmant: eaque vera et antiqua nomina.”

But though the prolific forces were the same in kind, the results were
very different in degree, and the developing circumstances were more
different still.

First, the abundance, the beauty, and the long continuance of early
Grecian poetry, in the purely poetical age, is a phænomenon which has
no parallel elsewhere.

Secondly, the transition of the Greek mind from its poetical to its
comparatively positive state was self-operated, accomplished by its
own inherent and expansive force—aided indeed, but by no means either
impressed or provoked, from without. From the poetry of Homer, to
the history of Thucydidês and the philosophy of Plato and Aristotle,
was a prodigious step, but it was the native growth of the Hellenic
youth into an Hellenic man; and what is of still greater moment, it
was brought about without breaking the thread either of religious
or patriotic tradition—without any coercive innovation or violent
change in the mental feelings. The legendary world, though the ethical
judgments and rational criticisms of superior men had outgrown
it, still retained its hold upon their feelings as an object of
affectionate and reverential retrospect.

Far different from this was the development of the early Germans. We
know little about their early poetry, but we shall run no risk of
error in affirming that they had nothing to compare with either Iliad
or Odyssey. Whether, if left to themselves, they would have possessed
sufficient progressive power to make a step similar to that of the
Greeks, is a question which we cannot answer. Their condition, mental
as well as political, was violently changed by a foreign action from
without. The influence of the Roman empire introduced artificially
among them new institutions, new opinions, habits, and luxuries, and,
above all, a new religion; the Romanized Germans becoming themselves
successively the instruments of this revolution with regard to such of
their brethren as still remained heathen. It was a revolution often
brought about by penal and coercive means: the old gods Thor and Woden
were formally deposed and renounced, their images were crumbled into
dust, and the sacred oaks of worship and prophecy hewn down. But even
where conversion was the fruit of preaching and persuasion, it did not
the less break up all the associations of a German with respect to
that mythical world which he called his past, and of which the ancient
gods constituted both the charm and the sanctity: he had now only the
alternative of treating them either as men or as dæmons.[1028] That
mixed religious and patriotic retrospect, formed by the coalescence
of piety with ancestral feeling, which constituted the appropriate
sentiment both of Greeks and of Germans towards their unrecorded
antiquity, was among the latter banished by Christianity: and while
the root of the old mythes was thus cankered, the commemorative
ceremonies and customs with which they were connected, either lost
their consecrated character or disappeared altogether. Moreover, new
influences of great importance were at the same time brought to bear.
The Latin language, together with some tinge of Latin literature—the
habit of writing and of recording present events—the idea of a
systematic law and pacific adjudication of disputes,—all these formed
a part of the general working of Roman civilization, even after the
decline of the Roman empire, upon the Teutonic and Celtic tribes. A
class of specially-educated men was formed, upon a Latin basis and upon
Christian principles, consisting too almost entirely of priests, who
were opposed, as well by motives of rivalry as by religious feeling,
to the ancient bards and storytellers of the community: the “lettered
men”[1029] were constituted apart from “the men of story,” and Latin
literature contributed along with religion to sink the mythes of
untaught heathenism. Charlemagne, indeed, at the same time that he
employed aggressive and violent proceedings to introduce Christianity
among the Saxons, also took special care to commit to writing and
preserve the old heathen songs. But there can be little doubt that
this step was the suggestion of a large and enlightened understanding
peculiar to himself. The disposition general among lettered Christians
of that age is more accurately represented by his son Louis le
Debonnaire, who, having learned these songs as a boy, came to abhor
them when he arrived at mature years, and could never be induced either
to repeat or tolerate them.[1030]

  [1028] On the hostile influence exercised by the change of
  religion on the old Scandinavian poetry, see an interesting
  article of Jacob Grimm in the Göttingen Gelehrte Anzeigen, Feb.
  1830, pp. 268-273; a review of Olaf Tryggvson’s Saga. The article
  _Helden_, in his Deutsche Mythologie, is also full of instruction
  on the same subject: see also the Einleitung to the book, p. 11,
  2nd edition.

  A similar observation has been made with respect to the old
  mythes of the pagan Russians by Eichhoff: “L’établissement du
  Christianisme, ce gage du bonheur des nations, fut vivement
  apprécié par les Russes, qui dans leur juste reconnaissance, le
  personnifièrent dans un héros. Vladimir le Grand, ami des arts,
  protecteur de la religion qu’il protégea, et dont les fruits
  firent oublier les fautes, devint l’Arthus et le Charlemagne de
  la Russie, et ses hauts faits furent un mythe national qui domina
  tous ceux du paganisme. Autour de lui se groupèrent ces guerriers
  aux formes athlétiques, au cœur généreux, dont la poésie aime à
  entourer le berceau mystérieux des peuples: et les exploits du
  vaillant Dobrinia, de Rogdai, d’Ilia, de Curilo, animèrent les
  ballades nationales, et vivent encore dans de naïfs récits.”
  (Eichhoff, Histoire de la Langue et Littérature des Slaves,
  Paris, 1839, part iii. ch. 2. p. 190.)

  [1029] This distinction is curiously brought to view by Saxo
  Grammaticus, where he says of an Englishman named Lucas, that
  he was “literis quidem tenuiter instructus, sed historiarum
  scientiâ apprime eruditus” (p. 330, apud Dahlmann’s Historische
  Forschungen, vol. i. p. 176).

  [1030] “Barbara et antiquissima carmina (says Eginhart, in
  his Life of Charlemagne), quibus veterum regum actus et bella
  canebantur, conscripsit.”

  Theganus says of Louis le Debonnaire, “Poetica carmina gentilia,
  quæ in juventute didicerat, respuit, nec legere, nec audire, nec
  docere, voluit.” (De Gestis Ludovici Imperatoris ap. Pithœum, p.
  304, c. xix.)

According to the old heathen faith, the pedigree of the Saxon, Anglian,
Danish, Norwegian, and Swedish kings,—probably also those of the German
and Scandinavian kings generally,—was traced to Odin, or to some of
his immediate companions or heroic sons.[1031] I have already observed
that the value of these genealogies consisted not so much in their
length, as in the reverence attached to the name serving as primitive
source. After the worship attached to Odin had been extinguished, the
genealogical line was lengthened up to Japhet or Noah,—and Odin, no
longer accounted worthy to stand at the top, was degraded into one of
the simple human members of it.[1032] And we find this alteration of
the original mythical genealogies to have taken place even among the
Scandinavians, although the introduction of Christianity was in those
parts both longer deferred, so as to leave time for a more ample
development of the heathen poetical vein—and seems to have created a
less decided feeling of antipathy (especially in Iceland) towards the
extinct faith.[1033] The poems and tales composing the Edda, though
first committed to writing after the period of Christianity, do not
present the ancient gods in a point of view intentionally odious or
degrading.

  [1031] See Grimm’s Deutsche Mythologie, art. _Helden_, p. 356,
  2nd edit. Hengist and Horsa were fourth in descent from Odin
  (Venerable Bede, Hist. i. 15). Thiodolff, the Scald of Harold
  Haarfager king of Norway, traced the pedigree of his sovereign
  through thirty generations to Yngarfrey, the son of Niord,
  companion of Odin at Upsal; the kings of Upsal were called
  Ynglinger, and the song of Thiodolff, Ynglingatal (Dahlmann,
  Histor. Forschung. i. p. 379). Eyvind, another Scald, a century
  afterwards, deduced the pedigree of Jarl Hacon from Saming,
  son of Yngwifrey (p. 381). Are Frode, the Icelandic historian,
  carried up his own genealogy through thirty-six generations to
  Yngwe; a genealogy which Torfæus accepts as trustworthy, opposing
  it to the line of kings given by Saxo Grammaticus (p. 352).
  Torfæus makes Harold Haarfager a descendant from Odin through
  twenty-seven generations; Alfred of England through twenty-three
  generations; Offa of Mercia through fifteen (p. 362). See also
  the translation by Lange of P. A. Müller’s Saga Bibliothek,
  Introd. p. xxviii. and the genealogical tables prefixed to Snorro
  Sturleson’s Edda.

  Mr. Sharon Turner conceives the human existence of Odin to
  be distinctly proved, seemingly upon the same evidence as
  Euêmerus believed in the human existence of Zeus (History of the
  Anglo-Saxons, Appendix to b. ii. ch. 3. p. 219, 5th edit.).

  [1032] Dahlmann, Histor. Forschung. t. i. p. 220. There is
  a valuable article on this subject in the Zeitschrift für
  Geschichts Wissenschaft (Berlin, vol. i. pp. 237-282) by Stuhr,
  “Über einige Hauptfragen des Nordischen Alterthums,” wherein
  the writer illustrates both the strong motive and the effective
  tendency, on the part of the Christian clergy who had to deal
  with these newly-converted Teutonic pagans, to Euêmerize the old
  gods, and to represent a genealogy, which they were unable to
  efface from men’s minds, as if it consisted only of mere men.

  Mr. John Kemble (Über die Stammtafel der Westsachsen, ap. Stuhr,
  p. 254) remarks, that “nobilitas,” among that people, consisted
  in descent from Odin and the other gods.

  Colonel Sleeman also deals in the same manner with the religious
  legends of the Hindoos,—so natural is the proceeding of Euêmerus,
  towards any religion in which a critic does not believe:—

  “They (the Hindoos) of course think that the incarnation of
  their three great divinities were beings infinitely superior to
  prophets, being in all their attributes and prerogatives equal
  to the divinities themselves. _But we are disposed to think that
  these incarnations were nothing more than great men whom their
  flatterers and poets have exalted into gods,—this was the way
  in which men made their gods in ancient Greece and Egypt._—All
  that the poets have sung of the actions of these men is now
  received as revelation from heaven: though nothing can be more
  monstrous than the actions ascribed to the best incarnation,
  Krishna, of the best of the gods, Vishnoo.” (Sleeman, Rambles and
  Recollections of an Indian Official, vol. i. ch. viii. 61.)

  [1033] See P. E. Müller, Über den Ursprung und Verfall der
  Isländischen Historiographie, p. 63.

  In the Leitfaden zur Nordischen Alterthumskunde, pp. 4-5
  (Copenhagen, 1837), is an instructive summary of the different
  schemes of interpretation applied to the northern mythes: 1, the
  historical; 2, the geographical; 3, the astronomical; 4, the
  physical; 5, the allegorical.

The transposition above alluded to, of the genealogical root from Odin
to Noah, is the more worthy of notice, as it illustrates the genuine
character of these genealogies, and shows that they sprung, not from
any erroneous historical data, but from the turn of the religious
feeling; also that their true value is derived from their being taken
entire, as connecting the existing race of men with a divine original.
If we could imagine that Grecian paganism had been superseded by
Christianity in the year 500 B. C., the great and venerated gentile
genealogies of Greece would have undergone the like modification; the
Herakleids, Pelopids, Æakids, Asklêpiads, etc., would have been merged
in some larger aggregate branching out from the archæology of the Old
Testament. The old heroic legends connected with these ancestral names
would either have been forgotten, or so transformed as to suit the new
vein of thought; for the altered worship, ceremonies, and customs would
have been altogether at variance with them, and the mythical feeling
would have ceased to dwell upon those to whom prayers were no longer
offered. If the oak of Dôdôna had been cut down, or the Theôric ship
had ceased to be sent from Athens to Dêlos, the mythes of Theseus and
of the two black doves would have lost their pertinence, and died away.
As it was, the change from Homer to Thucydidês and Aristotle took place
internally, gradually, and imperceptibly. Philosophy and history were
superinduced in the minds of the superior few, but the feelings of
the general public continued unshaken—the sacred objects remained the
same both to the eye and to the heart—and the worship of the ancient
gods was even adorned by new architects and sculptors who greatly
strengthened its imposing effect.

While then in Greece the mythopœic stream continued in the same course,
only with abated current and influence, in modern Europe its ancient
bed was blocked up, and it was turned into new and divided channels.
The old religion—though as an ascendent faith, unanimously and publicly
manifested, it became extinct—still continued in detached scraps and
fragments, and under various alterations of name and form. The heathen
gods and goddesses, deprived as they were of divinity, did not pass
out of the recollection and fears of their former worshippers, but
were sometimes represented (on principles like those of Euêmerus) as
having been eminent and glorious men—sometimes degraded into dæmons,
magicians, elfs, fairies, and other supernatural agents, of an inferior
grade and generally mischievous cast. Christian writers, such as Saxo
Grammaticus and Snorro Sturleson, committed to writing the ancient oral
songs of the Scandinavian Scalds, and digested the events contained
in them into continuous narrative—performing in this respect a task
similar to that of the Grecian logographers Pherekydês and Hellanikus,
in reference to Hesiod and the Cyclic poets. But while Pherekydês and
Hellanikus compiled under the influence of feelings substantially
the same as those of the poets on whom they bestowed their care,
the Christian logographers felt it their duty to point out the Odin
and Thor of the old Scalds as evil dæmons, or cunning enchanters,
who had fascinated the minds of men into a false belief in their
divinity.[1034] In some cases, the heathen recitals and ideas were
modified so as to suit Christian feeling. But when preserved without
such a change, they exhibited themselves palpably, and were designated
by their compilers, as at variance with the religious belief of the
people, and as associated either with imposture or with evil spirits.

  [1034] “Interea tamen homines Christiani in numina non credant
  ethnica, nec aliter fidem narrationibus hisce adstruere vel
  adhibere debent, quam in libri hujus proœmio monitum est de
  causis et occasionibus cur et quomodo genus humanum a verâ fide
  aberraverit.” (Extract from the Prose Edda, p. 75, in the Lexicon
  Mythologicum ad calcem Eddæ Sæmund. vol. iii. p. 357, Copenhag.
  edit.)

  A similar warning is to be found in another passage cited by
  P. E. Müller, Über den Ursprung und Verfall der Isländischen
  Historiographie, p. 138, Copenhagen, 1813; compare the Prologue
  to the Prose Edda, p. 6, and Mallet, Introduction à l’Histoire de
  Dannemarc, ch. vii. pp. 114-132.

  Saxo Grammaticus represents Odin sometimes as a magician,
  sometimes as an evil dæmon, sometimes as a high priest or pontiff
  of heathenism, who imposed so powerfully upon the people around
  him as to receive divine honors. Thor also is treated as having
  been an evil dæmon. (See Lexicon Mythologic. ut supra, pp. 567,
  915.)

  Respecting the function of Snorro as logographer, see Præfat.
  ad Eddam, ut supra, p. xi. He is much more faithful, and less
  unfriendly to the old religion, than the other logographers
  of the ancient Scandinavian Sagas. (Leitfaden der Nordischen
  Alterthümer, p. 14, by the Antiquarian Society of Copenhagen,
  1837.)

  By a singular transformation, dependent upon the same tone
  of mind, the authors of the French Chansons de Geste, in the
  twelfth century, turned Apollo into an evil dæmon, patron of the
  Mussulmans (see the Roman of Garin le Loherain, par M. Paulin
  Paris, 1833, p. 31): “Car mieux vaut Dieux que ne fait Apollis.”
  M. Paris observes, “Cet ancien Dieu des beaux arts est l’un des
  démons le plus souvent désignés dans nos poëmes, comme patron des
  Musulmans.”

  The prophet Mahomet, too, anathematized the old Persian epic
  anterior to his religion. “C’est à l’occasion de Naser Ibn
  al-Hareth, qui avait apporté de Perse l’Histoire de Rustem et
  d’Isfendiar, et la faisait réciter par des chanteuses dans
  les assemblées des Koreischites, que Mahomet prononça le vers
  suivant (of the Koran): Il y a des hommes qui achètent des contes
  frivoles, pour détourner par-là les hommes de la voie de Dieu,
  d’une manière insensée, et pour la livrer à la risée: mais leur
  punition les couvrira de honte.” (Mohl, Préface au Livre des Rois
  de Ferdousi, p. xiii.)

A new vein of sentiment had arisen in Europe, unsuitable indeed to
the old mythes, yet leaving still in force the demand for mythical
narrative generally. And this demand was satisfied, speaking generally,
by two classes of narratives,—the legends of the Catholic Saints and
the Romances of Chivalry, corresponding to two types of character, both
perfectly accommodated to the feelings of the time,—the saintly ideal
and the chivalrous ideal.

Both these two classes of narrative correspond, in character as well
as in general purpose, to the Grecian mythes—being stories accepted
as realities, from their full conformity with the predispositions and
deep-seated faith of an uncritical audience, and prepared beforehand by
their authors, not with any reference to the conditions of historical
proof, but for the purpose of calling forth sympathy, emotion, or
reverence. The type of the saintly character belongs to Christianity,
being the history of Jesus Christ as described in the gospels, and
that of the prophets in the Old Testament; whilst the lives of holy
men, who acquired a religious reputation from the fourth to the
fourteenth century of the Christian æra, were invested with attributes,
and illustrated with ample details, tending to assimilate them to
this revered model. The numerous miracles, the cure of diseases, the
expulsion of dæmons, the temptations and sufferings, the teachings
and commands, with which the biography of Catholic saints abounds,
grew chiefly out of this pious feeling, common to the writer and
to his readers. Many of the other incidents, recounted in the same
performances, take their rise from misinterpreted allegories, from
ceremonies and customs of which it was pleasing to find a consecrated
origin, or from the disposition to convert the etymology of a name
into matter of history: many have also been suggested by local
peculiarities, and by the desire of stimulating or justifying the
devotional emotions of pilgrims who visited some consecrated chapel or
image. The dove was connected, in the faith of the age, with the Holy
Ghost, the serpent with Satan; lions, wolves, stags, unicorns, etc.
were the subjects of other emblematic associations; and such modes of
belief found expression for themselves in many narratives which brought
the saints into conflict or conjoint action with these various animals.
Legends of this kind, so indefinitely multiplied and so preëminently
popular and affecting, in the Middle Ages, are not exaggerations of
particular matters of fact, but emanations in detail of some current
faith or feeling, which they served to satisfy, and by which they were
in turn amply sustained and accredited.[1035]

  [1035] The legends of the Saints have been touched upon by M.
  Guizot (Cours d’Histoire Moderne, leçon xvii.) and by M. Ampère
  (Histoire Littéraire de la France, t. ii. cap. 14, 15, 16); but
  a far more copious and elaborate account of them, coupled with
  much just criticism, is to be found in the valuable Essai sur les
  Légendes Pieuses du Moyen Age, par L. F. Alfred Maury, Paris,
  1843.

  M. Guizot scarcely adverts at all to the more or less of
  matter of fact contained in these biographies: he regards them
  altogether as they grew out of and answered to the predominant
  emotions and mental exigences of the age: “Au milieu d’un déluge
  de fables absurdes, la morale éclate avec un grand empire” (p.
  159, ed. 1829). “Les légendes ont été pour les Chrétiens de ce
  temps (qu’on me permette cette comparaison purement littéraire)
  ce que sont pour les Orientaux ces longs récits, ces histoires
  si brillantes et si variées, dont les Mille et une Nuits nous
  donnent un échantillon. C’était là que l’imagination populaire
  errait librement dans un monde inconnu, merveilleux, plein de
  mouvement et de poésie” (p. 175, _ibid._).

  M. Guizot takes his comparison with the tales of the Arabian
  Nights, as heard by an Oriental with uninquiring and unsuspicious
  credence. Viewed with reference to an instructed European, who
  reads these narratives as pleasing but recognized fiction, the
  comparison would not be just; for no one in that age dreamed of
  questioning the truth of the biographies. All the remarks of M.
  Guizot assume this implicit faith in them as literal histories:
  perhaps, in estimating the feelings to which they owed their
  extraordinary popularity, he allows too little predominance to
  the religious feeling, and too much influence to other mental
  exigences which then went along with it; more especially as he
  remarks, in the preceding lecture (p. 116), “Le caractère général
  de l’époque est la concentration du développement intellectuel
  dans la sphère religieuse.”

  How this absorbing religious sentiment operated in generating
  and accrediting new matter of narrative, is shown with great
  fulness of detail in the work of M. Maury: “Tous les écrits
  du moyen âge nous apportent la preuve de cette préoccupation
  exclusive des esprits vers l’Histoire Sainte et les prodiges qui
  avaient signalé l’avènement du Christianisme. Tous nous montrent
  la pensée de Dieu et du Ciel, dominant les moindres œuvres de
  cette époque de naïve et de crédule simplicité. D’ailleurs,
  n’était-ce pas le moine, le clerc, qui constituaient alors les
  seuls écrivains? Qu’y a-t-il d’étonnant que le sujet habituel de
  leurs méditations, de leurs études, se reflétât sans cesse dans
  leurs ouvrages? Partout reparaissait à l’imagination Jésus et ses
  Saints: cette image, l’esprit l’accueillait avec soumission et
  obéissance: il n’osait pas encore envisager ces célestes pensées
  avec l’œil de la critique, armé de défiance et de doute; au
  contraire, l’intelligence les acceptait toutes indistinctement
  et s’en nourrissait avec avidité. Ainsi s’accréditaient tous
  les jours de nouvelles fables. _Une foi vive veut sans cesse de
  nouveaux faits qu’elle puisse croire_, comme la charité veut de
  nouveaux bienfaits pour s’exercer” (p. 43). The remarks on the
  History of St. Christopher, whose personality was allegorized by
  Luther and Melancthon, are curious (p. 57).

Every reader of Pausanias will recognize the great general analogy
between the stories recounted to him at the temples which he visited,
and these legends of the Middle Ages. Though the type of character
which the latter illustrate is indeed materially different, yet the
source as well as the circulation, the generating as well as the
sustaining forces, were in both cases the same. Such legends were
the natural growth of a religious faith, earnest, unexamining, and
interwoven with the feelings at a time when the reason does not need
to be cheated. The lives of the Saints bring us even back to the
simple and ever-operative theology of the Homeric age; so constantly
is the hand of God exhibited even in the minutest details, for the
succor of a favored individual,—so completely is the scientific
point of view, respecting the phænomena of nature, absorbed into the
religious.[1036] During the intellectual vigor of Greece and Rome,
a sense of the invariable course of nature and of the scientific
explanation of phænomena had been created among the superior minds, and
through them indirectly among the remaining community; thus limiting
to a certain extent the ground open to be occupied by a religious
legend. With the decline of the pagan literature and philosophy, before
the sixth century of the Christian æra, this scientific conception
gradually passed out of sight, and left the mind free to a religious
interpretation of nature not less simple and _naïf_ than that which
had prevailed under the Homeric paganism.[1037] The great religious
movement of the Reformation, and the gradual formation, of critical
and philosophical habits in the modern mind, have caused these legends
of the Saints,—once the charm and cherished creed of a numerous
public,[1038] to pass altogether out of credit, without even being
regarded, among Protestants at least, as worthy of a formal scrutiny
into the evidence,—a proof of the transitory value of public belief,
however sincere and fervent, as a certificate of historical truth, if
it be blended with religious predispositions.

  [1036] “Dans les prodiges que l’on admettait avoir dû
  nécessairement s’opérer au tombeau du saint nouvellement
  canonisé, l’expression, ‘Cæci visum, claudi gressum, muti
  loquelam, surdi auditum, paralytici debitum membrorum officium,
  recuperabant,’ était devenue plûtot une formule d’usage que la
  rélation littérale du fait.” (Maury, Essai sur les Légendes
  Pieuses du Moyen Age, p. 5.)

  To the same purpose M. Ampère, ch. 14. p. 361: “Il y a un certain
  nombre de faits que l’agiographie reproduit constamment, quelque
  soit son héros: ordinairement ce personnage a eu dans sa jeunesse
  une vision qui lui a révélé son avenir: ou bien, une prophétie
  lui a annoncé ce qu’il serait un jour. Plus tard, il opère un
  certain nombre de miracles, toujours les mêmes; il exorcise des
  possédés, ressuscite des morts, il est averti de sa fin par un
  songe. Puis sur son tombeau s’accomplissent d’autres merveilles
  à-peu-près semblables.”

  [1037] A few words from M. Ampère to illustrate this: “C’est
  donc au sixième siècle que la légende se constitue: c’est alors
  qu’elle prend complètement le caractère naïf qui lui appartient:
  qu’elle est elle-même, qu’elle se sépare de toute influence
  étrangère. En même temps, l’ignorance devient de plus en plus
  grossière, et par suite la crédulité s’accroit: les calamités du
  temps sont plus lourdes, et l’on a un plus grand besoin de remède
  et de consolation.... Les récits miraculeux se substituent aux
  argumens de la théologie. Les miracles sont devenus la meilleure
  démonstration du Christianisme: c’est la seule que puissent
  comprendre les esprits grossiers des barbares” (c. 15. p. 373).

  Again, c. 17. p. 401: “Un des caractères de la légende est de
  mêler constamment le puéril au grand: il faut l’avouer, elle
  défigure parfois un peu ces hommes d’une trempe si forte, en
  mettant sur leur compte des anecdotes dont le caractère n’est pas
  toujours sérieux; elle en a usé ainsi pour St. Columban, dont
  nous verrons tout à l’heure le rôle vis-à-vis de Brunehaut et des
  chefs Mérovingiens. La légende auroit pu se dispenser de nous
  apprendre, comment un jour, il se fit rapporter par un corbeau
  les gants qu’il avait perdus: comment, un autre jour, il empêcha
  la bière de couler d’un tonneau percé, et diverses merveilles,
  certainement indignes de sa mémoire.”

  The miracle by which St. Columban employed the raven to fetch
  back his lost gloves, is exactly in the character of the Homeric
  and Hesiodic age: the earnest faith, as well as the reverential
  sympathy, between the Homeric man and Zeus or Athênê, is
  indicated by the invocation of their aid for his own sufferings
  of detail, and in his own need and danger. The criticism of
  M. Ampère, on the other hand, is analogous to that of the
  later pagans, after the conception of a course of nature had
  become established in men’s minds, so far as that exceptional
  interference by the gods was understood to be, comparatively
  speaking, rare, and only supposable upon what were called great
  emergences.

  In the old Hesiodic legend (see above, ch. ix. p. 245), Apollo
  is apprized by a raven of the infidelity of the nymph Korônis
  to him—τῷ μὲν ἄρ᾽ ἄγγελος ἦλθε κόραξ, etc. (the raven appears
  elsewhere as companion of Apollo, Plutarch, de Isid. et Os.
  p. 379, Herod. iv. 5.) Pindar, in his version of the legend,
  eliminated the raven, without specifying _how_ Apollo got his
  knowledge of the circumstance. The Scholiasts praise Pindar much
  for having rejected the puerile version of the story—ἐπαινεῖ
  τὸν Πίνδαρον ὁ Ἀρτέμων ὅτι παρακρουσάμενος τὴν περὶ τὸν κόρακα
  ἱστορἱαν, αὐτὸν δι᾽ ἑαυτοῦ ἐγνωκέναι φησὶ τὸν Ἀπόλλω ... χαίρειν
  οὖν ἐάσας τῷ τοιούτῳ μύθῳ ~τέλεως ὄντι ληρώδει~, etc.—compare
  also the criticisms of the Schol. ad Soph. Œdip. Kol. 1378, on
  the old epic Thebaïs; and the remarks of Arrian (Exp. Al. iii. 4)
  on the divine interference by which Alexander and his army were
  enabled to find their way across the sand of the desert to the
  temple of Ammon.

  In the eyes of M. Ampère, the recital of the biographer of
  St. Columban appears puerile (οὔπω ἴδον ὧδε θεοὺς ἀνάφανδρα
  φιλεῦντας, Odyss. iii. 221); in the eyes of that biographer,
  the criticism of M. Ampère would have appeared impious. When
  it is once conceded that phænomena are distributable under two
  denominations, the natural and the miraculous, it must be left to
  the feelings of each individual to determine what is and what is
  not, a suitable occasion for a miracle. Diodôrus and Pausanias
  differed in opinion (as stated in a previous chapter) about the
  death of Actæôn by his own hounds,—the former maintaining that
  the case was one fit for the special intervention of the goddess
  Artemis; the latter, that it was not so. The question is one
  determinable only by the religious feelings and conscience of the
  two dissentients: no common standard of judgment can be imposed
  upon them; for no reasonings derived from science or philosophy
  are available, inasmuch as in this case the very point in dispute
  is, whether the scientific point of view be admissible. Those
  who are disposed to adopt the supernatural belief, will find
  in every case the language open to them wherewith Dionysius of
  Halicarnassus (in recounting a miracle wrought by Vesta, in
  the early times of Roman history, for the purpose of rescuing
  an unjustly accused virgin) reproves the sceptics of his time:
  “It is well worth while (he observes) to recount the special
  manifestation (ἐπιφάνειαν) which the goddess showed to these
  unjustly accused virgins. For these circumstances, extraordinary
  as they are, have been held worthy of belief by the Romans, and
  historians have talked much about them. Those persons, indeed,
  who adopt the atheistical schemes of philosophy (if, indeed, we
  must call them _philosophy_), pulling in pieces as they do _all_
  the special manifestations (ἁπάσας διασύροντες τὰς ἐπιφανείας
  τῶν θεῶν) of the gods which have taken place among Greeks or
  barbarians, will of course turn _these_ stories also into
  ridicule, ascribing them to the vain talk of men, as if none of
  the gods cared at all for mankind. But those who, having pushed
  their researches farther, believe the gods not to be indifferent
  to human affairs, but favorable to good men and hostile to
  bad—will not treat _these_ special manifestations as _more_
  incredible than others.” (Dionys. Halic. ii. 68-69.) Plutarch,
  after noticing the great number of miraculous statements in
  circulation, expresses his anxiety to draw a line between the
  true and the false, but cannot find where: “excess, both of
  credulity and of incredulity (he tells us) in such matters is
  dangerous; caution, and nothing too much, is the best course.”
  (Camillus, c. 6.) Polybius is for granting permission to
  historians to recount a sufficient number of miracles to keep up
  a feeling of piety in the multitude, but not more: to measure
  out the proper quantity (he observes) is difficult, but not
  impossible (δυσπαράγραφός ἐστι ἡ ποσότης, οὐ μὴν ἀπαράγραφός γε,
  xvi. 12).

  [1038] The great Bollandist collection of the Lives of the
  Saints, intended to comprise the whole year, did not extend
  beyond the nine months from January to October, which occupy
  fifty-three large volumes. The month of April fills three of
  those volumes, and exhibits the lives of 1472 saints. Had the
  collection run over the entire year, the total number of such
  biographies could hardly have been less than 25,000, and might
  have been even greater (see Guizot, Cours d’Histoire Moderne,
  leçon xvii. p. 157).

The same mythopœic vein, and the same susceptibility and facility of
belief, which had created both supply and demand for the legends of the
Saints, also provided the abundant stock of romantic narrative poetry,
in amplification and illustration of the chivalrous ideal. What the
legends of Troy, of Thêbes, of the Kalydônian boar, of Œdipus, Thêseus,
etc. were to an early Greek, the tales of Arthur, of Charlemagne,
of the Niebelungen, were to an Englishman, or Frenchman, or German,
of the twelfth or thirteenth century. They were neither recognized
fiction nor authenticated history: they were history, as it is felt and
welcomed by minds unaccustomed to investigate evidence, and unconscious
of the necessity of doing so. That the Chronicle of Turpin, a mere
compilation of poetical legends respecting Charlemagne, was accepted
as genuine history, and even pronounced to be such by papal authority,
is well known; and the authors of the Romances announce themselves,
not less than those of the old Grecian epic, as being about to recount
real matter of fact.[1039] It is certain that Charlemagne is a great
historical name, and it is possible, though not certain, that the
name of Arthur may be historical also. But the Charlemagne of history,
and the Charlemagne of romance, have little except the name in common
nor could we ever determine, except by independent evidence (which in
this case we happen to possess), whether Charlemagne was a real or a
fictitious person.[1040] That illustrious name, as well as the more
problematical Arthur, is taken up by the romancers, not with a view to
celebrate realities previously verified, but for the purpose of setting
forth or amplifying an ideal of their own, in such manner as both to
rouse the feelings and captivate the faith of their hearers.

  [1039] See Warton’s History of English Poetry, vol. i. dissert.
  i. p. xvii. Again, in sect. iii. p. 140: “Vincent de Beauvais,
  who lived under Louis IX. of France (about 1260), and who, on
  account of his extraordinary erudition, was appointed preceptor
  to that king’s sons, very gravely classes Archbishop Turpin’s
  Charlemagne among the real histories, and places it on a level
  with Suetonius and Cæsar. He was himself an historian, and has
  left a large history of the world, fraught with a variety of
  reading, and of high repute in the Middle Ages; but edifying and
  entertaining as this work might have been to his contemporaries,
  at present it serves only to record their prejudices and to
  characterize their credulity.” About the full belief in Arthur
  and the Tales of the Round Table during the fourteenth century,
  and about the strange historical mistakes of the poet Gower in
  the fifteenth, see the same work, sect. 7. vol. ii. p. 33; sect.
  19. vol. ii. p. 239.

  “L’auteur de la Chronique de Turpin (says M. Sismondi,
  Littérature du Midi, vol. i. ch. 7. p. 289) n’avait point
  l’intention de briller aux yeux du public par une invention
  heureuse, ni d’amuser les oisifs par des contes merveilleux
  qu’ils reconnoitroient pour tels: il présentait aux Français
  tous ces faits étranges comme de l’histoire, et la lecture des
  légendes fabuleuses avait accoutumé à croire à de plus grandes
  merveilles encore; aussi plusieurs de ces fables furent-elles
  reproduites dans la Chronique de St. Denis.”

  Again, _ib._ p. 290: “Souvent les anciens romanciers, lorsqu’ils
  entreprennent un récit de la cour de Charlemagne, prennent un ton
  plus élevé: ce ne sont point des fables qu’ils vont conter, c’est
  de l’histoire nationale,—c’est la gloire de leurs ancêtres qu’ils
  veulent célébrer, et ils ont droit alors à demander qu’on les
  écoute avec respect.”

  The Chronicle of Turpin was inserted, even so late as the year
  1566, in the collection printed by Scardius at Frankfort of
  early German historians (Ginguené, Histoire Littéraire d’Italie,
  vol. iv. part ii. ch. 3. p. 157).

  To the same point—that these romances were listened to as real
  stories—see Sir Walter Scott’s Preface to Sir Tristram, p. lxvii.
  The authors of the Legends of the Saints are not less explicit in
  their assertions that everything which they recount is true and
  well-attested (Ampère, c. 14. p. 358).

  [1040] The series of articles by M. Fauriel, published in the
  Revue des Deux Mondes, vol. xiii. are full of instruction
  respecting the origin, tenor, and influence of the Romances of
  Chivalry. Though the name of Charlemagne appears, the romancers
  are really unable to distinguish him from Charles Martel or
  from Charles the Bald (pp. 537-539). They ascribe to him an
  expedition to the Holy Land, in which he conquered Jerusalem from
  the Saracens, obtained possession of the relics of the passion
  of Christ, the crown of thorns, etc. These precious relics he
  carried to Rome, from whence they were taken to Spain by a
  Saracen emir, named Balan, at the head of an army. The expedition
  of Charlemagne against the Saracens in Spain was undertaken for
  the purpose of recovering the relics: “Ces divers romans peuvent
  être regardés comme la suite, comme le développement, de la
  fiction de la conquête de Jérusalem par Charlemagne.”

  Respecting the Romance of Rinaldo of Montauban (describing the
  struggles of a feudal lord against the emperor) M. Fauriel
  observes, “Il n’y a, je crois, aucun fondement historique: c’est
  selon toute apparence, la pure expression poétique du fait
  général,” etc. (p. 542.)

To inquire which of the personages of the Carlovingian epic were real
and which were fictitious,—to examine whether the expedition ascribed
to Charlemagne against Jerusalem had ever taken place or not,—to
separate truth from exaggeration in the exploits of the Knights of the
Round Table,—these were problems which an audience of that day had
neither disposition to undertake nor means to resolve. They accepted
the narrative as they heard it, without suspicion or reserve; the
incidents related, as well as the connecting links between them, were
in full harmony with their feelings, and gratifying as well to their
sympathies as to their curiosity: nor was anything farther wanting to
induce them to believe it, though the historical basis might be ever so
slight or even non-existent.[1041]

  [1041] Among the “formules consacrées” (observes M. Fauriel)
  of the romancers of the Carlovingian epic, are asseverations
  of their own veracity, and of the accuracy of what they are
  about to relate—specification of witnesses whom they have
  consulted—appeals to pretended chronicles: “Que ces citations,
  ces indications, soient parfois sérieuses et sincères, cela peut
  être; mais c’est une exception et une exception rare. De telles
  allégations de la part des romanciers, sont en général un pur et
  simple mensonge, mais non toutefois un mensonge gratuit. C’est
  un mensonge qui a sa raison et sa convenance: il tient au désir
  et au besoin de satisfaire une opinion accoutumée à supposer
  et à chercher du vrai dans les fictions du genre de celles où
  l’on allègue ces prétendues autorités. La manière dont les
  auteurs de ces fictions les qualifient souvent eux-mêmes, est
  une conséquence naturelle de leur prétention d’y avoir suivi des
  documens vénérables. Ils les qualifient de chansons de _vieille
  histoire_, de _haute histoire_, de _bonne geste_, de _grande
  baronnie_: et ce n’est pas pour se vanter qu’ils parlent ainsi:
  la vanité d’auteur n’est rien chez eux, en comparaison du besoin
  qu’ils ont d’être crus, de passer pour de simples traducteurs,
  de simples répétiteurs de légendes ou d’histoire consacrée. Ces
  protestations de véracité, qui, plus ou moins expresses, sont de
  rigueur dans les romans Carlovingiens, y sont aussi fréquemment
  accompagnées de protestations accessoires contre les romanciers,
  qui, ayant déjà traité un sujet donné, sont accusés d’y avoir
  faussé la vérité.” (Fauriel, Orig. de l’Epopée Chevaleresque, in
  the Revue des Deux Mondes, vol. xiii. p. 554.)

  About the Cycle of the Round Table, see the same series of
  articles (Rev. D. M. t. xiv. pp. 170-184). The Chevaliers of the
  Saint Graal were a sort of _idéal_ of the Knights Templars: “Une
  race de princes héroïques, originaires de l’Asie, fut prédestinée
  par le ciel même à la garde du Saint Graal. Perille fut le
  premier de cette race, qui s’étant converti au Christianisme,
  passa en Europe sous l’Empereur Vespasien,” etc.; then follows
  a string of fabulous incidents: the epical agency is similar to
  that of Homer—Διὸς δ᾽ ἐτελείετο βουλή.

  M. Paulin Paris, in his Prefaces to the Romans des Douze Pairs
  de France, has controverted many of the positions of M. Fauriel,
  and with success, so far as regards the Provençal origin of
  the Chansons de Geste, asserted by the latter. In regard to
  the Romances of the Round Table, he agrees substantially with
  M. Fauriel; but he tries to assign a greater historical value
  to the poems of the Carlovingian epic,—very unsuccessfully, in
  my opinion. But his own analysis of the old poem of Garin de
  Loherain bears out the very opinion which he is confuting: “Nous
  sommes au règne de Charles Martel, et nous reconnaissons sous
  d’autres noms les détails exacts de la fameuse défaite d’Attila
  dans les champs Catalauniques. Saint Loup et Saint Nicaise,
  glorieux prélats du quatrième siècle, reviennent figurer autour
  du père de Pépin le Bref: enfin pour compléter la confusion,
  Charles Martel meurt sur le champ de bataille, à la place du roi
  des Visigoths, Theodoric.... _Toutes les parties de la narration
  sont vraies_: seulement _toutes s’y trouvent déplacées_. En
  général, les peuples n’entendent rien à la chronologie: les
  evènemens restent: les individus, les lieux et les époques, ne
  laissent aucune trace: c’est pour ainsi dire, une décoration
  scénique que l’on applique indifféremment à des récits souvent
  contraires.” (Preface to the Roman de Garin le Loherain, pp.
  xvi.-xx.: Paris, 1833.) Compare also his Lettre à M. Monmerqué,
  prefixed to the Roman de Berthe aux Grans Piés, Paris, 1836.

  To say that _all_ the parts of the narrative are true, is
  contrary to M. Paris’s own showing: _some_ parts may be true,
  separately taken, but these fragments of truth are melted down
  with a large mass of fiction, and cannot be discriminated unless
  we possess some independent test. The poet who picks out one
  incident from the fourth century, another from the fifth, and
  a few more from the eighth, and then blends them all into a
  continuous tale along with many additions of his own, shows that
  he takes the items of fact because they suit the purposes of his
  narrative, not because they happen to be attested by historical
  evidence. His hearers are not critical: they desire to have their
  imaginations and feelings affected, and they are content to
  accept without question whatever accomplishes this end.

The romances of chivalry represented, to those who heard them, real
deeds of the foretime—“glories of the foregone men,” to use the
Hesiodic expression[1042]—at the same time that they embodied and
filled up the details of an heroic ideal, such as that age could
conceive and admire—a fervent piety, combined with strength, bravery,
and the love of adventurous aggression, directed sometimes against
infidels, sometimes against enchanters or monsters, sometimes in
defence of the fair sex. Such characteristics wore naturally popular,
in a century of feudal struggles and universal insecurity, when the
grand subjects of common respect and interest were the Church and
the Crusades, and when the latter especially were embraced with an
enthusiasm truly astonishing.

  [1042] Hesiod, Theogon. 100—κλέα προτέρων ἀνθρώπων. Puttenham
  talks of the remnant of bards existing in his time (1589): “Blind
  Harpers, or such like Taverne Minstrels, whose matters are for
  the most part _stories of old time_, as the Tale of Sir Topaze,
  the Reportes of Bevis of Southampton, Adam Bell, Clymme of the
  Clough, and such other old Romances or _Historical Rhymes_.”
  (Arte of English Poesie, book ii. cap. 9.)

The long German poem of the Niebelungen Lied, as well as the Volsunga
Saga and a portion of the songs of the Edda, relate to a common
fund of mythical, superhuman personages, and of fabulous adventure,
identified with the earliest antiquity of the Teutonic and Scandinavian
race, and representing their primitive sentiment towards ancestors
of divine origin. Sigurd, Brynhilde, Gudrun, and Atle, are mythical
characters celebrated as well by the Scandinavian Scalds as by the
German epic poets, but with many varieties and separate additions to
distinguish the one from the other. The German epic, later and more
elaborated, includes various persons not known to the songs in the
Edda, in particular the prominent name of Dieterich of Bern—presenting,
moreover, the principal characters and circumstances as Christian,
while in the Edda there is no trace of anything but heathenism. There
is, indeed, in this the old and heathen version, a remarkable analogy
with many points of Grecian mythical narrative. As in the case of the
short life of Achilles, and of the miserable Labdakids of Thêbes—so
in the family of the Volsungs, though sprung from and protected by
the gods—a curse of destiny hangs upon them and brings on their ruin,
in spite of preëminent personal qualities.[1043] The more thoroughly
this old Teutonic story has been traced and compared, in its various
transformations and accompaniments, the less can any well-established
connection be made out for it with authentic historical names or
events. We must acquiesce in its personages as distinct in original
conception from common humanity, and as belonging to the subjective
mythical world of the race by whom they were sung.

  [1043] Respecting the Volsunga Saga and the Niebelungen Lied,
  the work of Lange—Untersuchungen über die Geschichte und das
  Verhältniss der Nordischen und Deutschen Heldensage—is a valuable
  translation from the Danish Saga-Bibliothek of P. E. Müller.

  P. E. Müller maintains, indeed, the historical basis of the tales
  respecting the Volsungs (see pp. 102-107)—upon arguments very
  unsatisfactory; though the genuine Scandinavian origin of the
  tale is perfectly made out. The chapter added by Lange himself,
  at the close (see p. 432, etc.), contains juster views as to the
  character of the primitive mythology, though he too advances
  some positions respecting a something “reinsymbolisches” in the
  background, which I find it difficult to follow (see p. 477,
  etc.).—There are very ancient epical ballads still sung by the
  people in the Faro Islands, many of them relating to Sigurd and
  his adventures (p. 412).

  Jacob Grimm, in his Deutsche Mythologie, maintains the purely
  mythical character, as opposed to the historical, of Siegfried
  and Dieterich (Art. _Helden_, pp. 344-346).

  So, too, in the great Persian epic of Ferdousi, the principal
  characters are religious and mythical. M. Mohl observes,—“Les
  caractères des personnages principaux de l’ancienne histoire de
  Perse se retrouvent dans le livre des Rois (de Ferdousi) tels
  que les indiquent les parties des livres de Zoroaster que nous
  possédons encore. Kaioumors, Djemschid, Feridoun, Gushtasp,
  Isfendiar, etc. jouent dans le poème épique le même rôle que dans
  les Livres sacrés: à cela près, que dans les derniers ils nous
  apparaissent à travers une atmosphere mythologique qui grandit
  tous leurs traits: mais cette difference est précisément celle
  qu’on devait s’attendre à trouver entre la tradition religieuse
  et la tradition épique.” (Mohl, Livre des Rois par Ferdousi,
  Preface, p. 1.)

  The Persian historians subsequent to Ferdousi have all taken
  his poem as the basis of their histories, and have even copied
  him faithfully and literally (Mohl, p. 53). Many of his heroes
  became the subjects of long epical biographies, written and
  recited without any art or grace, often by writers whose names
  are unknown (_ib._ pp. 54-70). Mr. Morier tells us that “the
  Shah Nameh is still believed by the present Persians to contain
  their ancient history” (Adventures of Hadgi Baba, c. 32). As
  the Christian romancers transformed Apollo into the patron of
  Mussulmans, so Ferdousi makes Alexander the Great a Christian:
  “La critique historique (observes M. Mohl) était du temps de
  Ferdousi chose presqu’inconnue.” (_ib._ p. xlviii.) About the
  absence not only of all historiography, but also of all idea of
  it, or taste for it among the early Indians, Persians, Arabians,
  etc., see the learned book of Nork, _Die Götter Syriens_,
  Preface, p. viii. _seqq._ (Stuttgart, 1842.)

Such were the compositions which not only interested the emotions,
but also satisfied the undistinguishing historical curiosity, of the
ordinary public in the middle ages. The exploits of many of these
romantic heroes resemble in several points those of the Grecian: the
adventures of Perseus, Achilles, Odysseus, Atalanta, Bellerophôn,
Jasôn, and the Trojan war, or Argonautic expedition generally, would
have fitted in perfectly to the Carlovingian or other epics of the
period.[1044] That of the middle ages, like the Grecian, was eminently
expansive in its nature: new stories were successively attached to the
names and companions of Charlemagne and Arthur, just as the legend
of Troy was enlarged by Arktinus, Leschês, and Stesichorus,—that of
Thêbes, by fresh miseries entailed on the fated head of Œdipus,—and
that of the Kalydônian boar, by the addition of Atalanta. Altogether,
the state of mind of the hearers seems in both cases to have been much
the same,—eager for emotion and sympathy, and receiving any narrative
attuned to their feelings, not merely with hearty welcome, but also
with unsuspecting belief.

  [1044] Several of the heroes of the ancient world were indeed
  themselves popular subjects with the romancers of the middle
  ages, Thêseus, Jasôn, etc.; Alexander the Great, more so than any
  of them.

  Dr. Warton observes, respecting the Argonautic expedition,
  “Few stories of antiquity have more the cast of one of the old
  romances than this of Jasôn. An expedition of a new kind is
  made into a strange and distant country, attended with infinite
  dangers and difficulties. The king’s daughter of the new country
  is an enchantress; she falls in love with the young prince, who
  is the chief adventurer. The prize which he seeks is guarded by
  brazen-footed bulls, who breathe fire, and by a hideous dragon,
  who never sleeps. The princess lends him the assistance of her
  charms and incantations to conquer these obstacles; she gives him
  possession of the prize, leaves her father’s court, and follows
  him into his native country.” (Warton, Observations on Spenser,
  vol. i. p. 178.)

  To the same purpose M. Ginguené: “Le premier modèle des Fées
  n’est-il pas dans Circé, dans Calypso, dans Médée? Celui des
  géans, dans Polyphème, dans Cacus, et dans les géans, ou les
  Titans, cette race ennemie de Jupiter? Les serpens et les
  dragons des romans ne sont-ils pas des successeurs du dragon
  des Hesperides et de celui de la Toison d’or? Les Magiciens! la
  Thessalie en étoit pleine. Les armes enchantées impénétrables!
  elles sont de la même trempe, et l’on peut les croire forgées
  au même fourneau que celles d’Achille et d’Enée.” (Ginguené,
  Histoire Littéraire d’Italie, vol. iv. part ii. ch. 3, p. 151.)

Nevertheless, there were distinctions deserving of notice, which render
the foregoing proposition more absolutely exact with regard to Greece
than with regard to the middle ages. The tales of the epic, and the
mythes in their most popular and extended signification, were the only
intellectual nourishment with which the Grecian public was supplied,
until the sixth century before the Christian æra: there was no prose
writing, no history, no philosophy. But such was not exactly the
case at the time when the epic of the middle ages appeared. At that
time, a portion of society possessed the Latin language, the habit of
writing, and some tinge both of history and philosophy: there were a
series of chronicles, scanty, indeed, and imperfect, but referring
to contemporary events and preventing the real history of the past
from passing into oblivion: there were even individual scholars, in
the twelfth century, whose acquaintance with Latin literature was
sufficiently considerable to enlarge their minds and to improve their
judgments. Moreover, the epic of the middle ages, though deeply imbued
with religious ideas, was not directly amalgamated with the religion
of the people, and did not always find favor with the clergy; while
the heroes of the Grecian epic were not only linked in a thousand ways
with existing worship, practices, and sacred localities, but Homer and
Hesiod pass with Herodotus for the constructors of Grecian theology. We
thus see that the ancient epic was both exempt from certain distracting
influences by which that of the middle ages was surrounded, and more
closely identified with the veins of thought and feeling prevalent in
the Grecian public. Yet these counteracting influences did not prevent
Pope Calixtus II. from declaring the Chronicle of Turpin to be a
genuine history.

If we take the history of our own country as it was conceived and
written from the twelfth to the seventeenth century by Hardyng,
Fabyan, Grafton, Hollinshed, and others, we shall find that it was
supposed to begin with Brute the Trojan, and was carried down from
thence, for many ages and through a long succession of kings, to the
times of Julius Cæsar. A similar belief of descent from Troy, arising
seemingly from a reverential imitation of the Romans and of their
Trojan origin, was cherished in the fancy of other European nations.
With regard to the English, the chief circulator of it was Geoffrey
of Monmouth, and it passed with little resistance or dispute into
the national faith—the kings from Brute downward being enrolled in
regular chronological series with their respective dates annexed. In
a dispute which took place during the reign of Edward I. (A. D. 1301)
between England and Scotland, the descent of the kings of England
from Brute the Trojan was solemnly embodied in a document put forth
to sustain the rights of the crown of England, as an argument bearing
on the case then in discussion: and it passed without attack from the
opposing party,[1045]—an incident which reminds us of the appeal
made by Æschinês, in the contention between the Athenians and Philip
of Macedôn, respecting Amphipolis, to the primitive dotal rights of
Akamas son of Thêseus—and also of the defence urged by the Athenians
to sustain their conquest of Sigeium, against the reclamations of the
Mitylenæans, therein the former alleged that they had as much right
to the place as any of the other Greeks who had formed part of the
victorious armament of Agamemnôn.[1046]

  [1045] See Warton’s History of English Poetry, sect. iii. p. 131,
  note. “No man before the sixteenth century presumed to doubt
  that the Francs derived their origin from Francus son of Hector;
  that the Spaniards were descended from Japhet, the Britons from
  Brutus, and the Scotch from Fergus.” (_Ibid._ p. 140.)

  According to the Prologue of the prose Edda, Odin was the
  supreme king of Troy in Asia, “in eâ terrâ quam nos Turciam
  appellamus.... Hinc omnes Borealis plagæ magnates vel primores
  genealogias suas referunt, atque principes illius urbis inter
  numina locant: sed in primis ipsum Priamum pro Odeno ponunt,”
  etc. They also identified _Tros_ with _Thor_. (See Lexicon
  Mythologicum ad calcem Eddæ Sæmund, p. 552. vol. iii.)

  [1046] See above, ch. xv. p. 458; also Æschinês, De Falsâ
  Legatione, c. 14, Herodot. v. 94. The Herakleids pretended a
  right to the territory in Sicily near Mount Eryx, in consequence
  of the victory gained by their progenitor Hêraklês over Eryx, the
  eponymous hero of the place. (Herodot. v. 43.)

The tenacity with which this early series of British kings was
defended, is no less remarkable than the facility with which it was
admitted. The chroniclers at the beginning of the seventeenth century
warmly protested against the intrusive scepticism which would cashier
so many venerable sovereigns and efface so many noble deeds. They
appealed to the patriotic feelings of their hearers, represented the
enormity of thus setting up a presumptuous criticism against the belief
of ages, and insisted on the danger of the precedent as regarded
history generally.[1047] How this controversy stood, at the time and
in the view of the illustrious author of Paradise Lost, I shall give
in his own words, as they appear in the second page of his History of
England. After having briefly touched upon the stories of Samothes son
of Japhet, Albion son of Neptune, etc., he proceeds:—

  [1047] The remarks in Speed’s Chronicle (book v. c. 3. sect.
  11-12), and the preface to Howes’s Continuation of Stow’s
  Chronicle, published in 1631, are curious as illustrating this
  earnest feeling. The Chancellor Fortescue, in impressing upon
  his royal pupil, the son of Henry VI., the limited character of
  English monarchy, deduces it from Brute the Trojan: “Concerning
  the different powers which kings claim over their subjects, I am
  firmly of opinion that it arises solely from the different nature
  of their original institution. So the kingdom of England had its
  original from Brute and the Trojans, who attended him from Italy
  and Greece, and became a mixed kind of government, compounded of
  the regal and the political.” (Hallam, Hist. Mid. Ages, ch. viii.
  P. 3, page 230.)

“But now of Brutus and his line, with the whole progeny of kings to the
entrance of Julius Cæsar, we cannot so easily be discharged: descents
of ancestry long continued, laws and exploits not plainly seeming to
be borrowed or devised, which on the common belief have wrought no
small impression: _defended by many, denied utterly by few_. For what
though Brutus and the whole Trojan pretence were yielded up, seeing
they, who first devised to bring us some noble ancestor, were content
at first with Brutus the Consul, till better invention, though not
willing to forego the name, taught them to remove it higher into a more
fabulous age, and by the same remove lighting on the Trojan tales, in
affectation to make the Briton of one original with the Roman, pitched
there: _Yet those old and inborn kings, never any to have been real
persons, or done in their lives at least some part of what so long hath
been remembered, cannot be thought without too strict incredulity._
For these, and those causes above mentioned, that which hath received
approbation from so many, I have chosen not to omit. Certain or
uncertain, be that upon the credit of those whom I must follow: _so far
as keeps aloof from impossible or absurd_, attested by ancient writers
from books more ancient, I refuse not, as the due and proper subject of
story.”[1048]

  [1048] “Antiquitas enim recepit fabulas fictas etiam nonnunquam
  incondite: hæc ætas autem jam exculta, præsertim eludens omne
  quod fieri non potest, respuit,” etc. (Cicero, De Republicâ, ii.
  10, p. 147, ed. Maii.)

Yet in spite of the general belief of so many centuries—in spite of
the concurrent persuasion of historians and poets—in spite of the
declaration of Milton, extorted from his feelings rather than from his
reason, that this long line of quasi-historical kings and exploits
could not be _all_ unworthy of belief—in spite of so large a body of
authority and precedent, the historians of the nineteenth century begin
the history of England with Julius Cæsar. They do not attempt either to
settle the date of king Bladud’s accession, or to determine what may
be the basis of truth in the affecting narrative of Lear.[1049] The
standard of historical credibility, especially with regard to modern
events, has indeed been greatly and sensibly raised within the last
hundred years.

  [1049] Dr. Zachary Grey has the following observations in his
  Notes on Shakspeare (London, 1754, vol. i. p. 112). In commenting
  on the passage in King Lear, _Nero is an angler in the lake of
  darkness_, he says, “This is one of Shakspeare’s most remarkable
  _anachronisms_. King Lear succeeded his father Bladud anno mundi
  3105; and Nero, anno mundi 4017, was sixteen years old, when he
  married Octavia, Cæsar’s daughter. See Funcii Chronologia, p. 94.”

  Such a supposed chronological discrepancy would hardly be pointed
  out in any commentary now written.

  The introduction prefixed by Mr. Giles, to his recent translation
  of Geoffrey of Monmouth (1842), gives a just view both of the
  use which our old poets made of his tales, and of the general
  credence so long and so unsuspectingly accorded to them. The
  list of old British kings given by Mr. Giles also deserves
  attention, as a parallel to the Grecian genealogies anterior to
  the Olympiads.

But in regard to ancient Grecian history, the rules of evidence still
continue relaxed. The dictum of Milton, regarding the ante-Cæsarian
history of England, still represents pretty exactly the feeling now
prevalent respecting the mythical history of Greece: “Yet those old
and inborn kings (Agamemnôn, Achilles, Odysseus, Jasôn, Adrastus,
Amphiaräus, Meleager, etc.), never any to have been real persons,
or done in their lives at least some part of what so long hath been
remembered, cannot be thought without too strict incredulity.” Amidst
much fiction (we are still told), there must be some truth: but how is
such truth to be singled out? Milton does not even attempt to make the
severance: he contents himself with “keeping aloof from the impossible
and the absurd,” and ends in a narrative which has indeed the merit
of being sober-colored, but which he never for a moment thinks of
recommending to his readers as true. So in regard to the legends of
Greece,—Troy, Thêbes, the Argonauts, the Boar of Kalydôn, Hêraklês,
Thêseus, Œdipus,—the conviction still holds in men’s minds, that there
must be something true at the bottom; and many readers of this work may
be displeased, I fear, not to see conjured up before them the Eidôlon
of an authentic history, even though the vital spark of evidence be
altogether wanting.[1050]

  [1050] The following passage, from the Preface of Mr. Price
  to Warton’s History of English Poetry, is alike just and
  forcibly characterized; the whole Preface is, indeed, full of
  philosophical reflection on popular fables generally. Mr. Price
  observes (p. 79):—

  “The great evil with which this long-contested question appears
  to be threatened at the present day, is an extreme equally
  dangerous with the incredulity of Mr. Ritson,—a disposition to
  receive as authentic history, under a slightly fabulous coloring,
  every incident recorded in the British Chronicle. An allegorical
  interpretation is now inflicted upon all the marvellous
  circumstances; a forced construction imposed upon the less
  glaring deviations from probability; and the usual subterfuge
  of baffled research,—erroneous readings and etymological
  sophistry,—is made to reduce every stubborn and intractable text
  to something like the consistency required. It might have been
  expected that the notorious failures of Dionysius and Plutarch,
  in Roman history, would have prevented the repetition of an
  error, which neither learning nor ingenuity can render palatable;
  and that the havoc and deadly ruin effected by these ancient
  writers (in other respects so valuable) in one of the most
  beautiful and interesting monuments of traditional story, would
  have acted as sufficient corrective on all future aspirants.
  The favorers of this system might at least have been instructed
  by the philosophic example of Livy,—if it be lawful to ascribe
  to philosophy a line of conduct which perhaps was prompted by
  a powerful sense of poetic beauty,—that traditional record can
  only gain in the hands of the future historian by one attractive
  aid,—the grandeur and lofty graces of that incomparable style in
  which the first decade is written; and that the best duty towards
  antiquity, and the most agreeable one towards posterity, is to
  transmit the narrative received as an unsophisticated tradition,
  in all the plenitude of its marvels and the awful dignity of its
  supernatural agency. For, however largely we may concede that
  real events have supplied the substance of any traditive story,
  yet the amount of absolute facts, and the manner of those facts,
  the period of their occurrence, the names of the agents, and the
  locality given to the scene, are all combined upon principles so
  wholly beyond our knowledge, that it becomes impossible to fix
  with certainty upon any single point better authenticated than
  its fellow. Probability in such decisions will often prove the
  most fallacious guide we can follow; for, independently of the
  acknowledged historical axiom, that ‘le vrai n’est pas toujours
  le vraisemblable,’ innumerable instances might be adduced, where
  tradition has had recourse to this very probability to confer
  a plausible sanction upon her most fictitious and romantic
  incidents. It will be a much more useful labor, wherever it can
  be effected, to trace the progress of this traditional story in
  the country where it has become located, by a reference to those
  natural or artificial monuments which are the unvarying sources
  of fictitious events; and, by a strict comparison of its details
  with the analogous memorials of other nations, to separate
  those elements which are obviously of a native growth, from the
  occurrences bearing the impress of a foreign origin. _We shall
  gain little, perhaps, by such a course for the history of human
  events_; but it will be an important accession to our stock of
  knowledge on the _history of the human mind_. It will infallibly
  display, as in the analysis of every similar record, the
  operations of that refining principle which is ever obliterating
  the monotonous deeds of violence that fill the chronicle of a
  nation’s early career, and exhibit the brightest attribute in
  the catalogue of man’s intellectual endowments,—a glowing and
  vigorous imagination,—bestowing upon all the impulses of the
  mind a splendor and virtuous dignity, which, however fallacious
  historically considered, are never without a powerfully redeeming
  good, the ethical tendency of all their lessons.”

I presume to think that our great poet has proceeded upon mistaken
views with respect to the old British fables, not less in that which
he leaves out than in that which he retains. To omit the miraculous and
the fantastic, (it is that which he really means by “the impossible
and the absurd,”) is to suck the lifeblood out of these once popular
narratives,—to divest them at once both of their genuine distinguishing
mark, and the charm by which they acted on the feelings of believers.
Still less ought we to consent to break up and disenchant in a similar
manner the mythes of ancient Greece,—partly because they possess the
mythical beauties and characteristics in far higher perfection, partly
because they sank deeper into the mind of a Greek, and pervaded both
the public and private sentiment of the country to a much greater
degree than the British fables in England.

Two courses, and two only, are open; either to pass over the mythes
altogether, which is the way in which modern historians treat the
old British fables, or else to give an account of them as mythes;
to recognize and respect their specific nature, and to abstain from
confounding them with ordinary and certifiable history. There are good
reasons for pursuing this second method in reference to the Grecian
mythes; and when so considered, they constitute an important chapter
in the history of the Grecian mind, and indeed in that of the human
race generally. The historical faith of the Greeks, as well as that
of other people, in reference to early and unrecorded times, is as
much subjective and peculiar to themselves as their religious faith:
among the Greeks, especially, the two are confounded with an intimacy
which nothing less than great violence can disjoin. Gods, heroes, and
men—religion and patriotism—matters divine, heroic, and human—were
all woven together by the Greeks into one indivisible web, in which
the threads of truth and reality, whatever they might originally have
been, were neither intended to be, nor were actually, distinguishable.
Composed of such materials, and animated by the electric spark of
genius, the mythical antiquities of Greece formed a whole at once
trustworthy and captivating to the faith and feelings of the people;
but neither trustworthy nor captivating, when we sever it from these
subjective conditions, and expose its naked elements to the scrutiny
of an objective criticism. Moreover, the separate portions of Grecian
mythical foretime ought to be considered with reference to that
aggregate of which they form a part: to detach the divine from the
heroic legends, or some one of the heroic legends from the remainder,
as if there were an essential and generic difference between them, is
to present the whole under an erroneous point of view. The mythes of
Troy and Thêbes are no more to be handled objectively, with a view to
detect an historical base, than those of Zeus in Krête, of Apollo and
Artemis in Dêlos, of Hermês, or of Promêtheus. To single out the Siege
of Troy from the other mythes, as if it were entitled to preëminence
as an ascertained historical and chronological event, is a proceeding
which destroys the true character and coherence of the mythical world:
we only transfer the story (as has been remarked in the preceding
chapter) from a class with which it is connected by every tie both of
common origin and fraternal affinity, to another with which it has
no relationship, except such as violent and gratuitous criticism may
enforce.

By drawing this marked distinction between the mythical and the
historical world,—between matter appropriate only for subjective
history, and matter in which objective evidence is attainable,—we
shall only carry out to its proper length the just and well-known
position long ago laid down by Varro. That learned man recognized three
distinguishable periods in the time preceding his own age: “First, the
time from the beginning of mankind down to the first deluge; a time
wholly unknown. Secondly, the period from the first deluge down to the
first Olympiad, which is called _the mythical period_, because many
fabulous things are recounted in it. Thirdly, the time from the first
Olympiad down to ourselves, which is called _the historical period_,
because the things done in it are comprised in true histories.”[1051]

  [1051] Varro ap. Censorin. de Die Natali; Varronis Fragm. p. 219,
  ed. Scaliger, 1623. “Varro tria discrimina temporum esse tradit.
  Primum ab hominum principio usque ad cataclysmum priorem, quod
  propter ignorantiam vocatur ἄδηλον. Secundum, a cataclysmo priore
  ad Olympiadem primam, quod quia in eo multa fabulosa referuntur,
  _Mythicon_ nominatur. Tertium a primâ Olympiade ad nos; quod
  dicitur _Historicon_, quia res in eo gestæ veris historiis
  continentur.”

  To the same purpose Africanus, ap. Eusebium, Præp. Ev. xx.
  p. 487: Μέχρι μὲν Ὀλυμπιάδων, οὐδὲν ἀκριβὲς ἱστόρηται τοῖς
  Ἕλλησι, πάντων συγκεχυμένων, καὶ κατὰ μηδὲν αὐτοῖς τῶν πρὸ τοῦ
  συμφωνούντων, etc.

Taking the commencement of true or objective history at the point
indicated by Varro, I still consider the mythical and historical
periods to be separated by a wider gap than he would have admitted. To
select any one year as an absolute point of commencement, is of course
not to be understood literally: but in point of fact, this is of very
little importance in reference to the present question, seeing that the
great mythical events—the sieges of Thêbes and Troy, the Argonautic
expedition, the Kalydônian boar-hunt, the Return of the Hêrakleids,
etc.—are all placed long anterior to the first Olympiad, by those who
have applied chronological boundaries to the mythical narratives. The
period immediately preceding the first Olympiad is one exceedingly
barren of events; the received chronology recognizes four hundred
years, and Herodotus admitted five hundred years, from that date back
to the Trojan war.