Produced by Richard Hulse, Barry Abrahamsen, and the Online
Distributed Proofreading Team at http://www.pgdp.net (This
file was produced from images generously made available
by The Internet Archive)









                           THE JEWS AMONG THE

                           GREEKS AND ROMANS




------------------------------------------------------------------------




[Illustration: ARCH OF TITUS, ROME]




------------------------------------------------------------------------




                           THE JEWS AMONG THE
                           GREEKS AND ROMANS



                                   BY

                               MAX RADIN

                    [Illustration: Publisher’s Logo]

                              PHILADELPHIA

               THE JEWISH PUBLICATION SOCIETY OF AMERICA

                                  1915


------------------------------------------------------------------------




                          COPYRIGHT, 1916, BY

               THE JEWISH PUBLICATION SOCIETY OF AMERICA


                  ------------------------------------




                               MATRI MEÆ
                             PIETATIS ERGO
                             HOC OPUSCULUM
                                D. D. D.


------------------------------------------------------------------------




                                PREFACE


It is a counsel of perfection that any historical study should be
approached with complete detachment. To such detachment I can make all
the less claim as I freely admit an abiding reverence for the history of
my own people, and, for the life of ancient Greece and Rome, a
passionate affection that is frankly unreasoning. At no place in the
course of the following pages have I been consciously apologetic. It is
true that where several explanations of an incident are possible, I have
not always selected the one most discreditable to the Jews. Doubtless
that will not be forgiven me by those who have accepted the anti-Semitic
pamphlets of Willrich as serious contributions to historical research.

The literature on the subject is enormous. Very few references to what
are known as “secondary” sources will, however, be found in this book. A
short bibliography is appended, in which various books of reference are
cited. From these all who are interested in the innumerable
controversies that the subject has elicited may obtain full information.

There remains the grateful task of acknowledging my personal
indebtedness to my friend, Dr. Ernst Riess, for many valuable
suggestions. Above all I desire to express my indebtedness to President
Solomon Schechter, of the Jewish Theological Seminary of America, at
whose instance the preparation of this book was undertaken. Those who
share with me the privilege of his friendship will note in more than one
turn of expression and thought the impress of that rich personality.

                                                               MAX RADIN

    NEW YORK CITY,
      October, 1915


------------------------------------------------------------------------




                                CONTENTS


                                                        PAGE

                   Introduction                           13

                I. Greek Religious Concepts               21

               II. Roman Religious Concepts               40

              III. Greek and Roman Concepts of Race       48

               IV. Sketch of Jewish History between       56
                     Nebuchadnezzar and Constantine

                V. Internal Development of the Jews       66
                     during the Persian Period

               VI. The First Contact between Greek and    76
                     Jew

              VII. Egypt                                  90

             VIII. Jews in Ptolemaic Egypt               104

               IX. The Struggle against Greek Culture    118
                     in Palestine

                X. Antiochus the Manifest God            135

               XI. The Jewish Propaganda                 148

              XII. The Opposition                        163

             XIII. The Opposition in Its Social Aspect   176

              XIV. The Philosophic Opposition            191

               XV. The Romans                            210

              XVI. Jews in Rome during the Early         236
                     Empire

             XVII. The Jews of the Empire till the       257
                     Revolt

            XVIII. The Revolt of 68 C.E.                 287

              XIX. The Development of the Roman Jewish   304
                     Community

               XX. The Final Revolts of the Jews         328

              XXI. The Legal Position of the Jews in     350
                     the Later Empire

                   Summary                               368

                   Notes                                 373

                   Bibliography                          415

                   Index                                 417


------------------------------------------------------------------------




                             ILLUSTRATIONS


             ARCH OF TITUS, ROME                      Frontispiece

             RUINS OF THE AMPHITHEATER AT GERASA         62
               (JERASH), GILEAD, PALESTINE

             ANTIOCHUS (IV) EPIPHANES, AFTER A COIN     136
               (from a Drawing by Ralph Iligan)

             GREEK INSCRIPTION, FOUND ON SITE OF        186
               TEMPLE AREA, FORBIDDING GENTILES TO
               PASS BEYOND THE INNER TEMPLE WALLS AT
               JERUSALEM

             RUINS OF AN ANCIENT SYNAGOGUE AT MEROM,    216
               GALILEE, PALESTINE (Roman Period)

             TOMBS OF THE KINGS, VALLEY OF KEDRON,      268
               JERUSALEM (from Wilson’s “Jerusalem”)

             SYMBOLS AND INSCRIPTIONS FROM JEWISH       310
               CATACOMBS AND CEMETERIES IN ROME (from
               Garrucci)


------------------------------------------------------------------------




                              INTRODUCTION


The civilization of Europe and America is composed of elements of many
different kinds and of various origin. Most of the beginnings cannot be
recovered within the limits of recorded history. We do not know where
and when a great many of our fundamental institutions arose, and about
them we are reduced to conjectures that are sometimes frankly
improbable. But about a great many elements of our civilization, and
precisely those upon which we base our claim to be called
civilized—indeed, which give us the word and the concept of civic
life—we know relatively a great deal, and we know that they originated
on the eastern shores of the large landlocked sea known as the
Mediterranean.

We are beginning to be aware that the process of developing these
elements was much longer than we had been accustomed to believe. Many
races and several millennia seem to have elaborated slowly the
institutions that older historians were prepared to regard as the
conscious contrivance of a single epoch. But even if increasing
archeological research shall render us more familiar than we are with
Pelasgians, Myceneans, Minoans, Aegeans, it is not likely that the
claims of two historic peoples to have founded European civilization
will be seriously impugned. These are the Romans and the Greeks. To
these must be added another people, the Jews, whose contribution to
civilization was no less real and lasting.

The Greeks and Romans have left descendants only in a qualified sense.
There are no doubt thousands of individuals now living who are the
actual descendants of the kinsmen and contemporaries of the great names
in Greek and Roman history; but these individuals are widely scattered,
and are united by national and racial bonds with thousands of
individuals not so descended, from whom they have become wholly
indistinguishable. We have documentary evidence of great masses of other
races, Celtic, Germanic, Slavic, Semitic, entering into the territory
occupied by Greeks and Romans and mingling with them, and to this
evidence is added the confirmation of anthropological researches. This
fact has made it possible to consider Greek and Roman history
objectively. Only rarely can investigators be found who feel more than a
very diluted pride in the achievements of peoples so dubiously connected
with themselves. It is therefore with increasing clarity of vision that
we are ordering the large body of facts we already know about Greeks and
Romans, and are gathering them in constantly broadening categories.

That unfortunately is not the case with the Jews. Here, too, racial
admixture was present, but it never took place on a large scale at any
one time, and may always have remained exceptional. However that may be,
common belief both among Jews and non-Jews holds very strongly the view
that the Jews of to-day are the lineal descendants of the community
reorganized by Ezra, nor is it likely that this belief would be
seriously modified by much stronger evidence to the contrary than has
yet been adduced.[1] The result has been that the place of the Jews in
history has been determined upon the basis of institutions avowedly
hostile to them. It may be said that historians have introduced the Jews
as a point of departure for Christianity, and have not otherwise
concerned themselves with them.

There was a time when Greek and Roman and Jew were in contact. What was
the nature of that contact? What were its results? What were the mutual
impressions made by all three of them on one another? The usual answer
has been largely a transference of modern attitudes to ancient times. Is
another answer possible? Do the materials at our disposal permit us to
arrive at a firmer and better conclusion?

It is necessary first to know the conditions of our inquiry. The period
that we must partially analyze extends from the end of the Babylonian
Captivity to the establishment of Christianity—roughly from about 450
B.C.E. to 350 C.E., some seven or eight hundred years.

The time limits are of course arbitrary. The contact with Greeks may
have begun before the earlier of the two limits, and the relations of
the Jews with both Greeks and Romans certainly did not cease with either
Constantine or Theodosius. However, it was during the years that
followed the return from the Exile that much of the equipment was
prepared with which the Jew actually met the Greek, and, on the other
hand, the relations of Christian Rome to the Jews were determined by
quite different considerations from those that governed Pagan Rome. It
is at this point accordingly that a study of the Jews among the Greeks
and Romans may properly end.


                              THE SOURCES


Even for laymen it has become a matter of great interest to know upon
what material the statements are based which scientists and scholars
present to them. It is part perhaps of the general skepticism that has
displaced the abundant faith of past generations in the printed word.
For that reason what the sources are from which we must obtain the
statements that we shall make here, will be briefly indicated below.

First we have a number of Greek and Latin writers who incidentally or
specially referred to the Jews. However, as is the case with many other
matters of prime importance, the writings of most of these authors have
not come down to us completely, but in fragments. That is to say, we
have only the brief citations made of them by much later writers, or
contained in very late compilations, such as lexicons, commonplace
books, or manuals for instruction. Modern scholars have found it
imperatively necessary to collect these fragments, so that they may be
compared and studied more readily. In this way the fragments of lost
books on history, grammar, music, of lost poems and plays, have been
collected at various times. Similarly the fragments concerning the Jews
have been collected, and gathered into a single book by M. Théodore
Reinach, under the title of _Textes d’ auteurs grecs et latins relatifs
au judaisme_. Here the Greek and Latin texts and the French translation
of them are arranged in parallel columns, and furnished with explanatory
footnotes. M. Reinach’s great distinction as a classical scholar enables
him to speak with authority upon many of the controverted questions that
these texts contain. Often his judgment as to what certain passages mean
may be unquestioningly accepted, and at all times one disagrees with him
with diffidence.

Secondly, we have the Jewish literature of the period; but that
literature was produced under such various conditions and with such
diverse purposes that a further classification is necessary.

Most important for our purposes is that part of Jewish literature which
was a direct outcome of the contact we are setting forth—the apologetic
writings of the Jews, or those books written in Greek, only rarely in
Latin, in which Jewish customs and history are explained or defended for
non-Jewish readers. Most of these books likewise have been lost, and
have left only inconsiderable fragments, but in the case of two writers
we have very extensive remains. One of these men is the Alexandrian Jew
Philo, a contemporary of the first Roman emperors. The other was the
Palestinian Jew Joseph, who played an important, if ignoble, part in the
rebellion of 68 C.E.

An estimate of the character of Philo and Josephus—to give the latter
the name by which alone he is remembered—or of the value of their works,
is out of place here. Philo’s extant writings are chiefly concerned with
philosophic exposition, and are only indirectly of documentary value.
However, he also wrote a “Defense” of his people, of which large
portions have survived, notably the _In Flaccum_, a bitter invective
against the prefect of Egypt under Tiberius, and the _Legatio ad Gaium_,
a plea in behalf of the Alexandrian Jews made to the emperor Caligula by
an embassy of which Philo was himself a member.[2]

An apologetic purpose, for himself more than for his fellow-citizens, is
discernible in practically all the extant writings of Josephus. One of
them, however, the misnamed _Contra Apionem_, is avowedly a defense of
the Jews against certain misrepresentations contained in Greek books.
The importance of Josephus’ works is impossible to overrate. For many
matters he is our sole authority. But the character exhibited in his own
account of his conduct has impaired the credibility of much of what he
says, and has provoked numerous controversies. It is impossible to
disregard him, and unsafe to rely upon him. However, it is not unlikely
that fuller knowledge, which the sands of Egypt and Palestine may at any
time offer, will compel us to change our attitude toward him
completely.[3]

Besides the apologetic Jewish writings, directed to gentile readers,
there was a flourishing literature in Greek (and perhaps in Latin too)
intended for Greek-speaking Jews. It may be said that no branch of
literary art was quite neglected. The great majority of these books are
lost. Some, however, of a homiletic or parenetic tendency, attained
partial sanctity in some of the Jewish congregations, and were, under
such protection, transferred to the Christian communities that succeeded
them. They may now be found in collections of Apocrypha and
Pseudepigrapha, such as the German collection of Kautzsch and that
recently completed in English by Charles. Examples are the Wisdom of
Solomon, the Jewish Sibyl, the Letter of Aristeas, etc.

All these books were intended for Jewish readers, but for Jews whose
sole mother tongue was Greek. In Palestine and Syria the Jews spoke
Aramaic, and the educated among them used Hebrew for both literary and
colloquial purposes. There was consequently an active literature in
these languages. Some books so written were early translated into Greek,
and from Greek into Latin and Ethiopic, and have survived as part of the
Apocrypha. Judith, First Maccabees, Tobit, are instances. It was a rare
and fortunate accident that gave us the Hebrew original of such a book,
of Ben Sira, or Ecclesiasticus.

Again, the highly organized religious and legal institutions of the Jews
found literary expression in the decisions and comments upon them that
all such institutions involve. The exposition of the consecrated ancient
literature was also begun in this period. It was not, however, till
relatively late, 200 C.E. and after, that actual books were put
together, so that it is dangerous to accept uncritically references to
earlier dates.

The books referred to are primarily the Mishnah and the other extant
collections of Baraitot. Besides these, such works as the Megillat
Taanit and the Seder Olam must be grouped here. The earlier portions of
both Talmuds may be included, perhaps all of the Jerusalem Talmud.

One source of somewhat problematic character remains to be considered.
Biblical critics have been at some pains to assign as much as possible
of the Bible to the earlier centuries of the period we have delimited.
That more than a very slight portion can be so assigned is scarcely
probable, but some of it may, especially those books or passages in
which Greek influence is clearly noticeable. However, little profit can
be gained for our purposes from material that demands such a deal of
caution in its use.

Finally, besides literary evidences, which, as we have seen, have
wretchedly failed to substantiate the poet’s vaunt of being more lasting
than brass, we have the brass itself; that is, we have the stones,
coins, utensils, potsherds, and papyri inscribed with Hebrew, Aramaic,
Greek, Latin, Babylonian, and Egyptian words, which are the actual
contemporaries, just as we have them, of the events they illustrate. It
is the study of evidences like these that has principally differentiated
modern historical research from the methods it displaced, and in the
unceasing increase of these fragmentary and invaluable remains our hopes
of better knowledge of ancient life are centered.[4]


------------------------------------------------------------------------




                               CHAPTER I

                        GREEK RELIGIOUS CONCEPTS


The Jew is presented to the modern world in the double aspect of a race
and a religion. In a measure this has always been the case, but we shall
not in the least understand what the statement of the fact means without
a very close analysis of the concepts of race and religion formed by
both Greeks and Romans.

The word religion has a very definite meaning to us. It is the term
applied to the body of beliefs that any group of men maintain about
supernatural entities upon whom they consider themselves wholly
dependent. The salient fact of modern religions is that for most men the
group is very large indeed, that it vastly transcends all national
limits. Christianity, Islam, and Buddhism, all profess the purpose of
gaining the entire human race for their adherents, and have actively
attempted to do so. The fact that the religions with which we are most
familiar are “world-religions,” and the abstract character of the
predicates of the Deity in them, would seem to make religion as such
practically free from local limitation. However, that is not completely
true even for our time. In the first place, the bulk of Christians, as
of Muslims and Buddhists, are in all three cases bearers of a common
culture, and have long believed themselves of common descent. They
occupy further a continuous, even if very large, area. Religious maps of
the world would show solid blocks of color, not spots scattered
everywhere. Secondly, even within the limits of the religion itself
national boundaries are not wholly expunged. The common Christianity of
Spain and England presents such obvious differences that insistence upon
them is unnecessary; nor does the fact that Southern Germany, Belgium,
and Ireland are all Roman Catholic imply that all these sections have
the same religious attitude.

These are modern illustrations, and they represent survivals of a state
of things which in the Greek world was fundamental. As it seems to us
axiomatic that an abstractly conceived God cannot be the resident of a
limited area on the surface of the earth, just so axiomatic it seemed,
at one stage of Greek religious growth, that a god was locally limited,
that his activities did not extend—or extended only in a weakened
form—beyond a certain sharply circumscribed geographical area. That is
probably the most fundamental and thoroughgoing of the differences
between Greek religious feeling and that of our day. Opinions may differ
widely about the degree of anthropomorphism present at the contrasted
periods; and then, as now, the statements made about the nature and
power of the Deity were contradictory, vague, and confusing. But one
thing it is hard to question: the devoutly religious man of to-day feels
himself everywhere, always, in the presence of his God. The Greek did
not feel that his god was everywhere with him, certainly did not feel
that he was everywhere approachable.[5]

At another point too we are in great danger of importing modern notions
into ancient conditions. Judaism, Christianity, and Islam are all
book-religions. The final source of their doctrines is a revelation that
has been written down, and is extant as an actual and easily accessible
book. Moreover, it is the narrative portion of this book that is the
best-known part of it, and that is generally associated in the popular
mind with it. In the same way, we are prone to think of Greek religion
as a series of extraordinarily beautiful myths or narratives of gods and
heroes, which have likewise been written down, and are extant in the
poems and dramas of which they are the subject. This view has been
greatly strengthened by the unfortunate currency of the epigram that
Homer was the Greek Bible. No one would be inclined to force, except as
a paradox, the analogy upon which the statement rests; yet the phrase is
so terse and simple, and the elements of the comparison are so generally
familiar, that consciously and unconsciously current conceptions are
moulded by it.

Now if the epigram quoted is essentially true, we have at once a measure
of Greek religious feeling, since the Homeric poems are as accessible to
us as to the Greeks themselves. We should be compelled to reckon with
variety in the interpretation of the text, but in the literal
signification there would always be a point of departure. And we should
at once realize that for divine beings depicted as they are by Homer a
devotion of a very different sort is demanded from that which modern
faiths give their Deity. Nor does later literature represent the gods on
a loftier moral plane. When we read Aristophanes,[6] it becomes still
more difficult to understand how the gods could retain their divinity
not only when deprived of their moral character, but even when stripped
of their dignity. So far from raising the moral character of the divine
beings who are the actors in these legends, the later versions of many
quite unexceptionable myths deliberately debase them by subjecting most
actions to a foully erotic interpretation.[7] The less offensive
narrative, to be sure, survives as well, but it is to be noted that the
divinity of the personages in question seems to be as unquestioned in
the corrupt as in the purer form of the story.

How might an emotionally sensitive or mentally trained man pour forth
supplication before a guzzling braggart like the Aristophanic Heracles
or an effeminate voluptuary like the Apollo of Alexandrian poetry? It
seems hard to discover any other defense than the one Charles Lamb
offered for the dramatists of the Restoration—that the world the gods
moved in was a wholly different one from the human world; a world in
which moral categories had no existence, a Land of Cockayne without
vices, because it was without the sanctions which vice disregards. No
doubt some Greeks felt in this way toward the myths. But it was not a
satisfactory theory. It introduced a dualism into standards of conduct
that soon became intolerable, when men reflected seriously upon other
sides of the divine nature, and drew inferences from it.

As a matter of fact, the difficulty we find in addressing words of
prayer and praise to such unworthy gods as sat upon the Homeric Olympus
is modern, and was probably not felt at all by the vast majority of
Greeks, either in Homer’s time or later. Not that the fraud, cruelty,
faithlessness there exhibited seemed to the Greeks of any epoch
commendable or imitable qualities. Even the Homeric Greek was far from
being in a barbarous or semi-barbarous state. Civic virtues as between
men were known and practised. But the personality of the individual gods
in these stories could be disregarded in practice, because they were in
no sense a part of the Greek religion. The chastest of men might with a
clear conscience worship the lecherous Zeus, because worship did not at
all concern itself with the catalogue of his amours. In Homer’s time and
after, the Greek firmly believed that the Olympians were actually
existing beings, but he scarcely stopped to ask himself whether it was
literally true that Zeus had bidden Hera be silent under threats of
personal violence. What did concern him in his relation with his gods
was the disposition in which the god was likely to be toward him or his
people. And his religious activity was directed to the end of making
that disposition as good as possible.

The matter just set forth is far from being new doctrine; but for the
general reader it must be constantly re-emphasized, because it is
constantly forgotten. We continually find the Greek myths discussed in
terms that would be true only of the Gospel narratives, and we see the
Greek gods described as though they possessed the sharpness of personal
outline which the Deity has in the minds of believing Christians. It is
no doubt the extant literature—a florilegium at best—that is at fault in
the matter. This literature, it must be remembered, was not preserved
altogether by accident. To a large extent it represents a conscious
selection, made for pedagogic purposes. The relative coherence which
Greek myths have for us is due to the fact that the surviving poems and
dramas which contain them were selected, partially at least, by
Hellenistic and Byzantine schoolmasters in order to fit into a set cycle
or scheme. Even in what we have there is abundant evidence that the
myths about the gods could pretend to no sanctity for anybody, devout or
scoffer, for the simple reason that they negated themselves, that widely
differing and hopelessly contradictory stories were told of the same
event or person.

In reality the Greek myths were not coherent. It is hard to discover in
many of them a folkloristic kernel that had to be kept intact. Almost
everywhere we are dealing with the free fantasies of highly imaginative
poets. So fully was this understood that the stories most familiar to us
are generally alluded to in serious Greek literature with an apologetic,
ὡς oἱ ποιηταί φασι, “as the poets say,” or some similar
phrase. And as these stories were largely unrelated, so also were the
gods of whom they were told, even though they bore the same name. If
mythographers had taken the trouble to collect all the stories known of
any one god—Hermes, for example—there would be nothing except the common
name to indicate that they referred to the same chief actor, and much
that, except for the common name, would be referred to different gods.
Not even a single prominent trait, not a physical feature, would be
found to run through all the myths so collected.

So far we have been dealing with extant literature. But if the more
recondite notices of popular superstition are taken into account, as
well as the archeological discoveries, we meet such figures as Demeter,
Artemis, Apollo,[8] in various and curious forms and associations, so
that one might be tempted to suppose that these highly individualized
figures of poetry were, in the shrines in which they were worshiped,
hardly more than divine appellatives of rather vague content. And on the
islands of the Aegean, in Crete and Cyprus, where the continuity between
Aegean, Mycenean, and Hellenic civilization[9] was perhaps less
disturbed by convulsive upheavals, this seems especially to have been
the case.

For cult purposes, then—the primary purpose of Greek religion—there was
less difference between gods than we might suppose. Not even the
strongly marked personages that poetry made of them were able to fix
themselves in the popular mind. Sculptors had been busy in
differentiating types, and yet even here the process was not completed.
While in general we know of Poseidon-types, Zeus-types, etc., in art,
the most thoroughly equipped critics find themselves embarrassed if they
are required to name a statue that is wholly lacking in definite
external symbols or attributes, such as the thunderbolt, trident,
caduceus, and others.[10] Even the unrivaled artistic abilities of Greek
sculptors found it impossible to create unmistakable types of the Greek
gods, for the reason that the character of the god as portrayed in myth
and fable was fluid, and not fixed.

As among most peoples of the time, the essential religious act was that
which brought the god and his worshiper into contact—the sacrifice. What
the real nature of sacrifice was need not concern us here. The undoubted
fact is that sacrifice and prayer formed a single act;[11] that it was
during the sacrifice that the worshiper ventured to address his prayer
to the godhead he invoked. In doing so he must of necessity use the
god’s name, and, as we have seen, the name was of more general and less
specific connotation than is usually supposed. But the act of worship
itself was specifically occasioned. Even the fixed and annually
recurring festivals related to a specific, if recurring, occasion in the
life of the people. This was eminently the case in the irregular acts of
worship that arose out of some unforeseen contingency. Whatever the
divine name was that was used, the specific occasion of its use made it
necessary also to specify the function of the divinity of which the
intervention was sought. That was regularly done by attaching to the
name a qualifying epithet. When the rights of hospitality were
threatened with invasion, it was Ζεὺς Ξένιος, Zeus the
Protector of Strangers, that was addressed. In gratitude for a
deliverance, Zeus or Apollo or Heracles or the Dioscuri or many another
might be invoked as “the Savior.”[12] And it might well be argued that
the Greek who did so had scarcely anything more definite in mind than a
Roman who worshiped Salus, the abstract principle of safety. In very
many cases the particular function was especially potent in certain
areas, so that a local adjective applied as a divine epithet would sum
up the power desired to be set in motion.

In the actual moment of prayer or propitiation, it was often a matter of
courtesy to ignore the existence of other gods. This makes perhaps a
sufficiently definite phenomenon to justify the application to it of the
special name “henotheism” long ago devised by Max Müller;[13] and in
henotheism we have very likely the germ of monotheism. But when not
actually engaged in worship, the Greek was well aware that there were
many gods, and that there were differences among them, and this quite
apart from the myths, to which, as has been said, no very great
importance can be attached in this connection. The differences in power
and prominence of deities were perhaps not original, but they had arisen
quickly and generally.

One difference particularly, that between gods and heroes, seems to have
been real to the popular mind. A difference in the terminology that
described the ritual act, and a difference in the act itself, point to a
real distinction between the two divine conceptions.[14]

Who and what the heroes actually were is an extremely doubtful matter.
That some of them were originally men is a proposition with which legend
has made us familiar.[15] We shall recur later to the common heroization
of the dead. That some of them were undoubted gods has been amply
established.[16] It may well be that they were deities of a narrowly
limited territory, knowledge of whom, for one reason or another,
remained sharply circumscribed for a long time, so that when they came
later within the range of myth-making they could not be readily fitted
into any divine scheme. Often the name that appears in some legends as a
hero appears in others as an epithet or cult-title of a better-known
god. This fact may be variously interpreted. At least one interpretation
derives this fusion of names from the fact that the worshipers of the
later deity invaded the cult-home of the earlier, and ultimately
degraded the latter to accessory rank. Or it may be taken as a
compromise of existing claims. At any rate, in some of the heroes we
seem to reach an element somewhat closer to the religious consciousness
of the Greek masses. And if the gods, or most of them, are heroes who
owe their promotion to a fortunate accident rather than to any inherent
superiority, we may discover the fundamental divine conceptions of the
Greeks in the traits that especially mark the heroes: sharp local
limitation, absence of personal lineaments, adoration based upon power
for evil as well as for good.[17]

It was because of this last fact that Greek poets could deal freely with
gods and heroes in the narratives they created. The divine name
possessed none of the ineffable sanctity it has for us by thousands of
years of tradition. Except during the performance of the ritual act, the
god’s presence and power were not vividly felt, and it would have been
considered preposterous to suppose that he resented as compromising an
idle tale from which he suffered no impairment of worship. That the gods
really existed, and that honor was to be paid them after the ancestral
manner, was more than the essence, it was the totality, of popular Greek
theology. Speculation as to the real nature of gods and the world, the
mass of citizens would have regarded as the most futile form of
triviality.[18]

But there were some who thought otherwise. Many thoughtful men must have
felt the absurdities and immoralities of the myths as keenly as we do.
Xenophanes[19] protests, and no doubt not first of all men, against
them. Further, with the earliest stirrings of cosmic speculation in
Ionia, systems of theology are proposed that dispense with demiurges and
administrators. Intellectually developed men cannot have been long in
ridding themselves of popular conceptions that violated the most
elementary reflection. To be sure, the philosopher did not always feel
free to carry his conviction to the point of openly disregarding the
established forms. To do so would bring him into conflict with other
institutions that he valued, and with which religious forms had become
inextricably bound up. But his own beliefs took broader and broader
ground, and well before Alexander became monotheism, pantheism, or
agnosticism.[20]

All these standpoints must be kept in mind when we deal with the
conflict between Greek and Jew: the popular one, no doubt rooted in a
primitive animism, to which the gods were of indifferent and somewhat
shifting personality, but to which the ritual act was vital; the
attitude of poetry and folk-lore, in which divine persons appeared
freely as actors, but in which each poem or legend was an end in itself
unrelated to any other; and finally the philosophic analysis, which did
not notably differ in result from similar processes of our own day.

We find the Hellenic world in possession of very many gods. Some of them
are found practically wherever there were Greeks, although the degree of
veneration they received in the different Greek communities varied
greatly. However, such common gods did exist, and their existence
involves the consideration of the spread of worships.

It is of course quite possible that the common gods grew out of the
personification of natural phenomena, the solar-myth theory, on which
nineteenth-century scholars sharpened their ingenuity.[21] It may be,
too, that one or more of them are the national gods of the conquering
Hellenes, whensoever and howsoever such a conquest may have taken place.
Some may have been of relatively late importation. The Greeks lived in
territory open to streams of influence from every point of the compass.
Of one such importation we know some details—the worship of
Dionysus.[22] Of others, such as Aphrodite, we suspect a Semitic origin
by way of Cyprus.[23] It will be noticed that the names of most of the
common gods are difficult to trace to Greek roots, a fact in itself of
some significance.

We must remember that the wandering of the god is often merely the
wandering of a name. That is especially true in those cases in which an
old divine name becomes the epithet or cult-title of the intruding
deity. Here obviously there was no change in the nature of the god
worshiped and no interruption of his worship. It is very likely, too,
that very few deities ever completely disappeared, even when there was a
real migration of a god. The new god took his place by the side of the
old one, and relations of many kinds, superior or inferior, were
speedily devised. So at Athens, in the contest between Poseidon and
Athena, permanently recorded on the west pediment of the Parthenon, the
triumph of Athena merely gave her a privilege. The defeated Poseidon
remained in uninterrupted possession of shrine and votaries.

How did the worship of certain gods spread? One answer is obvious: by
the migration of their votaries. Locally limited as the operation of the
divinity was, in normal circumstances there never was a doubt that it
could transcend those limits when the circumstances ceased to be normal.
And that certainly took place when the community of which the god was a
member changed its residence. The methods of propitiation, as
crystallized into the inherited ritual, and the divine name, in which,
for the rank and file, the individuality of the god existed, would be
continued, though they were subject to new influences, and not
infrequently suffered a sea-change.

But migration of all or some of the worshipers of a given deity was not
the only way by which the god himself moved from place to place. Exotic
rituals, as soon as men became acquainted with them, had attractions of
their own, especially if they contained features that made a direct
sensational appeal. The medium of transference may have been the
constantly increasing commerce, which brought strangers into every city
at various times. In all Greek communities there was a large number of
“disinherited”—metics, emancipated slaves, suffrageless plebs—to whom
the established gods seemed cold and aloof, or who had only a limited
share in the performance of the established ritual. These men perhaps
were the first to welcome newer rituals, which it was safer to introduce
when they were directed to newer gods.[24] They were assisted in doing
this by the long-noted tolerance Greeks exhibited toward other religious
observances, a tolerance which Christian Europe has taught us to
consider strange and exceptional.

That tolerance was not altogether an inference from polytheism itself.
Polytheism, to be sure, takes for granted the existence of other gods in
other localities, but it does not follow that it permits the entrance of
one god into the jurisdiction of another. And it was not universal.
Among communities inhospitable in other respects it did not prevail. But
it was the general rule, because the conception of ἀσέβεια,
of “impiety,”[25] was largely the same everywhere. Impiety was such
conduct as prevented or corrupted the established forms of divine
communication. The introduction of new deities was an indictable offense
at Athens only so far as it displaced the old ones. Where no such danger
was apprehended, no charge would lie. The traditions that describe the
bitter opposition which the introduction of Dionysus encountered in many
places, are too uniform to be discredited.[26] But the opposition was
directed to the grave social derangements that doubtless attended the
adoption by many of an enthusiastic ritual. The opposition cannot have
been general nor of long duration, since the worship of Dionysus spread
with extraordinary rapidity, and covered the whole Greek world.

Religious movements curiously like the “revivals” of medieval and modern
times visited Greece as they visit most organized communities. One of
the most important of these, which gradually spread over Greece during
the sixth and fifth centuries B.C.E., must be reserved for later
treatment. We may note here merely that there had been present from very
early times the nuclei of a more intense religious life than any that
could be experienced through the rather perfunctory solemnities of the
state cults. These were the mysteries, of which the most famous were the
Eleusinian in Attica. Some assign the latter to an Egyptian origin.[27]
Wherever they came from, they had assumed a large place in the
imagination of Greeks as early as the eighth century;[28] and they
gained their adherents not so much by wrapping themselves in
impenetrable secrecy as by promising their participants an otherwise
unattainable degree of divine favor. Other mysteries existed elsewhere,
possibly modeled upon the Eleusinian. All, however, made similar claims.
It was in the form of mysteries that the emotional side of religion was
deepened. Further, the organization of these mysteries exercised a
profound influence upon all propagandizing movements, whether religious
or not. It is not unlikely that the earliest organization of the
Christian ecclesiae was, at least in part, influenced by the
organization of the mysteries, whether of Eleusis or of some other sort.

It has been said that one commonly worshiped group of heroes were
frankly and concededly dead men. It needs no demonstration to make clear
that such worship of the dead must of necessity be very old; but at many
places in the Greek world this ancient worship of the dead had become
much weakened. The Homeric poems, for example, know it only in a very
attenuated form.[29] At many other places, on the other hand, it
flourished vigorously and continuously from the earliest times. The
application of the word ἥρως, “hero,” to the dead may have
had very ancient sanction. In later times, the term appears very
commonly,[30] and undoubtedly claims for the persons so qualified the
essential characteristics of other heroes—_i.e._ immortality, the
primary divine quality in Homer, and greatly increased power. It
involved no difficulty to the Greek mind to make this claim, for it was
a very common, perhaps universal, belief that gods and men were akin,
that they were the same in nature. Perhaps the very oldest of
transcendental beliefs is that the all-overwhelming phenomenon of death
is not an annihilation, and that something survives, even if only as a
shadow in the House of Hades. When men began to speculate actively upon
the real results of bodily death, it must have occurred to many that the
vaguely enlarged scope of such life as did survive was a return to a
former and essential divinity.[31]

But from a hero, limited and obscure, to a god, seated in full
effulgence at the table of Zeus, was a big step, and bigger yet was the
deification of living men. It may even be that the latter conception was
not Greek, but was borrowed from Egypt or Mesopotamia. There is no
indication of its presence before Alexander. That a man in the flesh
might be translated from mortality to immortality—_entrückt_—was a very
ancient conviction. The son-in-law of Zeus, Menelaos, had been so
privileged.[32] A poetic hyperbole claimed as much for the tyrannicide
Harmodius.[33] There were others, of no special moment, who by popular
legend had walked among men and were not found, as in later times
happened to Arthur and Barbarossa. But they became as gods only by their
translation. We do not meet in Greece for centuries men who ventured to
claim for themselves in the visible body that measure of divinity. In
Egypt, however, and Mesopotamia the conception was not new. Certainly
Pharaoh did not wait to receive his divine character from the hand of
the embalmer. He was at all times Very God. At both the Euphrates and
the Nile, Alexander found ample precedent for the assumption of divine
honors, to which he no doubt sincerely believed he had every claim. We
know how he derived his descent, without contradiction from his mother
Olympias. It was novel doctrine for Greeks, but the avidity with which
it was accepted and imitated showed that it did not absolutely clash
with Greek manner of thought.

After Alexander, every king or princelet who appeared with sufficient
force to overawe a town could scarcely avoid the formal decree of
divinity. The Ptolemies quietly stepped—though not at once—into the
throne and prerogatives of Ra. Seleucus adopted Apollo as his ancestor,
and his grandson took Θεός, “the God,” as his title. His
line maintained a shadowy relation with Marduk and Nebo of Babylon.
Demetrius the Besieger had only to show himself at Athens to be advanced
into Olympus.

The religion briefly and imperfectly sketched in this chapter was not
really a system at all. There is a deal of incoherency in it, of
cross-purposes and contradiction. There was no priestly caste among the
Greeks to gather into a system the confused threads of religious
thinking. Its ethical bearings came largely through the idea of the
state, in which religion was a highly important constituent. There was
also a personal and emotional side to Greek religion, and in particular
cases the adoration of the worshiper was doubtless the sacrifice of a
broken and contrite heart, and not the blood of bullocks. But the
crudities of animism cropped out in many places, and in the loftiest of
Greek prayers there is no note like “Thou shalt love the Lord thy God
with all thy heart, and all thy soul, and all thy might.” In its most
developed form a Greek’s dependence on his god was resignation, not
self-immolation.


------------------------------------------------------------------------




                               CHAPTER II

                        ROMAN RELIGIOUS CONCEPTS


Roman religious ideas were in many respects like those of the Greeks,
partly because they were borrowed from the Greeks and partly because
they were common to all the nations of the Mediterranean world. It may
even be that some of these common forms are categories which the human
mind by its constitution imposes upon some classes of phenomena,
_Grundideen_, as ethnologists call them.[34] Among both Romans and
Greeks we shall find deities sharply limited in their spheres, we shall
find the religious act exhausted in the ritual communion, we shall find
evanescent personalities among the gods. But all these things will be
found in a far different degree, and at various periods many other
matters will demand consideration which the Greeks did not know at all
or knew to a slighter extent.

The differences in national development would of themselves require
differences of treatment. Greek religion grew up in countless
independent communities, which advanced in civilization at very
different rates. Roman religion was developed within a single civic
group, and was ultimately swamped by the institutions with which it came
into contact. Again, it is much more necessary among the Romans than
among the Greeks to distinguish clearly between periods. Roman political
history passed through points of obvious crisis, and many institutions
were plainly deflected at these points into quite new paths of
development.

Real comprehension of Roman religion is a matter of recent growth.
During the vogue of comparative mythology, the Roman myths were
principally discussed, and the patent fact that these were mere
translations from the Greek seemed a complete summing up of Roman
religion. It is only when the actual Roman calendar, as recorded on
stone during the reign of Augustus, came to be studied that the real
character of Roman religion began to be apprehended.[35]

The results of this study have made it clear that during the highest
development of the Roman state the official religious ritual was based
upon pastoral and agricultural conditions that could scarcely be reached
even in imagination. Propitiatory and dramatic rites carried out with
painful precision, unintelligible formularies carefully repeated,
ceremonial dances in which every posture was subject to exact
regulation, all these things indicate an anxious solicitude for form
that is ordinarily more characteristic of magic than of religion. Now,
magic and religion have no very definite limits in anthropological
discussions, but most of those who use the terms will probably agree
that magic is coercive, and religion is not. We shall see at various
points in Roman religion that a coercive idea was really present in the
Romans’ relation with the gods, and that it followed in a measure from
the way the gods were conceived.[36]

The personality of the Greek gods was not so sharply individualized as
the myths we happen to know would indicate, but the gods were persons.
That is, during the act of prayer and sacrifice there was conjured up in
the mind of the worshiper a definite anthropomorphic figure, who dealt
with him somewhat as a flesh and blood man would do. But what was
present in a Roman’s mind in very early times—those of the kingdom and
the early republic—was probably not at all like this. The name of his
deity was often an abstraction, and even when this was not verbally the
case, the idea was an abstract one. And this abstraction had so little
plastic form that he was scarcely certain of the being’s sex to which he
addressed words of very real supplication, and wholly uncertain what, if
any, concrete manifestation the god might make of his presence.[37]

But it will be well to understand that this abstraction, which the Roman
knew as Salus, or Fortuna, or Victoria, was not a philosophic
achievement. It was not a Platonic “idea.” No one could doubt the fact
that in times of danger safety was often attained. The means of
attainment seemed frequently due to chance; that is, to the working of
unintelligible forces. It was to evoke these forces and set them in
operation that the Roman ritual was addressed, and whether these forces
acted of their own mere motion, or whether the formularies contained
potent spells, which compelled their activity, was not really of moment.
That was the nature of the “abstraction” which such words as Fides,
Concordia, and the rest signified to Roman minds.

In the early days a great deal of the religious practice was borrowed
from the Etruscan neighbors, conquerors and subjects of Rome. The
Etruscans, as far as anything can be said definitely about them, were
especial adepts in all the arts by which the aid of deities, however
conceived, could be secured. How much of actual religious teaching they
gave the Romans, that is, how far they actually influenced and trained
the emotions which the sense of being surrounded by powerful and
unaccountable forces must excite, is not yet determinable. But they gave
the Romans, or increased among them, the belief in the efficacy of
formulas, whether of the spoken word or of action.

Although most of the Roman deities were abstractions in the sense just
indicated, many others and very important ones bore personal names.
These names could not help suggesting to intelligent men at all times
that the god who bore one of them was himself a person, that his
manifestations would be in human form, and that his mental make-up was
like their own. Genetic relations between themselves and the gods so
conceived were rapidly enough established. It is very likely, too, that
some of these deities, perhaps Jupiter himself, were brought into Italy
by kinsmen of those who brought Zeus into Greece, although the kinship
must have been extremely remote. And when the gods are persons, stories
about them are inevitable, arising partly as folk-lore and partly from
individual poetic imagining. There are accordingly traces of an
indigenous Roman or Italic mythology, but that mythology was literally
overwhelmed, in relatively early times, by the artistically more
developed one of the Greeks, so that its very existence has been
questioned.[38]

The openness of the Romans to foreign religious influences is an outcome
of a conception, common enough, but more pronounced among the Romans
than anywhere else. In most places the gods were believed to be locally
limited in their sphere of action, and in most places this limitation
was not due to unchangeable necessity but to the choice of residence on
the part of the deity. Since it was a choice, it was subject to
revocation. The actual land, once endeared to god or man, had a powerful
hold upon his affections, vastly more powerful than the corresponding
feeling of to-day, but for either god or man changes might and did
occur.

Both Greeks and Romans held views somewhat of this kind, but the
difference in political development compelled the Roman to face problems
in the relations of the gods that were not presented to the Greeks.
Greek wars were not wars of conquest. They resulted rather in the
acknowledgment on the part of the vanquished of a general superiority.
With barbarians, again, the struggles were connected with colonizing
activity, and, when they were successful, they resulted in the
establishment of a new community, which generally continued the ancient
shrines in all but their names. Roman wars, however, soon became of a
different sort. The newly conquered territory was often annexed—attached
to the city, and ruled from it. To secure the lands so obtained it was
frequently found necessary to destroy the city of which they were once a
part, and that involved the cessation of rites, which the gods would not
be likely to view with composure. The Romans drew the strictly logical
inference that the only solution lay in bringing the gods of the
conquered city to Rome. The Roman legend knew of the solemn words with
which the dictator Camillus began the sack of Veii: “Thou, Queen Juno,
who now dwellest in Veii, I beseech thee, follow our victorious troops
into the city that is now ours, and will soon be thine, where a temple
worthy of thy majesty will receive thee.”[39] But besides this legendary
incident, we have an actual formula quoted by Macrobius from the book of
a certain Furius,[40] probably the contemporary of the younger
Africanus. The formula, indubitably ancient and general, is given as
Africanus himself may have recited it before the destruction of Carthage
in 146 B.C.E., and it is so significant that we shall give it in full:

    Whoever thou art, whether god or goddess, in whose ward the people
    and city of Carthage are, and thou above all, who hast accepted the
    wardship of this city and this people, I beseech, I implore, I beg,
    that ye will desert the people and city of Carthage, that ye will
    abandon the site, the consecrated places and the city, that ye will
    depart from them, overwhelm that people and city with fear, dread,
    and consternation, and graciously come to Rome, to me and my people:
    that our site, our consecrated places, and our city be more
    acceptable and more pleasing in your sight, and that ye may become
    the lords of myself, the Roman people, and my soldiers. Deign to
    make known your will to us. If ye do so, I solemnly promise to erect
    temples in your honor and establish festal games.[41]

What might happen as an incident of warfare could be otherwise effected
as well. We have very old evidence of the entry of Greek deities into
the city of Rome. The Dioscuri came betimes; also Heracles and Apollo,
both perhaps by way of Etruria. And in historical times we have the
well-known official importation of the Great Mother and of
Asclepius.[42]

These importations of Greek gods were at the time conscious receptions
of foreign elements. The foreign god and his ritual were taken over
intact. Greek modes of divine communion, notably the _lectisternium_, or
sacrificial banquet,[43] and the games, were adopted and eagerly
performed by Romans. When Rome reached a position of real primacy in the
Mediterranean, the process of saturation with foreign elements was
accelerated, but with it an opposition movement became apparent, which
saw in them (what they really were) a source of danger for the ancient
Roman institutions. The end of the second Punic war, approximately 200
B.C.E., shortly after a most striking instance of official importation
of cults, that of the Phrygian Cybele, particularly marks a period in
this respect as in so many others. From that time on, the entry of
foreign religions went on apace, but it was somewhat surreptitious, and
was carried on in the train of economic, social, and political movements
of far-reaching effect.

When the Jews came in contact with the Romans, this point had been long
reached. As far, therefore, as the Jews were concerned, their religion
shared whatever feeling of repulsion and distrust foreign religions
excited among certain classes, and equally shared the very catholic
veneration and dread that other classes brought to any system of
worship.

The former classes correspond roughly to those of educated men
generally. Their intellectual outlook was wholly Greek, and all their
thinking took on a Greek dress. But they received Greek ideas, not only
through Homer and Sophocles, but also through Plato and Aristotle. Not
popular Greek religion, but sophisticated religious philosophy, was
brought to the intellectual leaders of Rome. One of the very first works
of Greek thought to be brought to Roman attention was the theory of
Euhemerus, a destructive analysis of the existing myths, not merely in
the details usually circulated, but in respect to the fundamental basis
of myth-making.[44] In these circumstances educated men adopted the
various forms of theism, pantheism, or agnosticism developed by the
Greek philosophical schools, and their interest in the ceremonial of
their ancestral cult became a form of patriotism, in which, however, it
was not always possible to conceal the consciousness of the chasm
between theory and practice.

The other part of the Roman population, which knew Greek myths chiefly
from the stage, could not draw such distinctions. What was left of the
old Italian peasantry perhaps continued the sympathetic and propitiatory
rites that were the substance of the ancient Roman cult. But there
cannot have been a great number of these. The mass of the later plebs, a
mixed multitude in origin, could get little religious excitement out of
the state ritual. What they desired was to be found in the Oriental
cults, which from this time on invaded the city they were destined to
conquer.


------------------------------------------------------------------------




                              CHAPTER III

                    GREEK AND ROMAN CONCEPTS OF RACE


During the nineteenth century a peculiar rigidity was given to the
conception of race through the application of somewhat hastily formed
biological theories. One or another of the current hypotheses on
heredity was deemed an adequate or even necessary explanation, and by
any of them racial characteristics became determined, fixed: race was an
unescapable limiting condition. The Ethiopian could not change his skin.
These ideas, when popularized, corresponded crudely to certain other
ideas already present in men’s minds—ideas that often had a very
different basis. Their lowest manifestation is that form of vicarious
braggadocio which is known as jingoism, racial or national, and is
expressed in the depreciation of everything that concerns other “races.”

Many historians have been influenced by this modern and unyielding
concept of race, and have permitted themselves to make rather large
promises about the destinies of existing groups of men on the basis of
it.[45] But as late as a hundred years ago it was not yet in existence.
The term race then denoted a sum of national and social traits which it
might be difficult to acquire in one generation, but which could readily
be gained in two. Even such disparate ethnic groups as Austrian and
Magyar knew of no impassable chasm that good-will on either side could
not bridge.

It is the latter racial feeling and not the modern one that classical
antiquity knew. Consequently, in the clash of races that took place
during the period with which this book deals, “race” must be understood
as the centuries before the nineteenth understood it. Racial prejudices,
pride of blood, contempt for “slave-nations,” existed and found voice,
but the terms are not coextensive with those of to-day.

It is well-known that a primary Greek distinction was that between
Hellene and barbarian, and it is equally familiar that the distinction
had not been fully formed in the time of Homer. There is no indication
that the Trojans were felt to be fundamentally different from the
Acheans, although it is likely enough that the allies who attacked the
great city of the Troad were of different descent from those that
defended it. The one instance found in Homer of the word
βάρβαρος is in the compound βαρβαρόφωνος, “of
barbarous speech” (Iliad ii. 867), which makes the original meaning of
the word apparent. A Greek was one whose speech was intelligible. All
others were barbarians, “jabberers.” And it is not only incidentally
that Homer fails to make the racial division clear. When he of set
purpose contrasts the two armies, as in Iliad iv. 422-437, it is the
contrast between the silent discipline of the Greeks and the loose,
noisy marshaling of the Trojans: “For all were not of one speech or of a
single language. Mixed were their tongues, since the men came from
far-off lands.”

It is probably in the course of just such expeditions as the Iliad tells
of, a joint movement against a common foe, that a sense of national
unity arose, and it is likely that it came to include many tribes of
different race. We do not know what real basis there is for the
traditional divisions of Ionians, Dorians, and Aeolians. These divisions
have not proved very valuable means of classification to modern students
of Greek dialects. The generic name of Greek to the East was Yavan,
obviously the same as Ionian,[46] and that name indicates where the
first contact took place. The struggles of Greeks to establish
themselves on the coast of Asia Minor probably created the three
traditional groups, by forcing them to combine against threatened
destruction. But there is nothing to show that any real feeling of
common origin and common responsibility existed even here.

On the continent, again, there were large groups of men whom the Greeks
found difficulty in classifying. There were some Epirotes and
Macedonians whose claim to be Greeks was admitted. On the whole,
however, Epirotes and Macedonians were classed as barbarians, though a
different sort of barbarians from Scythian and Phrygian. The first
realization of national unity came with the first great national danger,
the catastrophe that impended from the Persians.

Even then actual invasion did not succeed in combining the Greeks even
temporarily. That was due to the inherent difficulty in interesting
Thessalians or Boeotians in the quarrels of Ionians.[47] In spite of
them, the danger was at that time averted, but it did not therefore
become less real. The consciousness of this ever-present danger and the
bitter experiences of subjection created groups that coalesced more
solidly than ever before about certain leaders, Athenians or Spartans.
In the fifth and fourth centuries, the concept of a Greek race received
a real outline, and the feeling of a common race pride became highly
developed.

This race pride showed itself principally in an over-weening confidence
in the superiority of Greek arms. It is a false notion that represents
the Greek as careless or contemptuously indifferent of the races about
him. Never were men more eager for curious tales of out-of-the-way
peoples. Their earliest historians won their chief success in this way.
But Greeks had beaten back the conquerors of the world, and had
maintained themselves aggressively as well. It was very natural that
something of this attitude was apparent in dealing with barbarians even
on terms of comity. The Greeks had at least colorable ground for
believing that in military matters they were masters wherever they
chose.

One phrase of which some Greek writers were fond need not be taken too
seriously. Barbarians, we are told, are by nature slaves.[48] It would
be an error to attach much importance to the statement. Greeks did not
really believe that Darius or Datames or Hamilcar was servile in
character or in disposition. The expression was merely the facile
chauvinism that military prestige readily stirs up in any nation. So at
certain times some Englishmen were ready to call the French cowards, or
Frenchmen to call Prussians so. Among the Greeks the principal basis for
the statement was the fact that the activity of Greek merchants and
pirates filled every city with slaves of all foreign nations. Indeed the
phrase is no more than a generalized assertion of that state of things.

We shall have to qualify similarly the statement now and then
encountered of a natural and permanent hostility between Greeks and
barbarians. It is a commonplace of Athenian orators, but it practically
always concerns the real hereditary enemy of Greeks, and particularly of
Athens—the Persians. It is in calling the Greeks against their ancient
foe that Isocrates uses the phrase,[49] and in Demosthenes[50] it is
especially based upon the hostilities so long maintained between Athens
and Persia and the ancient grudge Athenians bore for the sack of their
city in 480 B.C.E.

The first achievement of united Hellas was the invasion of Persia,
although it was under Macedonian leadership that this was done, but
soldiers of Alexander appeared as Greeks to the East, and Alexander is
מלך יון, _melek Yavan_, “king of Greece,” in the Book of
Daniel.[51]

Just at this culminating point in the development of Greek nationality,
the process of blurring began. Greek and non-Greek were no sooner
sharply contrasted than by the conscious assimilation policy of
Alexander’s successors the lines tended to obliterate themselves. At
first Greek culture was dominant, but beneath it Syrian, Egyptian, and
Cappadocian obstinately survived, and ultimately, under Christian and
Mohammedan influences, regained their place. It is with one phase of
this specific problem—the threatened submergence of an Asiatic people by
Greek culture—that we are particularly concerned.

The attitude of Romans toward other nations was, as might be expected,
even more arrogantly that of masters and conquerors. But where we find
among Greeks a certain theoretical importance attached to purity of
Hellenic descent[52] (which, by the by, was largely ignored in
practice), the Romans scarcely understood what the term meant. A system
in which emancipated slaves were citizens, who in the second generation
were eligible to high civic honors,[53] and not infrequently attained
them—such a system did not tend to encourage claims to purity of blood.
That does not mean that foreign origin, real or suspected, could not at
any time become a handle for abuse. Cicero fastens on the Celtic strain
in Piso’s lineage with savage delight, just as Demosthenes’ enemies
rarely forgot to remind him of his Scythian grandfather.[54] But these
are not matters of real significance. The significant fact was that they
who were Liby-Phoenicians in one generation were descendants of Romulus
in the next.[55]

_Sumus Romani qui fuimus ante Rudini_, “We are Romans, we who formerly
were Rudinians,” says Ennius,[56] and the metamorphosis was as complete
and as easy if, instead of Italians, they were wholly barbarous elements
that were absorbed. In religious matters the Romans more than the Greeks
felt the efficacy of form. So in political matters the formula of
emancipation and the decree of citizenship were deemed operative of a
real change in the persons affected.

The Roman nobility, it is true, often made pretensions to a purity of
descent that felt every foreign admixture as a stain.[57] But such
claims were absurdly groundless, and cannot really have deceived even
those who maintained them. The great majority of Romans had no quarrel
with any who desired and tried to be Roman. Even Juvenal’s venom is
vented only on the avowed foreigners, who as Greeks, Egyptians, and
Syrians lolled at their ease, while the ragged Cethegi and Cornucanii
munched, standing, the bread of affliction and charity. The leveling
tendencies of the autocracy removed a great many of the reasons of this
friction, and in part succeeded in giving even the Greek-speaking East
and the Latin-speaking West a common culture to maintain. But by that
time new movements of population made such race-concepts as were based
on blood-kinship too plainly out of accord with the facts to be
seriously asserted. At the close of the period we are discussing, every
man was either a Roman citizen, with a pressingly heavy share of the
burden of maintaining the Roman system, or he was not. Who his ancestors
were was wholly forgotten. It had even ceased to be of moment whether he
spoke Greek or Latin or Syriac, Punic, or even Gallic,[58] which had
never completely died out in their ancient homes.

At no time did a feeling of racial kinship make a strong sentimental
appeal. That the whole human race was an extended family was taken as
axiomatic. Striking physical differences did not prevent similarity of
names from proving kinship between Egyptian and Greek and Persian and
Ethiopian. All through Greek history factions in Greek cities called
upon outsiders against their countrymen. The Phoenicians of Utica
preferred the foreign Romans to their Carthaginian kinsmen. Similarly
the Campanians of Capua chose to fraternize with the Libyans and
Phoenicians of Hannibal’s army rather than the closely related
Latins.[59] In these circumstances nothing will lend itself more easily
to distorting our view of the times than the importation into them of
the modern view of race—of that view, at least, in which the historians
of the nineteenth century found so easy and adequate an explanation of
everything they desired to debase or extol.


------------------------------------------------------------------------




                               CHAPTER IV

    SKETCH OF JEWISH HISTORY BETWEEN NEBUCHADNEZZAR AND CONSTANTINE


We have briefly sketched in the foregoing chapters the concepts of race
and religion that Greek and Roman applied to the world about them. These
concepts were not starkly rigid. They changed considerably and often
rapidly in the six centuries our subject covers. They are further to be
qualified by the social environment within which they operated. But it
was not only the Greeks or Romans who in blood and thought passed
through many and profound changes. The Jews, too, developed in many
directions, and this development can no more be lost sight of than the
corresponding one among their neighbors.

In 586 B.C.E. the kingdom of Judah, which had for some years been a
Babylonian dependency, was ended as a political institution, and the
majority of its people, at any rate of the nobles and wealthy men of
them, were forcibly deported to Babylon. The deportation though
extensive was not complete. Some, principally peasants and artisans,
were left, but in districts so long wasted by war their condition can
only have been extremely wretched. Since the whole region was part of
the same huge empire, the old boundary lines were probably obliterated,
and those who lived there subjected to the control of imperial governors
residing in one or another of the walled cities of Syria or Philistia.

Within the next two generations momentous political changes occurred.
The Babylonian empire gave way to a Persian, which, however, can at
first have changed nothing except the personnel of the actual
administrators. According to a very probable tradition, one of the first
acts of Cyrus was to permit, at any rate not to oppose, the remigration
of some of the Judean families or clans to their former homes. Within
the next hundred years a larger and larger number of the families
deported by Nebuchadnezzar likewise returned, though never all of them
and perhaps not even a majority of them. Much of the old territory must
have been found unoccupied, since otherwise conflicts must have arisen
with interests vested within the fifty years and more that had elapsed,
and of these we do not hear. But we do hear of immediate conflicts
between the returned exiles and those who professed to be the
descendants of the Israelites (and Judaites) left by Assyrians (and
Babylonians) on the soil. These latter were beginning to gather about
Shechem, where they must already have been a dominant element, and where
they had created a cult center on Mount Gerizim. The conflict tended to
become compromised in time, until the activities of the reformer Ezra,
backed by the civil governor Nehemiah, again and permanently separated
them.

The returned exiles had from the beginning made the ancient capital
their center, and had succeeded in obtaining permission to rebuild their
ancient shrine. But they were at an obvious disadvantage compared with
their rivals at Shechem, until the city of David could receive the
characteristic of a city—the walls which alone distinguished village or
somewhat more densely populated section of the open country from the
polis or city proper. These, too, were obtained through Nehemiah, and
the prohibition of connubium between the so-called Samaritans of Shechem
and the Jews of Jerusalem was the first aggressive act of the now
self-reliant community.

The system of government of the Persian empire was not oppressive. The
distant king of kings was mainly insistent upon recognition of his
sovereignty and regularity of tribute, less as a means of support than
as an acknowledgment of submission. Within the provinces the satrap was
practically king, and might make his domination light or burdensome as
he chose. We have excellent contemporary evidence that he took his
responsibilities lightly for the most part. In the mountains of Asia
Minor many tribes seem scarcely to have known that they were born
vassals of the Persian king.[60] The local satrap rarely attempted to
control in detail the administrative affairs of the communities in his
charge, particularly when such an attempt would precipitate a rebellion.

In Judea the open plains and low hills rendered it easier for the
governor to emphasize the king’s authority than it was among the
mountains of Cappadocia or the fiords of Cilicia, whose native
_syennesis_, or king, retained both title and authority. We have,
however, a confused and particularly fragmentary record of what actually
happened in the two hundred years that elapsed between Zerubbabel and
Alexander. Changes of great moment in the political, social, and
religious life of the Jews were undoubtedly taking place, since we find
those changes completed a few years later, but we can only conjecture
the stages of the process. On the whole our sources, till considerably
later, are very imperfect. The Persian period forms the largest gap in
the history of the Jews.

A great many Biblical scholars, particularly in Germany, assign to this
period an influence nothing short of fundamental. A large part of the
texts now gathered in the Bible are placed in this time. The extreme
view practically refers the beginning of Jewish history to this date,
and assumes that only a very small part of the older literature and
institutions survived the Babylonian exile. The new community began its
life, it is asserted, with elements almost wholly dependent upon the
civilization of Babylon and Persia.

It is extremely unlikely that this theory is correct. Every individual
assertion of course must be judged in the light of the evidence
presented for it. And on this point it may be sufficient to mention that
the evidence for almost every position is of the feeblest. It consists
largely in apparent inconsistencies of statements or allusions, for
which the theory advanced suggests a hypothetical reconciliation. If
these hypotheses are to be considered scientifically, they at best
present a possible solution and always only one of many possible
solutions. But the general theory suffers from an inconsistency much
graver than those it attempts to remove.

The inconsistency lies in this: The soil of Palestine, never of high
fertility, had greatly deteriorated by the frequent wars of the seventh
century and the neglect and desolation of the following centuries.
Commerce, because of the absence of ports, was practically non-existent.
Those who returned can scarcely have found time for anything else than
the bare problem of living. In these circumstances it is obviously
improbable that a literary activity rich and powerful enough to have
created the masterpieces often assigned to this period can have existed.
The conditions of pioneers do not readily lend themselves to such
activities. City life, an essential prerequisite of high achievements in
art, was being reconstructed very slowly and was confined almost wholly
to Jerusalem. The difficulty is a serious one, and is quite disregarded
by many scholars to whom the bleakness of our records of this time
affords a constant temptation.

Jewish soldiers fought in the armies of their Persian master wherever
these armies went. Some must have been among the Syrian contingent at
Marathon and Plataea.[61] The garrisons of the frontiers contained many
of them. Recently a fortunate accident has disclosed, at the upper
cataract of the Nile, a garrison community of Jews, of which the
records, known as the Assuan and Elephantine papyri,[62] have opened up
quite new vistas in Jewish history. Perhaps the most important point
established is the beginning of the Diaspora. The existence of
communities of Jews outside of Palestine, developing their own
traditions and assimilating their appearance and social customs to those
of their neighbors, is a matter of capital importance for the history of
later Jewry. When such communities multiplied, Jerusalem came more and
more to have a merely religious presidency over them, and the
constitution of Judea itself became determined by that fact, while the
foundations were being laid for the career of religious propaganda later
so successfully undertaken.

The virtual autonomy of the Persian period allowed the development of a
well-organized ruling caste of priests, in which were perhaps included
the Soferim, or Scribes, men learned in the Law, who had no definite
priestly functions. The scope of the high priest’s jurisdiction, the
extent of his powers, may not have been sharply defined as yet. In
itself the presence of a high priest as head of the state was not at all
unusual in that region. As has been said, the interference of the
representative of the Persian sovereign was a variable quantity. In the
second half of the fifth century a Jew, Nehemiah, held the office of
tirshatha, or viceroy, an accident that was of inestimable value to the
growing community, and may have finally secured the threatened political
existence of Jerusalem.

One other political event, of which we have dim and confused accounts,
was a rebellion—whether in or of Judea—under Artaxerxes Ochus (359-338
B.C.E.). The account of Josephus speaks of feuds in the high-priestly
family, the murder of a claimant in the temple precincts, and the
intervention of the all-powerful eunuch Bagoas.[63] That some such thing
happened there can be no reasonable doubt, although we cannot recover
the details. It is, however, unwarranted to make the incident in any way
typical of the fortunes of Judea during Persian rule. There was no
tradition in later times of Persian oppression, nor can even this
rebellion, if rebellion it was, have involved serious repressive
measures, since the Greek invasion a few years later found the Jews
loyal to their overlord.

When the Macedonian Alexander changed the face of the East, the Jews
were swept along with the rest of the loose-jointed empire built by
Cyrus and Darius. Upon Alexander’s death, after uncertainties which the
whole Levant shared, Palestine fell to Egypt, of which it was a natural
geographical appanage as it had been for millennia before. Under the
suzerainty of the Ptolemies the Jewish communities in Egypt received
very considerable reinforcements, and the home-country became a real
national expression, and rapidly attained a relatively high degree of
material well-being, since the practical autonomy of Persian days was
continued. Seized by Antiochus of Asia in the decrepitude of Egypt,
Judea entered with full national consciousness into the heterogeneous
kingdom ruled by a singularly fantastic royal house. A blunder in policy
of the peculiarly fantastic Epiphanes provoked a revolt that was
immediately successful in causing the prompt abandonment of the policy,
and was helped by dynastic chaos to a still larger measure of success.


[Illustration: RUINS OF THE AMPHITHEATER AT GERASA (JERASH) GILEAD,
PALESTINE
(c. _Underwood and Underwood_)]


The leaders of that revolt, the Hasmonai family, produced a succession
of able soldiers. Besides the old Mattathiah and his heroic son Judah,
Jonathan, Simon, and John, by selling their service dearly to this one
or that one of the Syrian pretenders, by understandings with the
ubiquitous Roman emissaries, above all by military skill of the first
order, changed the virtual autonomy of Persian and Ptolemaic times into
a real one, in which Syrian suzerainty was a tradition, active enough
under the vigorous Sidetes, non-existent under the imbecile
Cyzicenus.[64]

During all this time Jews, from personal choice and royal policy, had
extended their dispersion throughout the new cities founded by their
Seleucid masters. Until the battle of Magnesia, 190 B.C.E., Asia Minor
was the real center of the Seleucid monarchy; and in the innumerable
cities established there, Jews in large numbers settled. When Judea
became independent there were probably as many Jews outside of it as
within it.

With the Hasmonean princes—“high priest” is the title which the Hebrew
legend on their coins gives them[65]—the country entered upon a career
of conquest. Galilee, Idumaea, the coast cities of Philistia, portions
of Gilead were seized by John, or Aristobulus, or Alexander, so that
Judea rapidly became one of the important kingdoms of the East, with
which no one could fail to reckon who became active in the affairs of
that region. Rome had backed the Hasmoneans against Syria so long as
Syria presented the possibility of becoming dangerous. But that soon
ceased. By a strange paradox of history the Hellenized East found its
last champion against the Romans in the Persian kings of Pontus, and
when Mithradates was crushed, it could only be a question of the order
in which every fragment of Alexander’s empire would slip into the maw of
the eagles. The Roman liquidator, Pompey, appeared in Asia, and Antioch
became a suburb of Rome.

The pretext for clearing their way to Egypt by taking Judea presented
itself in a disputed succession. The sons of Alexander Jannai were
compelled to accept the arbitrament of the Romans, with the usual
result. The loser in the award, Aristobulus, attempted to make good by
arms what he had lost in the decision. A Roman army promptly invested
Jerusalem, moved by the patent injustice of allowing a capable and
vigorous prince to usurp the place of a submissive weakling. The Roman
general walked into the inner court of the temple, and peered into the
Holy of Holies. He found nothing for his pains, but his act symbolized
the presence of the master, and left a fine harvest of hate and distrust
for the next generations to reap.

From that time on, the history of Judea is the not uncommon one of a
Roman dependency. The political changes are interesting and dramatic but
not of particular importance: vassal kings, docile tetrarchs, finally
superseded by the Roman procurator with all the machinery of his office.
Judea was different only in that her rebellions were more formidable and
obstinate. But Rome had developed a habit of crushing rebellions. Simeon
bar Kosiba, known chiefly as Bar-Kochba, was the last Jew to offer armed
resistance. With his death the political history of Judea comes to an
end.

The religious and social history of the Jews had for many centuries
ceased to be identical with that of their country. It was a minority of
Jews then living that participated in the rebellion of 68, and perhaps a
still smaller fraction that took part in the rising under Trajan and
Hadrian. The interest of all Jews in the fortunes of Judea must at all
times have been lively and deep, but the feeling was different in the
case of non-Palestinian Jews from that of men toward their fatherland.

Meeting for the study of their ancient lore in their “guild-house,” the
_proseucha_, or _schola_, the Jewish citizens of the various cities of
the Roman empire or the Parthian kingdom did not present to their
neighbors a spectacle so unique as to arrest the latter’s attention at
once. They were simply a group of allied cult-communities, sometimes
possessing annoying exemptions or privileges, but not otherwise
exceptional. An exceptional position begins for them when their
privileges are abolished, and their civil rights curtailed, by the
legislation of the early Christian emperors.


------------------------------------------------------------------------




                               CHAPTER V

       INTERNAL DEVELOPMENT OF THE JEWS DURING THE PERSIAN PERIOD


The Jews took to Babylon a highly complicated body of civil law and
religious doctrine. The essence of the latter was an exclusive
monotheism, and that belief was not the possession of a cultured few,
but the accepted credo of the entire nation. No doubt, among the common
people, practices still existed that implied the recognition of
polytheism. No doubt, too, words and phrases occurred in common speech,
in poetry, and in ritual, which had arisen in polytheistic times, and
are fully intelligible only with a polytheistic background. But these
phrases and practices do not imply the survival of polytheism, either as
a whole or in rudimentary form, any more than using the names of the
Teutonic gods for the days of the week commits us to the worship of
those gods, or the various funeral superstitions still in vogue allow
the inference that our present-day religion is a worship of the Di
Manes.

Just as the Jewish religion was in a highly developed form at the time
of the Exile, so the Law was very fully developed. That the entire Law,
as embodied in the Pentateuch, was promulgated by Moses is not
altogether likely, but that any considerable fraction of it is later
than 586 B.C.E. is equally unlikely. Interpolations doubtless occurred
often. To insert into an authoritative text an inference from the words
which the interpolator honestly believed to be true, was not a generally
reprehended practice. Perhaps some of the emphasis upon sacerdotal
organization which parts of the Pentateuch show, may have so been
imported into the constituent codes of the Torah. But on how slight a
scale this was can be readily seen by comparing the Pentateuch with any
of the apocryphal books consciously designed to magnify the
priesthood.[66] The actual civil law bears every mark of high antiquity.
The religious law is at least not inconsistent with such antiquity.

Now neither in civil law nor in religious thought did the community that
slowly formed itself about the acropolis of Zion remain stationary. We
must suppose that the energies of the returning exiles were pretty well
concentrated upon the economic problems before them. But an actual
community they were from the start, and although the communal life was
far from attaining at once to the richness of former days, it contained
all the elements necessary. Without a common law, _i.e._ a regulation of
conflicting claims to property, and without a common cult, _i.e._ a
regulation of the communication between the divine and the human members
of a state, no state was conceivable to the ancient world. Changed
conditions will infallibly modify both, and some of these modifications
it will be necessary to understand.

We possess in the book known as Ben Sira, or Ecclesiasticus,[67] an
invaluable and easily dated record of life as it appeared to a cultured
and wealthy inhabitant of Jerusalem about the year 200 B.C.E. The
incidental references to past time and, above all, the inferences which
may legitimately be drawn about the origins of a society so completely
organized as that of Judea at that time, render recourse to the book a
necessity at many points of our investigation. While accordingly we find
it a convenient terminus in both directions, we must make large
individual qualifications. Ben Sira does not fully represent his time or
his people. He belonged to a definite social stratum. His own studies
and reflections had no doubt developed conclusions that were far from
being generally shared. But he is an eloquent and unimpeachable witness
that the Biblical books had already reached a high measure of sanctity,
and the division later perpetuated in the tripartite canon of Law,
Prophets, and Writings, already existed; and, if nothing else, the
single reference to Isaiah as the prophet of consolation renders it
probable that even so heterogeneous a corpus as the canonical Isaiah was
already extant much as we have it now.[68]

Opinions may differ as to the length of time necessary to permit this
development. But that a very few generations could have sufficed for it
is scarcely credible. Since even the Secondary Canon, that of the
prophets, had already become a rigid one, in which historical
differences in parts of the same book were ignored, the Law must have
been fixed for an even longer time, and the process of interpretation
which every living code requires must have gone on apace for very many
years indeed.

We know very little of the actual agencies by which this process was
effected. The second great code of the Jews was not finally fixed till
200 C.E. We are, however, measurably familiar with the organization of
the judiciary for some two centuries before, but even here there are
distressing gaps, and for the time before Hillel the tradition is
neither clear nor full. All, therefore, that concerns the organization
of the judicial bodies that framed and applied the Law must be
conjectured, and the earliest conjectures embodied in Talmudic tradition
are perhaps as good as any. The development of “houses of prayer” was a
necessity where so many Jewish communities were incapacitated from
sharing in the great cult ceremonies at Jerusalem, and these houses
became a convenience within Palestine and Jerusalem itself. But the
creation of houses of prayer demanded local organization, and with that
local organization gradations of members and the establishment of local
magistrates. There can be little doubt that the organization of the
Greek city-state, familiar to the East for many years, became a model
for these corporately organized communities. Now the judicial function
inherent in the character of ancient magistrates of all descriptions
might easily have been the means of originating that long series of
responsa from which the later Mishnah was finally winnowed. With every
increase of population, power, and governmental machinery, the judicial
system increased in complexity, and the intimate relation which the
civil code bore to the ancient sacred code, as well as the close
penetration of life by religion, tended to render the complexity still
more intricate.

But if the origin of the oral law, in its application at least, can be
made clear to ourselves only by means of such imaginative
reconstruction, we are helped on the side of Jewish religious
development by the possession of at least one fact of prime importance.
The religious system of the Bible knows of a life after death, in Sheol,
but does not know of a survival of personality. Warlock and witch, by
such incantations as were used by Odysseus at the mouth of the dread
cave, or by the wise woman at En-Dor, could give the shadowy ghost
enough outline to be recognizable under his former name, but for the
most part all these flitting spirits were equal and indistinguishable.
But about 100 B.C.E. there was current generally, although not
universally, a very different belief, to wit, that in Sheol, or the
grave, personality was not extinguished, but at most suspended; and that
under certain conditions it might, or certainly would, be permanently
continued. In other words, between the deportation to Babylon and the
culmination of the Hasmonean rule, the belief about life after death had
very considerably changed for most people. And the change was of a
nature that must inevitably have affected conduct, since the
acceptability of man’s life could no longer be proved by the naïvely
simple method of Eliphaz the Temanite,[69] nor yet by the austere
consciousness of rectitude that was the ideal of the prophets.
Transferred to a world beyond perception, reward and penalty gave the
Torah a superhuman sanction, which must have been far more powerful than
we can now readily imagine.

It is idle to look for the origin of this belief in any one series of
influences. For many generations poets and philosophers had swung
themselves in bolder and bolder imagery up to the Deity, which they, as
Jews, conceived in so intense and personal a fashion. Very many passages
in the Bible have seemed to imply a belief in personal immortality and
resurrection, and perhaps do imply such a belief. Nor is it necessary to
assume that these passages are of late origin. Some of them may be, but
one would have to be very certain of the limitations of poetic
exaltation to say just what definite background of belief metaphor and
hyperbole demand. We shall not go far wrong if we assume that even
before the Exile, individual thinkers had conceived, perhaps even
preached, the dogma of personal immortality. Its general acceptance
among the people occurred in the period previously mentioned. Its
official authorization took place much later in the final triumph of
Pharisaism.

Personal immortality and resurrection of the body are kindred, but not
identical, conceptions. Of the two, resurrection is probably the older,
and resurrection, we may note, implies a real suspension of personality,
when the body is dissolved in death. But the body may be recombined,
and, when that occurs, the personal life is renewed. The exact time must
have been very differently conceived by different men. A great many,
however, had already very definite fancies—one can hardly say beliefs—as
to the great day that would deliver the souls from Sheol. That such a
great day would come, on which the whole cosmos would be permanently
readjusted, is the essence of all eschatology. It was only natural that
all other hopes of the people should tend to be combined with it; and of
these hopes the principal one was the Messianic hope.

It is obvious that no adequate discussion of the development of this
hope can be given here, even if our fragmentary sources permitted such
discussion. The most that can be done is to state the situation briefly.
It is all the more important, as the Messianic idea was the source of
the most powerful political movements among the people, and the direct
occasion of at least one of the desperate insurrections of the Jews.

Many nations look back to a golden age of power and prosperity, and
forward to a future restoration of it. The Jews likewise never forgot
the kingdom of David and Solomon, and saw no reason to despair of its
return. As a matter of fact, the Hasmonean rule at its greatest extent
was practically such a restoration. But conditions and people had
radically changed between David and Alexander Jannai. In 1000 B.C.E. it
was a mighty achievement for the small tribal confederation to have
dominated its corner of the Levant, to have held in check the powerful
coast cities of Philistia, to have been sought in alliance by Tyre and
Egypt. In 100 B.C.E., men’s minds had long been accustomed to the rise
and fall of great empires. Assyria, Babylonia, Persia, Macedon, Egypt,
Syria, Athens, and Sparta, and in the distant west Carthage and Rome,
had at different times been lords of many lands. The Judean kingdom
itself had arisen from the wreckage of such an empire. It was
accordingly a different political ideal that filled the imagination of
every nation at this time. To secure and maintain the independence of a
few square miles of semi-arid soil between the Jordan and the Sea was no
deed to puff men with inordinate pride, however difficult of actual
accomplishment it was. As a step toward larger deeds, however, it was
notable enough.

What was the larger deed, and how was it to be accomplished? However
disproportionate it may seem to us, it was nothing else than the
dominion over the whole world, to be accomplished by sudden and
miraculous conversion of men’s souls for the most part, or by force of
arms, if it should prove necessary. And, as was natural enough, it was
in the ancient royal line, the stock of David, that the leader, the
Anointed of God, was to be found.

The family of David, which was still important and powerful when
Zechariah xii. was written (perhaps the fourth century B.C.E.), had
evidently since fallen on evil days. It cannot, of course, have entirely
disappeared, but no member of undoubted Davidic lineage arises to make
political pretensions. It is even likely that, in the absence of
adequate records, and with the loss of importance which the family
suffered during the fourth and third centuries B.C.E., it had become
impossible for anyone to prove descent from David.

None the less, perhaps because of the decline of the family, popular
imagination clung to the royal house. In the bitter days of exile, the
writer of Psalm lxxxix. loses no faith in the destiny of David’s line:

               I have made a covenant with My chosen,
               I have sworn unto David, My servant,
               Thy seed will I establish forever,
               And build up thy throne to all generations.

So the author of First Maccabees, a loyal supporter of a non-Davidic
dynasty, puts in the mouth of the dying Mattathiah the acknowledgment of
the ultimate sovereignty of the ancient house: “David for being merciful
possessed the throne of an everlasting kingdom” (I Macc. ii. 57).

The certainty of this high destiny grew inversely with the political
fortunes of the people. But when even the Hasmoneans fell, and Judea, so
far from increasing the possessions of Solomon, found herself a
hopelessly insignificant fraction of a huge empire, it was not merely
the political side of the Messianic idea that fed upon its
non-realization. Obscure economic and religious factors had long been
operative, and all these raised popular temper to a point of high and,
as it proved, destructive tension. It must always be remembered that
those who undertook to lead the people against the Romans did not aim at
the restoration of the Hasmonean or even Solomonic kingdom. The
establishment of a throne in Jerusalem was the first step of that
triumphant march through the world which would inaugurate the reign of
the God-anointed son of David. The Judean zealots fought for no mean
prize.

The Jews who came into contact with Greeks and Romans were a people
whose development had been continuous from the earliest times. The
cataclysms of their history had produced disturbances, but no break in
their institutional growth. To the civil codes of the ancient polity
they were in the process of adding a new body of law based upon judicial
decisions. To the ethical monotheism of their former development the
popular mind was adding a belief in personal immortality and bodily
resurrection. Folk-lore and superstitions on one side, and speculative
philosophy on the other, were busy here, as they were busy everywhere,
in modifying the attitude of the people toward the established religion.

Finally the Messianic idea was gaining strength and form. In essence a
hope for future prosperity, it had united in itself all the dreams and
fancies of the people, which had arisen in many ways. It became in the
end the dream of a world-monarchy, in which a scion of David’s line
would be king of kings and give law to the world from Jerusalem. The
ushering in of that era soon became a great day of judgment affecting
the whole universe and ardently desired to correct the oppressive evils
of actual life.


------------------------------------------------------------------------




                               CHAPTER VI

                THE FIRST CONTACT BETWEEN GREEK AND JEW


Jews came into the occidental horizon as part of a larger whole. That
whole was known as Syria. Unfortunately Syria itself is a very vague
term, and is without real ethnographic or geographic unity. It might
include Mesopotamia and all the intervening region between the Taurus
and Egypt. One might suppose that with such a people as the Phoenicians
Greek dealings had been so extensive and frequent that it was impossible
to call them out of their name, but Tyrians too are considered and
spoken of as branches of the Syrians. The name soon became practically a
descriptive epithet, more or less derogatory in its implication.[70]

The lower part of the region between the Taurus and Sinai was known to
Greeks as Syria Palaestina, a name almost certainly derived from the
Philistine cities whose position on the coast and whose origin made them
familiar to traders. The Greeks knew, of course, that variously
denominated tribes occupied the hinterland, but what little they knew
about them did not until somewhat later get into the literary fragments
that have come down to us. Perhaps they would not even have been
surprised to learn that here, as in Asia Minor, a very large number of
peoples had settled and fought and jumbled one another into what seemed
to superficial outsiders a common group of Syrians.

The particular section later occupied by the Jews had itself been the
scene of a racial babel. The Israelites were, by their tradition,
expressly commanded to dispossess Hittite, Girgashite, Canaanite,
Amorite, Perizzite, Hivite, and Jebusite.[71] The recurrence of this
enumeration indicates an historical basis for the tradition. It is very
likely that nations so named were actually subdued by the invading
Hebrews. The fact that the tribes dispossessed are seven in number makes
caution necessary in accepting the statement. Perhaps some of these
“nations” are different names for the same group. Some of them, _e.g._
Hittite or Amorite, may be vague descriptive terms, like Syrian or even
Hebrew.

Then there were the Phoenicians, representing perhaps the first Semitic
invasion of this territory. Below them, the Philistines, “from
Caphthor,” who are very plausibly identified with Cretans or “Minoans,”
the Keftiu of the Egyptians.[72] During Mesopotamian and Egyptian
sovereignty, Mesopotamian and Egyptian infiltration may be safely
assumed. The desert never ceased to contribute its share of tribes.
Permanent results of such nomad invasions were the settlement of the
various Hebrew tribes—Moab and Edom in the southeast and Israel on both
sides of the Jordan.

If the analogy of other times and places is to be followed, no one of
these groups was ever completely and literally exterminated. Jewish
tradition knows of an attempted extermination—that of the
Amalekites—only as a very exceptional thing. The resultant
nationalities, which in Greek times occupied Palestine, were likely
enough to have been of somewhat mixed origin. When the Greeks came to
know them well, however, the Jews had long been a well-defined group,
frowning upon intermarriage, although it is not likely that the
prohibition of connubium had its source in any importance attached to
racial purity, or that all Jews everywhere were equally strict in
enforcing it.[73]

As has been suggested, the first contact was probably military. Since
Jews served in the Persian armies as far south as Elephantine, they
probably were equally present in the battalions of Datis and of
Mardonius.[74] Another early contact was in the slave-mart, no doubt
both as buyers and the bought. Enterprising Tyrian traders had made
themselves comfortable in Jerusalem before Nehemiah (Neh. xiii. 6), and
human commodities formed the chief merchandise of most commerce. Before
him, perhaps before the Exile, Joel reproaches the Phoenicians with the
words, “The children also of Judah and the children of Jerusalem have ye
sold unto the Grecians.”[75] “Syrus” had become a common slave-name in
Greece in the fifth century, and Syrus might include anything.[76]

All these scattered and uncertain hints do not tend to present a very
clear picture. However, the time was rapidly coming when Greek contact
with “Syria” was to be vastly more intimate.

In the spring of 334 B.C.E., Alexander crossed the Hellespont to carry
out the cherished vision of Isocrates, a united Hellas drastically
stamping out the Persian peril. From the complete success of his efforts
we are wont to date the so-called Hellenistic epoch, the period in which
Greek influences in art, government, and society were dominant. But
Hellenization had in actual fact begun long ago in the domain of art. It
had penetrated central Asia Minor far back in the seventh century
B.C.E.,[77] and the magnificent “satrap-sarcophagus” at Sidon shows how
thoroughly it was appreciated at the very borders of Judea well in the
middle of the fifth century B.C.E.[78] A generation before Alexander the
king of Sidon bore a Greek name.[79]

So the “king of Yavan,” who received the submission of Jerusalem,
passed, on his way to Egypt, among a people to whom the name of Greek
was quite familiar—who had long known of Greek skill in craftsmanship,
Greek prowess on the field of battle, and Greek shrewdness in
bargaining. The new empire, on the dizzy throne of which Alexander
placed himself, seemed to all the East commensurate with the whole
world, and to the kinsmen of the new king of kings and lord of lords all
men were ready enough to grant the deference formerly owed to Persians.

At Alexander’s untimely death it could scarcely have seemed to men that
great changes were impending. On the contrary, the prestige of his
literally miraculous successes, the impress of his powerful and
fascinating personality, continued for a long time. It might be
doubtful—in fact, it must have immediately become uncertain—whether the
persons to whom the actual administration of affairs would fall, would
be of Alexander’s blood. The satraps of the old régime had to some
extent been displaced by the great king’s generals. Every one of these
was convinced that the coveted prize would fall to the strongest or
cleverest or quickest; but for a while a short and troubled truce was
maintained under the shadow of regal authority embodied in the poor fool
Arrhidaeus and the unborn child of Roxane. When the young Alexander was
born, the conditions at Babylon challenged the intriguing of every
court-parasite. Ptolemy, son of Lagos, satrap of Egypt, was the first to
disregard the confused and divided authority of the zany king and his
baby colleague. A general débâcle followed. Palestine suffered more than
others, because it was unfortunately situated on the road to Egypt. But
by about 300 B.C.E. the country was definitely settled as a province of
Egypt, and it entered upon a century of extraordinary and varied growth.

It is just about this time that unmistakable knowledge of the Jews
themselves, as a separate nationality of Syrians, is evidenced in extant
Greek writers. Histories of the nearer and of the remote East,
impressions of travel and concatenation of irresponsible gossip of all
sorts had long been written by Greeks. Some of these may well have
contained reference to the Jews. In the fifth century, Herodotus speaks
of the “Syrians of Palestine” in connection with the rite of
circumcision, which, he claims to know from the testimony of the Syrians
themselves, was derived from Egypt.[80] However, he obviously writes at
second hand, so that we have no means of knowing whether or not he
refers to Jews. That he knew the name Ἰουδαῖοι is not
likely, but the fact that his source was probably a literary one makes
it possible to date the acquaintance of Greeks with the practice of
circumcision in this region, and therefore perhaps with Jews, at least
to the beginning of the fifth century B.C.E.

The peculiar natural phenomena of the Dead Sea attracted the attention
of travelers from very early times. Aristotle discusses it, and after
him—no doubt before him, as well—the collectors of wonder-tales, of
which we have so many later specimens. Interest in the Dead Sea,
however, by no means implied interest in those who dwelt on its borders,
and the story of the bituminous formation on the water and the curious
manner in which it was collected could be and was told without so much
as a mention of the name of Jews.[81]

But they are mentioned, and for the first time in extant Greek writers,
by the famous pupil and successor of Aristotle, Theophrastus of Lesbos.
The passage does not occur in any one of the works of Theophrastus which
we have in bulk, such as the Characters or the Natural History. It is a
quotation made by the Neoplatonic philosopher Porphyrius, who wrote
somewhere about 275 C. E. The quotation may, in accordance with ancient
custom, be of substance rather than verbatim. Faulty memory may have
further diminished its value for our purposes. When we add to these
facts possible uncertainties in the transmission of the text of
Porphyrius, we are in a fair way of realizing from what dubious material
we must piece our knowledge together.

The passage is in itself, except perhaps for one casual phrase,
strangely unimportant, but as the earliest plain reference to Jews in a
Greek writer it deserves citation in full:

    As a matter of fact, if the Jews, those Syrians who still maintain
    the ancient form of animal sacrifice, were to urge us to adopt their
    method, we should probably find the practice repellent. Their system
    is the following: they do not eat of the sacrificial flesh, but burn
    all of it at night, after they have poured a great deal of honey and
    wine upon it. The sacrifice they seek to complete rather rapidly, so
    that the All-Seer may not become a witness of pollution. Throughout
    the entire time, inasmuch as they are philosophers by race, they
    discuss the nature of the Deity among themselves, and spend the
    night in observing the stars, looking up at them and invoking them
    as divine in their prayers.

As Reinach points out,[82] there is scarcely a correct word in this
description considered as an account of actual Jewish sacrificial rites.
If we have a correct, or even approximately correct, version of
Theophrastus’ report, he or his informant was curiously misinformed.
This informant obviously could not have been a Jew. No Jew could have
been so ignorant of the customs of his people. Nor did his statement
come directly from any one who had actually witnessed, from the Court of
the Gentiles, even a small part of a Jewish sacrifice. It may well be
that we have before us an inextricable confusion between Jewish and
other Syrian rites. We are left to wholly uncontrolled speculation, if
we are bent on knowing whence Theophrastus derived the assertions he
makes here.

The important words of the passage are found in the casual phrase
ἅτε φιλόσοφοι τὸ γένος ὄντες, “inasmuch as they are
philosophers by race.” The phrasing indicates that this aspect of the
Jews is not wholly new. Word had come to Theophrastus, and to others
before him, of a Syrian people not far from the coast, whose ritual in
some respects—though the transmission is confused as to what
respects—differed from that of their neighbors, but whose customs were
strikingly different in one particular, that part of their divine
observance was some form of theologic discussion. That, as we know, was
a fact, since “houses of prayer”—we may call them synagogues—already
existed. This reference to them is the one kernel of observed fact in
this whole description, however indirectly obtained.

Now the Greeks of the fourth century knew of esoteric religious
communities, and they knew of nations that professed to be especially
attached to religious practices. But groups of mystae engaged in rapt
spiritual converse were never coextensive with entire nations. And
“religious” nations might be simply those among whom an elaborate state
cult was punctiliously performed. Even theocracies were no unheard-of
thing. Sidon was such a theocracy; _i.e._ theoretically ruled by the god
and administered by his priest.[83] But that too was largely formal, not
strikingly different from the patronage of Athena over Athens. The
Jewish theocracy was a more intensely real matter than this, but that
fact could not have been apparent to either merchant or traveler, from
whom in the last analysis the information about Jews before 300 B.C.E.
must have come. If, therefore, Greeks found something in the religious
customs of the Jews that aroused immediate attention, it was the very
general interest and participation of the masses in the theological
discussion as it was carried on in the synagogues.

This fact alone would justify the use of the term
φιλόσοφοι, “philosophers.” Theology, the knowledge of the
high gods, was an accredited branch of wisdom which the Platonic
Socrates strove with a little too palpable irony to elicit from
Euthyphro.[84] Those who busied themselves with it were properly termed
philosophers, whatever may have been the conclusions they reached. If we
venture to assume that the conclusions which the Jews had long reached
were actually known, Theophrastus’ phrase could only have been
confirmed. An exclusive monotheism was in every sense a philosophic and
not a popular concept.

A contemporary of Theophrastus was Clearchus of Soli in Cyprus. Of his
writings none whatever has survived, except quotations in other books.
Among other works he wrote dialogues more or less after the Platonic
manner, in which his master Aristotle is interlocutor in place of
Socrates. One of these dialogues was marked, no doubt as a subtitle,
περὶ ὕπνου, “On Sleep,” and in this dialogue an encounter
of Aristotle with a Hellenized Jew is described.

We need not seriously consider the question whether such an encounter
actually occurred. It is not in the least likely that it did. The only
inferences that may be drawn from this passage are those that concern
Clearchus.

Aristotle is the narrator, and tells his story, as he takes pains to
say, according to the rules formulated in Rhetoric.[85] He had met a man
in Asia, a Jew of Coele-Syria by birth, but Grecized in speech and in
soul. This Greek or Jew voluntarily sought out Aristotle and his
associates, πειρώμενος αὐτῶν τῆς σοφίας, “to find out
whether they were really as wise as their reputation.” On the whole,
however, he had given rather than received edification.[86]

What it was in this man’s conversation that so strongly aroused the
approval of Clearchus we are not told. Josephus, in whose Contra Apionem
we find the passage, ends here, to tell us briefly that the rest of
Aristotle’s story described the man’s great strength of character and
the admirable self-control of his habits of life. It may be suspected
that Clearchus’ Jew is little more than a mouthpiece for his own ethical
doctrines, a sort of fourth century Ingénu, or Candide.[87] But what he
does actually say is of great interest.

We have here the first mention of the capital in the form Jerusalēmē,
introduced, it may be noted, for its outlandish sound. And we have the
statement, curious enough to our ears, that the Jews are descendants of
Hindu philosophers, who bear the name of Jews in Syria and Calani in
India. Elsewhere Clearchus asserts an exactly similar connection between
the Persian magi and the Hindu gymnosophists.[88] It is obvious that
Clearchus has the caste organization of the magi in mind, and that his
knowledge of Jews is as mediate and remote as that of Theophrastus.

The connection of the Jews with India was evidently a hasty conclusion,
arrived at when knowledge came to the Greeks of the existence of castes
whose function was principally religious. The statement is repeated by a
man who should have known better—Megasthenes, Seleucus’ ambassador to
India. “All that has been written on natural science by the old Greek
philosophers,” he tells us, “may also be found in philosophers outside
of Greece, such as the Hindu Brahmans and the so-called Jews of
Syria.”[89] He is of course quite wrong as to the facts. But his
statement is evidence of the wide currency of the opinion that the Jews
possessed a very special and very profound lore. Megasthenes, it may be
noted, does not state or imply that the Greeks were borrowers. If he had
done so, the writer in whose book we find the citation, Clemens of
Alexandria (about 180 C.E.), would have pounced upon it. Clemens was
eagerly searching for demonstration of the thesis set up by many Jews
and most early Christians, that all Greek science and philosophy were
derived from an imagined early communication between Moses and the first
Asiatic philosophers.[90]

Theophrastus, Clearchus, and Megasthenes, all of them belonging to the
generation of or immediately after Alexander, hold largely the same
views. Influence of one of them upon the others is practically excluded.
We may find in them accordingly such knowledge of the Jews as at about
300 B.C.E. had reached educated Greeks.

If we try to imagine how this information reached them, we are reduced
to pure speculation. It does not seem to have been a common literary
source, although it is likely enough that in the numerous histories of
the East, now lost, casual and inaccurate references were made to the
Jews. And again it is not likely that the vastly increased communication
that followed Alexander’s campaign, at once brought the Jews much more
prominently within the circle of Greek interest. In those days, the
land-passage hugged the sea as closely as the sea-passage hugged the
land. Judea was a little inland country, somewhat out of the line of
direct communication between the Euphrates and the Nile. If then the
current views, expressed as they are by Theophrastus and his
contemporaries, had neither a literary source nor one of direct report,
it can only have spread as an indirect, filtered rumor, perhaps by way
of Phoenicians, Syrians, and Egyptians.

As far as Phoenicians and Syrians are concerned, immediate contact with
the Jews must have existed. Tyrians and Sidonians and Philistines are
frequently mentioned in the post-Exilic books of the Bible.[91] This
contact was not wholly hostile, though it was often so; but if these
nations were the sources of Greek information about the Jews, the
hostility is not apparent. Perhaps in the generations between Zechariah
and Alexander it had disappeared. At all events, it would appear that
the Canaanite neighbors of the Jews really knew very little about them,
except that the Jews were the residents of the hills about Jerusalem,
and that they had highly characteristic religious rites—characteristic
principally in the earnestness with which they were performed.

In Egypt, a country that had never ceased to be in communication with
Greece from very early times, and particularly since the founding of a
Greek city at Naucratis, in Egypt itself, about the middle of the sixth
century B.C.E., there had been communities of Jews from times that
antedated the Persian conquest. Into the situation here, newly
discovered papyri at Assuan and Elephantine allow us a glimpse, but only
a glimpse. Even the little we know includes one case of bitter conflict
between Jews and Egyptians.[92] No doubt it was not the only case of its
kind. Egyptians, we may be sure, knew of the Jews in the communities in
which Jews lived, and one might suppose that Greek visitors to Egypt
would at some time stumble across Jews there. However, our extant
sources, which speak of Egyptians often enough, do not seem to have
recognized the presence of foreign elements in the Egyptian population.
It was reserved for the papyri to show us Persians, Syrians,
Babylonians, and Jews established in the land as individuals and in
groups.

The view of the Jews that represented them as a mystical sect did not
cease when Judea became an important political factor in the East. One
Greek thinker particularly had professed so strange and esoteric a
doctrine that his biographers and critics inevitably looked for the
source of it in non-Greek tribes and especially in those who had
otherwise obtained a reputation for wisdom of various kinds. This was
Pythagoras. Some seventy-five years after Theophrastus, Hermippus of
Smyrna, in his Life of Pythagoras, ascribed certain definite doctrines
of the latter to the Jews and Thracians.[93] Pythagoras as a matter of
fact had traveled extensively, and had brought to his Italian home
little fragments of exotic lore variously derived. That his philosophy
was influenced by them, there is no sufficient proof, much less based
upon them, and the general belief that he was so influenced had probably
no sounder foundation than the indubitable strangeness of the rites he
instituted and his personal mannerisms. But in later times Pythagoras
was a name to conjure with for those who were bent on establishing a
connection between the Jews and the Greeks. Hermippus had numerous
imitators among later Jewish and Christian writers.

We shall of course never be able to discover the particular moment that
marked the first meeting of Jew and Greek. The contact that is indicated
in the words of Theophrastus or Megasthenes is already of some duration.
The term Ἰουδαῖος has a definite meaning for educated
Greeks. It denoted a Syrian sect, living together about their
rock-citadel and akin in doctrine and probably in blood to the Persian
Magi and Hindu gymnosophists. More exact information was scarcely
available. The two non-Judean sections where Jews were to be found,
Babylon and Egypt, were themselves strange and only partially understood
regions to Greeks in spite of their long acquaintance with both of them.


------------------------------------------------------------------------




                              CHAPTER VII

                                 EGYPT


In the relations that subsisted between Jews and Greeks after Alexander,
Egypt plays an important part, so that particular attention must be
directed to that country.

The influence of Egypt upon Palestine is no new thing in its history.
For century after century the mighty empire across Sinai had been the
huge and determining fact in the political destiny of all Palestinian
nations. Indeed Palestine is much more properly within the Egyptian
sphere of culture than the Babylonian. The glamor lasted even when the
Pharaoh had become a broken reed. Men’s minds instinctively turned in
that direction, and the vigor of the relatively youthful Assyria could
not hold imaginations with half the force of the remembered glories of
Thutmose and Ramses.

Egypt had been in Persian times a turbulent province, subdued with
difficulty and demanding constantly renewed subjugation. Shortly before
Alexander’s conquest, Artaxerxes Ochus had reconquered it with brutal
severity. It offered no resistance to the victorious Macedonians. Upon
Alexander himself it exercised an undoubted attraction. The ancient gods
of this most ancient of countries were those best fitted to confirm his
rather raw divinity. From none else than Amon himself, in his isolated
shrine in the desert, he claimed to have received revelation of his
divine lineage. And at the mouth of the Nile he laid the foundation of
the greatest monument he was destined to have, the city of Alexandria.

When Alexander’s satraps proceeded to carve out portions for themselves,
Egypt was seized by Ptolemy, whose quick brain had grasped at once the
advantages accruing from the possession of an inexhaustible granary and
from the relative remoteness of his position. The first contests would
have to be fought in Asia. To attack Egypt meant a costly and carefully
planned expedition, with the hazards of a rear attack. It was attempted,
and it failed. Egypt might, as far as the country itself was concerned,
breathe freely for a while, and give itself the opportunity of
developing its extraordinary resources.

One of Ptolemy’s first aggressive campaigns was the seizure of
Palestine, the natural geographical extension. Judea and Jerusalem fell
into his hands. It is probable, as will be later discussed, that the
story of the capture of the city on the Sabbath is apocryphal. But there
can be no doubt that one of the immediate consequences of the annexation
of Palestine was a greatly increased emigration of Jews, and doubtless
of Palestinians generally, to Egypt. There is the tradition of a
deportation, but it is feebly supported. However, the emigration was
unquestionably vigorously encouraged and stimulated by the king. The new
city needed inhabitants, and Egyptians were as yet looked at askance by
their Macedonian rulers.

From the beginning, a great number of Greeks, Jews, Persians, Syrians,
and Egyptians dwelt side by side in Alexandria. Greeks who now spoke of
Jews could do so at first hand, and they could also obtain at first hand
accounts of Jews from other nations, especially from the Egyptians.
When, therefore, at about this time, Hecataeus of Abdera, a Greek living
in Egypt, wrote a history of that country, he had more to say of the
Jews than that they were a Syrian caste of strange ritual. Indeed his
account of them is so important that it will be briefly summarized.

A pestilence broke out in Egypt, which was popularly attributed to the
neglect of the national cult owing to the presence of foreign elements
in the population. To propitiate the gods, the strangers
(ἀλλόφυλλοι) were expelled. The most distinguished and
energetic, as some say, arrived in Greece led by famous chieftains, of
whom Danaus and Cadmus are the best known. The mass of the population
settled in the neighboring Palestine, which was then a desert.

This colony (ἀποικία) was led by a certain Moses, famous
for his wisdom and valor. He founded several cities, of which Ierosolyma
is now the best-known. Having organized cult and government, he divided
the people into twelve tribes, because he considered that number the
absolutely perfect one, and because it corresponded to the number of
months in the year.

He made no statues of gods, because he regarded as God and Ruler of all
things the heavens that encircled the earth, and accordingly did not
believe that the Deity resembled man in form. The sacrifices he
instituted, the manner of life he prescribed, were different from those
of surrounding nations. This was due to the expulsion they had suffered,
which induced Moses to ordain an inhospitable (μισόξενον)
and inhuman (ἀπάνθρωπον) form of living.

Since the nation was to be directed by priests, he chose for that
purpose men of the highest character and ability. These he instructed,
not merely for their sacerdotal functions, but also for their judicial
and governmental duties. They were to be the guardians of law and
morality.

It is for this reason that the Jews have never had a king, but appoint
as ruler the wisest and ablest of their priests. They call him high
priest (ἀρχιερεύς), and regard him as bearer of the divine
commands, which he announces at the public assemblies and other
meetings. In this matter the Jews are so credulous that they fall to the
ground and adore (προσκυνεῖν) the high priest when he
interprets the divine message. At the end of their laws is written,
“These words, which Moses heard from God, he states to the Jews.”

Moses showed much foresight in military matters, since he compelled the
young men to train themselves by exercises that involved courage and
daring and endurance of privations. In his campaigns he conquered most
of the surrounding territory, which was divided equally among all
citizens, except that the priests received larger shares, so that they
might enjoy greater leisure for their public duties. These allotments
the possessors were forbidden to sell, in order to prevent depopulation
by the creation of great estates. As an additional means to that end he
compelled every one to rear his children, an arrangement that involved
little expense and made the Jews at all times a very populous nation.
Marriage and funeral rites were likewise quite different from those of
their neighbors.

However, many of these ancient customs were modified under Persian, and
more recently under Macedonian, supremacy.[94]

So far Hecataeus of Abdera. The fragment is interesting, not merely as
the first connected account of Jews by a Greek, but also from a number
of facts that are contained implicitly in his narrative.

We have seen, in the previous chapter, what general knowledge of the
Jews educated Greeks had in the latter half of the fourth century.
Hecataeus could scarcely avoid being familiar with that version before
he came to Egypt. That he ever was in Judea there is no evidence. If he
followed his master Ptolemy, he might easily have been there. But the
information he gives was almost certainly obtained in Egypt, and the
sources of that information will be more closely examined.

It is evident at once that some of his facts must have come from
contemporary Jewish sources. His statement of conditions among the Jews
is markedly accurate for the time in which he wrote, although to be sure
these conditions do not date to Moses. The absence of a king, the
presence of a priestly nobility, the judicial functions of the priests,
the compulsory military service, the supremacy of the high priest, and
the veneration accorded to him, are all matters of which only a resident
of Judea can have been cognizant.

Was the source a literary one? Did Hecataeus, writing at about 300
B.C.E., have before him a translation of the Bible or of the Pentateuch
or a part of it? In the first place there is very little reason to
believe that such a translation was current or was needed at this time.
Secondly, the matters mentioned are just those that do not stand out at
all in such a rapid reading of the Bible as a curious Greek might have
given it. To obtain even approximate parallels, single verses of the
Bible must be cited. But the statements of Hecataeus do correspond to
actual conditions in the Judea of his time. We may therefore plausibly
suppose that Hecataeus’ informant was a Greek-speaking Jew, perhaps a
soldier. Certain inaccuracies in the account would not militate against
such a supposition. Whoever it was from whom the information came,
cannot himself have been especially conversant with his national
history. The glorious period of Jewish history was that of the kings, of
David and Solomon. For any Jew to have asserted that no king ever
reigned over them is scarcely conceivable. But that may be an inference
of the Greek and not a statement of the Jew, and that in Egypt there
were Jews crassly ignorant of everything but the facts of their own
time, we can readily enough imagine.[95]

Was there any other source of information? Obviously no Jew told
Hecataeus that his people were descendants of Egyptian outcasts, at
least in the way in which they are here described; no Jew qualified the
institutions of his people as “inhospitable and inhuman”; no Jew
represented his kinsmen as credulous dupes. Plainly these stories are
told from the Egyptian point of view. The first almost surely is. It
constitutes in outline what has often been called the “Egyptian version
of the Exodus.”

As to that version this question at once arises: What are its sources?
Is it a malicious distortion of the Biblical story, or has it an
independent origin in Egyptian traditions?

The former supposition is the one generally accepted. We have seen that
there is little likelihood that a Greek translation of the Pentateuch
existed as early as 300 B.C.E. If then the Egyptian version is
consciously based upon the Jewish story, that story must have been known
to the Egyptians by oral transmission only. Until recently, imagined
difficulties in the way of assuming such a transmission seemed weighty
objections, but all these difficulties have disappeared in the light of
the Assuan and Elephantine papyri. The existence of Jewish communities
in Egypt from pre-Persian times is established by them, and particular
interest centers upon one of them, which alludes to the Passover
celebration and represents the Egyptian Jewries as referring certain
questions to the Palestinian community.[96]

It must be clear that if Passover had been celebrated in Egyptian
surroundings for two centuries, the Egyptian neighbors of the Jews knew
of the feast’s existence and of the occasion it was intended to
celebrate. In those two centuries the elements that make this version an
Egyptian one may easily have arisen. Indeed, it would have been strange
if stories representing the Exodus as anything but the Jewish triumph it
is depicted in the Pentateuch had not circulated widely among Egyptians.

The mere celebration of Passover was apt to make permanent a certain
hostility between the two nations. When we compare Deut. xxiii. 7, “Thou
shalt not abhor an Egyptian,” with Ezra ix. 1, where the customs of the
Egyptians are classed as abominations, and where Egyptian, Moabite, and
Edomite are added to the list of peoples (Deut. vii. 1) to be shunned
and avoided, it is plain that the attitude toward Egyptians had
undergone considerable change in the intervening centuries. It requires
a long period of antagonism to explain the later Alexandrian
anti-Semitism.

At the same time the papyri show other phases of life as well. They
offer instances of amicable relations, even of intermarriage, as well as
instances of hostility, such as that which resulted in the destruction
of the shrine of Yahu at Elephantine. The latter incident is too obscure
to permit us to draw inferences from it. But it is clear that it can no
more be considered typical than the other examples, which show perfectly
free and friendly intercourse.

The story as it appears in Hecataeus, however, does not imply, even in
its unflattering aspects, hostility on the part of the Egyptians. It may
be remembered that the founders of several Greek nations as well as the
Jews were expelled from Egypt on the occasion mentioned. It is easy to
see how Egyptians, learning of Greek and Jewish legends that ascribed
the origin of those nations to themselves, would accept the ascription,
and make it a part of their own stories in a way to flatter the national
vanity.

While therefore the supposition that Egyptians based their version on
the Jewish story of the Exodus as it became known to them is much the
more probable view, the possibility of an independent Egyptian tradition
on the subject is not to be dismissed cavalierly.

The Egyptian records that have come down to us do not often mention
Jews. Careful study has made it plain that the Pharaoh of the oppression
or the Exodus cannot be identified so readily as was formerly done, but
they have shown that the popular traditions about the Hyksos had at
least so much foundation in fact, that about 1580 B.C.E. Ahmose I did
actually drive out the Semitic or half-Semitic conquerors of the
country, and these conquerors are quite plausibly identified with the
Hyksos. Now during the Hyksos period we hear of a ruler named Jacob-Her,
or Jacob-El, and a few centuries after the inscriptions of Mer-ne-ptah
show Israel already established in Palestine. If, in the casual
selection of inscriptions that has been made by the lapse of thirty-five
centuries, these facts appear, it is surely not impossible that in 300
B.C.E. a great many more facts were known. It is not likely that every
Egyptian priest could read the hieroglyphics, but some could, and the
knowledge of a few could easily become common possession.

When Greeks came to Egypt in the train of Alexander and Ptolemy, they
not only brought Jews there, but they found them, as well as the story
just discussed, whether two hundred or twelve hundred years old.

When we meet the Egyptian version again, it is in a form unmistakably
malevolent. A very few years after Hecataeus, an Egyptian priest named
Manetho wrote the history of his people in Greek. His sources were
popular traditions much more than the monuments, but they were at least
partly documentary. Manetho’s book has been lost, and its “fragments,”
as usual, appear in the form of quotations in much later books, where we
must estimate the probabilities of wilful and careless error.

The fragments of especial interest to us are contained in Josephus’
apologetic work known as _Contra Apionem_ (§1, 26-27), where
unfortunately one cannot always distinguish between the statements of
Josephus and those of Manetho.

The essential part of Manetho’s story, as far as we can piece it
together, is that the Exodus of the Jews from Egypt was nothing more nor
less than the defeat and expulsion of certain rebellious Egyptians.
These latter had been isolated from their fellow-men as lepers and
criminals, and had treasonably summoned to their aid the Bedouin Hyksos
from Jerusalem. The Egyptian outcasts were led by a Heliopolitan priest
named Osarsiph, who afterwards changed his name to Moses. After a short
domination over Egypt, they were defeated and expelled, and pursued to
the frontiers of Syria.

If the very indefinite words of Josephus are to be trusted (_Contra
Apionem_, i. 26), Manetho expressly asserts that this account is based
upon what is popularly told of the Jews (τὰ μυθευόμενα καὶ
λεγόμενα περὶ τῶν Ἰουδαίων). Whether Manetho really said so or
not, it is extremely unlikely that it was the case. The account seems
too finished and detailed to have such an origin. It is much more likely
that it is a deliberate invention of Manetho himself, following the
Jewish story with a certain amount of care. As has been suggested, the
name Osarsiph is simply an Egyptian version of Joseph, the name of
Osiris (which often appears as Osar- or Osor- in names)[97] being
substituted for the assumed theophoric element Jo-, a syllable that
would be familiar to all Egyptians in such very common Jewish names as
Johanan and Jonathan.

The “Egyptian version” as we found it in Hecataeus is far from
malevolent. In Manetho it is plainly inspired by hatred. The Jews are
represented as the mongrel offspring of Egyptian outcasts and
half-civilized Bedouins. The vice of unsociability is reasserted,
coupled with a charge of “atheism,” a term we shall have to deal with
later in detail. Moses, or Osarsiph, forbade the Jews “to have any
dealings with anyone whatsoever except their confederates”
(συνωμοσμένοι). That is, of course, more precise than the
words “inhospitable and inhuman manner of life” of Hecataeus, and formed
in ancient times a more serious indictment than in our own.

Now Josephus, of course, is roused to considerable heat by the “silly
lies” of Manetho, although as testimony to the antiquity of his people
the story is grist to his mill. He points out very clearly and correctly
that many of the incidents are admissions that the corresponding
incidents of the Jewish story are essentially true. These admissions do
not prove that Manetho read these matters from the hieroglyphic records,
but merely that he knew the Jewish story, and, except for the confusion
of Moses and Joseph, that he knew it well.

Nearly all Manetho’s details are suggested in some way by the Biblical
story. The leprosy of Osarsiph is probably derived from the story of
Moses (Exodus iv. 7); the convicts in the quarries (οἱ ἐν ταῖς
λᾳοτομίαις), from the bondage which the Jews acknowledged of
themselves (Exodus i. 12-14). Manetho cannot accept Joseph’s rule nor
Pharaoh’s discomfiture at the Red Sea, but, as many other ancient and
modern writers did, he will not absolutely deny what he wishes to avoid,
but prefers to present it in a form less galling to his pride. Osarsiph
did rule over Egypt, but his rule was a chastisement of the Egyptians
for the impiety of King Amenophis, and was effected only by the aid of
foreign mercenaries. Pharaoh did advance to “the river” with a picked
army and then withdraw before the enemy, but it was a voluntary
withdrawal, impelled by his fear of the offended gods.[98]

It is by no means impossible that all the facts implied may have been
learned by Manetho through oral acquaintance with the Jewish story of
the Exodus. But if Manetho acquired his information so, we should expect
confusion in the sequence of events. We should find anachronisms of
various sorts. It is therefore more likely that he had an actual book
before him. Tradition of strong intrinsic probability assigns the
translation of the Pentateuch into Greek to the reign of Philadelphus.
Writing at about 270 B.C.E., Manetho may well have read the Pentateuch,
at least cursorily. Indeed it would be easy to suppose that it was the
circulation in Greek of stories so offensive to Egyptians that specially
moved him to publish his own interpretation of those stories. He was
hardly likely to have made so much of them, if they were merely legends,
scarcely known except to the Jews themselves and their closest
neighbors.

The “Egyptian version” may be said to have been the more successful. The
leprosy of Moses, the founder of the nation, was constantly girded at by
later writers. Tacitus repeats Manetho faithfully in the matter,[99] and
one of the latest pagan writers of whom we have fragments concerning the
Jews, Helladius, makes allusion to the same thing.[100] The point does
not seem to us of capital importance, but among peoples that regarded
bodily defects as obvious signs of divine displeasure in the person
afflicted, it was likely to have weight.

It may, however, be well to remember that both versions were in equal
circulation. To many the Jewish story seemed the more probable. But it
is significant that at the very beginning of the period when the Jews
took a larger share in the life of the Mediterranean world we find Jews
and Egyptians distinctly in conflict. That conflict was destined to
become embittered, but it must not be taken as an epitome of Jewish
relations generally with other nations.


------------------------------------------------------------------------




                              CHAPTER VIII

                        JEWS IN PTOLEMAIC EGYPT


Greek civilization was essentially urban. The city-state, or polis, was
its highest governmental achievement. When, therefore, under Alexander
and Ptolemy, Egypt was to be transferred wholly within the sphere of
Greek culture, it was by means of a polis that this was to be effected.

The same was still more largely true for the other parts of Alexander’s
empire. In Asia and Syria the “Successors” were busy founding, wherever
convenient, cities diversely named. However, in these regions they were
merely continuing, in a somewhat accelerated fashion, a practice begun
long before. In Egypt, on the contrary, it was plain that a modification
of that policy was necessary. There was, to be sure, an ancient Greek
city at one of the western mouths of the Nile, the city of Naucratis.
But that had been founded as an emporium, and due care was taken that it
should be essentially nothing more, that it should acquire no supporting
territory in Egypt. And however important and wealthy Naucratis became,
it remained confined to its foreign trade for its subsistence.[101]
Besides, it had considerably dwindled in 330 B.C.E., so that its claims
could never have been seriously considered by Alexander, in comparison
with his desire to found a new city and in comparison with the much
superior location of Alexandria.

It is not likely that Alexander himself completed the plans for the
organization of the city. That was left to Ptolemy, and it was
accomplished with a modification of the Greek system that illustrates
both the wariness and the foresight of this most astute of Alexander’s
officers.

The essential part of the polis was its organization as a commonwealth,
_i.e._ as a group of citizens, each of whom had a necessary function to
perform in the state. From time immemorial the administration of affairs
was assigned to a boulē, or senate, the actual executives being little
more than committees of the boulē; but at all times an essential element
of the constitution was the confirmation, real or constructive, of all
acts of the boulē by the dēmos, or mass of citizens. The manner in which
the boulē was selected, as well as the extent to which the check
exercised by the dēmos was real, determined the measure of democracy
each polis obtained. However, even in cities which, like Sparta, were in
theory permanent camps, the same view was held of the necessity of these
parts and of their respective functions, so that everywhere, in legal
contemplation, sovereignty resided in the dēmos.[102]

It must not be supposed that all men who lived within the walls of the
city were members of the dēmos. That is a conception of democracy wholly
alien to ancient ideas. The participation of the individual in the state
was a privilege, acquired in the first instance by birth. Side by side
with the citizens was the slave, who was wholly devoid of legal rights,
and the metic, or resident foreigner, who had, as a result of a direct
compact with the state, acquired the right of residence and personal
protection upon the payment of certain specified taxes.

The privilege of citizenship was a complex of rights, to which were
attached certain very definite and sharply emphasized obligations. What
those rights were depended upon the constitution of the given polis.
Where they were fullest, as at Athens, they included voting in the
public assembly, the holding of public office, service on the jury, and
a claim for certain personal privileges, such as admission to the
dramatic performances at the Dionysiac festivals. In other states they
were not quite so extensive, but the obligations were everywhere the
same, _i.e._ payment of taxes and military service. The state was in the
habit of remitting from time to time certain or all of these taxes and
other compulsory services, so that we may say that various grades of
citizens and metics generally existed.

Now Naucratis was just such a polis as this. So were the various
Apameas, Antiochias, Seleucias, Laodiceas, established in Asia and
Syria. It is true that the boulē and dēmos of these cities were the
merest shadows; and actually the despotism of the monarch was as
undoubted as it had been in Persian times. But the shadows were at least
a concession to the Hellenic spirit, and as such were immensely
treasured; nor can it be denied that as long as they remained the
remembrance of free institutions remained as well. At Pergamon, which
the Attalids created, no public act was done except as the deliberate
choice of senate and people.[103]

But when Ptolemy constituted Alexandria, he deliberately departed from
this plan. As has been said, Naucratis had boulē and dēmos and all the
other appurtenances of a well-regulated polis. So had Ptolemais somewhat
later; and many years later, when the emperor Hadrian founded an
Antinois in memory of his dead minion, he likewise made it a full and
complete Greek city. In Alexandria, on the other hand, there is no
trace, till late in Roman times, of a boulē; and of a dēmos as little.
In the great mass of Greek papyri that have come from Egypt there is
nowhere any indication that a senate ever met, or a people ever
assembled, to parody the deliberations of the Athenian ecclesia. In
other words Alexandria was much less a polis than it was a royal
residence, _i.e._ the site of the king’s palace amidst a more densely
gathered group of his subjects.[104]

In externals Alexandria was every inch a city. It had the high walls,
which, as Alcaeus tells us, do _not_ constitute a state. It had the
tribe and deme, or district division, and it had its various grades of
citizens, determined by the duties and imposts to which they were
subjected.

Of its tribe and district division we know some details. There were
probably five tribes, each of which consisted of twelve demes, or
districts, which in turn had twelve phratries, or wards. The tribes were
known by the first five letters of the Greek alphabet. In the absence of
even formal political rights, this division can have been made simply in
the interests of the census and the police. The obligations to pay taxes
and perform military service were very real ones, and their proper
enforcement necessitated some such organization of the city.[105]

Different classes of citizenship were at once created by the
establishment of special taxes and special exemptions. The peculiar
Greek fiscal arrangement known as the liturgy, which made the
performance of certain services to the state a means of compounding for
taxes, was also in vogue. We have records of certain of these classes of
citizens, or inhabitants, and it is at least probable that there were
other classes of which we know nothing.

First of all, there were the Macedones, or Macedonians. These form a
specially privileged group, whose residence was probably by no means
confined to Alexandria. Just what their privileges were we do not know,
but that they lay chiefly in fiscal exemptions of one sort or another,
is almost certain.

Then there were the Alexandreis, or Alexandrians. We know that there
were at least two groups—those that were enrolled in a given tribe, or
deme, and those not so enrolled. We can only conjecture the purpose of
this division, and one conjecture will be mentioned later.

Besides these, there were other men whose legal right to residence was
unquestioned. They were variously designated. We find Persians, Jews,
and other nationalities, qualified with the phrase τῆς
ἐπιγονῆς, which means literally “of the descent,” but the exact
force of which is unknown. This classification procured for those so
termed certain very much valued exemptions. Native Egyptians also were
present, paying a special poll-tax, and no doubt a very large number of
metics and transient foreigners. Greek publicists regarded the presence
of a large number of metics and foreign merchants as a sign of great
prosperity.[106] We may be sure that no burdensome restrictions made the
settling of these classes difficult at Alexandria.

Were the Jews in Alexandria citizens? A great many heated controversies
have been fought on this subject, some of which would surely not have
been entered into if a clearer analysis had been available of what
constituted Alexandrian “citizenship.” As we have seen, the question can
only be framed thus: Did the Jews of that city appear on the census
books as “Alexandreis,” with or without the deme and tribe adjective
after them, or were they classified as Jews, and did they form a
distinct fiscal class by themselves?

The denial of their citizenship is principally based upon distrust of
Josephus, who asserts it. But distrust of Josephus may be carried to an
extravagant degree. Modern writers with pronounced bias may, of course,
be disregarded, but saner investigators have equally allowed themselves
to be guided by disinclination to credit Josephus, and have come to the
conclusion that the Jews were not citizens of Alexandria.

There were of course very many Jews in Alexandria who were not legally
Alexandrians. Josephus’ assertion did not and could not mean that every
Jew in the city was, by the very fact of his residence, an Alexandrian.
Nowhere in the ancient world could citizenship be acquired except by
birth or by special decree. Jews who emigrated from Palestine to
Alexandria, and were permitted to remain there, were metics, and became
Alexandrians only if they were specially awarded that designation. But
that was just as true for a foreign Greek or a foreign Macedonian, since
at Alexandria “Macedonian” was a class of citizenship, not an ethnic
term. Those who assisted in the founding of the city were undoubtedly
classified either as “Macedones” or “Alexandreis,” and the tradition
that Jews were among them is based upon other authority than Josephus.
It is not enough, therefore, if one desires to refute Josephus, to show
that there were Jews in Egypt who were not “Alexandreis.” Undoubtedly
there were thousands of them. But if, in the papyri, we do find Jews
among the “Macedones” and others among the “Alexandreis,” the statements
of Josephus on the subject are strikingly confirmed, for he says no more
than that there were Jews in both these categories.[107]

Of the two classes of Alexandrians, those enrolled in demes and those
not so enrolled, it is likely that the Jewish “Alexandreis” belonged to
the latter class. The former either paid a special district tax, or,
more likely, were charged with the performance of certain district
duties, either religious in their nature, such as the burying of the
pauper dead, or of police character. When Alexandrians were constituted,
not registered in demes, the purpose can only have been to secure
exemption from these local duties, and the example quoted would of
itself indicate why the Jews may have been so exempted.

It was not, however, merely in Alexandria that the Jews settled,
precisely as it was not merely in Greek cities that Greeks were to be
found. That part of Egypt which lay outside the definite civic
communities as they were founded from time to time, was organized in
nomes, in large agricultural districts containing many villages or even
cities. In every instance, however, the administrative unit was the
nome.

These nomes had themselves a history of immemorial antiquity. Some of
them were surely in boundary coincident with the petty nationalities
that antedated the first dynasties. The mass of the population in them
had practically always been peasant-serfs, and continued to be so.
Beside them, in the villages and towns, there lived in Greek times
motley groups of men, whose legal status was determined in a number of
ways. Some were citizens of Alexandria, Ptolemais, etc., and merely
resident in the nome. Others enjoyed certain military and fiscal
privileges, which involved the right of residence. But in all
circumstances, in the elaborate financial organization of Egypt every
resident had certain precise dues to pay, and was marked by a certain
designation.

The military and other settlers whom the Greeks found in Egypt, whether
they were Persians, Jews, Syrians, or Babylonians, retained their
status, _i.e._ they paid taxes and performed services differing from
those of the native Egyptians in part, although no doubt certain taxes
were levied upon all. The foreigners whom Ptolemy invited or brought
into Egypt must have been settled either in the cities or the nomes, and
were given a definite fiscal status. And besides all these various
grades, there were metics—a term which may have included emancipated
slaves, and of course slaves as well—in huge numbers. There can be
little doubt that Jews were to be found in all classes, from the highly
privileged nobility of “Macedones” to the slaves.[108]

In most large Greek cities metics of foreign birth or ancestry existed.
There were Phoenicians and Egyptians in Athens in very early times. But
they were all, together with non-Athenian Greeks, gathered into the
general group of metics, and no one group ever became numerically so
preponderant that a special class had to be legally constituted of them.
In Egypt, however, the general term metic was rarely used. For the nome
organization of the country it seemed scarcely applicable. Instead,
those foreigners who had acquired legal residence and other rights were
known by their national name. So there was a group of Egyptian residents
known as Ἰουδαῖοι, as “Jews,” which was in their case a
legal designation, whereas, when the “Macedones,” “Alexandreis,” etc.,
of the same nationality were referred to as Ἰουδαῖοι, the
term was merely descriptive.

We do not know whether the Ἰουδαîοι that had no other
classification were more numerous or less numerous than those who had.
But it was shortly found advisable to organize the Jewish metics to the
extent of superadding upon their own cult-organizations certain royal
officers responsible to the king. Of these the chief was the ethnarch,
and it is evident that the ethnarch would assume an importance in
proportion to the number under his jurisdiction. The right to have an
ethnarch seems to have been a prized privilege and was not confined to
the Jews. What the relation of the later alabarch[109] was to the
ethnarch is not clear. The two terms may perhaps designate the same
office.

But a complete understanding of the condition of the Jews in Egypt and
Alexandria necessitates some account of the synagogue organization.

There is no reason to question the Jewish tradition that the synagogue
was Exilic or pre-Exilic in origin. In fact, it is not easily
conceivable that it could have been otherwise. Worship was a social act
in the ancient world, and properly to be performed in concert. It was
inevitable therefore that just as soon as the Jews were removed from
those places where the ancestral and traditional ritual was performed
without any conscious organization for that purpose, they would combine
themselves in groups in order to satisfy the strongly marked religious
emotion that characterized them.

Corporate organization, based upon the performance in common of some
religious act, characterized the whole ancient world. The state was
itself a large corporation of this kind, and the local divisions rapidly
assumed, or always possessed, the same form. Obviously members of the
same nationality residing in a foreign city would be specially prone to
organize themselves into such corporations, and as a rule make the
religious bond, which seems to have been a formal requisite, the common
worship of one of their own gods. The merchants of Citium at Athens
formed a guild for the worship of the Cyprian Aphrodite. It was in this
way that Egyptian merchants and artisans made Isis known to the Roman
world.[110]

It has been said that the state itself was such a corporation, of which
the formal basis was the common performance of a certain ritual act.
When new states were founded or new men admitted into old states, a
great deal was made of the act. It follows therefore that when Jews were
admitted into the newly founded civic communities of Asia, as we know
they were, some relation would have to be entered upon between
themselves and the religious basis of the state. In most cases, special
exemption from participation in these religious acts seems to have been
sought and obtained.

In Egypt the conflict between the exclusive worship of Jehovah and the
less intolerant worship of the Nile-gods had been in existence for
centuries before the Greeks. The pre-Greek Jewish immigrants were
perhaps not of the sort that sought to accentuate the conflict, though
friction was unavoidable. At the Greek conquest, it must be remembered,
no great disposition was shown by the first Ptolemies to accept the
native institutions or the native gods. The new god of Alexandria, the
mighty Sarapis, was not, as has been generally supposed, a composite of
Osiris and Apis, but an out and out Greek god, imported from his obscure
shrine in direct opposition to the indigenous gods.[111] Membership in
the civic communities, or residence in the country districts, can have
involved no obligation to share the ritual localized there. Every group
of foreigners might freely disregard it, and maintain unimpaired their
own ancestral forms.

We accordingly find Jewish synagogues—in the sense of
cult-organizations, each having its own meeting-house, schola, or
proseucha, and organized with magistrates and council, like miniature
states—not only in Alexandria but in insignificant little towns of Upper
and Lower Egypt.[112] Nor was the legal basis of such organization
wanting, _i.e._ the corporate personality, since we find these
synagogues enjoying the rights of property and subject to the imposts
levied upon it.[113] The extent of each synagogue was limited by the
physical capacity of the schola. There must have been in Alexandria very
many of them.

Who were members of them? The various classes of Jews in the city and
country were divided by social and legal lines. In the synagogue social
distinctions cannot have disappeared, but there can be no doubt that in
many, if not in all, there would be found Jews representing every class
of the community. In other parts of the Greek world it was no strange
thing to see citizens, metics, foreigners, slaves, claiming membership
in the same cult-organization, and jointly worshiping a native or
foreign god. The synagogue likewise contained among its members nobles
and slaves. The tendency for the wealthier classes to become completely
Hellenized, and so completely to abandon the synagogue, did not show
itself prominently for some time.

We may readily suppose that the native Egyptians regarded all the
foreign invaders with scarcely discriminating hatred. In most cases,
when Greeks and Jews dwelt in the nomes, they were both exempt from
local dues, and both paid the same special tax. What the attitude of the
Egyptians was to their Greek and Macedonian masters, we have no need to
conjecture.[114] As under Persian rule, they rose in bloody riots; and
after a century of Greek domination, they were so far successful that a
complete change in the policy of the Ptolemies was effected. The house
had very rapidly degenerated—a process perhaps hastened by the Egyptian
custom of brother and sister marriage, which they adopted. From the
weaker kings of the close of the third century B.C.E., the Egyptian
priests received a complete surrender. Continuity with the Pharaohs was
consciously sought. The ancient titles in a modified form were adopted
in Greek as well as Egyptian for the rulers. The hieroglyphics
represented Ptolemy as the living god, sprung from Ra, just as they had
done for Amen-hem-et thousands of years before.[115]

But a Hellenizing process had gone on as well as an Egyptizing process.
The irresistible attractions of Greek culture had converted even the
fiercest nationalists into Greeks outwardly, and in the horde of Greek
names that the papyri exhibit we have sometimes far to seek, if we wish
to discover unmistakably Greek stock. Intermarriage and concubinage must
have given Egypt a large mixed-blood population, which no doubt called
itself Greek. Evidences of Greek aloofness on the subject of marriage
have been sought in the denial of connubium by the city of Ptolemais to
foreigners.[116] But that applied to foreign Greeks as well, and was a
common regulation in most Greek cities.

The Hellenizing process affected the Jews even more. In Alexandria the
Jewish community had begun to show signs of the most active intellectual
growth, and the results of that growth, naturally enough, wore a Greek
dress. But that process had been active in Palestine as well, where the
consequences were somewhat more important. It is there that we shall
turn for a study of the first conflicts between Judaism and Hellenism.


------------------------------------------------------------------------




                               CHAPTER IX

            THE STRUGGLE AGAINST GREEK CULTURE IN PALESTINE


While Palestine was a Greco-Egyptian province, the influences at work
over the whole Levant had been as effectually operative there.

In the matter of government no change had been made that was at all
noticeable. The internal autonomy of Persian times had been maintained;
the claims of the tax-collector and recruiting sergeant were dealt with
by the whole community, not by the individual.

Socially and economically, relative peace had permitted considerable
progress. At the close of this period the work of Ben Sira is the best
of all possible evidence, both of the literary productivity out of which
the book arose and of the society which it implies. We are given
glimpses of settled and comfortable life, which could scarcely have been
attained unless the preceding century had been one of constantly
increasing well-being. It is a well-equipped table at which Ben Sira
bids us sit. The graces and little luxuries of life are present, and
equally the vices that went with these luxuries.[117]

Nor had the character of the whole spiritual culture essentially
changed. The language of daily intercourse was Aramaic, the _lingua
franca_ of the whole region. But the literary language was still Hebrew.
It must have been constantly spoken among educated men, for the changes
it continued to exhibit are not such as would occur if it had been quite
divorced from life. And the literary activity, which took its forms from
the established and already canonical literature, took its substance
from the life about it. That this life had been impregnated with Greek
elements, there can of course be no manner of doubt.

Not only the old Philistian and Phoenician cities of the coast had
acquired a Greek varnish, but Judea was being surrounded by a closer and
closer network of new Greek foundations. Ptolemais, Anthedon, Apollonia,
Arethusa, and the cities of the Decapolis across the Jordan, brought the
external forms of Greek culture so near that even the peasant who went
no great distance from his furrow must have encountered them.

What made up the fascination of Greece for the nations she dominated? In
the first place it must be insisted upon that there was a national
resistance, whether or not it took the form of insurrection. Indeed,
insurrection was a thing quite apart from resistance to Hellenism. As we
have seen in the case of Egypt, national resistance to the political
domination of Greeks did not by any means imply national resistance to
the spread of Greek culture. The latter resistance generally took the
form of a dull and obstinate clinging to ancestral ritual and language.
At Antioch in the fourth century C.E., some men and women still spoke
Aramaic, and knew no Greek.[118] It is only within the rather narrow
limits set by wealth and education that the Hellenization was really
effective. Unfortunately most of our available evidence is concerned
with this class.

Among these men, who were naturally open to cultural impressions, the
attraction of Hellenism was undoubted, and had been growing slowly for
years before Alexander, and it had meant for them all the charm of an
intellectual discovery. The mere fact that what the Greeks had was new
and different could have been of no real influence. There must have been
an actual and evident superiority in Greek life or culture to have drawn
to itself so quickly the desires and longings of alien peoples.

In one field that superiority was evident, in the field of art. Whatever
may have been the origins of Greek art, from the seventh century on no
one seriously questioned that Greek workmen could produce, in any
material, more beautiful objects than any other people. Artistic
appreciation is no doubt a plant of slow growth, but the pleasure in
gorgeous coloring, in lifelike modeling, in fine balances of light and
shade, in grouping of masses, is derived immediately from the visual
sensation. No peasant of Asia could fail to be impressed by his first
glimpse of such a city as the Ephesus and Miletus of even the sixth or
fifth century. After the extraordinary artistic progress of the fifth
century had vastly increased the beauty of Greek cities, every foreigner
who visited them must have found greater and greater delight, as his
knowledge grew broader and deeper.

In other branches of art, in music, poetry, dancing, the wealthier
Asiatic had a training of his own. But it is likely that even a slight
acquaintance with Greek taught him to depreciate the achievements of his
own people. Doubtless, in poetic capacity and imagination, Phrygian,
Lydian, or Lycian was the equal of Greek. Yet we have no choice but to
believe that in sheer sensuous beauty of sound, which made a direct
appeal to any partly cultivated ear, no one of the languages could
compare with Greek. Nor is it likely that any written literature existed
in Asia that could be ranked with Greek.

With the appeal to eye and ear there went an appeal to the intellect.
Greek mental capacity was not demonstrably greater than that of the
Asiatic peoples to whom the Greeks were perhaps akin, but both
imagination and reflection had framed their results in systematic form.
The rich narrative material found in every race was available in Greek
in dramatic and finished pieces. The philosophic meditation in which
others had long anticipated the Greeks was among the latter set forth in
clearer and simpler phrasing.

The allurement of all these things was intensified by a franker and
fuller exploitation of all physical instincts, and the absence of many
tabus and forms of asceticism that existed among non-Greek peoples. A
vastly increased freedom over one’s body seemed a characteristic of
Greek life, and a vastly greater freedom of political action was
characteristic of the Greek polis.

It is small wonder therefore that the upper classes of Asia and Syria
had for two or three centuries before the conquest succumbed to a
culture that possessed so visible a sorcery. Then, with the conquest,
came a new factor. To be a Greek was to be a _Herrenmensch_, a member of
the ruling caste, a blood-kinsman of the monarch. Syrians, Asiatics, and
Egyptians found themselves under the direct sway of a Greek dynasty,
supported by a Greek court and army. All the tendencies that had made
Greek cultural elements attractive for certain classes were intensified
by the eager desire of the Greeks to identify themselves with the
dominant race, and this identification seemed by no means impossible of
achievement.

What had to be given up? As far as language was concerned, a smattering
of Greek was the common possession of many men. Every trading-post had
for generations swarmed with Greek merchants. Greek mercenaries were to
be found in most armies. It was no especially difficult matter for those
classes which knew a little Greek to increase their familiarity with it,
to multiply the occasions for its use, to sink more and more the soon
despised vernacular. The latter, we must repeat, was not and could not
be suppressed, but it became the language of peasants. In the cities men
spoke Greek.

But there were other things—the ancestral god and the ancestral
ritual. These were not so readily discarded. However, the attitude of
the Greeks in this matter made it unnecessary to do so. The gods of
Greece were often transplanted, but rarely more than the name. In
Syria and Asia particularly it was only in wholly new foundations that
Greek gods and Greek forms were really established. Generally the
sense of local divine jurisdiction was keenly felt. Greeks had a
wholesome awe of the deity long in possession of a certain section,
and in many cases erected shrines to him, invoking him by the name of
some roughly corresponding Hellenic god. Frequently the old name was
retained as an epithet. Thus Greek and Syrian might approach the
ancient lord of the soil in the ancient manner and so perpetuate a
bond which it was ἀσέβεια, “impiety,” to break.

Since the essentials were maintained, the only step necessary to turn a
Syrian into a Greek was to purchase a himation, change his name of
Matanbal to Apollodorus, and the transformation was complete. He might
be known for several years as “ὁ καὶ _Matanbal_”—“_alias_
Matanbal”; he might suffer a little from the occasional snobbishness of
real Greeks, but, especially if he was wealthy, such matters would be of
short duration. The next generation would probably escape them
altogether, and their children, the young Nicanors, Alexanders,
Demetriuses, would talk glibly of the exploits of their ancestors at
Marathon or under the walls of Troy.

But there was also no inconsiderable group that combined adoption of the
new with loyalty or attempted loyalty to the old. Many Syrians,
Egyptians, Phoenicians, and others, conscious of a history not without
glory, desired to acquire the undeniably attractive Hellenic culture,
while maintaining their racial ties, of which they felt no real reason
to be ashamed. That was particularly true of the Seleucid dominions
where Alexander’s assimilative policy was consistently pursued. Persian
or Lydian or Phoenician descent was a thing many men boasted of. It was
with a sense of adding something to the culture of the world that
natives with Greek training prepared to transmit in Greek forms the
history of their people to Greeks and to interpret their institutions to
them. And they found a ready enough audience. On many points, especially
in religion and philosophy, the Greeks were willing enough to concede a
more profound acquaintance to barbarians than they themselves possessed;
and often the weariness of civilization made Greeks search among fresher
peoples for a sound social life, since that life was tainted, in Greek
communities, by many grave diseases.

But people of this class found themselves in a delicate situation, an
unstable equilibrium constantly disturbed. It was hard to remain a
Grecized Syrian. Generally the temptation to suppress the Syrian was
well-nigh irresistible. Now and then, the rise of national political
movements would claim some of the younger men, so that the fall was on
the native side. In general, the older conservative attitude expressed
itself naturally in avoidance of Greeks as far as possible, and
precisely in proportion to the value set upon the national and
indigenous culture.

The situation of the Jews was only in so far unique that there could be
no question among them of gradual steps in the acquisition of Greek
culture, but only of partial acceptance of it. The final step of
interchanging gods—of accepting the Greek name and maintaining the old
rite and of exercising that reciprocity of religious observance which
was a seeming necessity for those who lived in the same region—that, as
every Jew was aware, could never be taken. The religious development
among the Jews had been fuller than elsewhere, and had resulted in a
highly specialized form, which by that fact had none of the elasticity
of other cult-forms. It was easy to make any one of the Baalim of local
Syrian shrines into Zeus Heliopolitanus, Zeus Damascenus, etc. It was
not possible to turn the Lord Zebaoth of Zion, the awful and holy God of
psalm and prophecy, into an epithet of Zeus or of another.

Consequently Jews who felt the pull of Greek art and literature, who,
like other subjects of Greek sovereigns, were eager to gain the favor of
their masters, had to realize to themselves the qualifications of their
Hellenism, or determine to discard wholly their Judaism. And this latter
step, even to enthusiastic Philhellenes, was intensely difficult. For so
many generations “Thou shalt have no other gods” had been inculcated
into men’s hearts that it was no simple thing to undertake in cold blood
to bow before the abominations of the heathen.

He who could not do that—and there were many—might feel free to adopt
Greek language and dress and name; but, even more than Babylonian and
Egyptian, he was conscious of making a contribution of his own to the
civilization of the East. An inherited wisdom, which was in effect
closer communion with the Absolute, he believed he had, and, as we have
seen, he was generally credited with having. He felt no need therefore
of yielding unreservedly to the claims of Greeks, but might demand from
them the respect due to an independent and considerable culture.

Barriers to mutual comprehension were created by the Jewish dietary
regulations as well as by ritual intolerance. Courtesy and good breeding
however might soften and modify what they could not remove, and social
intercourse between Greek and Jew certainly existed. Nor need we
exaggerate the embarrassments these relations would suffer from the fact
that while a Greek might, and doubtless would, assist at the little
ceremonies of his Jewish neighbor’s household, the Jew might not without
sin reciprocate. By judicious absence on occasion—perhaps by little
compromises—the average easy-going Jewish citizen of an Asiatic or
Egyptian community need not have found himself in constant conflict.

As in the case of other nations, the first Greek-speaking Jews that
desired to emphasize their origin while accepting the all-pervading
Greek culture, wished primarily to convey to Greeks the facts of their
history and institutions. The Septuagint, at least the Pentateuch, was
probably written in the early part of the third century B.C.E., and
although primarily intended for Jews, no doubt came within the knowledge
of Greeks as well. But its purpose was utilitarian. The Greek-speaking
synagogues absolutely needed it. If others were to be acquainted with
the history of the Jews, some other means had to be devised.

About 225 B.C.E., an Egyptian Jew named Demetrius wrote the history of
his people in Greek. Unfortunately we have only such fragments of his
work as Eusebius, the church historian, and Josephus have chosen to
quote; but what we have, permits the conjecture that he wrote in a
concise and simple style, without oratorical embellishment, and
obviously without apologetic motives. It seems to have been a sober and
dignified narrative, the loss of which is a serious gap in our
records.[119]

The name of this man, Demetrius, is not without significance. It
contains the name of a Greek deity, Demeter, so that religious
precisians might find in it an honor—even if only a verbal one—to the
Abomination. But Alexandrian Jews were not likely to be religious
precisians, and we may readily suppose that these names, attrited by
constant use, did not immediately convey the suggestion of being
theophoric. In 238 B.C.E., an Arsinoite slave is named Apollonius or
Jonathas, and about the same time a Jewess is found with the name of
Heraclea.[120]

In the case of Demetrius it was rather the redoubtable Besieger than the
goddess that was honored, just as the very first Jew whom we know by a
Greek name, Antigonus of Socho, is probably named after Demetrius’
father, the one of Alexander’s officers who became so nearly a real
Successor. It is to be noted that Antigonus of Socho is one of the
earliest doctors of the law, whose fine saying is recorded in Abot
i.,[121] and, although we know no Hebrew name for him, there can be no
question here of Hellenizing or partly Hellenizing tendencies.

Otherwise Jews in adopting Greek names were prone to translate them
approximately. The common Jonathan and Nathaniel became Theodotus,
Dositheus, Theodorus, and the like. Phoenicians had long done the same,
but there would be of course no difficulty in the case of the latter if
they chose to turn Meherbal into Diodorus. That the Jews were scarcely
more scrupulous in this matter is a little surprising. It fits in well
however with the conclusion that friction in unessentials was rather
avoided than invited by the average Jew.[122]

The conflict that was preparing itself in Palestine was not one between
Greek and Jew, but between Hellenizing and reactionary elements among
the Jews themselves. And the term reactionary is chosen advisedly. In
the many centuries that had witnessed the slow spread of Hellenism, and
the hundred years or so in which that progress had been immensely
accelerated by the political domination of Greeks, a resistance was also
preparing itself. In the early years of the movement, before and after
Alexander, the numbers affected had been too few to justify active
opposition. But the number became constantly greater, and the imminence
of a real peril became vividly present to thinking men. The method of
opposition was at once indicated. It could be only a conscious
restoration of such national institutions as had lapsed into comparative
disuse, a recultivation of ancient national practices, and a more
intense and active occupation with the traditional sacred literature.

In just this way opposition to the orientalizing of the imperial
religion produced the reactionary reforms of Augustus, and much later
opposition to an excessive clerical interference with life expressed
itself in the very real paganism of the Italian Renaissance. In all
these instances the attempt was deliberately made to rebuild with
material still present, even if largely discarded, a structure that had
fallen into ruins. The success of such movements depends wholly on the
amount of material still present. If it has to be painfully gathered and
swept together from forgotten corners, success is more than problematic.
The Jewish reactionaries were fortunate in that the ancient institutions
still held their ground, and in having no huge gap of disuse to fill.

They were also fortunate that the actively Hellenizing party was limited
in numbers, and the line of demarcation was the easily noticeable one of
wealth and position. Not all men of wealth were in this class. Such a
man as Ben Sira, in whose book some have detected Greek elements,
betrays no Hellenizing tendencies.[123] He is Jew to the marrow, and he
can be no isolated phenomenon. But there had been a rapid growth of a
moneyed class, and this not so much composed of great landowners as of
the newer class of capitalists, who grew rich through the various forms
of financial speculation then open, particularly the tax-farmers, of
whom that magnificent vulture, the Tobiad Joseph, is a permanent
type.[124] The life of these men involved such an association with king
and court that marked discrepancies of social custom, such as dietary
regulations, or any form of abstinence, as well as differences in dress,
were not to be thought of.

It is unfortunate that any discussion of the nature and character of the
opposition involves a controversial question of the first magnitude,
that which concerns the Hasidim, or ʽAssidaei. It were idle to
enumerate, much less to examine critically, the theories that have been
advanced. Our evidence is so scanty that it can be made to fit into many
different schemes, all of which can be shown to be conceivable. The
simplest interpretation of the extant sources however is by far the
best, and it has further the merit of being the longest-established and
most widely current.

Now concerning the Hasidim we have only three passages that can be
considered even approximately contemporary, two in the First Book of
Maccabees and one in the Second.

The first passage, I Macc. ii. 41, states that after the martyrdom of
the loyal Jews who had taken refuge in the desert, there united with
Mattathias the συναγωγὴ Ἁσσιδαίων, “the congregation of
Hasidim, a body of great power and influence in Israel, containing all
those who were devoted to the Law.” In the second passage, I Macc. vii.
12, we read that when the renegade high priest Alcimus and the Greek
prefect Bacchides entered Judah with peaceful overtures, they were met
by the congregation of scribes, who brought their lawsuits to him, and
then recognized his authority. “And the ‘Asidaei were the first among
the children of Israel, and they also sought peace from them. For they
said, ‘A priest has come of the seed of Aaron with a powerful army, and
he will not injure us.’”

Taken together, these passages are best understood to mean that at the
beginning of the Hasmonean revolt an already existing and powerful
group, known as the “ʽAsidaei,” or “Hasidim,” gave their official
support to the Modin rebels, but that upon the arrival of the duly
ordained high priest they, or at any rate their officials, put
themselves under his authority, to their own undoing. The author of I
Maccabees speaks in terms of the highest respect of them, and applies to
the treacherous murder of their leaders the words of Psalm lxxix.

In II Macc. xiv. 6, Alcimus replies to the question of King Demetrius as
follows: “The so-called ʽAsidaei among the Jews, of whom Judas Maccabeus
is the leader, maintain the war and sedition, and will not permit the
realm to secure peace.” It will be seen that this passage is not
necessarily in contradiction with those of I Maccabees, since it is here
put into the mouth of Alcimus, and is meant to be a wilful
misrepresentation of the facts on his part. Like the other passage, it
implies that such a definite body with a distinct name existed before
the Hasmonean revolt.

To find in Psalms xii., lxxxix., cxlix., and others references to the
same group of men is quite gratuitous. The ordinary sense of “righteous”
or “saintly” amply satisfies every one of the occurrences of the word
Hasid in the Psalms. And the figurative קהל חסידים (Ps.
cxlix. 1) no more implies an organized body than קהל מרעים
of Psalm xxvi. 5 implies a formal association of evil-doers, a Camorra.
We shall be compelled to rely wholly on the passages in Maccabees for
any information about the ʽAssidaei, or Hasidim, in the sense of a
definite organization bearing that title.

Who were these ʽAssidaei? That admirable writer and sturdy patriot, the
author of I Maccabees, says they were a body of great power and
influence in Israel, ἰσχυρὰ δυνάμει, the leaders of the
Jews, and, as has been seen, organized before the revolt. Nothing is
clearer than that they are not identical with the “scribes,” with whom
they are grouped in I Macc. vii., among those who acknowledged Alcimus.
It is equally clear that they are not at all the same as the Hasmonean
partisans, for they join Mattathiah later, and abandon Judah, at least
temporarily, early in the struggle. They are characterized by their zeal
for the Law, a zeal which naturally manifested itself in strong
opposition to Hellenism.

In Palestine, accordingly, for at least a generation before the revolt,
the disintegrating tendencies of Hellenism, as evidenced in the apostasy
of many wealthy Jews and in the neglect of many traditional customs on
the part of others, provoked an organized opposition. Forming themselves
into a fraternity or groups of corporate bodies, to which they applied
the name of “saints,” the opponents of the Greeks directed their efforts
to the exact fulfilment of the Torah, and no doubt carried on a violent
polemic against Greek innovations, however harmless and valuable. At
about the same time an exactly similar movement among Egyptians had
brought the Ptolemies to terms. It was not of course to be expected that
a single province of the Syrian-Babylonian monarchy would accomplish the
same result. In the eyes of the Antiochene court their programme was no
doubt treasonable fanaticism. But it was not, as in the case of Egypt,
directly political in its scope, and it might never have led to armed
conflict.

According to Jewish tradition a pupil of Antigonus of Socho, José ben
Joëzer, was a member of this sect of “saints.”[125] And it is
significant that, although he is represented as especially rigorous in
all religious requirements that had a separatist tendency, he was
strikingly liberal in all matters of what might be called internal
religious practice. It is likely enough that the tradition is accurate
and the “saints” were not at all precisians or fanatics, but that their
cohering bond was simply opposition to Hellenism. As has been said, it
was against the Hellenizing Jews more than the Greeks that their attack
was directed. These latter had on their side the advantages of wealth
and social position, but they lacked just that which made their
opponents strong, a compact organization. There was no συναγωγὴ
Ἑλλήνων, no congregation or fraternity of Philhellenes. They
included all shades of Greek sympathizers, from out and out apostates to
parvenus, to whom speaking Greek was a mark of fashion. No doubt the
feeling between the two groups ran high, and neither side spared bitter
abuse and invective.

The conflict was finally precipitated by an act that was one of the
commonest occurrences of ancient political struggles. The party
defeated, or in danger of defeat, does not scruple to invite foreign
intervention. In this case the irreconcilable Hellenists, evidently
losing ground in face of the rapid growth of Hasidic conventicles,
appeal to the Greek king, whose policies their own efforts were
furthering, and of whose sympathy they were assured. That king happened
to be the bizarre Antiochus Epiphanes.


------------------------------------------------------------------------




                               CHAPTER X

                       ANTIOCHUS THE MANIFEST GOD


“And there arose from them [the companions of Alexander] a root of sin,
to wit, Antiochus Epiphanes, son of King Antiochus, he who had been
hostage in Rome.” That to the writer of I Maccabees is a complete
characterization of the king whose reign was to be of fateful
consequences to the Jews, a ῥίζα ἁμαρτωλός, an ill sapling
of a noble tree. Perhaps the writer had in mind the שרש פרה ראש
ולענה (Deut. xxix. 17), “a root bearing gall and wormwood.” And
he had been a hostage in Rome; a man, that is, of no usual character and
no usual career.

Except in this general way, he can scarcely be said to have a
personality at all to the writers of the Books of Maccabees. He is
merely the type of tyrant, proud and presumptuous, unduly exalting
himself above God because of his vain and transitory successes, and
dying in agony, after an edifying deathbed repentance. No more than the
Nebuchadnezzar of the Book of Daniel, is he anything other than an
instrument of the wrath of God. It is hard to believe that there was any
real feeling on the writer’s part.

But Antiochus had a real personality and an especially interesting one.
Both in modern and in ancient times characterization of this strange
figure has been attempted, and the verdicts have been so widely
different that the summary may be given in Livy’s words: _Uti nec sibi
nec aliis, quinam homo esset, satis constaret_, “So that neither he
himself nor anyone else could clearly state what manner of man he was.”

The freakish outbursts, which amazed and scandalized his contemporaries,
amply justified the common parody of his title Epiphanes by Epimanes,
“the madman.”[126] Some there were—perhaps his royal nephew and
biographer, Ptolemy of Egypt, among them—who regarded him as
unqualifiedly demented.[127] It is likely enough, if the stories about
him are even partly true, that he had periods of real derangement. But
it seems evident that he was a right royal personage, of unusual charm
of manner, of undoubted military capacity, quick and decisive in action,
fostering a dream of empire whose rude shattering must have been an
important contributing cause to his death.


[Illustration: ANTIOCHUS (IV) EPIPHANES
AFTER A COIN
(From a drawing by Ralph Iligan)]


His was a strange blend. Various epochs met in him, and it is not
surprising that many incongruities resulted from that fact. First of all
he was in every sense a Macedonian despot. Macedonians had always been
accustomed to the concentration of supreme power in the hands of a
single individual. For four or five generations Antiochus’ immediate
ancestors had wielded such power over a rabble of nations stretching
from the Aegean to the frontiers of India.[128] The emotional reactions
which the existence and the possession of this power must have, were
present in him. One constant result of it, the absence of any real
social life, is an especially fertile source of deterioration, but the
worst effects are noticed chiefly in those born to the purple.
Antiochus’ exile saved him from them. Yet nothing could save him from
the consciousness that he might, if he chose, gratify every whim, and
yield to every impulse, and his associates found quickly enough that his
_bonhomie_ and engaging simplicity were moods, which might be succeeded
by bursts of quite incalculable and murderous rage.

There was the additional fact that the monarchy founded by Alexander
was in legal contemplation the reign of a god made flesh. Seleucus,
we may remember, entered almost at once into the titularies of Sumer
and Akkad.[129] The second Antiochus was styled “the God,”
Θεός, _tout simple_. Our Antiochus called himself
Epiphanes—which, it need scarcely be said, is to be translated “the
Manifest Deity,” and not “the Illustrious.”[130] And, at any rate at
certain moments, the designation was doubtless a real one to him and
not a conscious pose. Worship of the king, the foundation of the
later Augustus-cult, was an apparent unifying element in the
hopeless jumble of gods and rituals. For that purpose it might be
encouraged even by hard-headed peasants like Vespasian, or
philosophers like Marcus, who had no illusions about the character
of their divinity. But that Alexander in all sincerity believed
himself to be god can scarcely be questioned, and Epiphanes may
often have similarly impressed himself.

Secondly, he was a Greek. Hellenism was to him a real and profound
enthusiasm. His early life as a Roman hostage must have immensely
stimulated this side of his character. At Rome his associates were the
Scipionic circle, to whom Greek culture had come as a revelation. The
distinguished Roman families with whom the young prince lived read
Greek, spoke Greek, discussed Greek, and were eager to act as the
interpreters of Hellenism to their slower-witted countrymen. In these
surroundings anyone boasting not only Greek but regal blood must have
found his racial self-esteem flattered to an extraordinary degree.
Antiochus’ first act on his release was to betake himself to the
intellectual capital of Greece, to Athens, in whose citizenry he eagerly
enrolled himself. In fact, he was an Athenian magistrate—στρατηγὸς
ἐπὶ τὰ ὅπλα[131]—when news came to him of the assassination of
his brother Seleucus and of the opportunities waiting one who could act
quickly.

When he was king, so much of his policy as did not look to the
aggrandizement of his empire was directed to the rehabilitation of Greek
cities and temples. Megalopolis, Tegea in Arcadia, Delos, Rhodes, were
the beneficiaries of his Philhellenic enthusiasm. The truckling
Samaritans—at least the Hellenizing party among them—knew that nothing
would make a quicker appeal to him than to rename the sanctuary on
Gerizim in honor of Zeus Hellenius.[132] He would probably have found it
difficult to understand that anyone could seriously maintain the claims
of any other culture against that of the Greeks, and no doubt received
as a matter of course the representations of the Jewish Hellenizers that
a little impetus would greatly expedite the Hellenizing process in
Palestine.

When we find Antiochus, king of kings, Manifest God, soliciting
the suffrages of the Antiochene burghers for the office of
“market-commissioner,” or of “district mayor,”[133] we are not to
regard it as an eccentricity of the same sort that set him
wrangling in the public squares with Hob and Dick, or pouring
priceless ointments on his fellow-bathers in the public
baths.[134] The maintenance of the structure of the Greek polis
was an expression of Hellenic pride in a characteristically
Hellenic institution. No one, to be sure, was deceived by it into
thinking that Citizen Antiochus could not incontinently change
into an irresponsible master at will, but, comedy as it was, it
had a real significance, which did not escape even the scoffers
and, least of all, the king.

Finally there was an ultra-modern side in him. Antiochus was also a
cultivated gentleman, to whom skepticism was an index of education and
sacrilege a concrete instance of skepticism. He lived in a very
unsettling age. As has been said before, the Greek culture that found
its way into Rome after the Hannibalic wars was a sophisticated,
disintegrating culture, to which the ancient institutions had at best a
practical utility, and which acknowledged theoretically no binding
principles in the physical or moral world. It was in this culture that
the young Antiochus was reared. He was not alone in it. Many of the
incidents of this period show a revolting cynicism on the part of the
actors. One Greek commander erected altars to “Impiety and Illegality.”
A Spartan brigand called himself “Hybristas,” “the Outrager.”[135]

Indeed it was as a wanton desecrater of shrines that Antiochus gained an
unenviable notoriety. His pillaging of the temple at Jerusalem was only
one of a series of similar acts. At Hierapolis, as well as at many other
Syrian shrines, and finally at Elymaea, he coolly appropriated the
temple treasures, which in most cases involved violence on his part. But
it needed his outrageous “marriage” to Diana to set the seal upon his
derisive attitude toward his fellow-gods. The sober Polybius attributes
his death to his impiety, a conclusion which naturally is warmly
supported by Josephus.[136]

It is idle to attempt to reconcile this sort of cynicism with the
pretensions to actual divinity which he probably made in all
seriousness. The two are of course quite irreconcilable, and represent
merely the shifting moods of a complex and slightly abnormal
personality. Under almost any king such an outbreak as the Hasmonean
revolt might have taken place. Perhaps the conflict was inevitable. But
the form the conflict took, the high degree of religious and national
enthusiasm which it evoked, and the powerful aid that enthusiasm gave to
the propaganda which was preparing itself, were directly consequent upon
the character of Antiochus the God Manifest. The rigor and thoroughness
with which he strove to suppress the Jewish cult were characteristic of
him. His indifference to sacred traditions made his violation of the
temple almost a casual act on his part, his Hellenism justified his
plans, and his despotic nature, raging under the humiliating rebuff he
had received from Rome, found an outlet in the punishment of a
disobedient province.

The writer of I Maccabees places the responsibility for the persecution
by Antiochus directly upon the Jews themselves. Many, he tells, were
persuaded to identify themselves wholly with the Greeks.[137] The first
offense to Jewish religious sentiment did not come from the king at all.
The men who waited upon Antiochus, and obtained permission to set up a
gymnasium at Jerusalem, acted quite of their own volition. Antiochus’
direct action in the matter begins with his return from Egypt.
“Embittered and groaning,” Polybius says, he left Egypt and returned to
Syria. Now, just what happened in Judea is not quite clear. First
Maccabees tells of an unprovoked pillage of the temple and a massacre of
the people. Second Maccabees reports a furious struggle between the two
pretenders, Menelaus and Jason, upon a rumor of the king’s death. In all
likelihood the fight ended with the discomfiture of Antiochus’
appointee, Menelaus, and the king immediately proceeded to rescue him.
The sack of Jerusalem and a massacre followed. No doubt the massacre was
no worse than befell any captured city, since of a special policy of
extermination there can as yet have been no question.

Menelaus was restored, the temple treasures were surrendered to the
king, and, either directly or after an interval of two years, the
programme of forcible suppression of the Jewish cult was announced.

It is for this programme that an adequate explanation is wanting. There
is nothing really quite like it in Greek history. Not that religious
persecution, or the suppression of an obnoxious cult, was an unheard-of
undertaking. The establishment of the worship of Dionysus had
encountered vigorous opposition in continental Greece. A probable
tradition recounts the attempts at thorough repression with which
several Greek communities, notably Thebes, met the intruder.[138] But
this movement had as its object the preservation of an ancestral
religion, not its destruction. To compel anyone to abjure his national
customs, to forsake τὰ πάτρια, must have seemed monstrous
to all people in whom the sense of kinship with the deity, and the
belief in the god’s local jurisdiction, were as strong as they were
among the Greeks.

Somewhat later, among the Romans, a successful attempt was made to
extirpate the Druidic ritual in Cisalpine Gaul. As far as this was an
effort to destroy root and branch an ancient and established form of
worship, it presents many analogies to the project of Antiochus. But the
persecution of the Druids was based on specific charges of immoral and
anti-social practices associated with their ritual, especially that of
human sacrifices. That may have been a pretext. The Druids may not after
all have been guilty of these enormities. However, the pretext was at
least advanced, and the exile of Druidic brotherhoods and the
destruction of their sanctuaries were publicly justified only by
that.[139]

In the case of the Jews no such assertions are to be discovered.
Antiochus, instigated by renegade Jews, sets about a systematic
obliteration of the distinctively Jewish ritual. The synagogue services
were to be checked by the destruction of the Torah. Perhaps periodic
reunions in the synagogue were forbidden altogether, since meetings of
citizens were proverbially looked at askance in monarchies.[140] The
temple was rededicated to the Olympian Zeus, and the ceremony of
circumcision was made a capital offense. Observance of the Sabbath was
construed as treason. No detail was overlooked.

This complete scheme is not to be explained by the existence of a strong
animosity toward the Jews. There is, in the first place, none of the
evidence that was met with in Egypt, that such animosity existed. And,
secondly, animosity between racial groups expressed itself in bloody
riots, not in a carefully prepared plan for extirpating a religion while
sparing its professors. Nor can we find in the personal character of
Antiochus a sufficient cause for the persecution. He undoubtedly
exhibited the gusts of passion common enough among those who wield
irresponsible power, but the sustained and bloody vindictiveness of such
a programme is a very different thing.

It has been frequently suggested that his cherished policy was the
thorough Hellenization of his empire, that among the Jews only was there
a determined resistance, that upon learning that the basis of their
resistance was a devoted attachment to their ancestral superstition, he
determined to root out the latter. The difficulties with this view are,
first, that opposition was not confined to the Jews, but was met with
everywhere—a dull and voiceless opposition, which, however, unmistakably
existed. Secondly, among the Jews a very large number, we are told,
“were persuaded”; and it is highly likely that Antiochus came in direct
contact wholly with the latter, or almost wholly, so that the situation
in Judea cannot have impressed him as radically different from that of
Syria or Babylonia.

But, above all, it is the conclusion that the obstacles to his policy
would lead to persecution on his part, which is more than doubtful. No
one could have known better than he did himself that ancestral religious
customs are not to be eradicated by violence. The Egypt which was so
nearly in his grasp might have taught him that, if nothing else could.
There the indigenous religion had triumphed. He himself, upon his entry
into the kingdom, had crowned himself _more Aegyptico_, “after the
Egyptian fashion,”[141] that is, with full acknowledgment of the
sovereignty of Ptah and Isis over their ancient demesnes.

We shall probably have to look to the Hellenizing Jews not only for the
initiation, but for the systematic carrying out, of the policy of
persecution. And, as has been suggested, it is one of the commonest
phenomena of ancient life. There was scarcely a Greek city in which a
defeated faction had not at some time summoned the public enemy into the
city, and by their aid taken a cruel vengeance on their opponents. If
the Hellenizing faction in Judea found its influence waning, its action
was from the point of view of ancient times natural enough. It appealed
to foreign aid and strove systematically to stamp out the institutions
it opposed, just as at Athens the Athenian oligarchs, placed in power by
Spartan arms, tried to maintain themselves by wholesale proscription and
by systematically removing all the democratic institutions that had
developed since Clearchus.[142]

It is likely too that the impelling motive was not solely the rancor
which apostates feel for the faith or nation they have quitted. They saw
themselves in the presence of a real danger. Among them was to be found
most of the wealth of the community, and no doubt a great deal of the
intellectual culture. Many of them were already in the third or fourth
generation of Hellenistic Jews. The ancient ritual had for these men no
personal associations whatever. In the various communes they enjoyed the
position which wealth necessarily, and in those days especially,
brought. That there was any virtue in poverty or privation in themselves
had not yet been preached to the world, and would have seemed a wild
paradox; and although the vanity of wealth without wisdom was a
philosophic truism, ordinary wits would not always trust themselves to
make the distinction.

When these men, who formed almost a hereditary nobility, and already
cherished a superb aloofness from the mass, felt their influence and
power challenged, perhaps saw themselves outvoted in the governing
councils of the synagogues and communes, and the foundations of their
petty glory sapped, they were roused to a counter-effort, of which the
results have been indicated. The danger in which they found themselves
came from the Hasidim, the group of brotherhoods that made a conscious
opposition to Hellenism their bond of union. In Egypt the opposition had
found its organs in the caste-like corporations of priests. In Judea the
organs had to be created. And that they were successful, the words of I
Maccabees testify. They contained the leaders of the nation; their
position was already one of dominating influence.

It is unnecessary to detail the course of the Hasmonean revolt. Even the
brilliant successes of Judas in the field, and the less splendid but
equally solid triumphs of his brothers, would have had fewer political
consequences than they had except for the chaos in the Seleucid
succession. But of the permanent triumph of the movement there was never
any doubt. If the revolt had ended with the death of Judas, the
discomfiture of the Hellenists would have been complete. No Macedonian
king would ever be tempted to provoke another revolt by a similar
project. It could never be a part of a sane ruler’s policy to sacrifice
valuable military material in order to gratify a local faction. And it
must never be forgotten that the Greek rule of the Syrian kingdom was
the domination of a military class. Every diminution of the army was a
dead loss.

The suggestion may be hazarded that not merely the Hellenistic Jews, but
also the Greeks themselves, viewed the progress of the Hasidim with real
alarm. We are far as yet from the epoch of real propaganda, but to some
extent it may already have begun. Where and when we can only speculate.
Perhaps the fervor of Hasidic preaching had touched non-Jewish Syrians;
perhaps some of the younger men of the Hellenists “relapsed” under
Hasidic stimulation into Judaism. However the case may be, Greeks of
influence may have noted that the Grecizing of Coele-Syria was not
merely hindered by obstacles in Judea, but that the Judaizing of
portions already won was a possibility that was attaining a constantly
greater vividness. If this was the case, the persecution by Antiochus
was a precaution, insensate and futile, but less at variance with Greek
methods than it seems in the usual interpretation of the facts we know.


------------------------------------------------------------------------




                               CHAPTER XI

                         THE JEWISH PROPAGANDA


The preaching of a gospel seems to us as natural as the existence of a
religion. That is because the religions we know best are universal ones,
of which the God is a transcendent being, in whose sight human
distinctions are negligible. But for the Mediterranean world that was
not the case. The religions were not universal; many of the gods were
concretely believed to be the ancestors of certain groups of men, and
not always remote ones. Local associations played a determining part. If
we find an active propaganda here, it cannot be because the spread of a
ritual or faith is an inherent characteristic. On the contrary, in
normal circumstances there seems to be no reason why one community
should change its gods or forms of worship for those of another.

But, as a matter of fact, they did change them. And the change was often
effected consciously by the planned efforts of a group of worshipers,
and in all the ways that have been used since—preaching, emotional
revivals, and forcible conquest. One such carefully planned effort was
that of the Jews, but only one of them. The circumstances in which this
propaganda was carried out need close investigation.

In discussing Greek religion (above, p. 34) it has been suggested that
there was in every community a large number of men who found no real
satisfaction in the state cult, and that it was chiefly among them that
the proselytes of new and foreign religions were to be found. But that
does not make us understand why these foreign religions should have
sought proselytes, why they should have felt themselves under
obligations to assume a mission. The stranger within the gates might
reasonably be expected to do honor to the divine lord of the city: if he
remained permanently, his inclusion in the civic family in some way is
natural. But what was it that impelled Isis to seek worshipers so far
from the Nile, where alone she could be properly adored, or the
mysterious Cabiri to go so far from the caves where their power was
greatest and most direct?[143]

The movement of which these special missions are phases was old and
extensive. It covered the entire Eastern Mediterranean, and went perhaps
further west and east than we can at present demonstrate. Its beginnings
probably antedated the Hellenes. The religious unrest of which Christian
missionaries made such excellent use was a phenomenon that goes back
very far in the history of Mediterranean civilization. At certain
periods of that history and in different places it reached culminating
waves, but it is idle to attempt to discover a sufficient cause for it
in a limited series of events within a circumscribed area of Greece or
of Asia.

The briefest form in which the nature of this unrest can be phrased is
the following—the quest for personal salvation.

We shall do well to remember that the ancient state was a real
corporation, based not upon individuals but upon smaller family
corporations. The rights of these corporations were paramount. It was
only gradually that individuals were recognized at all in law.[144] The
desire for personal salvation is a part of the growing consciousness of
personality, and must have begun almost as soon as the state corporation
itself became fixed.

Within a state only those individuals can have relatively free play who
are to a certain extent the organs of the state; that is, those
individuals who by conquest, wealth, or chance have secured for
themselves political predominance in their respective communities. But
these could never be more than a small minority. For the great majority
everyday life was hemmed in by conventions that had the force of laws,
and was restricted by legal limits drastically enforced. And this narrow
and pitifully poor life was bounded by Sheol, or Hades, by a condition
eloquently described as worse at its best than the least desirable
existence under the face of the insufferable sun.[145]

The warrior caste, for whom and of whom the Homeric poems were written,
were firmly convinced that the bloodless and sinewless life in the House
of Hades was the goal to which existence tended. But they found their
compensation in that existence itself. What of those who lacked these
compensations, or had learned to despise them? In them the prospect of
becoming lost in the mass of flitting and indistinguishable shadows must
have produced a profound horror, and their minds must have dwelt upon it
with increasing intensity.

It is one of the most ancient beliefs of men in this region that all the
dead become disembodied spirits, sometimes with power for good or evil,
so that their displeasure is to be deprecated, sometimes without such
power, as the Homeric nobles believed, and the mass of the Jews in the
times of the monarchy. These spirits or ghosts had of themselves no
recognizable personality, and could receive it only exceptionally and in
ways that violated the ordinary laws of the universe. Such a belief is
not strictly a belief in immortality at all, since the essence of the
latter is that the actual person of flesh and blood continues his
identity when flesh and blood are dissolved and disappear, and that the
characteristics which, except for form and feature, separated him from
his fellows in life still do so after death. The only bodiless beings
who could be said to have a personality were the gods, and they were
directly styled “the Immortals.”

However, the line that separated gods and men was not sharp. The
adoration offered to the dead in the Spartan relief[146] is not really
different from the worship of the Olympians. From the other side, in
Homer, the progeny of Zeus by mortal women are very emphatically
men.[147] Whether the Homeric view is a special development, it is
demonstrably true that a general belief was current in Greece not long
after the Homeric epoch, which saw no impossibility in favored men
securing the gift of immortality; that is, continuing without
interruption the personal life which alone had significance. This was
done by the translations—the removal of mortal men in the flesh to
kinship with the gods.[148]

This privilege of personal immortality was not connected, in the myths
that told of it, with eminent services. It was at all times a matter of
grace. In the form of bodily translation it always remained a rare and
miraculous exception. But the mere existence of such a belief must have
strongly influenced the beliefs and practices that had long been
connected with the dead.

We cannot tell where and when it was first suggested to men that the
shadow-life of Hades might by the grace of the gods be turned into real
life, and a real immortality secured. It may be, as has been supposed,
that the incentive came from Egypt. More likely, however, it was an
independent growth, and perhaps arose in more than one place. The favor
and grace of the gods, which were indispensable, could obviously be
gained by intimate association, and in the eighth and perhaps even the
ninth pre-Christian century we begin to hear in Greece of means of
entering into that association. One of these means was the “mystery,” of
which the Eleusinian is the best-known. In these cult-societies, of the
origin of which we know nothing, a close and intimate association with
the god or gods was offered. The initiated saw with their own eyes the
godhead perform certain ceremonial acts; perhaps they sat cheek by jowl
with him. It is obvious that such familiarity involved the especial
favor of the gods, and it is easy to understand that the final and
crowning mark of that favor would not be always withheld. The communion
with the god begun in this life would be continued after it. To the
mystae of Eleusis, and no doubt elsewhere, and to them only, was
promised a personal immortality.[149]

It may not have been first at Eleusis. It may have been in the obscure
corners of Thrace where what later appeared as Orphic societies was
developed. But there were soon many mysteries, and there was no lack of
men and women to whom the promise was inexpressibly sweet. The spread of
Orphism in the sixth and fifth centuries B.C.E. bears witness to the
eagerness with which the evangel was received.

Outside of Greece, in Persia, India, and Egypt, perhaps also in
Babylonia, there were hereditary groups of men who claimed to possess an
arcanum, whereby the supreme favor of the gods, that of eternal
communion with them, was to be obtained. These hereditary castes desired
no extension, but jealously guarded their privileges. But among them
there constantly arose earnest and warm-hearted men, whose humanity
impelled them to spread as widely as possible the boon which they had
themselves obtained by accident. Perhaps many attempts in all these
countries aborted. Not all Gotamas succeeded in becoming Buddhas.

The Jews seemed to the Greeks to possess just such an arcanum, and
whatever interest they originally excited was due to that fact. The
initiatory rite of circumcision, the exclusiveness of a ritual that did
not brook even the proximate presence of an uninitiate, all pointed in
that direction, even if we disregard the vigorously asserted claims of
the Jews to be in a very special sense the people of God.

The Jews too had as far as the masses were concerned developed the
belief in a personal immortality during the centuries that followed the
Babylonian exile (comp. p. 70), and as far as we can see it developed
among them at the same time and somewhat in the same way as elsewhere.
That is to say, among them as among others the future life, the _Olam
ha-bo_, was a privilege and was sought for with especial eagerness by
those to whom the _Olam ha-zeh_ was largely desolate. Not reward for
some and punishment for others, but complete exclusion from any life but
that of Sheol for those who failed to acquire the _Olam ha-bo_, was the
doctrine maintained, just as the Greek mystae knew that for those who
were not initiated there was waiting, not the wheel of Ixion or the
stone of Sisyphus, but the bleak non-existence of Hades.[150]

But there was a difference, and this difference became vital. Conduct
was not disregarded in the Greek mysteries, but the essential thing was
the fact of initiation. Those who first preached the doctrine of a
personal salvation to the Jews were conscious in so doing that they were
preaching to a society of initiates. They were all mystae; all had
entered into the covenant: all belonged to the congregation of the Lord,
קהל יהוה. To whom was this boon of immortality, the _Olam
ha-bo_, to be given? The first missionaries, whether they did or did not
constitute a sect, had a ready answer. To those to whom the covenant was
real, who accepted fully the yoke of the Law.

The sects of Pharisees and Sadducees, whose disputes fill later Jewish
history, joined issue on a number of points. No doubt there was an
economic and social cleavage between them as well. But perhaps the most
nearly fundamental difference of doctrine related to the _Olam ha-bo_.
The Pharisees asserted, and the Sadducees denied, the doctrine of
resurrection. It is stated by Josephus,[151] that the Sadducees called
in question the _Olam ha-bo_ itself. When and where these sects took
form is uncertain. The Pharisees at least are fully developed, and form
a powerful political party under John Hyrcanus.[152] It is very unlikely
that they are related to the Hasidim or are a continuation of them. The
latter were a national, anti-Hellenic organization, and contained men of
all shades of beliefs and interests. But the Pharisees, like the
Hasidim, began as a brotherhood or a group of brotherhoods, however
political their aims and actions were in later times. The fact is
indicated by the name _Haber_, “comrade,” which they gave themselves,
and the contemptuous _Am ha-aretz_, “clod,” οἱ πολλοί, with
which they designated those who were not members of their congregations.

Now the _Haberim_, who preached the World-to-Come, were not in a
primitive stage of culture, but in a very advanced one. Their God was
not master of a city, but Lord of the whole earth. And they had long
maintained the principle that merit in the eyes of God was determined by
conduct, both formal and moral, a distinction less profoundly separating
than seems at first to be the case. If that were so, anyone, Jew or
Gentile, might conceivably acquire that merit. How was the _Olam ha-bo_
to be refused to anyone who had taken upon himself the yoke of the Law,
who did all that the Lord required at his hands? Jewish tradition knew
of several eminently righteous gentiles, such as Job, in whom God was
well pleased. It was an untenable proposition to men whose cardinal
religious doctrine had for centuries been ethical and universal that all
but a few men were permanently excluded from the beatitude of life after
death.[153]

Since, however, the promises of the sacred literature were addressed
primarily to Israel, those who were not of Abraham’s seed could become
“comrades” only by first becoming Jews. That conception involved no
difficulty whatever. The people of the ancient world had empirically
learned some of the more elementary facts of biological heredity; but
membership in a community, though determined by heredity in the first
instance, was not essentially so determined. In earlier times, when the
communities were first instituted, not even the pretense of kinship was
maintained. The essential fact was the assumption of common _sacra_.

That a man might by appropriate ceremonies—or without ceremonies—enter
into another community, was held everywhere. If, as has been suggested
(above, p. 147), the Hasidim found some of their members among the
non-Jewish population of Syria,[154] it is not likely that the process
of becoming Jews was rendered either difficult or long. Abraham, a late
tradition stated, brought many gentiles under the wings of the Shekinah,
the Effulgence. If this tradition is an old one, it indicates that
proselytizing was in early times held to be distinctly meritorious.[155]

The first conquests of the Hasmonean rulers brought non-Jewish tribes
under immediate political control of the Jews. Most of them, notably the
Idumeans, were forcibly Judaized, and so successfully that we hear of
only one attempted revolt.[156] There can of course have been no
question here of elaborate ceremonies or lengthy novitiates. The
Idumeans were dealt with as shortly as Charlemagne’s Saxons, and gave
the most convincing demonstration of their loyalty in the time of the
insurrections.[157]

This drastic way of increasing the seed of Abraham must have been viewed
differently by different classes of Jews. To the Haberim the difference
between a heathen and a Jewish aspirant to their communion lay in the
fact that the heathen had undergone the fearful defilement of worshiping
the Abomination, while the Jew had not. For the former there was
accordingly necessary an elaborate series of purgations, of ceremonial
cleansing; and until this was done there was no hope that he could be
admitted into the congregation of the Lord. But it might be done, and it
began to be done in increasing numbers. It would have been strange if,
among the many gentile seekers for salvation, Greek, Syrian,
Cappadocian, and others, some would not be found to take the path that
led to the conventicles of the Jewish Haberim. This was especially the
case when, instead of an obscure Syrian tribe, the Hasmoneans had made
of Judea a powerful nation, one of the most considerable of its part of
the world.

All the mysteries welcomed neophytes, but none made the entrance into
their ranks an easy matter. In some of them there were degrees, as in
those of Cybele, and the highest degree was attained at so frightful a
cost as practically to be reserved for the very few.[158] In the case of
the Jews, one of the initiatory rites was peculiarly repellent to Greeks
and Romans, in that it involved a bodily mutilation, which was performed
not in the frenzy of an orgiastic revel, but in the course of a solemn
ritual of prayer. That fact might make many hesitate, but could not
permanently deter those who earnestly sought for the way of life.

The Jewish propaganda was not confined to receiving and imposing
conditions on those who came. Some at least sought converts, although it
is very doubtful that the Pharisaic societies as a class planned a real
mission among the heathen. The methods that were used were those already
in vogue—methods which had achieved success in many fields. Books and
pamphlets were published to further the purpose of the missionaries;
personal solicitation of those deemed receptive was undertaken. Actual
preaching, such as the _diatribe_ commenced by the Cynics, and before
them by Socrates, was probably confined to the synagogue, or meeting
within the proseucha, and reached only those who were there
assembled.[159]

The literary form of the propaganda was especially active in those
communities in which Jews and Greeks spoke a common language and partly
shared a common culture. Even books intended primarily for Jewish
circulation contain polemics against polytheism and attacks upon heathen
custom, which the avowed purpose of the book would not justify.

It is not to be supposed that the literary propaganda was the most
effective. It was limited by the very field for which it was intended.
Such a book as the Wisdom of Solomon was both too subtle and too
finished a product to appeal to other than highly cultivated tastes, and
men of this stamp are not readily reached by propagandizing religions.
The chief object of attack was the Greek polytheism. “Wisdom” ventures
even on an historical explanation of polytheism, which is strangely like
that of Herbert Spencer.[160] Now, just for the Greeks, who might read
and understand such a book, to refute polytheism was destroying a man of
straw. No one of them seriously believed in it. Those who were not
agnostics or atheists believed in the unity of the Divine essence, and
at most maintained the existence of certain subordinate ministerial
beings, who might or might not be identical with the names of the actors
in the myths. But many Jews would be ready to admit so much. Indeed that
there were subordinate _daemonia_, helpful and harmful, was a widespread
belief in Judea, even if without authoritative sanction. Very often the
heathen gods were conceived to be not absolute nullities, but demons
really existing and evil—a belief which the early Christian church
firmly held and preached.[161]

Accordingly the polished society of a Greek city did not need the
literary polemics against polytheism to be convinced that monotheism was
an intellectually more developed and morally preferable dogma. On the
other hand, it was a very difficult task to convince it that the
ceremonies of the official cult, granting even their philosophic
absurdity, were for that reason objectionable. To make them seem so,
there would have to be present the consciousness of sin, and that was
not a matter which argumentation could produce.

One other point against which Jewish writers of that time address
themselves is the assumed viciousness of Greek life. How much one people
has with which to reproach another in that respect in ancient or in
modern times need not be considered here. The fact remains that in many
extant books sexual excesses and perversions are made a constant
reproach to the heathen—which generally implies the Greek—and the extant
Greek and Latin literature gives a great deal of color to the
charge.[162] This is due not so much to the actual life depicted as to
the attitude with which even good men regarded these particular
incidents. It is true that we have contemporary evidence that many Jews
in Greek communities were no paragons of right living or self-restraint.
But it is at least significant that this accusation, continually
repeated by the Jews, is not met by a retort in kind. The anti-Jewish
writings are not especially moderate in their condemnations. But with
viciousness in their lives they do not charge the Jews, and they cannot
have been unaware of what the Jews wrote and said.

Polytheism and immorality, the two chief counts in the indictment which
Jewish writers bring against heathendom, were not things Greeks were
disposed to defend. But it is doubtful whether the books that inveighed
against them were valuable weapons of propaganda. We have practically no
details of how the movement grew. In the last century before the
Christian era it had reached the extraordinary proportions that are
evidenced by the satire of Horace as well as by the opposition which it
encountered. Jewish apocalyptic literature confidently expects that all
the heathen on the rapidly approaching Judgment Day will be brought
within the fold.[163] The writers may be forgiven if the success of
their proselytizing endeavors made them feel that such a result was well
within the range of possibility.

Within the same period the worships of Cybele, of Sabazios, and of Isis,
had perhaps even greater success in extending themselves over the Greek
and Roman world. The communities they invaded only rarely welcomed them.
Even at Rome the official introduction of Cybele was the last desperate
recourse of avowed superstition, and it was promptly restricted when
success and prosperity returned to the Roman arms. But in all the
communities great masses of men were thoroughly prepared in mind for the
doctrines the Asiatic religions preached. A public preaching, such as
the Cynics used, was rarely permitted. But if we recall how many slaves
and ex-slaves as well as merchants and artisans were of Asiatic stock,
the spread of these cults, including that of the Jews, by the effective
means of personal and individual conversion is nothing to be wondered
at. The state was perforce compelled to notice this spread. Individuals
had noticed it long before.


------------------------------------------------------------------------




                              CHAPTER XII

                             THE OPPOSITION


The ancient state was based on community of _sacra_, of
cult-observances. Anything that tended to destroy them or impair general
belief in their necessity, went to the very roots of the state, was
therefore a form of treason, and was punished as such. The state rarely
was interested in the honor of the gods themselves. Roman law had a
maxim, which was very seriously stated, but which makes upon us the
impression of a cynical witticism: _Deorum iniuriae dis curae_, “Let the
gods attend to their own wrongs.” Since the kinship of members of the
state was generally known to be a legal fiction, the bond that took its
place was common worship. The state could not look without concern upon
anything that threatened to weaken its formal structure.

Most Greek states made ἀσέβεια, “impiety,” a criminal
offense. But just what acts or omissions constituted impiety was in each
case a question of fact, to be determined specially in every instance.
At Athens various persons of greater and less distinction were
prosecuted under that indictment—Socrates, Theophrastus, Phryne. In
every one of these cases, the gravamen of the charge was that the
defendant did not regard as gods those whom the state so regarded
(μὴ νομίζειν θεοὺς οὓς ἡ πόλις νομίζει, Plat. Apol. 24B and
26B), and taught so. In general, individual prosecutions such as these
were deemed sufficient to repress the spread of dangerous doctrines. It
was not believed necessary to consider membership in any sect or
community as _prima facie_ evidence of such impiety, punishable without
further investigation. In later times, however, even this step was
taken. Certain philosophic sects—which, we may remember, were
corporately organized—were believed to be essentially impious. The city
of Lyctos in Crete forbade any Epicurean to enter it under penalty of
the most frightful tortures.[164]

We shall have to distinguish these police measures, which, when aimed at
religious bodies, constitute an undoubted religious persecution, from
the mutual animosity with which hostile races in any community regarded
each other and the bloody riots that resulted from it. In the new city
of Seleucia in Babylonia, the Syrians, Jews, and Greeks that lived there
were very far from realizing the purpose of the city’s founder and
coalescing into a single community. Sanguinary conflicts, probably on
very slight provocation, frequently took place. Sometimes the Jews and
Syrians combined against the Greeks; sometimes the Greeks and Syrians
against the Jews, as recounted by Josephus.[165] The situation in
Alexandria, where Egyptians hated Greeks, Jews, and doubtless all
foreigners with a scarcely discriminating intensity, is peculiar only
because we are well informed of conditions there by the papyri. When any
one of these nationalities gained the upper hand, there was likely to be
a bloody suppression of its foes, often followed by equally bloody
reprisals. Salamis, in Cyprus, is a grim witness of the frenzy with
which neighbors could attack each other, when years of hostility
culminated in a violent outbreak.[166]

The attitude of Greek states toward the Jewish congregations in their
midst was certainly not uniformly hostile. But in many cases there could
not help being a certain resentment, owing to the fact that these
congregations were by special grant generally immune from prosecution
for impiety, although as a matter of fact they very emphatically “did
not regard as gods those whom the state so regarded.” Of itself this
circumstance might have been neglected, but the active and successful
propaganda they undertook made them a source of real danger to the
state. We therefore hear of attempts made sporadically to abrogate the
immunity, to compel the Jewish corporations to conform to the local law
of ἀσέβεια. Nearly always, however, the immunity was a
royal grant, and therefore unreachable by local legislation, a fact that
did not tend to alleviate friction where it existed.[167]

At Rome police measures to suppress irreligion were long in existence.
However, the Roman attitude toward any form of communion with gods or
_daemonia_ was so uniformly an attitude of dread, that prohibition of
religious rites and punishment of participants in them were not a task
lightly assumed by a Roman magistrate. The suppression of the
Bacchanalia in 186 B.C.E. was nothing short of a religious persecution,
but the utmost care was taken to make it appear to be directed against
certain licentious practices alleged against the Bacchae, and the
senate’s decree expressly authorizes the Bacchic rites, under certain
restrictions deemed necessary to insure their harmlessness.[168] Very
early the Isiac mysteries and other Eastern cults came within the
animadversion of the urban police.[169] Here too the theory was that the
crimes and immorality of the communicants were the sole objects of
punishment, especially that species of fraud which took the form of
magic and unofficial fortune-telling. In reality, however, all these
pretexts covered the fact that the Romans felt their state ritual
endangered, not by the presence, but by the spread, of such rituals
among Romans; and in this their alarm was very well grounded indeed. But
to proceed openly and boldly against any manifestation of a divine
numen, was more than the average Roman board of aediles ventured to do.

If the official attitude of various communities toward outside cults and
toward the Jews in particular can be brought under no general rule, we
may be sure that the personal attitude of individual Greeks toward them
varied from enthusiastic veneration to indifference and determined
antagonism. In certain cities the Jews as foreigners could not hope to
escape odium nor the jealousy of competing individuals and
organizations. In Egypt particularly, the feud between Egyptians and
Jews existed before the coming of the Greeks there, and grew in
intensity as time went on. As far as definite attacks upon the Jews and
their institutions went, many of them had an Egyptian origin, and many
others were wholly confined to that country.

These attacks are not essentially different from the methods that
generally obtained when one group of men found itself in frequent
opposition to another group on the field of battle or otherwise. The
populace needs no rhetorical stimulation to represent its enemies as
wicked, cowardly, and foolish. That is a human weakness which exists
to-day quite as it has existed for many centuries. However, even for the
populace, such phrases were accepted conventions. They were not quite
seriously meant, and could be conveniently forgotten whenever the former
foe became an ally.

Among professional rhetoricians this particular method of argumentation
formed a set rhetorical device, one of the forms of _vituperatio_[170]
as classified in the text-books. Certain τόποι,
“commonplaces,” were developed concerning all nations, and used as
occasion required. Historical facts, popular gossip, freely imagined
qualities, were all equally used to support the statements made or to
illustrate them. Now it is in the works of professional rhetoricians
that most of the attacks on the Jews are to be found. Further, we have
their works wholly in the form of citations taken from the context. We
cannot even be sure to what extent the authors themselves were convinced
of what they said. Wherever we meet what is plainly a rhetorical
τόπος, we have little ground for assuming that it
corresponds to any feeling whatever on the writer’s part. Often it was
mechanically inserted, and has all the effect of an exercise in
composition.

With a laughter-loving people one of the first resources in controversy
is to render the opponent ridiculous. It was especially on the side of
religion that the Jews maintained their difference from their neighbors,
and claimed a great superiority to them. A Greek enemy would be much
inclined to heap ridicule, first on the pretensions to superiority, and
then on the religious form itself. That may be the basis of a story,
which soon became widely current, to the effect that the Jews worshiped
their god in the form of an ass.

The story is of Egyptian origin. Just where and when it began, cannot be
discovered. Josephus in combating Apion refers to a writer whose name
the copyists have hopelessly jumbled. It is not unlikely that he was a
certain Mnaseas, perhaps of Patara in Lycia, or Patras in the
Peloponnesus, a highly rhetorical historian of the second century
B.C.E.[171] He wrote therefore before the establishment of the Maccabean
state. Wherever he was born, he was a pupil of Eratosthenes, and
therefore a resident of Alexandria.[172]

We have his words only at third hand, in Josephus’ account of Apion’s
reference. Each citation is of substance, not the _ipsissima verba_;
and, besides, of this part of Josephus we have only a Latin translation,
not the original. The story, whether it is Mnaseas’ or Apion’s, is to
the effect that a certain Idumean, named Zabidus, duped the Jews into
believing that he intended to deliver his god, Apollo,[173] into their
hands, and contrived to get into the temple and remove “the golden head
of the pack-ass.”

The uncertainty and indirectness of the citation makes it dubious
whether Mnaseas understood this ass to be the actual divine symbol or,
as others said, merely one of the figures of a group. The absurdity of
the story seems so patent that its existence is almost incredible. It
indicates the extreme strictness with which gentiles were excluded from
even the approach to the temple at Jerusalem that the baselessness of
the ass-legend was not immediately discovered.[174]

Josephus’ indignation and his frequent reference to the “pretended wit”
of Apion or of Mnaseas make the tone and intention of the story quite
plain. It can have had no other purpose than that of holding the Jews up
to ridicule. But just what the point of the jest is, is by no means
quite so easy to discover. We cannot reconstruct even approximately the
words of Mnaseas. It is, however, at least likely that if he had
attributed the adoration of an ass to the Jews, a somewhat less
equivocal statement to that effect would appear. Other writers do make
that statement plainly enough. The point of Mnaseas’ raillery seems
rather to be the easy credulity of the people, a characteristic that was
at all times attributed to them in the ancient world, from the earliest
references, as they are found in Hecataeus, to the latest. It is curious
that this quality, which to Greeks and Romans seemed the most striking
trait of the Jews, is the very last that modern observers would ascribe
to them.

If we follow the story as it appears in later writers, we shall meet it
next in the history of the Syrian Posidonius, who lived about 100 B.C.E.
Again, we have his statement only in quotation, this time in a fragment
of the work of Diodorus, a Sicilian contemporary of Augustus. Posidonius
does no more than make the assertion that the innermost shrine of the
temple contained the statue of a long-bearded man, assumed to be Moses,
riding on an ass (λίθινον ἄγαλμα ἀνδρὸς βαθυπώγωνος
καθήμενον [sic] ἐπ’ ὄνου.[175] This is very far from
accusing the Jews of worshiping an ass. Indeed it is likely enough that
nothing was further from the mind of the writer. Perhaps Mnaseas too
told the same or a very similar story, since his anecdote would fit in
just as well with the account of Posidonius as with the later version.

The story appears again in the writings of Molo, the tutor of Caesar and
Cicero; but Molo’s statement is wholly lost. In the next generation we
find it in the writings of the Egyptian Apion, and in Damocritus, of
whom we know nothing, but who, it is likely enough, was a resident of
Alexandria.[176]

Here the statements are unmistakable. According to Damocritus, if he is
accurately cited by the late Byzantine lexicographer Suidas, the Jews
adored the gilded head of an ass (χρυσ ῆ ν ὄνου κεφαλὴν
προσεκύνουν). Apion, in the Latin translation of Josephus,
asserts that the Jews “adored this ass’ head, and worshiped it with much
ceremony” (_id_ [_i.e. asini caput_] _colere ac dignum facere tanta
religione_).[177]

Probably from Apion it got to Tacitus, 120 C.E., who in his Histories
(v. 4) uses the words, _effigiem_ [_asini_] _penetrali sacravere_, “they
consecrated the figure of an ass in their inner shrine.” Tacitus
expressly avoids the allegation of worshiping this statue. He probably
intentionally modified the words of Apion to fit the statement into the
then abundantly proven fact that the Jews worshiped an imageless and
abstract deity (Hist. v. 5).

The Greek essayist Plutarch, almost a generation before Tacitus, makes a
similar reference, though in his case without the least hostile or
satiric intention. The ass is according to him the animal most honored
among the Jews (τὸ τιμώμενον ὑπ αὐτῶν μάλιστα θηρίον), a
statement which, it may be said incidentally, is by no means without
foundation.[178]

It is generally assumed that the use of an ass as an object of adoration
necessarily aroused derision. That would probably be true of our own
times in Europe or in America, but it would not obtain in the ancient
world. Veneration of an ass was no more extraordinary to a Greek than
veneration of any other animal symbol. Nor was the ass associated in
men’s minds only with contemptuous and derisive images. He played a
large part in the economy of the people, and was in many places
correspondingly esteemed. The very first reference to him in Greek
literature is in the Iliad (xi. 558), where Ajax’s slow retreat is
compared to the stubborn and effectual resistance of an ass in the
fields—surely no dishonoring simile. The ass was a part of the sacred
train of Dionysus,[179] long before the latter was identified with the
Phrygian Sabazios. Again, the ass was transferred to heaven, where he
still shines as a constellation. At Lampsacus and Tarentum he was a
sacrificial animal.[180] At Rome he was associated with Vesta, and
crowned at the Consualia.

Among the Jews, as among all the people of that portion of Asia, his
importance is such as to justify in a large measure the words of
Plutarch. Generally in the Bible he is preferred to the horse (Prov.
xxvi. 3; Psalm xxxii. 9). In the ancient song of Deborah (Judges v. 10)
those who sit on white asses are the princes of the people. The Anointed
of God would ride into the city upon an ass. It is not without meaning
that asses, but not horses, appear on Assyrian sculpture.

In Egypt, however, the ass was a symbol of evil. He was associated with
the demoniac Typhon, and was an object of superstitious fear and
hatred.[181]

For most of the Mediterranean nations the worship of an ass was only in
so far contemptible as the worship of any animal was so considered.
Romans and Greeks take very lofty ground indeed when they speak of
Egyptian theriolatry, although innumerable religious practices of their
own were associated in some way or other with animals.[182] It is not
likely accordingly that the allegation of this form of fetichism against
the Jews arose among Greeks or Romans or Syrians or Palestinians. For
Egyptians, on the contrary, this particular story would charge the Jews
with “devil-worship,” or, at least, the veneration of a deity hostile to
them. In Egypt, and in Egypt alone, the story would have a special
point.

It may further be noted that in Manetho’s account the Jews are brought
to Avaris, a site consecrated to Typhon.

As it appears in Posidonius, perhaps in Mnaseas and Molo, and certainly
in Plutarch, the story is based upon a real Jewish tradition and actual
custom. In Damocritus and Apion, on the other hand, it is a malicious
slander, needing no basis in observed fact. It is one of the many
developments of the mutual hatred of Jew and Egyptian, of which there is
such a wealth of other evidence.

This story has been dealt with in some detail because it illustrates in
very many ways the character, sources, and methods of the literary
anti-Semitism of ancient times. Wholly without basis from the beginning,
it becomes almost an accepted dogma, as well grounded as many another
facile generalization in those days and ours. Further, it will be
observed that it does not everywhere necessitate the inference of
hostility on the part of the writer. The historians of those days were
_ex professo_ rhetoricians. Every form of literary composition had as
its prime object a finished artistic product. Since the subject of
literature, or artistic verbal expression, was human life, history,
which is the record of human life, was eminently the province of the
word-fancier, the rhetorician. The trained historian has no words of
sufficient contempt for the mere logographer whose object is the
recording of facts. That “pretty lies” do not in the least disfigure
history, is the opinion of the Stoic Panaetius and his pupil and admirer
Cicero. And that was particularly the case when the history was, as it
often became, an expanded plea or invective, in which case the tricks of
trade of the advocate were not only commendable but demanded.[183]

Most of the accounts of the Jews or the fragments of such accounts come
to us from just these rhetorical historians. If the whole book were
extant in any case, we should be in a position to determine the occasion
for the account and the source of its color. As it is, we are on
slippery ground when we endeavor to interpret the fragments in such a
way as to discover the facts of which they present so distorted an
image.

Not all historians, however, were of this type. Even among the rhetors,
many had, or at any rate professed to have, a passion for truth. And
among the others there is manifested from time to time a distinct
historical conscience, a qualm as to the accuracy of the assertion so
trippingly written.

It is for this reason an especially painful gap in our sources to find
that portion of Polybius missing in which he promised to treat at length
of the Jews. Polybius of Megalopolis, a Greek who lived as an Achean
hostage in Rome, in the second third of the second century B.C.E., was
the nearest approach the ancient world had to an historian in the modern
sense, one whose primary object was to ascertain the truth and state it
simply. Polybius could, for example, feel and express high admiration
for Roman institutions and at the same time do justice to the bitter
hater of the Romans, Hannibal. And this too in the lifetime of men who
may themselves have heard the dreadful news of Trasimene and Cannae.

In his sixteenth book, Polybius briefly relates the conquest of Judea
among other parts of Coele-Syria, first by Ptolemy Philometor’s general,
then by Antiochus the Great. “A little while after this, he [Antiochus]
received the submission of those of the Jews who lived around the temple
known as Jerusalem. About this I have much more to tell, particularly
because of the fame of the temple, and I shall reserve that narrative
for later.”

An evil chance has deprived us of that later narrative. If we possessed
it, we should probably have a very sane and, as far as his sources
permitted, an accurate account of the condition of the Jews during the
generation between Antiochus the Great and the Maccabees. Polybius,
however, wrote before the establishment of the Jewish state and the
spread of its cult had focused attention upon the people, and roused
opposition. And he wrote, too, at the very beginning of Roman
interference in the East, which reduced Egypt to a protectorate before
another generation. When he speaks therefore of the “great fame of the
temple” (ἡ περὶ τὸ ἱερὸν ἐπιφάνεια), he is an especially
important witness of what the name meant to the Romans and Greeks, for
whom he wrote.[184]


------------------------------------------------------------------------




                              CHAPTER XIII

                  THE OPPOSITION IN ITS SOCIAL ASPECT


If the rivals and opponents of the Jews had nothing more to say of them
than that they worshiped the head of an ass, it is not likely that their
opposition would have been recorded. But they would have put their
training to meager use, if they could not devise better and stronger
terms of abuse.

The very first Greek historian who has more than a vague surmise of the
character and history of the Jews is Hecataeus of Abdera (comp. above,
p. 92). As has been seen, his tone is distinctly well-disposed. But he
knows also of circumstances which to the Greek mind were real national
vices. He mentions with strong disapproval their credulity, their
inhospitality, and their aloofness.

Credulity is not a vice with which the Jews were charged in later times.
That may be due to Christian tradition, in which of course the sin of
the Jews is that they did not believe enough, as stated in Christian
controversial writings. But Greeks and Romans were quite in accord, that
the Jews were duped with extraordinary facility; especially that they
were the victims of the deception of their priests, so that they
attached importance to thousands of matters heartily without importance.
We may remember Horace’s jibe, _Credat_ _Iudaeus Apella_, “Tell it to
the Jew Apella”;[185] and nearly two hundred years later Apuleius
mentions the _Iudaei superstitiosi_, “the superstitious Jews.”[186]

Among the Greeks particularly the quality of εὐήθεια,
“simplicity,” had rapidly made the same progress as the words “silly”
and “simpleton” have in English.

Sharpness and duplicity were the qualities with which non-Greek nations
credited the Greeks, and whether the accusation was true or not,
“naïveté,” εὐήθεια, excited Greek risibilities more quickly
than anything else. The εὐήθεια of the Jews lay of course
not in their beliefs about the Deity. On that point all educated men
were in accord. But it lay in believing in the sanctity of the priests,
and in the observance of the innumerable regulations, particularly of
abstention, which had already assumed such proportions among the Jews.
The line of Meleager of Gadara, about his Jewish rival,

       ἕστι καὶ ἐν ψυχροῖς σάββασι θερμὸς Ἔρως,[187]

       Even on the cold Sabbaths Love makes his warmth felt,

contains in its ψυχρὰ σάββατα “cold Sabbaths,” an epitome
of the Greek point of view, ψυχρός, “cold,” was almost a
synonym for “dull.” That a holiday should be celebrated by abstention
from ordinary activities and amusements seemed to a Greek the essence of
unreason. Their own religious customs were, like those of all other
nations, full of tabus, but they were the less conscious of them because
they were wholly apart from their daily life. Jews avoided certain
foods, not merely as an occasional fast, but always. Their myths were
not irrelevant and beautiful stories, but were firmly believed to be the
records of what actually happened. The precepts of their code were
sanctioned, not merely by expediency, but by the fear of an offended
God.

An excellent example of how the rhetorical τόπος of
“naïveté” was handled is presented by Agatharchidas of Cnidus, who wrote
somewhere near 150 B.C.E.[188]

He tells us of Stratonice, daughter of Antiochus Soter and wife of
Demetrius of Macedon, who was induced by a dream to remain in a
dangerous position, where she was taken and killed. The occasion is an
excellent one to enlarge upon the topic of superstition, and
Agatharchidas relates in this connection an incident that is said to
have happened one hundred years before Stratonice, the capture of
Jerusalem by Ptolemy Soter through the fact that the Jews would not
fight upon the Sabbath. “So,” says Agatharchidas, “because, instead of
guarding their city, these men observed their senseless rule, the city
received a harsh master, and their law was shown to be a foolish
custom.” One cannot reproduce in English the fine antitheses of the
related words φυλάττειν τὴν πόλιν balanced by
διατηρούντων τὴν ἄνοιαν, νόμος answering to
ἐθισμόν; but, besides the artificiality of the phrases, the
total absence of any attempt to make the words fit the facts is shown by
the conclusion to which Agatharchidas, by rule of rhetoric, had to come.
Now a “harsh master” is just what Ptolemy was not to the Jews, and
Agatharchidas of all men must have been aware of that fact, for he wrote
not only at Alexandria, but at the court of Philometor, an especial
patron of the Jews individually and as a corporation.

The practice of the Sabbath was one of the first things that struck
foreigners. It is likely that the congregations of Sabbatistae in Asia
Minor were composed of Jewish proselytes.[189] The name of the Jewish
Sibyl Sambethe,[190] the association of Jewish worship with that of the
Phrygian Sabazios,[191] were based upon this highly peculiar custom of
the Jews. But its utter irrationality seemed to be exhibited in such
instances as Agatharchidas here describes, the abstention from both
offensive and defensive fighting on the Sabbath.

Whether the incident or others of the same kind ever occurred may
reasonably be doubted. The discussion of the question in Talmudic
sources is held at a time when Jews had long ceased to engage in
warfare.[192] Their nation no longer existed, and their legal privileges
included exemption from conscription, if they chose to avail themselves
of it. In the Bible there is no hint in the lurid chronicles of wars and
battles that the Sabbath observance involved cessation from hostilities
during time of war, and the supposition that no resistance to attack was
offered on that day is almost wholly excluded. It is not easy to imagine
one of the grim swordsmen of David or Joab allowing his throat to be cut
by an enemy because he was attacked on the Sabbath.

That any rule of Sabbath observance which demanded this had actually
developed during the post-Exilic period is likewise untenable. The Jews
served frequently in the army under both Persian and Greek rule. This is
amply demonstrated by the Aramaic papyri of Elephantine and the
existence of Jewish mercenaries under the Ptolemies.[193] The
professional soldier whose service could not be relied upon one day in
seven would soon find his occupation gone.

Several passages in the Books of Maccabees have often been taken to
imply that the strict observance of the Sabbath was maintained before
the Hasmonean revolt, and deliberately abrogated by Mattathiah (I Macc.
ii. 30-44; II Macc. viii. 23-25). But upon closer analysis it will be
seen that the incidents there recorded do not quite show that. The
massacre of the loyal Jews in the desert was a special and exceptional
thing. They were not rebels in arms, but hunted fugitives. Their passive
submission to the sword was an act of voluntary martyrdom (I Macc. ii.
37). ἁποθάνωμεν οἱ πάντες ἐν τη ἁπλότητι ἡμων: μαρτυρει ἐφ’ ἡμας
hὃοὐρανος καὶ ἡ γἡ ὁτι ακριτως ἀπόλλυτε ἡμας, “Let us all die in
our innocence. Heaven and earth bear witness for us that ye put us to
death wrongfully.”

Again, it is not Mattathiah, but the sober reflection of his men, that
brings them to the resolution that such acts of martyrdom, admirable as
they are in intention, are futile. The decision is rather a criticism of
their useless sacrifice than anything else.

Similar acts of self-devotion on the part of inhabitants of doomed
cities were not uncommon. As final proofs of patriotism on the part of
those who would not survive their city, they received the commendation
of ancient writers.[194] But to kill oneself or allow oneself to be
killed for a fantastic superstition, could have seemed only the blindest
fanaticism.

Now there is no reason for doubting the essential accuracy of the report
in I Maccabees, to the effect that one group of Jewish zealots chose
passive resistance to the attempt of Antiochus, and by that nerved the
Hasmoneans to a very active resistance. And it is very likely that in
this event we have the basis for the stories that related the capture of
Jerusalem—almost in every case—on the Sabbath. The story is told of the
capture by Nebuchadnezzar, by Artaxerxes Ochus, by Ptolemy, and by
Pompey. It is a logical inference from the non-resistance of the
refugees mentioned in I Maccabees. The conditions of ancient warfare
make it highly improbable that it was more.

The rationalist Greek or Roman felt it a point of honor to hold in equal
contempt the “old-wives’ tales” of his own countrymen as to the
supramundane facts with which the myths were filled,[195] and the vain
and foolish attempts by which barbarians, and Greeks and Romans too,
sought to dominate the cosmic forces or tear the secret from fate. These
attempts generally took the form of magic, not, however, like the
primitive ceremonies, of which the real nature had long been forgotten,
but in the elaborate thaumaturgic systems which had been fashioned in
Egypt, Persia, and Babylon. In their lowest forms these were petty and
mean swindling devices. In their more developed forms they contained a
sincerely felt mysticism, but under all guises they aroused the contempt
of the skeptic, to whom the most ancient and revered rites of his own
cult were merely ancestral habits which it did no harm to follow. The
tone such men adopted toward the complicated Oriental theologies and
rituals was very much like that of modern cultivated men toward the
various “Vedantic philosophies,” which at one time enjoyed a certain
vogue. Those who seriously maintained that by the rattling of a sistrum,
or the clash of cymbals, or by mortifications of the flesh, influences
could be exerted upon the laws that governed the universe, so as to
modify their course or divert them, were alike insensate fools, whose
chatter no educated man could take seriously. The Jews, who observed,
even when they were less rigorous, a number of restrictive rules that
gravely hampered their freedom of action, who seriously maintained that
they possessed a direct revelation of God, were fanatics and magicians,
and exhibited a credulity that was the first sign of mental inferiority.

“Senseless,” “nonsense,” ἀνοητός, ἄνοια, are terms that are
principally in the mouths of the Philopator of III Maccabees and the
Antiochus of IV Maccabees, in whose words we may fairly see epitomized
all the current abuse as well as criticism which opponents to the Jews,
from philosophers to malevolent chauvinists, heaped upon them.

Hecataeus says of Moses that he instituted an “inhospitable and strange
form of living.”[196] The two words μισόξενον and
ἀπάνθρωπον form a _doublette_, or rhetorical doubling of a
single idea. That idea is “inhospitality,” lack of the feeling of common
humanity, a term which for Greeks and Romans embodied a number of
conceptions not suggested by the word to modern ears.

The word ξένος, which is the root of the words for
“hospitality” and its opposite, has no equivalent in English. A
ξένος was a man of another nation, who approached without
hostile intent. The test of civilization was the manner in which such a
ξένος was dealt with. The Greek traditions, even their
extant literature, have a very lively recollection of the time when
hospitality was by no means universal, when the ξένος was
treated as an enemy taken in arms or worse. The one damning epithet of
the Cyclops is ἄξενος, “inhospitable.”[197] The high
commendation bestowed upon the princely hospitality of the Homeric
barons itself indicates that this virtue was not yet a matter of course,
and that boorish nations and individuals did not possess it.

Legally, of course, the ξένος had no rights. Such claim as
he could make for protection rested upon the favor of the gods,
especially of Zeus, who was frequently addressed by the cult title of
Ξένιος, the Protector of Strangers. The uncertain aid of
the gods was soon displaced by personal relations between individuals
and groups of individuals in different states, who were mutually
πρόξενοι to each other, a title that always created a very
definite moral obligation and soon a legal one as well. So, when
Alexander destroyed Thebes, he spared the πρόξενοι of his
own family and of the Macedonians in general.[198]

The institution and the development had practically gone on in similar
ways all through the Mediterranean world. The Bedouins still maintain
the ancient customs of their fathers in that respect. The Romans had the
word _hospes_, of which the history is a close parallel to that of
ξένος.

Of the Jews the same thing may be said. The Bible enjoins the protection
of strangers as a primary obligation. They were the living symbols of
the Egyptian bondage. So Exodus xxiii. 9, “Also thou shalt not oppress a
stranger, for ye know the heart of a stranger, seeing ye were strangers
in the land of Egypt.” One of Job’s protests of righteousness is his
hospitality (Job xxxi. 32).

In these circumstances just what could the charge of
μισοξενία, of “inhospitality,” have meant? We shall look in
vain in Greek literature for an injunction to hospitality as finely
phrased as the passage just quoted from Exodus. To understand the term
as applied to the Jews we shall have to examine the words that are used
for the acts connected with hospitality.

In Homer the word ξεινίζω[199] is frequently found.
Strictly of course it means simply “to deal with a stranger,” but it is
used principally in the sense of “entertain at dinner.” The wandering
stranger might as such claim the hospitality of the people among whom
chance had brought him, and claim it in the very concrete sense that
food and lodging at the master’s table were his of right. Indeed it
would almost seem that he became _pro hac vice_ a member of the family
group in which he partook of a meal, protected in life and limb by the
blood-vengeance of his temporary kinsmen.

That however seems to have been the general rule in the older
communities of the East, in Palestine just as in Greece and Asia. There
was no feeling against entertaining a stranger at table among the Jews,
although the relation could not well be reversed. And there was the rub.
It was not in Palestine (where the Jew was likely always to be the
host), but in the communities in which Jew and non-Jew acknowledged the
same civic bond that the refusal of the Jew to accept the hospitality of
his neighbor would be a flagrant instance of μισοξενία, of
dislike of strangers. We need not suppose that it needed careful
investigation and the accumulation of instances to produce the
statement. A few incidents within anyone’s experience would suffice. We
shall have to remember further that we are dealing with a literary
tradition in which many statements are taken over from the writer’s
source without independent conviction on his own part.

However, among the great masses the general feeling that the
Jews disliked strangers, and so were properly to be termed
μισόξενοι, was in all likelihood based on an observation of
more obvious facts than dietary regulations. It is principally in meat
diet that the separation is really effective, and meat diet was the
prerogative of the rich. Then, as now, the great majority of the people
ate meat rarely, if at all, and surely could take no offense at a man’s
squeamishness about the quality or nature of the food he ate. But what
everybody was compelled to notice was that the Jews deliberately held
aloof from practically all public festivities, since these were nearly
always religious, and that they created barriers which seemed as
unnecessary as they were foolishly defended. That in itself could be
interpreted by the man in the street only as a sign of deep-rooted
antipathy, of μισοξενία.

This accusation, as has been shown, was more than the reproach of
unsociability. The vice charged by it was of serious character. Those
individuals who in Greek poetry are called inhospitable are nothing
short of monsters. It implied not merely aloofness from strangers, but
ill-usage of them, and that ill-usage was sometimes assumed to be
downright cannibalism. So Strabo (vii. 6) tells us that the
“inhospitable” sea was called so, not only because of its storms, but
because of the ferocity of the Scythian tribes dwelling around it, who
devoured strangers and used their skulls for goblets. That was of course
to be inhospitable with a vengeance, but the term covered the extreme
idea as well as the milder acts that produced at Sparta and Crete
frequent edicts of expulsion (ξενηλασίαι)[200] and a
general cold welcome to foreigners.

In very many cases, especially in the rhetorical schools,
“inhospitality,” “hatred of strangers,” was a mere abusive tag,
available without any excessive consideration of the facts. And when
intense enmity was to be exhibited, the extreme form of “inhospitality”
was naturally enough both implicitly and expressly charged against the
objects of the writer’s dislike.


[Illustration: GREEK INSCRIPTION, FOUND ON SITE OF TEMPLE AREA,
FORBIDDING GENTILES TO PASS BEYOND THE INNER TEMPLE WALLS AT JERUSALEM
(Now In the Imperial Ottoman Museum Constantinople)]


There are many instances in which the hereditary enemy was credited with
human sacrifice or cannibalism. Indeed it was currently believed that
cannibalism had universally prevailed at one time, and with advancing
civilization was gradually superseded.[201] As far as human sacrifice
was concerned, many highly civilized states knew of vestiges or actual
recurrences of it in their own practice. Rome is a striking example. But
in Rome such things were rare exceptions, employed in times of unusual
straits to meet a quite unusual emergency.[202] In Greece there were
many traces frankly admitted to be such—if not actual instances of such
sacrifices. But here, as at Rome, the act was admittedly something out
of the ordinary, a survival of primitive savagery.[203]

Accordingly when Greeks and Romans spoke of human sacrifices, it was not
of an inconceivable form of barbarity, which placed those who took part
in it quite out of the human pale, but as a relic of a condition from
which they had themselves happily grown, and to which they reverted only
in extremities. Its presence among other tribes was a demonstration that
they were still in the barbarous stage, and especially was it deemed to
be so when all strangers who chanced to come upon the foreign shore were
the selected victims of the god.

That charge, as we know, was made against many Scythian and Thracian
tribes. The story of Iphigenia in Tauris is an example of it. It was
made against the Carthaginians, at least in the early stages of their
history. The Gauls, according to both Greek and Roman writers, had made
of it a very common institution.[204] We do not know very much of the
evidence in the case of the Thracians, Scythians, and Gauls. It is not
impossible that customs like certain symbolic rites found in many places
were misinterpreted. Or it is highly likely that, if human sacrifices
existed, they were, as among Greeks and Romans, a rare form of
expiation. For the Carthaginians the story is almost certainly a
by-product of national hatred, and rests upon the same foundations as
the “cruelty” and “perfidy” of Hannibal.

Human sacrifices, similar to those of Greece and Rome, existed in
Palestine. Children were sacrificed to the nameless god or gods that
bore the cult title of _melech_, _i.e._ “king.” As in the rest of the
Mediterranean world such sacrifices were exceptional and grisly forms of
expiation, used when ordinary means had failed. Among the Jews, on the
other hand, they seem to have been prohibited from the very beginning of
their history as a community. It is a purely gratuitous theory that
makes _melech_, or _molech_, a cult-title of Yahveh in Israel. There is
simply no evidence of any kind that it was so. On the contrary, the
oldest traditions of the Jews represent the abolition of human
sacrifices as one of the first reforms instituted by the founders of
their faith. The Mosaic code made these sacrifices a capital offense
(Lev. xviii. 21; xx. 2). The very name _molech_ indicates an intense
abhorrence, if, as has been plausibly suggested, it is simply
מלך, or “king,” with the vowels of בשת, “the
Abomination.”[205]

With so old a tradition on the subject, the Jews must have felt, as
peculiarly irritating, the transference of this vituperative tag to
them. That it might be so applied was of course an inevitable expansion
of the belief that the Jews were μισόξενοι, “haters of
strangers.” However, it must not be supposed that the statement was
widely current. On the contrary, we have only two references to it.
Damocritus, who lived perhaps in the first century B.C.E., as quoted by
the late Byzantine compiler Suidas,[206] asserts that the Jews captured
a stranger every seven years, and sacrificed him to their god; and
Apion, in the first century C.E., relates the circumstantial story of
the captured Greek who was found immured in the temple by Antiochus
Epiphanes.

The latter story is an amusing instance of rhetorical method. Of its
baselessness of course no proof need be adduced. It is almost certainly
the concoction of Apion himself, perhaps based upon some such statement
as this just quoted from Damocritus. Its melodramatic features, the
fattening of the stranger, the oath sealed by blood, are highly
characteristic of Apion’s style.

It cannot be said that this particular charge against the Jews had any
real success. The later writers do not mention it. Tacitus and Juvenal,
both of whom are very likely to have read Apion, pass by the story in
silence. And Juvenal, who in his Fifteenth Satire expresses such
detestation of a similar act among the Egyptians he abominated,[207]
would certainly not have let off the Syrian fortune-tellers, whom he
equally disliked, with an allusion to their unsociability.

_Non monstrare vias nisi eadem sacra colenti_,[208] “They are instructed
not to point out a road except to those who share their rites.” It might
almost seem as though even rhetorical animosity demanded more for its
terms of abuse than the authority of Apion.

The tragic importance of the “ritual murder” in the modern history of
the Jews since the Crusades has given the account of Apion a
significance to which it is by no means entitled. The least analysis
will show that the “ritual murder” of modern times is not really like
the ancient story at all. The latter is simply an application to the
Jews of the frequent charge of ξενοθυσία, “sacrifice of
strangers,” such as was made against the Scythians. And Apion’s fable
found practically no acceptance. There is of course no literary
transmission between Apion and the chroniclers of Hugh of Lincoln, but
we cannot even suppose that there was a popular one. In the fearful
struggles of the rebellions under Hadrian and Trajan, it is impossible
to believe that the mutual hatred, which found such expression as the
massacre at Salamis and the reprisals of the Greeks, would have failed
to register this charge against the ἀνόσιοι Ἰουδαῖοι, “the
wicked Jews,” if it were known.

The early Middle Ages, at any rate from the Crusades on, devised the
“ritual murder” without the aid of older authorities. It is one of the
many cases in which parallel developments at different times and in
different places produce results that are somewhat similar, although
only superficially so.[209]


------------------------------------------------------------------------




                              CHAPTER XIV

                       THE PHILOSOPHIC OPPOSITION


A favorite adjective in describing the Jews was “superstitious.”
Strangely enough, another, perhaps even more general, was “irreligious.”
The Jews were frequently stigmatized as ἄθεοι, a word
generally translated “atheist,” and undoubtedly often used in the sense
of the modern term. It remains to be seen whether the term meant, in its
application to the Jews, all that the corresponding modern term implies.
That is particularly necessary here, since to the modern world the
devotion of the nation to its Deity is its most striking characteristic,
and at least one of the key-notes of its historical development. Upon us
it has almost the effect of a paradox to read that this people impressed
some Greeks as a nation of “atheists” or “godless.”

The modern term and the ancient partly cover each other. Both often
denote the speculative negation of a supernatural direction of the
world. Now it simply cannot be, in view of the wide distribution of the
Jews and their successful propaganda, that even the unthinking could
associate the people whose claims to direct divine guidance were so many
and so emphatic, with a term that implied the non-recognition of any
god. We may remember how even the very first contact had seemed to
emphasize the religious side of the Jewish communal life.

The usual explanations will not bear analysis. It is frequently asserted
that “atheist” was applied to the Jews because of their imageless cult.
The natural inference, we are told, from the fact that there were no
statues was that there were no gods. But that is to assign to the statue
a larger importance in ancient religious theory than in fact belonged to
it. We meet, to be sure, cases where the identification of the statue
and the resident deity seems to be complete. Especially in such scoffers
as Lucian,[210] or in the polemics of the philosophic sects, or in those
of Jews and Christian writers, Romans and Greeks are often charged with
the adoration of the actual figure of stone or bronze. That, however,
was surely not the general attitude of any class. The passages that seem
to show it are generally figurative and often imply merely that the god
had taken his abode within the statue, and might leave it at will.

Indeed, just for the masses, the most intense and direct religious
emotions were always aroused, not by the great gods whose statues were
the artistic pride of their cities, but by the formless and bodiless
spirits of tree and field and forest that survived from pre-Olympian
animism. And these latter, if adored in symbolic form, were represented
generally by pillars or trees, and not by statues at all.

Nor were the Jews the only imageless barbarians whom the Greeks and
Romans encountered. Most of the surrounding nations can scarcely have
possessed actual statues at first. And the Greeks or Romans drew no such
inference as atheism from the fact that they found no statues of gods
among Spaniards, Thracians, Germans, or Celts. On the contrary, we hear
of gods among all these nations, many of them outlined with sufficient
clearness to be identified promptly with various Greek deities. What a
Greek would be likely to assume is rather that these barbarians lacked
the skill to fashion statues or the artistic cultivation to appreciate
them. If it occurred to him to explain the imageless shrine at Jerusalem
at all, he would no doubt have offered some such statement, especially
as it was quite common to assume lack of artistic skill in barbarians.

Atheism as a philosophic doctrine was relatively rare. Diagoras of
Melos, a contemporary of Socrates, and Theodore of Cyrene,[211] a
contemporary of the first Ptolemy, were said to have held that doctrine,
and the former was known from it as “the Atheist.” However, even in this
case we cannot be quite sure of our ground. Some of the poems of
Diagoras seem to have a distinct, even a strong, religious feeling.
Josephus asserts that Diagoras’ offense in Athenian eyes was scoffing at
the mysteries.[212] If that is true, he received his sobriquet less from
atheism, as we understand it, than from the same facts that brought
Protagoras, Anaxagoras, and Socrates himself within the ban of the
Athenian police. That is, he was charged rather with contempt of the
actually constituted deities of the Athenian state than with a general
negation of a divinity. The term itself, ἄθεος, is not
necessarily negative. In fact, Greek had very few purely negative ideas.
In Plato’s Euthyphro[213] the only alternatives that are admitted are
θεοφιλές and θεομισές, _i.e._ what the gods
hate and what the gods love. So the various Greek adjectives compounded
with “α privative,” ανωφελής, “useless,”
ἄβουλος, “thoughtless,” are really used in a positive sense
contrary to that of the positive adjective. So ανωφελής is
rather “harmful” than merely “useless”; ἄβουλος is
“ill-advised”; etc. The word ἄθεος would, by that analogy,
rather denote one that opposed certain gods than one who denied them. A
man might be ἄθεος in one community and not in another.
Indeed his “atheism” might be an especial devotion to a divine principle
which was not that recognized by the state.

In ordinary literary usage ἄθεος is denuded even of this
significance. It means little more than “wicked.” It is used so by
Pindar, by Sophocles, and in general by the orators. Often it runs in
pairs with other adjectives of the same character. Xenophon calls
Tissaphernes (An. II. v. 29) ἀθεότατος καὶ πανουργότατος,
“most godless and wicked,” in which the superlative is especially
noteworthy. As a matter of fact it is often used of a man whom the gods
would have none of, rather than one who rejects the gods. Ἄθεος,
ἄφιλος ὀλοίμαν, cries the chorus in Oedipus Rex, “May I die
abandoned by gods and men.”[214]

When it is first used of the Jews by Molo, it is as part of just such a
group; ἄθεοι καὶ μισάνθρωποι, he calls the Jews, “hateful
to gods and men,” and other rhetoricians follow suit. As a term of
abuse, ἄθεος was as good as any other.

But there may have been a more precise sense in which the Jews might by
an incensed Greek be properly stigmatized as ἄθεοι. To the
thoroughgoing monotheists, the gods of the heathen are non-existent.
They are not evil spirits, but have no being whatever. The prophets and
the intellectual leaders of the Jews held that view with passionate
intensity. But even they used language which readily lends color to the
view that these gods did exist as malignant and inferior _daemonia_. The
“devils” of Leviticus xvii. 7 are undoubtedly the gods of other
nations.[215] The name “Abomination,” which for the Jew was a
cacophemism for “god,” equally implies by its very strength a common
feeling of the reality of the being so referred to. Likewise the other
terms of abuse which the Jews showered upon the gods of the heathen
indicate a real and fiercely personal animosity.

Hatred and bitterness formed almost a religious duty. An implacable war
was to be waged with the abominable thing, and it is not likely that
dictates of courtesy would stand in the way. The retort of
ἄθεοι would mean no more than a summary of the fact that
the Jew was the declared enemy of the constituted deity, whose anger he
provoked and whose power he despised.[216]

Something of this appears in the statement of the Alexandrian
Lysimachus, that the Jews were enjoined to overturn the altars and
temples which they met (Josephus, Contra Ap. i. 34), and in the phrase
of the elder Pliny (Hist. Nat. XIII. iv. 46), _gens contumelia numinum
insignis_, “a race famous for its insults to the gods.”

Most of the phrases that have been quoted have been taken from works
where they were little more than casual asides imbedded in matter of
different purport. Rhetoricians, in attempting to establish a point, use
some phrase, either current through popular usage or a commonplace in
their schools. In this respect the Jews fare no better and no worse than
practically all nationalities of that time. Individual writers disliked
or despised various peoples, and said so in any manner that suited them.
Slurs against Romans, Athenians, Boeotians, Egyptians, Cappadocians are
met with often enough. The Cretans were liars, the Boeotians guzzlers,
the Egyptians knaves, the Abderitans fools; antiquity has furnished us
with more than one entertaining example of national hate and
jealousy.[217] The epithets which the Acheans showered on their Aetolian
rivals certainly leave nothing to be desired as far as intensity is
concerned.[218] The various panders of Roman comedy often are
represented as particularly choice specimens of Agrigentine
character.[219] Cicero particularly knew from his rhetorical masters how
to use national prejudices in the conduct of his business. If Celts are
the accusers of his client, as they were in the case of Fonteius, they
are perjurers, murderers, enemies of the human race. “Tribes,” he says,
“so far removed from other races in character and customs that they
fight, not for their religion, but against the religion of all
men.”[220] If they are Sardinians, these are a “tribe whose
worthlessness is such that the only distinction they recognize between
freedom and slavery is that the former gives them unlimited license to
lie.”[221]

To take this seriously is to misconceive strangely both the functions of
an advocate and the license of rhetoric. Now the abusive paragraphs
directed against the Jews are quite of this type. And it is in the
highest degree extraordinary that these phrases, which, in the instances
just cited, are given no weight in determining national attitude, should
be considered of the highest importance in the case of the Jews. Whether
it was Syrian, Greek, or Celt that was attacked, the stock epithet means
no more than the corresponding terms of our own day mean.

But besides these occasional flings there were whole books directed
against the Jews, and to that fact a little attention may be given.

It is a relatively rare thing that a writer should nurse his bile
against a particular people to the extent of expanding it into a whole
book. We must of course remember that a “book” was sometimes, and
especially in this polemical literature, a single roll, and we are not
to understand it in the sense of a voluminous treatise. However, there
were such books and these we must now consider.

What such a book was like, recent anti-Semitism has made it very easy to
imagine. There is no reason to suppose that this type of pamphlet was
appreciably different in those days. It consisted of a series of bitter
invectives interspersed with stories as _pièces justificatives_. Now and
then an effort is made to throw it into the form of a dispassionate
examination. But even in very skilful hands that attitude is not long
maintained.

Of several men we know such treatises. All have already been
mentioned—Apollonius Molo, Damocritus, and probably Apion.

Apollonius, either son of Molo, or himself so named, was one of the most
considerable figures of his day. He taught principally, but not
exclusively, at Rhodes, and numbered among his pupils both Cicero and
Caesar. As a rhetorician he enjoyed an extensive and well-merited
influence. It was during his time that the reaction against the florid
literary style of Asia culminated in the equally artificial simplicity
of the Atticists—a controversy of the utmost importance in the history
of Latin literature no less than Greek. The doctrine of _mediocritas_,
“the golden mean,” set forth by Molo, moulded the style of Cicero and
through him of most modern prose writers. The refined taste and good
sense which could avoid both extremes justify his repute and power.

He was a voluminous writer on historical and rhetorical subjects. Only
the smallest fragments remain, not enough to permit us to form an
independent estimate of his style or habits of thought. Just what was
the incentive for the pamphlet he wrote against the Jews it is
impossible to conjecture. But it is not likely that it contained many of
the specially malignant charges. To judge from Josephus’ defense, it
seems to have concerned itself chiefly with their unsociability, and may
have been no more than a sermon on that text. Josephus’ charge against
him is that of unfairness. There is none of the abuse in Josephus’
account of Molo which he heaps upon Apion. We may accordingly infer that
Molo’s pamphlet was considerably less offensive. It may have been, in
effect, a mere _declamatio_, a speech in a fictitious cause, or the
substance of an oration delivered in an actual case. Or perhaps a single
instance of personal friction produced it as an act of retaliation. The
rhetoricians of those days were essentially a _genus irritabile_, and
their wrath or praise was easily stirred.

Of Damocritus we know almost nothing. Suidas, a late Byzantine
grammarian, mentions a short work of his on Tactics, and one as short,
or shorter, on the Jews. The reference to human sacrifice (above, p.
189), might be supposed to indicate a strong bias. While it is likely
enough that it was hostile in character, that single fact would not
quite prove it, since we do not know whether Damocritus represented
these human sacrifices as an ancient or a still-existing custom.

The third name, Apion, has become especially familiar from the apology
of Josephus. The latter refers to him throughout as an Egyptian, and in
spite of certain very warm and modern defenders, he very likely was of
Egyptian stock. From the Oasis where he was born, he came to Alexandria,
where he established a great reputation. Undoubtedly possessed of
fluency and charm as a speaker, he was a most thoroughgoing charlatan, a
noisy pedant wholly devoid of real critical skill. He boasted of magical
power, through which he was enabled to converse with the shade of Homer.
His vanity prompted the most ludicrous displays of arrogance. Tiberius
Caesar dubbed him the _cymbalum mundi_, “the tom-tom of the world,” a
characterization that seems to have been generally accepted.[222]

In the appeal of the Jewish residents of Alexandria against the
maladministration of the prefect Flaccus, argued before the emperor, he
represented the Alexandrian community, whose acts were the basis of the
charge made by the Jews. As such he no doubt delivered an anti-Jewish
invective, and it is at least likely that this speech formed the
substance of his book on the subject, just as the defense of the Jews
and the attack upon Flaccus are contained in the two extensive fragments
of Philo, the _Legatio ad Gaium_, and the _In Flaccum_.

It has been doubted whether he really wrote such a book, although there
are express statements that he did. It is true enough that those who
assert it may easily have been misled by the fact that certain books of
his History of Egypt may have contained these anti-Jewish passages or
most of them. None the less, the fact that he must have prepared a set
speech in the case mentioned, coupled with the statements of Clemens of
Alexandria and Julius Africanus, renders the older view the more
probable.[223] There would of course be nothing strange if the books of
the History of Egypt and a special monograph contained essentially the
same material.

As to other similar pamphlets, we hear of a περὶ Ἰουδαίων
by a certain Nicarchus, son of Ammonius, which may have had an
“Egyptian” bias, in that Moses is said to have been afflicted with white
scales upon his body—an assertion that seems to be a revamping of
Manetho’s “leprous outcasts.” But the title of the book does not point
to a wholly hostile attitude, nor does the passage referred to
necessarily imply such an attitude.[224]

Taking all these passages together, from Manetho to Apion, one thing
must be evident: Manetho himself, Mnaseas, Agatharchidas, Chaeremo,
Lysimachus, Apion, are either Egyptians or are trained in Alexandria,
and represent the Egyptian side of a bitter racial strife, as intense
and lasting as was generally the case when the same community contained
several compact groups of different political rights and privileges.

The conditions of the population of Alexandria have been previously
discussed. It was the great market center of the East, and as such of
the Mediterranean world, since the commercial and intellectual hegemony
was always east of the Aegean Sea. The population had been a mixed one
since its foundation. The warped notions that have often been held of
the position of the Jews there are due to a failure to realize
concretely how such a city would be likely to grow. The Greeks and
Macedonians that were originally settled there undoubtedly constituted a
real aristocracy, and made that attitude very thoroughly felt. One thing
further is clear, that the native Egyptians, who probably formed the
mass of the populace, looked upon these Greeks as they did upon all
foreigners, with intense dislike. We have a document in which a Greek
suitor in court impugns the credibility of Egyptian testimony against
him because of the well-known hatred Egyptians bear toward Greeks.[225]

Egyptian animosity toward Jews had been of longer standing simply
because intercourse in close proximity was much older. Further, the
Jewish colonies from early Persian times had always represented the
foreign master. It was as natural, therefore, for this animosity to
express itself in street-conflicts in Alexandria as for anti-Greek
feeling to be manifested there. Those modern investigators who have
confidently asserted that Alexandrian “anti-Semitism” was of Greek
origin and leadership have permitted the rattle of the _cymbalum mundi_
to confuse their minds. For it is Apion and Apion alone that makes the
claim that the Jews are especially embittered against Greeks, and seeks
to create a general Greek feeling against them. His motives are too
apparent to need comment, and there is no evidence whatever that he was
successful.

Further, it is the Egyptians Manetho and Apion whose tirades have a
fiercely personal coloring. The Greek Alexandrians make their
anti-Jewish polemics on the basis of general theories, and particularly
lay stress on what was to them the perfectly irrational separatism which
the Jews had made a part of their religion. As has been frequently
shown, the relatively small fragments of these writers do not enable us
to say how far this Jewish characteristic is used to point a moral, much
as the modern clergy takes chauvinistic commonplaces to illustrate the
evil results of doctrines they are attacking.

In the case of two Greeks, Posidonius of Apamea in Syria, and Molo, no
Egyptian influence can be shown. Both were among the most influential
men of their time. Molo’s career and importance have been briefly
sketched. To Posidonius must be assigned a still more powerful
intellectual influence over his generation and those that followed.[226]
The leader of the Stoic school or, as it may well be called, sect, he so
reorganized its teaching that the Stoa became nothing else than the
dominant faith among cultivated men, a situation perhaps paralleled by
Confucianism in China, which is also an ethical philosophy that finds it
possible to dwell on terms of comity with various forms of cruder
popular belief.

What Molo’s philosophic affiliations were is not easy to determine. The
Stoics were nearer than most other schools to rhetoricians and
grammarians, but many men of these professions acknowledged allegiance
to the Academy, to Epicureanism, or even to the revived Pythagoreanism
of the first century B.C.E. Of the extensive writings of the Rhodian
rhetorician there is not enough left to give even a probable answer.

But most philosophic sects laid stress on the universality of their
teachings, and were marked by an intense intellectual rationalism. The
crude psychology of those days made the formation of categories a simple
thing. Thinkers could scarcely be expected to admit that inherited
instincts could qualify the truth of a philosophic dogma. More
particularly, the philosophic movements were powerful solvents of
nationalism. Even the distinction between Greek and barbarian did not
exist in theory for them.[227] The notion of the state and the
maintenance of its ancestral rites became for them a meaningless but
innocuous form, which men of common sense would not despise, but to
which one could attach no great importance.

Face to face with congregations like those of the Jews, which enforced
their separation by stringent religious prohibitions, the Stoics more
than others found their opposition roused. More than others, because
many Stoics adopted from the Cynical school the methods of the diatribe,
the popular sermon, and, indeed, made an active attempt to carry the
universality of their principles into practice. And the Stoics, more
than others, would find the height of irrationality in the stubborn
insistence on forms for which only an historical justification could be
found.

A highly interesting document, which gives a certain phase of the
controversy, or perhaps even fragments of an actual controversy, between
the general philosophic and the Jewish doctrine, has come down to us in
the tract known as the Fourth Book of Maccabees. The author announces
his purpose of setting forth a most philosophic thesis, to wit, whether
the pious reason is sovereign over the passions. The philosophic
argument, which fills the first three chapters, is Stoic in form and
substance. Then, to illustrate his point, he cites certain vaguely
remembered stories of II Maccabees, which he expands into highly
detailed dramatic forms. In the mouth of Antiochus Epiphanes are placed
the stock philosophic arguments against the Jews, which are triumphantly
refuted by the aged Eleazar and the seven sons of Hannah.

So we hear Epiphanes reasoning with Eleazar and urging him to partake of
swine’s flesh (IV Macc. v. 8 seq.):

    For it is obviously a senseless proceeding to refrain from enjoying
    those pleasures of life which are free from shame: it is even wicked
    to deprive oneself of the bounties of nature. And it seems to me
    that your conduct will be still more senseless, if you provoke my
    anger because of your zeal for some fancied principle. Why do you
    not rid your mind of the silly doctrine of your people? Discard that
    stupidity which you call reason. Adopt a form of thought that suits
    your age, and let your philosophic principle be one that actually
    serves you.... Further consider this: If in the Deity you adore
    there is really a power that oversees our deeds, it will grant you
    full pardon for all transgressions which you have been forced to
    commit.

To a Greek, and no doubt to many modern men, the reasoning is
conclusive. It presents the Greek point of view very well indeed, and is
doubtless the epitome of many conversations and even formal disputes in
which these matters were discussed between Greek and Jew. And just as
the argument of Epiphanes seems strangely modern in its appeal to common
sense and expediency, so the answer of Eleazar rings with a lofty
idealism that is both modern and ancient:

    We, whose state has been established by God, cannot admit that any
    force is more powerful than that of the Law. Even if, as you assume,
    our Law were not divine, yet, since we suppose that it is, we durst
    not set it aside without gross impiety.

Eleazar then proceeds to elaborate upon the Stoic paradox that the
slightest and the greatest transgressions are equally sinful;[228] and
that in so far as abstention is a form of self-control, it is an
admirable and not a contemptible act. After a detailed account of the
hideous sufferings heroically endured by the priest, the author breaks
out into a panegyric of him as a maintainer of the Law, in which the
fundamental Stoic proposition with which he begins is less prominent
than his intense Jewish piety.

For us, however, the prime importance lies in the sharp contrast between
the Greek and the Jewish attitude. Upon the philosophically cultured
man, the reasoning of Epiphanes could not fail to produce a certain
impression. In the case of the seven sons of Hannah, while many elements
are repeated (IV Macc. viii. 17 seq.), the writer has in mind the appeal
to the flesh, which Hellenism made. “Will you not change your mode of
life for that of the Greeks and enjoy your youth to the full?” asks
Antiochus (_ibid._ viii. 8); and that no doubt was the whisper that came
to the heart of many a young man, surrounded by the bright and highly
colored life of the Hellenic communities in which he dwelt. There is no
exchange of vituperation. The denunciations hurled against Antiochus are
impersonal, indeed are generic. He is the type of tyrant, another
Busiris or Phalaris, a bowelless despot. And the one word which
alternates with “senseless” in the mouths of Antiochus and his
executioners is “mad.”

The actual events described are of course quite unhistorical. But we do
not find here any of the various forms in which racial animosity or
personal spleen exhibited itself against the Jews. In spite of the
setting, the controversy is, judged by disputation standards, quite
decorous. The terms that qualify the Jewish doctrine as “irrational” are
almost controversial commonplaces. The martyrs do not resent the
epithet. They seem to accept it as the logical inference of the carnal
philosophy of their oppressors and claim to be justified by a higher
wisdom.

Jewish and Greek life began to touch each other at many points in the
six or seven generations that intervened between Alexander and Caesar.
Hellenism dominated the political and social culture of the Eastern
Mediterranean, although the nationalities it covered were submerged
rather than crushed. In Egypt the indigenous culture maintained itself
successfully, and forced concessions from the conqueror, which made the
Hellenism of that country a thing quite different from that of the other
lands within the sphere of Greek influence. The resistance of the Jews
also took the form of successful insurrection, and in their case enabled
an independent political entity to be constituted.

The dispersal of the Jews was already considerable at this time. It
differed from the dispersal of the Syrians in the fact that the bond of
union of the Jewish congregations existed in the common cult and the
common interest in the fortunes of the mother-country. On the other
hand, the Syrians of Rome and of Naples shared nothing except the
quickly effaced memory of a common racial origin.[229]

The propaganda of the Jews was also well under way. Since it was
believed that they possessed a mystery, initiation into which gave
promise of future beatitude, they were strong rivals of the Greek and
Oriental mysteries that made similar claims. It was chiefly among the
half-educated or the wholly unlettered that these claims would find
quickest belief. However, the Jewish propaganda had also its philosophic
side, and competed with the variously organized forms of Greek
philosophic thought for the adherence of the intellectually advanced
classes as well.

Through the Diaspora and this active propaganda an opposition was
invited. In Egypt the opposition was older, because the presence of Jews
in Egypt was of considerably earlier date than the period we are
considering. The occasions for its display were various, but the
underlying cause was in most cases the same. That was the fact of
religious separatism, which in any given community was tantamount to
lack of patriotism. It does not appear, however, that this opposition
found voice generally except in Egypt. Elsewhere racial friction was
relatively rare.

The literature of the opposition falls into two classes: first, that
which scarcely knows the Jews except as a people of highly peculiar
customs, and uses these customs as illustrations of rhetorical theses;
and second, that which is inspired by direct animosity, either personal
or, in the case of the Egyptians, racial in its character.


------------------------------------------------------------------------




                               CHAPTER XV

                               THE ROMANS


We have been concerned so far almost wholly with Greeks and the Greek
attitude toward the Jews. It will be necessary at this point to turn our
attention to a very different people, the Romans.

If we desire to trace the development of this all-overwhelming factor in
our reckoning, it will not be possible to go back very far. During the
fifth century B.C.E., in which Greek genius is believed to have reached
its apogee, it is doubtful whether even the faintest whisper had reached
Greeks that told of the race of Italic barbarians destined so soon to
dominate the world. Little as was known of the Jews by Greeks of this
period, the Romans were still less known. The eyes of men were
persistently turned east.

Rome, however, even then was not wholly insignificant. Many centuries
before, there had grown up, on the south bank of the Tiber, a town of
composite racial origin. It is possible to consider it an outpost of the
Etruscans against Sabine and Latin, or a Latin outpost against the
Etruscans. Whatever its origin, at an indeterminate time, when the
Etruscan hegemony over central Italy was already weakened, this town of
Rome became a member of the Latin Confederation, a group of cities of
which the common bond was the shrine of Jupiter Latiaris on the Alban
Mount.

There may have been rude hamlets upon this site from times very ancient
indeed. But from the beginning of its existence as a real city Rome must
have been a considerable community. Her strategic position upon seven
hills, the commercial advantages of her location upon a navigable
stream, conspired to this end.

The Latin Confederation had long been under the real or titular
presidency of the city of Alba. At some time before our records become
reliable, Rome had obtained a decidedly real leadership in the league,
and unscrupulously used the latter’s resources for the furtherance of
her own power and wealth. Without a definite programme of conquest, and
with military skill and personal hardihood very little, if at all,
superior to that of their neighbors, the Romans had, by steadfastness
and native shrewdness, developed a policy which it is difficult to put
in precise terms, because it was never even approximately formulated,
but which may be said to consist of unremitting vigilance and long
memory, combined with special alertness to profit by the mistakes or
division of the foe. It may be that the indubitably mixed character of
Rome’s population produced that result. Certainly in these respects no
other ancient community was its equal.

The legendary history of Rome is as generally familiar as the commonest
household stories of the race. Modern investigators have abandoned the
attempt to find out even partially the line at which its history ceases
to be legendary. Fairly correct accounts of Rome begin with the
permanent contact of the city with a literate community of which the
records have survived, namely, the Greeks.[230]

The Greeks had founded cities along the southern coast of Italy and the
eastern half of Sicily as early as the ninth century B.C.E. With some of
these cities it was inevitable that Rome should be in frequent
communication, but the communication did not impress itself for many
years upon that class of Greeks which, in the extant books, speaks for
the whole people. Not till the time of Alexander (330 B.C.E.) do our
Greek records begin to deal with Romans. At that time Rome was already
the dominant power in central and in the interior of southern Italy,
succeeding roughly to the empire of that great Tuscan League of which
she was once the subject. And yet, Alexander’s teacher, the
encyclopedically learned Aristotle, had only vaguely heard of Rome as an
Italian city overrun by marauding Gauls.[231]

The position occupied then by Rome would of itself have made active
participation in Mediterranean affairs a necessity. The embroilment of
Romans in the conflicts in which international politics is expressed was
precipitated by the ambition of the restless Diadochi and their
successors. It was a kinsman of the lurid Demetrius the Besieger, the
Epirote prince Pyrrhus, who undertook to save the Greek civilization of
the coast cities from the Italian barbarians. Pyrrhus ultimately retired
with his tail between his legs, after having dragged the Romans into
Sicily and brought them face to face with the Carthaginians. The
succeeding three generations were occupied in the mortal grapple between
these two. It ended with the triumph of Rome.

So far Rome had dealt only with the West, but with the permanent
eastward bent of men’s minds the lord of the Western Mediterranean was,
as such, a power in the East as well. Scarcely a single generation
passed before it became the sole power in the East, so that future
political history becomes the act of officially recording successive
realizations of that fact. And yet, this extraordinary people, which had
in an astoundingly short time secured the primacy over a considerable
fraction of the earth, was apparently possessed of slighter intellectual
endowments than many of its subjects. It had not succeeded in giving
such culture as it had developed any artistic form. And before it had
taken any steps in that direction, it came into immediate contact with
nations of much older culture, which had done so; in one case, a nation
which had carried artistry of form to a degree never subsequently
attained by any single people. First, the Etruscans had given in bulk a
mass of finished cultural elements, especially in religion and
constructive crafts, and had otherwise exercised an influence now wholly
undeterminable. Secondly, by Etruscan mediation and afterwards directly,
the Romans became the intellectual vassals of the Greeks, a fact that
lends some justification to the modern tendency to treat classical
antiquity as a single term.

The Romans obtained their very earliest knowledge of the Jews when the
political and social development just outlined was practically complete.

The treaty cited in I Macc. viii. 22 seq. is perhaps apocryphal, but the
substantial accuracy of the chapter is scarcely doubtful. “And Judas had
heard the name of the Romans,” we read, and this statement is followed
by a lengthy recital of the recent conquests of Rome. After the first
Hasmonean successes the little knowledge that Roman and Jew had of each
other may be so summed up. On the Roman side, the responsible senatorial
oligarchy learned with undisguised satisfaction that a previously
unknown tribe of Syrian mountaineers, grouped about a famous temple-rock
not far from the Egyptian frontier, had successfully maintained
themselves against a troublesome and unaccountable tributary king. On
the Jewish side, the leaders of the victorious rebels, conscious of the
precarious nature of their success, turned at once to that mighty
people—known as yet scarcely by report—which from far off directed men’s
destinies. Even at that time the Roman policy of _divide et impera_,
“divide and rule,” was well understood and consciously exploited by all
who could do so. The embassy sent by Judas—there is no real reason for
questioning its authenticity—presented to curious Romans in 162 B.C.E.
an aspect in no way different from that of other Syrian embassies long
familiar to the capital. And if it is true that some of that train or of
a later embassy of Simon took up permanent residence in Rome, that fact
was probably scarcely noticed from sheer lack of novelty.

Generally speaking, the Roman attitude to the Jews, as to all other
peoples, was that of a master: the attitude of the Goth in Spain, the
Manchu in China, the English in India. No one of these analogues is
exact, but all have this common feature, that individuals of the
dominant race can scarcely fail to exhibit in their personal relations
with the conquered an arrogance that will vary inversely with the man’s
cultivation. It is so very easy to assume for oneself the whole glory of
national achievements. No doubt every Italian peasant and artisan
believed that it was qualities existing in himself that commanded the
obedience of the magnificent potentates of the East. The earliest
attitude of Roman to Jew could not have been different from that toward
Syrians or foreigners in general. If in 150 B.C.E. the term _Iudaei_ had
reached the ears of the man in the street, it denoted a Syrian
principality existing like all other principalities at sufferance and
upon the condition of good behavior.

For nearly a hundred years this state of things remained unchanged. Then
the inevitable happened. Syria became Roman, and the motives that had
won Roman support for the Jews no longer existed. Roman sufferance was
withdrawn, and Judea’s good behavior ceased. That Gnaeus Pompey
encountered serious resistance on his march from Antioch to Jerusalem is
doubtful. The later highly-colored versions of his storming of the
temple are probably rhetorical inventions. The Psalms of Solomon, which
are very plausibly referred to this period, are outbursts of passionate
grief at the loss of the national independence; for no recognition of
Hyrcanus’ rank could disguise the fact of the latter’s impotent
dependence upon the senate, and the limitations openly placed upon the
vassal-king’s authority show that the Romans were at no pains to
disguise the fact.[232]

When the Romans added Asia to their dominions, as they had in the
generation preceding the occupation of Jerusalem, they annexed with Asia
many hundreds of Jewish synagogues in the numerous cities of Asia. Jews
lived also in Greece, in Italy and Rome itself, and in Carthage. Egypt,
which contained many hundreds of thousands, was still nominally
independent. Roman officials had long known how to distinguish the
_Iudaei_ from others of those ubiquitous Syrians who, as slaves,
artisans, physicians, filled every market-place of the empire. More than
one provincial governor must have collected a few honest commissions
from a people indiscreet enough to collect sums of considerable
magnitude, as the Jews did for the support of the temple.


[Illustration: RUINS OF AN ANCIENT SYNAGOGUE AT MEROM, GALILEE,
PALESTINE
(Roman Period)
(_c. Underwood and Underwood_)]


That they were classed as Syrians did not raise the Jews in general, and
particularly in Roman, esteem. The Syrians, to be sure, were one of the
most energetic, perhaps mentally the quickest, of the races then living,
but they were the slave race _par excellence_; _i.e._ the largest number
of slaves were and had long been derived from among them. The vices of
slavery, low cunning, physical cowardice, lack of self-respect, were
apparent enough in those Syrians who were actually slaves, and were
transferred to all men of that nation. “Syri” is nothing less than a
term of contempt applied to any people of unwarlike habits.[233]

Unwarlike the Jews of that day were not. All that had commended them to
Roman notice was their military successes over the troops of Antiochus
and Demetrius. Pompey may not have found Aristobulus and his Nabatean
allies really formidable, but he did have to fight, and did not meet
that docile crawling at his feet which he had encountered elsewhere.
That made considerable difference in Roman eyes, and may have caused the
unusual tenderness they manifested as a rule for what they loftily
termed the Jewish superstition.

As has been said, we have reason to believe that a Jewish community
already existed at Rome, and we shall see that it must have been fairly
numerous. As a city, Rome was probably the least homogeneous in the
world. It may have contained at this time something less than a million
people, perhaps much less; but this population was of the most diverse
origin. Not only had the capital of the world attracted to it all manner
of adventurers; not only was it teeming with slaves of every imaginable
blood and speech; but the thronging of the city with the refuse of the
world had been a conscious policy of the democratic and senatorial
rings, to whom modern “colonization” was a familiar and simple process.
When we recall that the accepted governmental theory was still that of
the city-state, we shall see that mere residence made to a certain
extent a Roman of everyone who lived within the walls. Various measures
of expulsion, such as the Lex Junia Penni and the Lex Papia of 65
B.C.E., were wholly ineffective.

As a matter of fact, the governmental apparatus of the city-state was
quite unable to cope with the situation that presented itself. Until 200
B.C.E., the turning-point in Roman history, the city was small and mean;
the population, though composite, was still almost wholly Italian in
character. A rapid increase in wealth and a consequent increase in
glaring inequalities of fortune began at this point. The governing
council of ex-magistrates, whose office had in practice become almost
hereditary, found itself confronted by a needy and exigent proletariat,
which it could neither overawe nor purchase.

The urban tendency of the population of Italy was due largely to the
failure of the small farms to support their man. Free labor was
subjected to the constant drain of military levies, and temporary
suspension of cultivation was ruinous. The obvious remedy was a forced
and unprofitable sale to the agrarian capitalists, whose leasehold
interest in the great public lands had long been so nearly vested that
it was almost sacrilege to attack it. To migrate to the city was then
the only course open to the peasant, but in the city the demand for free
labor was never great. The new arrivals joined the great mass of
landless rabble, sinking soon into an idle and pauperized mob.

But at the same time infusions of foreign blood came into the city. The
rapid rise in wealth and power had poured into Rome a constant stream of
the commonest of wares, viz. human chattels. These slaves, Greek,
Thracian, but above all Syrian, were directly consequent upon the
imperative demand for skilled labor, which they alone could satisfy. But
the very number of these slaves, and the changes in personal fortunes,
which were then even more frequent than now, made them often a liability
rather than an asset to their master.

Enfranchisement was encouraged by another consideration. The Roman law,
determined by a very ancient patriarchal system, was apparently very
rigid as to the extent of the master’s _dominium_. The slave was, in law
and logic, a sentient chattel indistinguishable from ox and ass. But in
other respects the Roman law was extraordinarily liberal. For practical
purposes the slave could and did acquire property, the so-called
_peculium_, and could and did use it to purchase his freedom.

Further, the newly-made freeman became a full citizen, a _civis
Romanus_. His name was enrolled in the census books; he possessed full
suffrage, and lacked only the _ius honorum_, the right of holding
office. Even this, however, his children acquired. Sons of slaves who
held magistracies are frequent enough to furnish some notable examples;
_e.g._ Cn. Flavius, the secretary to Appius Claudius; P. Gabinius, the
proposer of the Lex Tabellaria of 139 B.C.E.[234] It is for this reason
that indications of servile origin have been found in names nothing less
than illustrious in Roman history.[235]

With this steady influx of dispossessed peasants and enfranchised Greek
and Asiatic slaves, the urban population was a sufficiently
unaccountable quantity; and in this motley horde, constantly stirred to
riot by the political upheavals, which quickly followed each other from
the Gracchan period onward, all manner of strange and picturesque
foreigners lived and worked. To the Roman of cultivation they were
sometimes interesting, more often repellent, especially if he found
himself compelled to reckon with them seriously on the basis of a common
citizenship. Even for foreigners Roman citizenship was not very
difficult to acquire, and was, as we have seen, obtained with especial
facility through slavery. The emancipated slave was as such a _civis
Romanus_. His son had even the _ius honorum_; he might be a candidate
for the magistracy. This process had been accelerated after the Social
War, which admitted an enormous and quite unmanageable number into
citizenship. The popular leaders were especially lavish, and no doubt
many ward politicians took it upon themselves to dispense with the
formalities when a few votes were needed.

We are very fortunate in possessing for this period records of quite
unusual fulness and variety. The last century of the Roman republic was
rich in notable men, with some of whom we are especially familiar. In
literary importance and in permanent charm of personality, no one of
them can compare with the country squire’s son, Marcus Tullius Cicero,
who achieved the impossible in his lifetime, and attained posthumous
fame far beyond his wildest dreams. He was consul of Rome in the very
year in which Jerusalem was captured, and was in the throes of the same
political uncertainty that marked his whole later career. The most
brilliant pleader in the city or the world, he was feared, loved, and
hated for his mordant wit, his torrential fluency of speech, and his
remarkable power and skill in invective. Although his personal instincts
had always inclined him to the gentlemanly aristocracy that made up the
majority of the senate, he had won his first successes in politics on
the other side, and reached the summit of his ambition, the consulship,
as a popular candidate, receiving the support of the senate only because
he was deemed the least dangerous of three.

In the year 59 B.C.E. Cicero, concededly the leader of the Roman bar and
still more concededly the social lion of the day, undertook the defense
of Lucius Valerius Flaccus, former governor of Asia, who was charged
with maladministration and oppression. The counts in the indictment were
numerous. Among them was the following allegation: That Flaccus as
praetor had seized certain sacred funds; to wit, the moneys which
Asiatic provincials, Jews in origin, had, in accordance with ancient
custom, collected and were about to transfer to the temple at Jerusalem.
By so doing Flaccus had doubled embezzlement upon sacrilege, for the
sanctity of the temple was established by its antiquity, and confirmed
by the conduct of Pompey, who had ostentatiously spared it and its
appurtenances.

It will be necessary to examine in some detail the circumstances of the
entire case. Flaccus was a member of the reactionary wing of the
senatorial party, which until recently had held Cicero aloof as an
upstart provincial. His birth and training were those of an aristocrat.
A certain portion of Cicero’s defense is occupied in descanting on the
glories of the Valerian house, to which Flaccus belonged. The
prosecution of Flaccus, again, was a political move of the popular
opposition, now at last, after the futile essays of Lepidus and
Catiline, finding voice and hand in the consummate skill of Gaius Julius
Caesar.

Shortly before this date a powerful combination had been made, which
enlisted in the same scheme the glamour of unprecedented military
success in the person of Gnaeus Pompey, the unlimited resources of the
tax-farmers and land-capitalists represented by Marcus Crassus, and the
personal popularity of the demagogue Caesar. Each no doubt had his own
axe to grind in this coalition, and the bond that held them was of an
uncertain nature, opposition to the senatorial oligarchy. Further, only
in the case of Caesar was the opposition a matter of policy. In the case
of the other two, it was the outcome of nothing loftier than pique. None
the less, when the strings were pulled by Caesar, this variously
assembled machine moved readily enough.

In 59 B.C.E. this cabal had been successful in winning one place in the
consulship, that of Caesar himself. Lucius Flaccus had earned Caesar’s
enmity by his vigorous action against the Catilinarians in 63 B.C.E. E.,
so that when an influential financier, C. Appuleius Decianus, complained
of Flaccus’ treatment of him, the democratic leader found an opportunity
of gratifying his allies, of posing as the protector of oppressed
provincials, and wreaking political spite at the same time. A certain
Decimus Laelius appeared to prosecute the ex-governor of Asia.

Of Flaccus’ guilt there seems to be no reasonable question. He was
plainly one of the customary type of avaricious nobles to whom a
provincial governorship was purely a business proposition. No doubt he
was no worse than his neighbors. His guilt seems to have been especially
patent. “Cicero,” says Macrobius, “secured the acquittal of Flaccus by
an apposite jest, although the defendant’s guilt of the charges made was
perfectly apparent.”[236] And indeed on the principal counts Cicero has
no evidence except exaltation of Flaccus’ personal character, and abuse
of the witnesses against him, whom he characterizes as lying and
irresponsible Greeks. His peroration is a flaming denunciation of the
prosecution and an appeal to the jury not to permit the supporters of
the dead traitor Catiline to win a signal triumph.

The speech was successful. Flaccus was acquitted, and the acquittal may
have hastened Cicero’s own banishment. But for us the particularly
interesting part of this brilliant effort is contained in §§ 66-69.
After he has disposed of the various charges of peculation and
extortion, he turns to the charges made by the Jews:

    Next comes the malicious accusation about the gold of the Jews. No
    doubt that is the reason why this case is being tried so near the
    Aurelian terrace. It is this count in the indictment, Laelius, that
    has made you pick out this place, and that is responsible for the
    crowd about us. You know very well how numerous that class is, with
    what unanimity they act, and what strength they exhibit in the
    political meetings. But I shall frustrate their purpose. I shall
    speak in a low tone, just loud enough for the jury to hear. There is
    no lack of men, as you very well know, to stir these fellows up
    against me and every patriotic citizen; and I have no intention of
    making the task of such mischief-makers lighter by any act of mine.

    The facts are these: Every year it has been customary for men
    representing the Jews to collect sums in gold from Italy and all our
    provinces for exportation to Jerusalem. Flaccus in his provincial
    edict forbade this to be done in Asia.

    Now, gentlemen, is there a man who can honestly refuse commendation
    to this act? That gold should not be exported is a matter which the
    senate had frequently voted, and which it did as recently as my own
    consulship. Why, it is a proof of Flaccus’ vigorous administration
    that he took active steps against a foreign superstition, as it is
    an indication of a lofty sense of duty that he dared defy, where the
    public weal was concerned, the furious mass of Jews that frequently
    crowd our meetings.

    But, we are told, when Jerusalem was captured, the conqueror Gn.
    Pompey touched nothing in that shrine. And that was very wisely done
    on Pompey’s part, as in so many other acts of that commander. In so
    suspicious and slanderous a city as ours, he would leave nothing for
    his detractors to take hold of. But I do not believe, and I cannot
    suppose you do, that it was the religion of such a nation as the
    Jews, recently in arms against Rome, that deterred our illustrious
    general. It was rather his own self-respect.

    In view of these considerations, just wherein does the accusation
    lie? You do not anywhere charge theft; you do not attack the edict;
    you admit due process of law; you do not deny that the moneys were
    openly confiscated upon official investigation. The testimony itself
    discloses that the whole matter was carried on by men of rank and
    position. At Apamea, Sextus Caesius, a Roman knight and a gentleman
    of whose honor and integrity there can be no question, openly seized
    and weighed out in the forum at the feet of the praetor a little
    less than a hundred pounds of gold. At Laodicea an amount somewhat
    more than twenty was seized by Lucius Peducaeus, a member of this
    very jury; at Adramytus, ... by the governor’s representative, L.
    Domitius. A small quantity was also seized at Pergamon. The accounts
    of the gold so seized have been audited. The gold is in the
    treasury. There is no charge of theft. The purpose of the charge is
    to excite odium against my client. It is not the jury that the
    prosecution is addressing, but the audience, the crowd about us.

    Religious scruples, my dear Laelius, are primarily national
    concerns. We have our own, and other states have theirs. And as a
    matter of fact, even while Jerusalem was standing, and the Jews were
    at peace with us, there was very little in common between the
    religious customs of which their rites are examples and those which
    befit an empire as splendid as ours, or a people of our character
    and dignity. Our ancestral institutions are as different from theirs
    as they well can be. Now, however, there surely can be all the less
    obligation upon us to respect Jewish religious observances when the
    nation has demonstrated in arms what its feelings are toward Rome,
    and has made clear how far it enjoyed divine protection by the fact
    that it has been conquered, scattered, enslaved.

There are a number of difficulties with the passage. The text of the
final sentence is doubtful—but the discussion of that point will be
reserved for the Notes.[237]

We cannot suppose that Cicero was guilty of deliberate misstatement on
matters about which he could be immediately confuted. We must therefore
accept his assertion that this count in the indictment did not charge
theft or malversation, but merely public confiscation of the funds in
question. It is undoubtedly a fact that the exportation of the precious
metals had been frequently forbidden, although the senatorial resolution
to this effect was far from being a law, but with this precedent and
even without it no one could very well deny that it was within the
imperium of a proconsul to make such a regulation if he saw fit.[238]

One may well ask with Cicero, _Ubi ergo crimen est?_ The point seems to
be that previous officials had interpreted the rule to refer to
exportation for commercial purposes, and had exempted from its operation
contributions for religious purposes. Doubtless the self-imposed temple
tax of the Jews was not the only one of its kind. If custom had
sanctioned that exemption, Flaccus’ act would be felt as an act of
oppression, since the strict or lenient enforcing of the edict on this
point was purely a matter of discretion.[239] Flaccus’ successor,
Quintus Cicero, a brother of the orator, seems to have reverted to the
former practice.

In one other respect the seizure of these sums may have seemed an act of
arbitrary tyranny. The sum seized at Apamea was said to be one hundred
pounds of gold—about 72 English pounds—and must have equaled about
75,000 Roman denarii or Athenian drachms. As the temple tax was a
didrachm, that would imply over 35,000 heads of families, or a total
Jewish population for Apamea of 170,000. That number is quite
impossible. It is, however, very likely that the Jews of the various
synagogae paid their didrachm with their other dues to the corporation
_arca_, or treasury, and that it was the whole treasury that was seized.
That would give the Jews of these cities a very real grievance, and make
their animus against Flaccus easy to explain.

The importance of the passage, however, is in no way concerned with the
justice or injustice of the accusation against Flaccus. It lies first in
its picture of the Jewish community at Rome, and secondly in its
indication of Cicero’s personal views.

The very insertion of the charge proves that a considerable Jewish
element existed, whose aid the prosecution was anxious to enlist.
Cicero’s own statements show this directly. Here and here only in his
speech he refers to the popular odium sought to be incited against his
client, and speaks of the number and power of the Jews _in
contionibus_,[240] “in the political meetings,” and in the crowd about
him. Part of this, the _summissa voce_, “low tone,” for example, is the
veriest acting. Cicero was really not afraid to say loudly what he
wished to say, and if the jury could hear him, part of the crowd could
hear as well. But although the Roman Jews were probably not so
redoubtable as Cicero would have his jury believe, they must have formed
a large contingent. Where did they come from?

We have the statement of Philo that it was not until the capture of
Jerusalem by Pompey in 63 B.C.E. that Jews were brought to Rome in large
numbers.[241] These, it is supposed, were enfranchised shortly after,
and are the people here referred to. That may be said to be the general
view.

There are, however, serious difficulties in it that escape those who
hold it, simply because they fail to follow in detail the implications
of their view. Pompey did not arrive in Rome till January, 61 B.C.E. His
army had been previously dismissed, but was to assemble for the great
triumph that took place in September, 61. The trial of Flaccus was held
in August, 59 B.C.E. Some months must have been spent in preparing the
case against him. Accordingly we are to suppose that thousands of Jewish
captives were brought to Rome, sold there, enfranchised, learned Latin,
became politically organized, and developed formidable voting strength,
all within less than two years! The mere question of language makes the
hypothesis impossible. Pompey’s captives were Palestinian Jews, of most
of whom the native language was Aramaic, not Greek.[242] Without command
of Greek or Latin the ready acquisition of either was nothing short of
miraculous, and the immediate political activity is only less so.

But the chief difficulty lies in another matter. The phrase “taken
prisoners” immediately suggests the conditions of modern warfare, in
which whole armies are surrendered and transferred in bulk great
distances for safe-keeping. It is to be feared that some such idea was
suggested to modern writers by the words of Philo. But that is not at
all what occurred in ancient times. Prisoners taken on the field of
battle were sold immediately at the nearest market. Slave-dealers
followed the army. Caesar’s account of his campaign in Gaul affords
numerous instances of this immediate disposal of captured foes; _e.g._
the case of the Atuatuci and Veneti.[243] If they were assigned as loot
to individual soldiers, they were disposed of in the same way. Here and
there a soldier would, for one reason or another, retain his prisoner as
a personal slave, but in general he had almost no facilities for
providing or caring for a number of them. A few of the distinguished
captives were reserved by the commander for a triumph.

Now Pompey’s army had just finished a five years’ campaign. It had
marched through Asia and Syria, winning battles that were not very
bloody, but must have been immensely lucrative. The Jews formed only a
small portion of the total prisoners taken. If all those prisoners
actually accompanied their captors to Rome, the question of
transportation and provision for such a horde must have been tremendous.
What could have induced a general or private to assume this enormous
expense and care, when the greatest slave-market in the world, viz. that
at Alexandria, was relatively near by, is inconceivable. If they got to
Rome, the city’s population must have swelled visibly under the process.
There is no record that it did, and it could scarcely have escaped
notice, had such a thing taken place.

And finally, even if we assume that such a wholly unprecedented and
inexplicable incident occurred, how are we to explain the immediate and
wholesale enfranchisement of so large a number? Ransom by wealthy
coreligionists at Rome is excluded by the hypothesis. Similar action by
Jews outside the city would demand a much longer time. The reasons
generally assigned are based upon the assumed uselessness of Jewish
slaves for ordinary purposes because of their dietary laws and religious
intransigeance. But that is a purely dogmatic assertion. Papyri and
inscriptions have shown that in spite of a bitter racial opposition and
perhaps economic strife as well, Jew and non-Jew could live quite
peaceably together. The dietary laws would not render his master’s meals
obnoxious to a Jewish slave, because he did not eat at his master’s
table, and might consume his scanty vegetable food where and how he
pleased. If a master actually chose to force attendance at the
sacrifice, the compulsion of necessity would have been a valid excuse
for all but those of martyr stuff, and we cannot suppose that every
Jewish soldier had in him the zeal of a martyr. Besides, for such
compulsion the slave would in no sense be responsible, and it is with
disadvantages moving from him that we are concerned.

It is simply impossible to imagine what could have induced Pompey’s
soldiers or those who purchased from them to enfranchise immediately
slaves transported from such a distance and at such expense.

Philo’s statement is at best a conjecture, made without any better
acquaintance with the facts than we ourselves possess, and contradicted
by the necessary inference from Cicero’s words.

We must therefore assign the settlement of Jews in Rome to a much
earlier date. The tradition that some of the train of Simon’s embassy
had remained in Rome is, as we have seen, probable enough. To that
nucleus there was added, by a perfectly natural and even inevitable
infiltration, a group of Jewish freedmen, artisans, and merchants who
were establishing themselves all over the Mediterranean. Jews are met
with at Delphi a hundred years before the delivery of this speech.[244]

We have therefore, in 59 B.C.E., an established Jewish community,
necessarily organized in synagogues and chiefly of servile origin. The
use of foreigners at the polls by the political leaders had led to the
Lex Junia Penni of 80 B.C.E. and the Lex Papia of 65 B.C.E., which
ordered foreigners to leave the city. But these measures were wholly
ineffective, and in any case could have only partly served those who
proposed them, since the mass of the democratic strength lay in the
proletariat, and the proletariat was largely composed of undoubted
citizens, although freedmen. The Jews formed, as we see, an active and
troublesome element in the turbulent city populace. Their attachment to
the democratic leader, Caesar, is well attested, and Caesar’s marked
favor toward them has all the appearance of the payment of a political
debt, as in the case of the Cisalpine Gauls.[245]

As far as Cicero was concerned personally, we may assume that his
attitude was the contempt which he no doubt honestly felt for the
_infima plebs_ and for Syrian barbarians in particular. He probably
voices the sentiments of the optimates,[246] with whom, though still
hesitant, he had already cast his fortunes. The abuse arises from the
necessities of the case. As previously pointed out, it is in this very
speech that we have fine examples of the device of abusing your
opponent’s witnesses when arguments give out. These phrases show no
special animus. Just as Greeks are liars if they are on the other side,
and men of honor on his own, as exhibited almost in successive
paragraphs of this speech,[247] so we may be sure if Cicero were
prosecuting Flaccus, a few eloquent periods would extol the character of
those ancient allies and firm friends of Rome, the Jews.

How much Cicero really knew of the Jews is not certain. He is aware that
in point of religious observance the Jews are strikingly different from
other tribes. The contrast he emphasizes in his speech may be an
allusion to the imageless cult of the Jews and the inference of meanness
and poverty of ceremonial which Romans would draw from it. And the taunt
_quam dis cara_, “how dear to the gods,” seems an unmistakable fling at
the claim of the Jews, loudly voiced in their propaganda, to possess in
a high degree the favor of the Divinity, or even a special communion
with the Deity in their mysteries.

All this Cicero might have learned from his surroundings. It is doubtful
that he learned it from Posidonius and Molo, both of whom he knew well.
In these two appear stories which Cicero could hardly have overlooked if
he knew them. When we remember what he says of Sardinians in the
Scauriana, of Gauls in the Fonteiana,[248] he surely would not have
omitted to catalogue the tales treasured up by these two Greek teachers
of his; to wit, the ass-god, the scrofulous prophet, the savage
inhospitality and absurd fanaticism Molo and Posidonius ascribe to the
Jews.

One other phrase which Cicero applies to Jews would deserve little
attention if it were not for the extraordinary general inferences some
have drawn from it. In May, 56 B.C.E., Cicero has an opportunity to vent
his venom on his enemy Gabinius, consul in 59 B.C.E., whom he held
personally responsible for the humiliation of his exile. Gabinius, in
56, was governor of Syria, and seems to have been rather short with the
tax-farmers, whom, to the delight of the provincials, he treated with
contumely and no doubt with gross injustice. The persistent favor he
showed to all provincial claims against these men, many of them Cicero’s
personal friends and at all times his supporters, caused the orator to
exclaim:

    As far as the unfortunate tax-farmers are concerned—and I count
    myself equally unfortunate to be compelled to relate their
    misfortunes and sufferings—Gabinius made them the chattel-slaves of
    Jews and Syrians, races themselves born to be slaves.

The concluding phrase is simply the application of the rhetorical
commonplace of Greeks that barbarians as such were slaves by nature. It
was applied to Syrians with a certain justice, as the slave name Syrus
testifies. From that standpoint, however, it was obviously absurd to
assert that it was true of Jews. Cicero’s inclusion of them is due to
the fact that, as governor of Syria, Gabinius would have had many
occasions to favor Jewish litigants against the publicans, probably in
pursuance of his party’s policy. Gabinius, we may recall, was a very
obedient servant of his masters, the triumvirs, and the interest of the
leading spirit of the coalition in the provinces has been previously
pointed out.[249]

Allusions of this type made in the course of vehement advocacy or
invective are really of little meaning even as an indication of personal
feeling. It is true, however, that Cicero shows very little sympathy in
general with the Roman masses or with the provincials, despite the
Verrine prosecution. That he could have felt any interest or liking for
Syrian barbarians in or out of the city is very improbable.

None the less, within Cicero’s own circle, the same elements in Jewish
customs which had impressed Greeks, such as Theophrastus and Clearchus,
could not fail to strike such Romans as made philosophic pretensions.
The fame of the shrine at Jerusalem had reached Rome a century earlier,
as we have seen from Polybius. Pompey’s capture of the city formed no
inconsiderable item in his exploits. Cicero refers to him jestingly as
_noster Hierosolymarius_, “Our Hero of Jerusalem.”[250] We can tell from
Cicero’s own words the emphasis that Laelius had laid on the fame of the
temple and its sanctity when he denounced Flaccus. As a matter of fact
it was a constant practice of Romans to find, in those institutions of
barbarians which could be called severe or simple, the image of their
own golden age of simplicity, before the advent of Greek luxury. So
Cicero’s learned friend and correspondent Varro is quoted by
Augustine[251] as referring to the Jews among others as a people whose
imageless cult still maintains what the Romans had abandoned. There may
be very little sincerity in this regret of a simple-living past, but it
is an indication that the exceptional character of Jewish religious
customs might in Cicero’s own _entourage_ be characterized in terms
somewhat different from those of the Flacciana.

We shall have reason to distinguish very sharply between the attitude of
Romans of rank and cultivation and that of the great mass. However, that
is true not only in this relatively minor detail but in thousands of
other matters as well. The Roman gentleman was distinct from the mass,
not merely in political principles, but in his very speech. In the
following generations social readjustments of all sorts frequently
modified the position of the Jews in Rome, but until the increasing
absolutism of the monarchy practically effaced distinctions, the
cleavage just indicated largely determined the point of view and even
the terms used.


------------------------------------------------------------------------




                              CHAPTER XVI

                  JEWS IN ROME DURING THE EARLY EMPIRE


We are all familiar with the assertion that both Greeks and Romans of
the last pre-Christian century were in a state of complete moral and
religious collapse, that polytheism had been virtually discarded, and
that the worn souls of men were actively seeking a new religious
principle to take its place. This general statement is partly true, but
it is quite inadequate, if it is made to account for the situation at
Rome at that time.

The extant literature of the time makes it quite clear that there was no
belief in the truth of the mythology. But it is doubtful whether there
ever had been, and mythology was no part of religion. This was
particularly true at Rome. For some thousands of years the inhabitants
of central Italy had performed ceremonies in their fields in connection
with their daily life. A great many of these ceremonies had become
official and regulated in the city of Rome and many other Italic civic
communities. It was the practice of educated Italians to devise
aetiological stories for these practices and to bring them into
connection with Greek myths. In this way a Roman mythology was created,
but more even than in the case of the Greeks it was devoid of a
folkloristic foundation.[252] For the masses these stories can scarcely
be said to have existed. But the ceremonies did, and their punctilious
performance and the anxious care with which extraordinary rites of
purgation were performed satisfied the ordinary needs of ordinary men.

Mention has been made of the religious movement which from the seventh
century B.C.E. spread over the Eastern Mediterranean, and which was
concerned with the demand for personal salvation and its corollary, a
belief in personal immortality. In the Greek-speaking world the carriers
of that movement were the Orphic and Dionysiac mysteries. In the
non-Greek East there was abundant occasion for beliefs of this kind to
gain ground. The great world monarchies introduced such cataclysms in
the smaller nations that a violent readjustment of relations with the
divinity was frequently necessitated, since the god’s claim to worship
was purely national. No such profound political upheavals occurred in
Greece. Here, however, a fertile field for the spread of mysteries and
extra-national means of divine relations was found in the rapid economic
degeneration caused by the slave system. Attachment to the state was
confined to those who had a stake in it. The maxim that a man’s
fatherland was where his fortune brought him seemed less a bold and
cynical aphorism than the veriest commonplace for all but a few
idealists.[253] To save the personality that individual misfortunes
threatened to overwhelm, recourse was had to every means and especially
to the vague and widespread doctrine of other and fuller existences
beyond the confines of mortality.

In Rome the obvious hinge in the destinies of the people from almost
every point of view was the Hannibalic war. For a short time disaster
seemed imminent, and the desperate reaching out to the ends of the earth
for divine support could not fail to make a deep impression upon
thousands of men. In that moment of dreadful stress, it was not the
Etruscan Triad on the Capitol nor Father Mars, but the mystic Ma, the
Ancient Mother of Phrygia with her diadem of towers, her lion-chariot
and her bloody orgies, that stayed the rush of the Carthaginian. It is
true that the city’s ultimate triumph caused a reaction. An increased
national self-consciousness made Romans somewhat ashamed of their
weakness, but they could not blot out the memory of the fact.

The city’s increase in total well-being went on with tremendous strides,
but the disintegrating forces of a vicious economic system were present
here too. Besides, the special circumstances that tended to choke the
city with people of diverse origin were intensified. In the next few
generations we hear of the threatening character of foreign mysteries,
of surreptitious association with the Cybele worshipers, of Isis
devotees gaining ground. Shortly after the Second Punic War occurs the
episode of the Bacchic suppression. One can scarcely help noticing how
strikingly similar were the accusations directed against the Bacchanales
and those later brought against the Christians, and wondering whether
they were any truer in the one case than in the other. The whole
incident can easily be construed as an act of governmental persecution,
which, it may be noted, was as futile as such persecution generally is.
The orgiastic Dionysus was not kept from Italy, though he always
remained an uncomfortable god for Romans of the old type. One reason has
already been referred to; viz., the constant recruiting of the _infima
plebs_ from enfranchised foreign slaves. The lower classes were becoming
orientalized. The great Sicilian slave revolt of 134 B.C.E. was almost a
Syrian insurrection, and was under the direct instigation of the Syrian
goddess Atargatis.[254]

During the civil wars and the periods of uncertainty that lay between
them, all political and social life seemed as though conducted on the
edge of a smouldering volcano. Innumerable men resorted to magic, either
in its naïve form or in its astrological or mathematical refinements.
Newer and more terrific rites, stranger and more outlandish ceremonials,
found a demand constantly increasing. And the Augustan monarchy brought
only a temporary subsidence of this excitement. Order and peace
returned, but Augustus could not cure the fundamentally unsound
conditions that vitiated Roman life, nor did he make any real attempt to
prevent Roman society from being dissolved by the steady inpour of
foreign blood, traditions, and non-Roman habits of mind. The need of
recourse to foreign mysteries was as apparent as ever.

In this way the internal conditions of Roman society impelled men to the
alien forms of religion. And external impulses were not lacking. There
were present professional and well-equipped missionaries. Our
information about them is fullest with reference to the philosophic
schools, which consciously bid for the support of educated Romans. These
groups of philosophers were nearly all completely organized, and formed
an international fraternity as real as the great International Actors
Association and the similar Athletic Union.[255] It was scarcely
feasible to stand neutral. A man was either an Academic, or Stoic, or
Epicurean, or Neopythagorean, and so on. So skilful a trimmer as
Cicero’s friend, the astoundingly shrewd Atticus, was enrolled as an
Epicurean.[256] Even skepticism classified a man as an Academic, as
Cicero himself was classed despite occasional exhibitions of sympathy
for the Stoa. And the combat was as intense and as dogmatic as that
between competing religious sects. That is precisely what they were, and
they bandied their shibboleths with the utmost zeal and unction.

Some of these philosophic sects, the Cynics and Stoics, reached classes
of lower intellectual level. And there they came in conflict with
astrologer and thaumaturg, with Isis and with Atthis devotees, and with
Jews. The popular sermon, the diatribe, was an institution of the
Cynics, and was directed to the crowd. Indeed the chief object of Cynic
jibes was the pretension of philosophers to possess a wisdom that was in
any way superior to the mother-wit of the rudest boor.[257] The Stoics
too used the diatribe with success. It must not, however, be supposed
that either Stoic or Cynic was a serious rival of the dramatic and
sensationally attractive rites of the Eastern cults. The latter counted
their adherents by the hundreds where the preaching philosopher might
pick up an occasional adherent. The importance of the philosophers for
the spread of non-Roman beliefs lies chiefly in the fact that they
reached all classes of society, and, different as they seem from the
cult-associations of the various foreign deities, they really
represented the same emotional need as the latter.

These had literary support as well. We have recently had restored to us
some astrological pamphlets, such as that of Vettius Valens,[258] and we
can only guess from what arsenal Isiac or Mithraist drew those arguments
with which he boasted of confuting even Stoics and Epicureans. But we
may safely assume that tracts existed of this sort.

As far as the Jews are concerned, their propaganda may have begun with
their first settlement in Rome. Cicero does not mention it, but Cicero
was not interested in what went on among the strata of society in which
the Jews then moved. In the next generation their propaganda was so wide
and successful that it must have been established for a considerable
time.

Further, from what has been said it is clear that this propaganda must
have been directed primarily to the plebs, to the same classes, that is,
as those who received Isis and Cybele, Mithra and the Cabiri. At first
it practically did not reach the intellectually cultivated at all. But
the Jews possessed an extensive literature, which in Egypt and the East
generally had assumed the form of “most philosophic” treatises. Indeed,
it is quite clear that the Wisdom of Solomon could be enjoyed by none
but cultured men.[259] Books of this sort, as well as the Bible, were
accessible, and were read by some. The synagogue service was an
exposition of Jewish doctrine upon topics that ranked as philosophic.
While therefore it was mainly from among the masses that Jewish converts
came, here and there men of education must have found the Jewish
preachers as convincing as the philosophic revivalists, who boasted of
no more respectable credentials.

The Roman state had found itself obliged to take cognizance of the
foreign religious movements at an early date. The official acceptance of
Cybele had promptly been surrounded by restrictions. Cybele was always
to remain a foreign goddess. Romans were stringently forbidden to take
part in her ceremonies.[260] Toward the forms of worship themselves, the
Roman attitude was tolerant enough. As long as they were confined to
Egyptians, Syrians, Cappadocians, the participants would be secure from
molestation. But that the foreign rites might displace the ancestral
forms was a well-grounded fear, and drastic precautions were taken
against that. The Bacchanalian incident of 186 B.C.E. is the first of
these instances.

In the same way the Roman police found it necessary at various times to
proceed against astrologers, Isis-worshipers, and philosophers. The
statement frequently occurring, that these groups were banished, is
constantly misunderstood. It can apply only to foreigners in these
classes, not to Roman citizens affected by these strange beliefs; but it
implies that the Roman citizens so affected were sufficiently numerous
to make the desertion of the national religion a probable contingency.
Of course Roman citizens could not violate the laws that regulated
religious observances with impunity. These laws, however, were
ostensibly never directed against the religious observances, but against
abuses and acts that were connected with them. That was true even in the
case of the Bacchanalia, when the decree of the senate expressly
permitted the celebration of the rites under proper restrictions.

Whether honestly or not, the Roman government aimed its measures solely
at certain indubitably criminal acts, which, it was alleged, were
associated with the practice of the foreign cults. These acts were often
offenses against public morality. Conditions of high religious
excitement often sought a physical outlet in dancing or shouting, and no
doubt often enough, when the stimulation of wine or drugs or
flagellation was added, in sexual excesses. Instances that were perhaps
isolated and exceptional were treated as characteristic, and made the
basis for repressive legislation.[261]

Another and better founded objection to many of the forms of foreign
religion was the opportunities they offered for swindlers. As early as
139 B.C.E. the astrologers were banished from Rome, not because of the
feeling that the astrological system was baseless, but because of the
readiness with which professed astrologers defrauded the simple by
portentous horoscopes, which they alone could interpret or avert.[262]
The “Chaldeans” or _mathematici_ included many men who were neither the
one nor the other. It was obviously easier for a Syrian or Oriental
generally to make these claims than for either Greek or Italian. Syrians
in the city accordingly found the profession of quack tempting and
profitable, and doubtless many Jews as well entered it.

We have evidence too that many of the mushroom political associations
were grouped about some of these foreign deities. The possession of
common _sacra_ was, in a sense, the distinguishing mark of any organized
body of men, and organization of the masses in all forms was the
commonest device of the agitators of the revolutionary period. Clodius
had his mobs grouped in _decuries_ and _curiae_.[263] It is likely
enough that in some of these groups, consisting largely of freedmen of
foreign birth, various foreign deities were worshiped in the communal
_sacra_, so that the various police measures restricting or forbidding
these rites may have had strong political motives as well.

When Caesar reconstituted the state after Pharsalia, he knew from direct
experience the danger that lay in unrestricted association ostensibly
for religious purposes. The θίασοι, “cult-associations,”
which he dissolved were undoubtedly grouped about some Greek or Oriental
deity. The Jews were specially exempted, for reasons easy to guess at,
but which we cannot exactly determine.[264] This striking favor cannot
but have immensely increased their influence. We need not suppose that
Caesar’s orders were any more effective than previous decrees of this
character had been. But even a temporary clearing of the field gave the
active propagandists among the Jews an opportunity which they fully
utilized.

We have sketches of Jewish activities in Rome during the following years
drawn by master hands. In every instance, of course, the picture is
drawn with distinct lack of sympathy, but it is none the less valuable
on that account. Easily of first importance is the information furnished
us by the cleverest of Roman poets, Horace.

Quintus Horatius Flaccus was the son of a former slave. His racial
origin accordingly may have been found in any corner of the
Mediterranean in which we choose to look for it. That fact, however, is
of little importance, except that the consciousness of servile ancestry
must have largely influenced his personal intercourse, and his
patriotism must have been somewhat qualified, despite some vigorously
Roman sentiments. Suave, obese, witty, a thoroughly polished gentleman
of wide reading and perfect manners, both sensual and shrewdly
practical, Horace had early reached the point at which one descants on
the merits of frugality and simplicity at the end of a seven-course
dinner. His star was in the ascendant. His patron Maecenas was the
trusted minister of Augustus; and to Augustus, and not Antony, fell the
task of rebuilding the shattered framework of the state. Secure in the
possession of every creature comfort, the freedman’s son could loaf and
invite his soul.

That he did so in exquisite verses is our good fortune, and that he
chose to put his shrewd philosophy and criticism of life into the form
of sketches that are medleys of scenes, lively chat, satirical attacks,
and portraits of types and individuals, makes the period in which he
lived and the society in which he moved almost as vivid to us as that
depicted in the letters of Cicero.

In one of his Satires—“Chats,” as he called them—he tells the story of
his encounter with a pushing gentleman, of a type familiar to every age.
Horace cannot escape from the infliction of his presence, and miserably
succumbing to the inane chatter of the bore, he comes upon his friend
Titus Aristius Fuscus. But his hopes in that quarter are doomed to
disappointment.

“Surely,” says Horace, nudging Fuscus, “you said you had something you
wanted to speak to me about in private.”

“Yes, yes, I remember,” answers Fuscus, “but we’ll let that go for some
more suitable time. To-day’s the thirtieth Sabbath. Why, man, would you
want to offend the circumcised Jews?”

“I can’t say that I feel any scruples on that score.”

“But I do. I haven’t your strength of mind. I’m only a humble citizen.
You’ll excuse me. I shall talk over our business at some other time.”

The little scene is so significant that we shall have to dwell on it.
One unescapable inference is that the Jews in Rome were numerous, and
that a great many non-Jews participated wholly or partially in their
observances. Fuscus need not be taken seriously about his own beliefs,
but his excuse would be extravagant in the highest degree if the
situation of the Jews were not such as has been suggested. Indeed, the
terms of intentional offensiveness which Fuscus uses indicate the
serious annoyance of either himself or Horace that that should be the
case.

The “thirtieth Sabbath” will probably remain an unsolved riddle.[265]
And whatever the day was, the extreme veneration expressed by Fuscus in
declining even to discuss profane affairs is of course absurdly out of
keeping with the words he uses. Fuscus is simply assuming the tone of a
demi-proselyte, a _metuens Sabbata_, whose superstitious dread of the
rites he has half embraced would make him carry his devotion to an
excess. Horace thus obtains an opportunity of sketching a new type of
absurdity, in the very act of girding at the one which is the subject of
the _sermo_.

And Horace makes still another reference to the proselytizing activities
of the Jews. “You must allow me my scribbling,” he writes to Maecenas in
another Satire. “If you don’t, a great crowd of poets will come to help
me. We far outnumber you, and, like the Jews, will compel you to join
our rout.”[266]

This is explicit enough in all conscience, and gives a very vivid
picture of the public preaching that must have brought the Jews to the
unwelcome notice of every saunterer in the Roman streets. Horace,
despite his slave grandfather, is a gentleman, the associate of Rome’s
aristocracy, a member of the most select circle of the city’s society.
The Jewish proselytes, whether fully converted or “righteous strangers,”
must have been very numerous indeed, if he was forced to take such
relatively frequent notice of them. Horace has no pictures, like those
of Juvenal, of presumptuous Syrians, Egyptians, or Greeks swaggering
about the city. It is only these Syrian _Iudaei_ whom he finds
irritating, and wholly because of their successful hunt for souls.

It is true that all this may be due to personal circumstances in his own
surroundings. Some of his acquaintances, or men whom he occasionally
encountered, may have been proselytes; others may have been impressed by
certain Jewish forms or ideas. Horace is taking his fling at them in his
usual light manner. There is something ludicrous to a detached
philosopher in the eager striving to save one’s soul, and still more
absurd in the earnest attempt to gain adherents for an association that
promises salvation.

Once he takes a more serious tone. In the famous journey he made with
Maecenas to Brundisium Horace is told of an altar-miracle at Egnatia.
The incense melts of itself, it seems, in the local temple. “Tell it to
the Jew Apella,” says Horace, “not to me. I have always been taught that
the gods live free from every care, and if anything wonderful occurs in
nature, it is not because it has been sent down from heaven by
meddlesome divinities.”[267]

This Jew Apella—a dialect-form of Apollas or Apollonius[268]—is no doubt
a real person, who may perhaps have recounted to Horace some of the
miracles of the Bible. Horace’s raillery is directed plainly enough at
the credulity that will accept these stories, and equally at the
troublesome theology which makes the god a factor in daily life. Life
was much simpler if no such incalculable quantity were injected into it.
And to keep life free from harassing and unnecessary complications was
the essence of his philosophy.[269]

At about the same time another writer, the geographer Strabo, of Amasea
in Cappadocia, makes a statement of special interest. As quoted by
Josephus (Ant. XIV. vii. 2) he says: “These people have already reached
every city, and it would be difficult to find a place in the whole world
that has not received this tribe and succumbed to it.”

Obviously the statement is a gross exaggeration, and at most applicable
to the cities of Egypt and Cyrene, in connection with which it is made.
But that such a statement could be made at all is excellent evidence
that it was at least partially true, and that there were Jewish
communities practically everywhere, although it can hardly be the case
that they were everywhere dominant. However, the sketches by Horace are
an eloquent commentary upon the statement of Strabo. Not merely the East
or Africa, but the capital itself, was overrun with Jews, and their
number was constantly increasing.

Horace, it has been said, wrote of and for a cultured aristocracy. So
did the other poets of the age, Propertius, Tibullus, Ovid. But all of
them were more than ordinarily familiar with the _bas-lieux_ where
disreputable passions might be gratified. The voluptuary Ovid was
especially prone to go down into the sewer for new sensations, and just
as Horace met Jews in the boulevards, so Ovid knew them in the slums.

In his salad-days Ovid had written a manual of debauchery, which he
called the “Art of Love.” He was destined to regret bitterly the
facility of verse and of conscience that gave birth to this bold
composition. But written it was, and in his advice to the dissolute
young Romans he enumerates the time and place for their amours.

    Rome [he says in _Ars Amatoria_, i. 55 seq.] is the place for
    beauties. Venus has her fixed abode in the city of her Aeneas.
    Whatever you desire you may find. All you have to do is to take a
    walk in the Porticus of Pompey or of Livia,... Do not pass by the
    place where Venus mourns Adonis, or where the Syrian Jew performs
    his rites each seventh day. Nor overlook the temples of the
    linen-clad heifer from Memphis. She makes many what Jove made her.
    Even the fora favor Love, ... those where the Appian aqueduct gushes
    forth near the marble temple of Venus.... But above all stalk your
    game in the theaters.

In these instances Ovid refers to place, not to time, and it is only as
part of the passages as a whole that the individual references can be
understood. It will be seen that all the localities, beginning with the
Porticus of Pompey in the Campus Martius, are merely casual. It is at
the theater and circus where Ovid’s pupils are chiefly to pick out the
ladies of their light loves. For that reason the other places specified
are also, to a certain extent, show places. The mention of the
law-courts is especially noteworthy in this connection.

We must therefore assume that in the Jewish proseucha and in the temple
of the Egyptian Isis there were to be found a certain number of curious
onlookers, particularly women, and while many of them became ardent
converts, a certain number were innocent of any intentions except to
while away an idle hour, and were easy game for the professional
“mashers” for whom Ovid writes. Isis and Judaism were the two Oriental
cults which at this time had the greatest success in Rome. And we can
easily imagine how the unoccupied of all classes thronged to every new
fashion in religious stimulation as in others.

Ovid is as explicit in the selection of time as of place.

    Do not disregard time,... Avoid the first of April. Then the rainy
    season begins, and storms are frequent. But begin the day of the
    defeat at the Allia, or the day on which the Sabbath feast comes
    again, which the Syrian from Palestine celebrates. That’s a day on
    which other business ought not to be done. (Ars. Am. i. 413 seq.)

Again, in his palinode, with which he vainly hoped to regain his
shattered reputation, “The Cure for Love” (vv. 214 seq.), he brings the
same things together:

    Off with you; take a long journey to some distant land.... The less
    you want to go, the more you must; remember that! Be firm and make
    your unwilling feet run. Do not pray for rain. Let no imported
    Sabbaths hinder you, nor the day on which we remember the disaster
    on the Allia.[270]

As far as Ovid is concerned, and we must assume he is speaking for
Fuscus’ _multi_, a certain Jewish feast, whether it is the Sabbath or
some special holiday, such as the Day of Atonement, is ranked with the
Dies Alliensis, the fifteenth of July, the day on which, in 390 B.C.E.,
the Romans suffered their great defeat at the hands of the Gauls, and
which was in consequence an ill-omened day from that time forth. Again,
the Sabbath is classed with the rainy season as a day that might
ordinarily incline a man to put off serious business.

As stated in the Notes, it is a common error to suppose that the
generally ill-omened character of these days makes them eminently proper
for flirtation. No Roman, however cynical, could flout superstition to
that extent. The advice is given for purely practical considerations.
The rainy season at the time of the equinoxes is an inauspicious time to
begin a courtship, which, as we have seen in the previous passage, must
be carried on almost wholly in the open air. Social gatherings in the
houses of friends in the society of ladies were not common. There was
nothing among the Romans to correspond to modern five-o’clock’s or
receptions, at which court might be paid to anyone who had caught the
fancy of the Roman man about town. It is in the porticoes, in the idle
crowds at the theater or circus, where the steps of ingratiating are to
be carried out, and for these the rising of the Pleiades (Ars. Am. i.
409) is distinctly unpromising.

This is especially borne out by the passage immediately following the
one quoted from the “Art of Love” (Ars. Am. i. 417 seq.). The most
inauspicious day to attempt the beginning of an intrigue is the lady’s
birthday. Gifts are in order then, and they undoubtedly deplete one’s
pocket-book. Ovid is jocose here, but the point is the same throughout.
The hints and suggestions are as practical and direct as the formula of
Ovid’s face-powder, which he also sets forth in the unfinished verses
called _Medicamina Faciei Femineae_.

That which makes the Dies Alliensis and the Jewish Sabbath desirable is
the fact that the former is in mid-July and the latter in the early
fall, the most delightful of Italian seasons. Then an unbroken series of
cloudless skies is almost assured; and the Roman fop could count on
meeting his fair one day after day in one of the places of assignation
so conveniently enumerated by Ovid.

The phrase _rebus minus apta gerendis_, “unsuitable for transacting
business,” is best taken as given in the translation (above, p. 251).
Ovid knows that undertakings are rare on that day, and that causes its
insertion. If it were merely that cessation of ordinary business made it
easier for idlers to pursue their amours, it must be remembered that the
_jeunesse dorée_ had no other ordinary business than falling in love.

The reference in the “Cure for Love” (above, p. 251) is of quite a
different character. It will be noted that _pluviae_, “the rainy
season,” which in the first case is particularly contrasted with the
Sabbath and the Allia day, is here associated with them. “Let nothing
hinder you,” says Ovid, “neither a good excuse nor a bad one; neither
the weather nor superstition.” The point of the reference in the two
cases is accordingly not at all the same. In the first instance the
accidental fact that the Allia day and a certain Jewish festival occur
during pleasant weather singles them out for mention. In the second it
is the religious association of the day that Ovid has in mind.

As far as Ovid is personally concerned, there is no more than in Horace
a trace of sympathy for the Jewish cult. We have seen that in every
instance this cult is only one of several illustrations. The adjective
_peregrina_, “foreign,” applied to the Sabbath, gives the tone of all
the passages. Ovid is a collector of light emotions. Of serious beliefs
he has no vestige. But the presence of these Syrians in the city
interests him as anything else picturesque would. He takes cognizance of
the part they play in the life of the city, and is a valuable witness on
that point.

The same inference may be drawn from the letter of Augustus to Tiberius
(Suet. Aug. 76): “There is no Jew, my dear Tiberius, who keeps his fast
on the Sabbath as I kept it to-day.” If the considerations advanced in
Note 269 are valid, the Sabbath here is the Day of Atonement. But the
significant fact is the use of the illustration at all. It confirms
Strabo’s statement of the extent and success of the propaganda of the
Jews that all these writers in some way mention their presence.

That the preaching of the Jews was vigorous and aggressive is almost a
necessary inference. We know no less than three of their synagogues by
name, Augustenses, Volumnienses, Agrippenses,[271] and we have no reason
to assume that these three exhausted the list. To many Romans the ardor
of their proselytizing was offensive. It seemed a systematic attempt to
transform the ancestral faith of the state. A casual reference in
Valerius Maximus, a contemporary of Tiberius, charges the Jews with
having attempted “to contaminate Roman beliefs by foisting upon them the
worship of Jupiter Sabazios.”[272] Valerius goes on to say that the
praetor Hispalus expelled the Jews for that reason as early as 139
B.C.E. If such a thing took place, it was undoubtedly an act similar to
an expulsion under Tiberius (below, p. 306), and was based on definite
infractions of law, perhaps the law against unlicensed fortune-telling.
The Jews in both cases were associated with the Chaldeans, a fact that
makes the supposition more likely. But Valerius has in mind the
conditions of his own day, when the success of the Jewish propaganda was
bitterly resented, as we have seen, by Horace and Fuscus, and, as we
shall later see, by Seneca and his associates generally.

If we try to imagine what the Jewish Roman communities of that day were
like, we shall have to think of them as a proletariat. Freedmen in the
second or third generation must have constituted a large part of them,
and later references make it likely that many earned their livelihood by
the proscribed arts of divination and fortune-telling. As in Alexandria,
the bulk were probably artisans. Some were physicians, a profession then
ranking in social degree with the manual trades, and usually exercised
by slaves or freedmen.[273] The Roman encyclopedist Celsus mentions two
Jewish medical authorities (De Med. V. xix. 11; xxii. 4). But the
majority must have formed part of the pauperized city mob, turbulent and
ignorant, and no doubt only moderately acquainted with their own laws
and literature, so that we cannot be surprised to find indications of
many things among them that were regarded as sacrilege in Jerusalem,
such as carved animal figures on tombstones.[274]

However, there must at least have been some of a different type, whose
command of their controversial literature enabled them to meet the
competing philosophies upon their own ground and impress themselves upon
some of the men of Augustus’ own circle.


------------------------------------------------------------------------




                              CHAPTER XVII

                 THE JEWS OF THE EMPIRE TILL THE REVOLT


One of the great determining events in ancient and modern history took
place on January 1, 27 B.C.E., when Gaius Caesar Octavianus, returned
from his successful campaigns in the East, was solemnly invested with
the civil and military primacy of the Roman world. The importance of
that particular historic moment is due of course not to anything in
itself, but to the fact that it was the external and overt stamp put
upon the development of centuries. The basic governmental scheme of
ancient society—the city-state—was bankrupt. Its affairs were being
wound up, and the receiver was in possession.

The reconstitution by Augustus appeared to the men of his day as the
inauguration of an epoch. Poets hailed the dawn of a new day, and
unqualifiedly saluted its great figure as a living god.[275] But we
shall receive a false impression of the time and its condition, if we
assume it to resemble an empire of modern type.

The Roman empire as founded by Augustus was simply the expression of the
fact that between the Euphrates and the Ocean, between the Danube and
the great African Desert, all the various forms of constituted authority
were subject to revision by the will of the Roman people, _i.e._ those
who actually lived, or had an indefeasible right to live, within the
walls of the Roman city. The _populus Romanus_ had chosen to delegate
functions of great extent and importance to a single man, to Augustus;
but the power wielded by Augustus was not in any sense the power of an
unrestrained master, nor was the rule of the Roman people the actual and
direct government of the nations subject to it.

It would be quite impossible to enumerate the various communities which,
under Augustus, as they had before, maintained their customs as the
unbroken tradition of many centuries. In the mountains of Asia Minor it
is likely that such a people as the Carduchi, whom Xenophon encountered
there, were still under Augustus determining their mutual rights and
obligations by rules that were either the same as those of Xenophon’s
time or directly derived from those rules.[276] So the cartouches on the
Egyptian monuments might have been read by the clerks of Amen-hem-et,
and would have excited no queries from them. The communities of the
Mediterranean enforced their law—that is, the rules which constrained
the individual member to respect the claims of his fellows—without
noticeable break. The difference was that there was a limit to which it
might be enforced, and that limit was set by the caprice of another and
a paramount people.

Although the sovereignty of the Roman people was limitless, it was not,
as a matter of fact, capriciously exercised. During the republic the
theory of provincial organization had been somewhat of the following
nature. Within any given territory contained in the limits of the
province, there existed a certain number of individual civic units,
which might take the form of city-states, territorial states of varying
extent, leagues of communities, kingdoms, tetrarchies, or hieratic
religious communes. Any or all of these might be gathered within a
single province, a word which is essentially abstract, and denoted a
magisterial function rather than a territory. Into the midst of these
_civitates_, this jumble of conflicting civic interests, there was sent
a representative of the sovereign Roman people, invested with
_imperium_, or supreme power, a term in which for Romans was the essence
of the higher magistracies. Since the provincial magistrate had no
colleagues, and since the tribunician check upon him was inoperative
beyond the first milestone from the city, the wielder of the _imperium_
outside of Italy was at law and often in fact an absolute despot for the
period of his office.

However, in theory his functions were divided as follows: first, he was
the only officer with jurisdiction over the Roman citizens temporarily
resident in the province; secondly, he kept the peace; thirdly, he
guaranteed the treaty rights of those communities that had treaties with
Rome; and fourthly, he enforced and maintained the local customary law
of all these communities. His judicial functions might include cases of
all these kinds, so that in rapid succession the praetor or propraetor
might be called upon to enforce the Twelve Tables and an ancient tribal
usage of the Galatian Tectosages.

The checks upon the holder of _imperium_ at Rome consisted in the
peculiar Roman theory of magistracy, one of the corollaries of which was
the right of any other equal or superior magistrate, or of any tribune,
to veto any administrative act. A second check lay in the right of
appeal in capital cases to the people. A third was found in the
accountability for every illegal or oppressive action. This
accountability however existed only after the magistracy had expired.

Outside of Rome only the last check existed. For everything done beyond
the functions enumerated above, it was possible, even usual, to attempt
to make the governor responsible after his term of office was over. We
know how frequently that attempt was futile, and how constantly and
flagrantly corrupt juries acquitted equally corrupt governors. “Catiline
will be acquitted of extortion,” writes Cicero in 65 B.C.E., “if the
jury believes that the sun does not shine at noon.”[277] The jury
evidently thought so, since he was acquitted. But upon occasion, and
generally when there were personal and political motives at work as
well, these governors were convicted, so that there was always a certain
risk attached to any attempt at playing the tyrant for the brief period
of a governor’s authority.[278]

The Augustan monarchy brought no real change into the theory of
provincial organization, except as to relatively unimportant details.
But one great reform was instituted. The responsibility of the governor
became a real one, and was sharply presented to those officials. For the
provinces, accordingly, the advent of Augustus was an unmixed blessing,
since, except for a few sentimentalists, the presence of the Roman
representative as the final court of appeal was not at all resented. We
can accordingly understand the extravagance with which the rich and
populous East, always the center of wealth and civilization, received
the Reformer, and the unanimity and perhaps sincerity with which he was
hailed as living god.[279]

We cannot be certain that this was encouraged by Augustus himself. There
is nothing in his character that indicates any special sympathy with the
point of view demanded by it; nothing of that daemonic strain noticeable
in Alexander, which makes it easy to believe that the latter was one of
the first to be convinced by the salutation of the priests of Ammon. But
Augustus recognized at once the value for unity that the tendency to
deify the monarch possessed. The reverence for the living monarch, to be
transformed into an undisguised worship at his death, was, however, to
be superimposed upon existing forms. Nothing was more characteristically
Roman than Augustus’ eagerness to make it clear that the vast domain of
the empire was to remain, as before, a mass of disparate communities of
which the _populus Romanus_ was only one, although a paramount one, and
that in each of these communities every effort was to be made to
maintain the ancestral ritual in government and worship. What he added
was simply the principle that to keep the community together, to prevent
the chaos and anarchy of a dissolution of the empire, it was necessary
to bestow on the _princeps_, on the First Citizen of the paramount Roman
people, such powers and functions as would assure the coherence of the
whole. These powers he selected himself. Such a step as that taken by
the Constitution of Caracalla, which attempted to enforce a legal
merging of all the communities into a single state, would have been
nothing else than abhorrent to Augustus.[280] And, indeed, it was a
distinctly un-Roman idea.

In Rome Augustus was chiefly intent upon a restoration of everything
that could well be restored in the social, religious, and political life
of the people. Certain of the political elements, such as the actual
sovereignty of the _populus_, as far as it could be physically assembled
in the Campus Martius, had to be abandoned, as demonstrably inconsistent
with the larger purpose which Augustus had set himself. But in every
other respect, he did not, as Julius Caesar had done, compel the Romans
to face the unpleasant fact that a revolution had taken place, but
professed to be simply a restorer of the ancient polity. Perhaps he did
not face the facts himself. At any rate he seems sincerely to have
believed that morality and sobriety could be reconstituted by statute,
and that one, by dint of willing, might live under Caesar as men lived
under Numa—barring such un-Sabine additions as marble palaces and purple
togas.

With his mind full of these views, Augustus could hardly be expected to
regard favorably those tendencies in his own time which inevitably made
for real unity of the empire in speech, blood, and religion. He was
quite aware that this unity would not be produced by a coalescing of
everything into new forms, but by the conquest of all or most of the
existing elements by the one most powerful or most aggressive.
Unchecked, it was likely that Greek speech would drive out Latin, Syrian
blood dominate Roman, or any one of the various Oriental worships
dislodge the Capitoline Triad.

On the last point he had even a definite policy of opposition. His
sagacious adviser Maecenas had laid great stress upon the ease with
which foreign religions introduce a modification of habits of life, in
his last words:[281]

    Take active part in divine worship, in every way established by our
    ancestral customs, and compel others to respect religion, but avoid
    and punish those who attempt to introduce foreign elements into it.
    Do so not merely as a mark of honor to the gods—although you may be
    sure that anyone who despises them, sets little value upon
    anything—but because those who introduce new deities are by that
    very act persuading the masses to observe laws foreign to our own.
    Hence we have secret gatherings and assemblies of different sort,
    all of which are inconsistent with the monarchical principle.

His commendation of Gaius’ avoidance of sacrifice at Jerusalem was of a
piece with this policy.[282]

The Jews in Rome, who had been directly favored by Caesar, had to be
contented, as far as Augustus was concerned, with freedom from
molestation. However, this freedom was real enough to enable their
situation in Rome to reach the development hinted at in the Augustan
poets, although their activities militated strongly against the most
cherished plans of Augustus.

In the rest of the empire the Jews of the various communities found
their situation unchanged. Even the obnoxious privileges which they had
in several cities of Asia continued unimpaired,[283] and here the
orthodox Jewish propaganda and a few generations later the heterodox
Jewish propaganda made rapid strides.[284]

Judea belonged, in spite of the quasi-independence of Herod, to the
province of Syria, which meant that such dues as Herod, the Jewish king,
owed Rome would be enforced, if he were recalcitrant, by the Roman
legate at Antioch. Herod’s name throughout the empire was as much a
synonym for wealth as it is now for cruelty. And his wealth and power
advertised the Jews notably, a fact which their propaganda could
scarcely help turning to account.[285]

The attitude of the various Jewish synagogues and communes toward
Judea was one that appeared to the men of the day as that which bound
various colonies of a city to the mother-city. Indeed the Jewish
communities outside of Palestine were styled explicitly colonies,
ἀποικία. Such a tie, however, was conceived in the Greek
fashion and not in the Roman. The Greek colony was bound to its
mother-city by sentiment only, not, as in the case of the Romans, by
law. That sentiment might be powerful enough at times, but it was not
inconsistent with the bitterest warfare. Consequently such movements
as appear in Palestine need not at all have been reflected in the
synagogues of the East and West, and there is nothing to indicate that
the active and successful proselytizing of the Asiatic and Roman
synagogues was either directed or systematically encouraged by the
Pharisaic majority in the Sanhedrin at Jerusalem. It will at all times
create a wholly false impression to speak of the Jews of that period
as of a single community bound by common interests and open to
identical influences. The independence of the Jewish congregations of
one another was quite real, and was even insisted upon. Neither the
high priest nor the Nasi of the Sanhedrin pretended to any authority
except over those legally resident in Judea; and often, when the
reverence for the temple and the holy city was most strongly
emphasized, intense contempt might be manifested for those who were at
the moment the holders of the supreme authority in the mother-country.

Another matter that is apt to be lost sight of in this connection is the
fact that not all Jews of the time lived within the Roman empire. The
Persian kingdom, which Alexander had conquered, and which the Seleucidae
had with varying success attempted to maintain, had fallen to pieces
long before the Roman occupation of Syria. Media, Babylonia, Bactria
resumed a quasi-independence, which however was soon lost when the
obscure province of Parthia—as Persis had done five centuries
before—assumed a dominance that ended in direct supremacy. The Roman
limits were set at the river Euphrates, leaving Armenia a bloody,
debatable ground. The one great moment in the history of this new
Parthian empire was the decisive defeat of Crassus at Carrhae in 58
B.C.E., a victory that gave the Parthians sufficient prestige to
maintain themselves under conditions of domestic disorder that would
ordinarily have been fatal. The Augustan poets and courtiers might
magnify the return of the Roman standards by King Phraates to their
hearts’ content. They might, as they did, exultantly proclaim that the
Crassi were avenged, that the known world to the shadowy confines of the
Indus bowed to the will of the living god Augustus. The fact remained
that, after Carrhae, the conquest of the country beyond the Euphrates
ceased to be a part of the Roman programme, and, except for the
transient successes of Trajan, was never seriously attempted.

In this Parthian kingdom, of which the capital was the ancient and
indestructible city of Babylon, Jews had dwelt since the time of
Nebuchadnezzar. There is even every reason to believe that those who
remained at Babylon were decidedly not the least notable of the people
in birth or culture. And between Babylon and Judea there was constant
communication. When Babylon became the seat of the only power still
existing that seemed formidable to Rome, it is obvious that the
uninterrupted communication between the Jews of that section and the
mother-country would create political situations of no slight delicacy,
and may have played a much more important part in determining the
relations of the governing Romans to the Jews than our sources show.

That there was at all times a Parthian party among the Palestinian Jews
there can be no doubt. We know too little of the history of Parthia to
speak confidently on the subject, but Parthian rulers seem to have
brought to the Jewish religious philosophy a larger measure of sympathy
and comprehension than most Roman representatives. While the existence
of Parthian sympathizers may date almost from the beginning of Parthian
supremacy, their presence was very concretely manifested when Jannai’s
son, Aristobulus, appealed to Parthia as Hyrcanus had appealed to Rome.
Indeed a Parthian army invaded and captured Palestine, and gave
Aristobulus’ son, Mattathiah-Antigonus, a brief lease of royal dignity.
Every instance of dissatisfaction with the Roman government was the
occasion for the rise of Parthian sympathies.

It may further be recalled that Parthia was the continuation of Persia.
Of all foreign dominations the Persian rule was the one most regretted
by the Jews, and the Persian king’s claim to reverence never died out in
the regions once subject to him. We may remember with what humility,
some years later, Izates of Adiabene dismounted and walked on foot
before the exiled Parthian king, although the latter had gone to him as
a suppliant, and had been prostrate in the dust before him. The prestige
of the Great King, diminished considerably to be sure, had still not
completely faded.[286]

The one general term that covered all the Jews of various types was
“race of the Jews,” _gens Iudaeorum_, γένος Ἰουδαίων. It
was meant to be a racial descriptive appellation, and was constantly
combined with other adjectives denoting nationality or citizenship. The
temptation to make an actual unit of any group that can be covered by a
single term is well-nigh irresistible, and it is strengthened for us by
the century-old associations that have made Palestine the embodiment of
an ideal. Varying as the Jews of that time were in temperament,
character, occupations, position, and mental endowments, the fate and
vicissitudes of the mother-country, and particularly of the holy
metropolis Jerusalem, went home vividly to all of them, scattered as
they were between the shores of the Caspian Sea and Spain. In this
respect the _gens Iudaeorum_ was a real unit. Their hearts were turned
to the Zion Hill.

Not all Palestine, however, formed this mother-country. The mere fact
that the Hasmoneans had brought a great deal of the surrounding
territory under subjection, and made the boundaries of their power
almost as extensive as those of David and Solomon, did not make a single
country of their dominions. The real metropolis was Jerusalem and its
supporting territory of Judea. In this predominance of the city in
post-Exilic Judaism, we may see either Greek influence or the
continuance of the ancient city-state idea, as much a general
characteristic of Eastern civilization as it is specifically of Greek.
Not even undoubted Jewish descent, or loyalty to the Jewish Law, made of
the adjacent lands an integral part of Judea. The Jews of Gaulonitis,
Galilee, Ituraea, Peraea, Trachonitis, Idumaea, were, like the Jews of
Rome, of Alexandria, or of Babylon, Jews of foreign nationality to
inhabitants of Jerusalem, although the association was notably closer
and the occasion of common performance of Jewish rites much more
frequent than was the case with the more distant Jews.


[Illustration: TOMBS OF THE KINGS, VALLEY OF KEDRON, JERUSALEM
(From Wilson’s “Jerusalem”)]


The Idumean Herod had been confirmed by Rome in the sovereignty of a
wide and miscellaneous territory, which included Greek cities, as well
as these territorial units enumerated above. The favor he enjoyed
granted him practically all the privileges that an independent sovereign
could hold, except that of issuing gold coins.[287] Further, the
authority was only for his life. The right of disposing of his dominions
was no part of his power. His will was merely suggestive, and carried no
weight beyond that.

His favor in the eyes of the Romans was based upon his scarcely
disguised Hellenic sympathies and his proven loyalty to his masters. The
Parthian invasion of 40 B.C.E. and the existence of Parthian
sympathizers made the maintenance of order in Palestine a matter of the
highest importance. The significance of these Eastern marches for the
stability and safety of Rome was even greater than those of the North
along the Rhine, where also constant turbulence was to be feared, and
eternal vigilance was demanded. In the East, however, there was not
merely a horde of plundering savages to be repelled, but the aggression
of an ancient and civilized power, bearing a title to prestige compared
with which that of Macedonian and Roman was of recent growth. And
Parthian successes here immediately jeopardized Egypt, already rapidly
becoming the granary of the Empire.

Quite in accordance with Roman policy, indeed with ancient policy in
general, Augustus vastly preferred to have the peace of this region
assured by means of a reliable native government than directly by Roman
administration. The Romans did not covet responsibility. If a native
prince was trustworthy, it was a matter of common sense to permit him to
undertake the arduous duty of policing the country rather than assume it
themselves. The difficulty was to discover such a man or government.
Experience and the suspiciousness that was almost a national trait
convinced the Romans that only very few were to be trusted, and these
not for long. In Herod, however, they seemed to have discovered a
trustworthy instrument, and while it is not strictly true that the
powers conferred upon him were of unexampled extent, they were
undoubtedly unusual and amply justified the regal splendor Herod
assumed. The readiness with which Herod’s loyalty to Antony was pardoned
demonstrated the clear perception on the part of Augustus of how
admirably Herod could serve Roman purposes here.

One of the motives that generally impelled Romans to permit native
autonomy was no doubt to gain credit for generosity with their subjects.
They might be forgiven for supposing that Roman rule would be more
acceptable if it came indirectly through the medium of a king that was
himself of Jewish stock. The distinction between Idumean and Jew proper
would hardly be recognized by a Roman, although the distinction between
the geographical entities of Idumaea and Judea was familiar enough.

But the Romans likewise knew and consciously exploited Herod’s
unpopularity. Strabo states that the humiliating execution of Antigonus
was intended to decrease the prestige of the latter and increase that of
Herod.[288] Josephus and the Talmud would be ample evidences themselves
of the hatred and the bitter antagonism with which Herod was
regarded.[289] None the less it may well be that the unpopularity was
largely personal, and produced by the violence and cruelties of which
Herod was guilty. It appears so in Strabo’s account. Idumean descent
cannot have been the principal reproach directed against Herod by his
subjects. On more than one occasion the Idumeans had evinced their
attachment to the Jewish Law.[290] Nor was Herod wholly without ardent
supporters. In the cities which he had founded there were many men
devoted to him. Even—or perhaps especially—among the priests, there was
a distinctly Herodian faction.[291] It is highly likely that hatred of
Herod was especially strong in those who hated Rome as well, either
through Parthian proclivities or because Rome seemed to present a danger
to the maintenance of their institutions. And among these men were, it
appears, most of those whose teachings have come down to us in the
course of later tradition.

To the Romans this devotion of the Palestinian Jews to their Law seemed
an excessive and even reprehensible thing. As we have seen, the Jews
were qualified as _superstitiosi_, “superstitious” (above, p. 177). In
general, to be sure, zeal for ancestral institutions was supported by
the Romans, and they were not particularly concerned that foreign
institutions should resemble theirs. However, if there were any from
which a breach of the peace was to be apprehended, they might be
regarded as practices to be suppressed.

The Romans had shown for certain Jewish customs a very marked respect.
The intense Jewish repugnance to images was at first difficult for
Romans to realize, since they had been training themselves for
generations to test the degree of civilization by the interest in the
plastic arts. That there might be among barbarians no statues was
natural enough: that the barbarians would refuse to take them when
offered, was incomprehensible. But, hard though it was to realize, the
Romans quickly enough did realize it. The capital concession of issuing
no Roman coins for Judea with anything but the traditional symbols on
them, of carefully eliminating those which bore the emperor’s effigy,
undoubtedly showed their good-will in the matter.[292] And the fact may
be noted that after the coins celebrating the triumph of Vespasian and
Titus, with the Latin and Greek legends Ἰουδαίας Ἑαλωκυίας,
_Iudaea capta_, “For the Conquest of Judea,” no Roman coins with
imperial effigies appear till the radical reorganization by Hadrian.
That indicated clearly enough the extent to which the Romans were
willing to respect what was to them a purely irrational prejudice.

One other matter was easier for Romans to comprehend, and that was the
inviolable sanctity of certain things and places. It was a common enough
conception that certain places were unapproachable to all but a few,
ἄδυτα; and that certain things, like the Palladium,
suffered profanation from the slightest touch. They submitted
accordingly with a good grace to exclusion from most of the temple
precincts, and Nero[293] readily gave his consent to the building of the
wall that prevented Agrippa II from turning the temple ceremonies into a
show for his courtiers. The punishment of a Roman soldier, who tore a
scroll of the Pentateuch, is another case in point. The soldier may have
been a Syrian enrolled from the section in which he served, and not
properly a Roman at all. None the less an arbitrary and distinctly
unsympathetic procurator felt his responsibility for threatened
disorders keenly enough to make this drastic example.[294]

Herod had kept order. He had done so with a high hand, and had met with
frequent rebellions. Himself wholly inclined to complete Hellenization,
he had made many efforts to conciliate his Jewish subjects. His lust for
building he gratified only in the pagan cities subject to him. His coins
bear no device except the inanimate objects and vegetable forms allowed
by law and tradition. With cautious regard to certain openly expressed
fears on the part of the Jews, he rebuilt the temple on a magnificent
scale. He spoke of the Israelites as “our ancestors.”[295] As has been
said, he did not wholly want adherents among priests and people. That he
died as an embittered and vindictive despot, conscious of being
generally detested, and contriving fiendish plots to make his death
deplored, is probable enough, and is amply explained by the domestic
difficulties with which he had to contend all his life.[296]

In some cases at least, it was his zeal for orderly administration that
caused friction with the people. His law sentencing burglars to foreign
slavery is an instance (Jos. Ant. XVI. i. 1). In general, however, the
mere suppression of more or less organized brigandage was a task that
took all his attention, but this “brigandage” was often a real attempt
at revolution, in which popular teachers were suspected of being
implicated, and every such suppression carried with it in its train a
series of executions that did not increase the king’s popularity.

These “robbers” or “brigands” were of different types. The distinction
which Roman lawyers made between war proper, _iustum bellum_, and
brigandage, _latrocinium_, was in Syria and the surrounding regions
rather quantitative than qualitative. So, after Herod’s first defeat by
the Arabians, “he engaged in robberies,” τοὐντεῦθεν ὁ μὲν Ἡρώδης
ληστείαις ἐχρῆτο (Jos. Ant. XV. v. 1), which meant only that he
made short incursions into the enemy’s country, until he had the
strength to attempt another pitched battle. So also of the Trachonitians
(_ibid._ XVI. ix. 3). Every one of the expeditions in which the
Hasmonean rulers had increased their dominions had been in the eyes of
the Syrian historians “robberies.” Itureans and other Syrians had been
guilty of them under the last Seleucids.[297] In the prologue to
Pompeius Trogus’ Thirty-ninth Book, as contained in Justin’s
epitome,[298] we hear the conquests of John Hyrcanus and Alexander
Jannai described as _latrocinia_. And again (xl. 2) we read that Pompey
refused the petition of Antiochus, son of Cyzicenus, to be called king
of Syria, on the ground that Antiochus had miserably shirked his
responsibilities for eleven years, and he, Pompey, would not give him
what he could not maintain, “lest he should again expose Syria to Jewish
and Arabian brigands,” _ne rursus Syriam Iudaeorum et Arabum latrociniis
infestam reddat_.

Herod had kept these robbers in check, and had effectually fulfilled his
tacit engagement to the _populus Romanus_. His death immediately removed
the strong hand. His son Archelaus found an insurrection on his hands
almost at once, which he suppressed with great bloodshed. The moment he
left for Rome to maintain his claims to a part of this inheritance, the
governor of Syria suppressed another revolt; and hardly had he turned
his back, when his procurator Sabinus found himself surrounded by a
determined band of rebels recruited principally from Galilee, Idumaea,
Jericho, and the trans-Jordan territory. In spite of a successful sortie
by the Romans, Sabinus was nothing less than besieged in the Tower of
Phasael.

Innumerable (μύριοι) disorders, Josephus tells us (Ant.
XVII. x. 4), occurred at about the same time. Some two thousand of
Herod’s soldiers engaged, as was so often the case, in plunder on their
own account. Sepphoris in Galilee was seized and plundered by Judah, son
of the highwayman (αρχιληστής) Hezekiah, who made the
neighboring country dangerous with his band of “madmen”
(ἀπονενοημένοι). At Jericho Simon, a former slave of Herod,
had himself proclaimed king and sacked the palace there. But more
serious than these was the band of outlaws commanded by four brothers,
of whom only Athronges is mentioned. These attacked both the local
troops and even Roman detachments and were not suppressed till much
later.[299]

All these disorders required the presence of Varus[300] once more. He
marched on Jerusalem at the head of an army, turning over the various
towns on his route to be sacked by his Arabian allies, precisely as both
British and French used their Indian allies during the colonial wars in
America.

The effect of such conditions in so critical a place as Judea, was to
call Roman attention to the country to a much greater extent than was
advantageous to the Jews. The region very naturally appeared to them
as a turbulent and seditious section, much as Gaul did to Julius
Caesar and largely for the same reason, the instinctive love of
liberty and the presence of “innovators,” νεωτερισταί,
_cupidi rerum novarum_, restless and ambitious instigators of
rebellion.[301] The Jerusalem Jews are, to be sure, very eager to
escape the reproach of disloyalty. The rebellion was the work of
outsiders (ἐπήλυδες), to wit, the Galileans and
Gileadites above-mentioned.[302]

Varus crucified two thousand men, and then disbanded his auxiliary army.
The latter, composed obviously of natives of the country, proceeded to
plunder on their own account. Varus’ prompt action brought them to
terms. The officers were seized and sent to Rome, where, however, only
the relatives of Herod, who had added impiety to treason, were punished.

But the reproach of being a seditious people was resented by other Jews
than those of Jerusalem. The Jews in Rome were largely descended from
those who had left the country before even Antipater, Herod’s father,
had become powerful there. On them, of course, the house of Herod could
make no claim, and for obvious reasons closer relations with Rome seemed
to them eminently desirable. The Jewish embassy which Varus had
permitted the Judeans to send—how selected and led we have no
information—was joined by an immense deputation from the Roman
synagogues. The substance of their plea was the petition that they be
made an integral part of the province of Syria. “For it will thus become
evident whether they really are a seditious people, generally impatient
of all forms of authority for any length of time” (Jos. Ant. XVII. ii.
2; Wars, II. vi. 2).

This plea, to be joined to Syria, is particularly significant if we
remember that the motive of the Jews in sending the embassy was, in the
words of the Wars (II. vi. 1), to plead for the autonomy of their nation
(cf. Ant. XVII. xi. 1). We see strikingly confirmed the theory of the
Roman provincial system, in which the proconsul or propraetor was only
an official added to, but not superseding, the local authorities.

The representative of Archelaus, Nicolaus of Damascus,[303] charged the
former’s accusers with “rebellion and lust for sedition,” with lack of
that culture which consists in observance of right and law. Nicolaus had
in view primarily the Jewish accusers of his employer, but no doubt made
his remarks general. In the earlier version of the embassy, as it
appears in the Wars (II. vi. 2), it is the whole nation that Nicolaus
charges directly with “a natural lack of submission and loyalty to royal
power.”

Augustus declined to continue the heterogeneous kingdom of Herod. A
brief trial of Archelaus as ethnarch of Judea proper convinced him of
the latter’s worthlessness. The request of the Jewish envoys was now
granted. Judea became a part of Syria—and the agent or _procurator_ of
the Syrian proconsul took up official residence at Caesarea. We find,
however, that this step, which the Jews themselves had suggested, almost
immediately provoked a serious rebellion in Galilee, led by one Judah of
Gamala in Galilee and by a Pharisee named Zadok, who, if we may believe
Josephus, were appreciably different from the various “robbers,”
ληστής, whom he had formerly enumerated, and, in his eyes,
even more detestable than they were. They placed their opposition on the
basis of a principle. This principle was that of the sinfulness of all
mortal government and the consequent rejection of Roman authority as
well. Accordingly they refused to pay tribute. These advocates of a pure
theocracy had of course obvious Scriptural warrant for their position,
but the relatively rapid spread of such a doctrine in the form of an
actual programme of resistance can be accounted for only by the
extremely unsettled state of the country and the still more unsettled
state of men’s minds.

That this Judah formed a fourth sect of the Jews in addition to the
three, Pharisees, Sadducees, and Essenes, already in existence, as
Josephus tells us, may not be quite true.[304] Men of his type are
scarcely founders of sects. But there can be no doubt that the doctrines
which these zealots espoused were those which Josephus has described.
The later history of Europe has abundant examples of such groups of
fanatic warriors maintaining one of many current religious dogmas,
especially in times of economic and political disorder. Of such
incidents the Hussite bands of Ziska and the Anabaptist insurrection are
examples. In this case the distress and uncertainty were largely
spiritual. The economic conditions, while bad, had not become
particularly worse. Indeed, if anything, more direct administration had
somewhat lightened the burdens, by making them less arbitrary and by
removing the heavy expense of a court and the need of footing the bill
for Herod’s building enterprises.

Josephus regards the great rebellion of 68 C.E. as the direct
consequence of this insurrection of Judah. He is therefore very bitter
against this “fourth philosophic system,” which spread among the younger
men and brought the country to ruin. It is at least curious that in his
earlier work, the Wars, in which the recollection of Jewish disaster
would be, one would suppose, vastly more vivid, he does not ascribe to
this rebellion any such far-reaching effects (Wars, II. viii. 1); nor is
it in any degree likely that this insurrection was after all more than
what it appears to be there, a sporadic outburst in that hotbed of
unrest, the Galilean hills, noteworthy only for the special zeal with
which the theocratic principles were announced.

No riots or disturbances are mentioned in Judea till the famous
image-riots of the time of Pontius Pilate. However we may wish to
discount the highly colored narrative preserved in Josephus, there can
be no doubt that these riots did take place. It may even be that the
representation of influential Jews induced the much desired concession
on Pilate’s part of removing the “images.” But what these images were
does not appear with any clearness from Josephus’ account, and of course
we are under no obligation to take literally the “five days and five
nights” during which the ambassadors lay prostrate, with bare necks, at
Pilate’s feet.

Josephus speaks of the “images of Caesar which are called standards”
(Wars, II. ix. 2; Ant. XVIII. iii. 1). The Roman standards, _signa_,
σημαίαι, often contained representations of the emperor.
But these were in the form of medallions in flat relief, hung upon the
standard. They would have been noticed only upon relatively close
inspection. There were also statues in the camp. But it is quite
unlikely that if the Roman provincial administrators were instructed to
issue no coins with the imperial effigy, they would be allowed to carry
into the city actual statues of the emperor. They may well have
forgotten that the military standards would be themselves offensive, if
they bore, as they always did, the representation of animal forms. All
legions at this time carried the eagle, and most of them had other
heraldic animals as well.[305]

Now it may be remembered that the chief legion permanently encamped in
Syria, of which detachments must have accompanied Pilate upon his
transference of the praetorium from Caesarea to Jerusalem, was the Tenth
Legion, called Fretensis (Leg. X Fretensis), and that its standards were
a bull and a pig.[306] To the mass of the Jews the carrying, as though
in triumph, of the gilded image of an unclean animal must have seemed
nothing less than derision, and can easily explain the fury of the
populace.

Another of the Syrian legions, of which certain divisions may have been
with Pilate, was the Third Gallic Legion (Leg. III Gallica). This
legion, like the X Fretensis, bore a bull as a standard, which, while
less stimulating to the mass of the population, must have seemed even
more than the pig the emblem of idolatry to those who had the history of
their people in mind.[307]

If this was the occasion of the disturbance, Pilate may well have been
innocent of any provocative intention. That can scarcely have been
altogether the case in the riots provoked by the aqueducts. Pilate
seized certain sacred funds for that purpose, and in this case no
official, Roman or Greek, could have failed to understand the nature of
the funds or the offense involved in using them for secular purposes.

A certain significance is attached to the Samaritan episode mentioned by
Josephus (Ant. XVIII. iv. 1). It is one of the incidents that become
more and more frequent. The promises of a plausible thaumaturg cause an
enormous throng to gather. It does not appear that he had any other
purpose than that of obtaining credit as a prophet or magician. But
Pilate, as most Roman governors would no doubt have done, held the
unlicensed assemblage of armed men to be sedition, and suppressed it as
such.

Shortly afterwards Palestine and the closely connected Egyptian
communities were thrown into a frenzy of excitement by the widely
advertised attempt of Gaius to set up his statue in the temple at
Jerusalem. The imperial legate at Antioch had no desire whatever to
arouse a rebellion in which all the forces of religious hatred would be
let loose upon him. He therefore temporized and postponed at his own
imminent peril. In view of the constantly threatening attitude of
Parthia, Petronius[308] may well have felt his responsibility with
especial force. Only a few years before, an invasion on the part of the
Parthian king Artabanus had been generally feared. Agrippa had even been
accused of complicity with the Parthians.[309] The governor of Syria had
every reason to hesitate to gratify the caprice of an obviously insane
emperor at so great a risk to the state. Luckily for him, the
assassination of Gaius saved him from the consequences of his
hesitation. His subsequent procedure against the people of Doris[310]
indicated a lively comprehension on his part of the inflammable
character of the people he had to govern and the particular importance
to be attached to this question of images.

To the Roman historian, the incident of Gaius’ attempted erection of his
statue in the temple is only an illustration of the readiness with which
this nation rebelled. Tacitus[311] treats the period between
insurrections as one of smouldering revolt. The incident of Gaius
precipitated an outbreak (Hist. v. 9), which his death calmed, and
enabled the Jews to suppress their inclinations a few years longer.
_Duravit tamen patientia Iudaeis_, he tells us, _usque ad Gessium Florum
procuratorem_, “The submission of the Jews lasted till the
procuratorship of Gessius Florus.”

The short reign of Herod’s popular grandson, Agrippa, “the great king
Agrippa, friend of Caesar and the Romans,” as he calls himself on his
coins and inscriptions,[312] rather confirmed Roman anxiety about the
loyalty of their Jewish subjects than lightened it. It was by a complete
adoption of Jewish customs—an adoption that can hardly have been
sincere—that Agrippa secured and maintained his hold on their
affections.[313] His deference to the religious leaders of the people
was unqualified. His dealing with the Pharisee Simon, who publicly
challenged his right to enter the temple precincts at all, is an
illustration.[314] The Pharisaic tradition of his reign as preserved in
the Talmud is that he was a pious and scrupulously observant Jew,
painfully conscious that his Idumean origin made him half a stranger in
Israel.

But to Rome Agrippa’s methods, in spite of their success, indicated only
that no real progress had been made in the subjugation of Palestine.
Rome was not without experience of lands difficult to subdue. Gaul,
Belgium, Germany, Britain, were all lands where insurrections might at
any time be feared through the devotion of an influential minority to
their ancestral customs. But in Palestine there was even less
appreciable increase in Romanization or Hellenization of customs than in
the countries mentioned. To an antiquary and scholar like the emperor
Claudius there might be something interesting and admirable in the
maintenance of an historic culture, but to the Roman administrative
official, accountable for the security of the East, there was little
that was admirable about it.

A quarrel between the Jews of Peraea and the neighboring city of
Philadelphia may have had only local significance. And the Ptolemy
executed by Fadus may have been only a common highwayman.[315] But a
very little later the success of a certain Theudas, an “impostor,”
γόης τις ἀνήρ, Josephus calls him, in gaining adherents as
a prophet is highly significant.[316] This Theudas undertook to divide
the Jordan, and pass across it with his followers. It is noteworthy that
every such claim to miraculous power immediately elicited drastic action
on the part of the Romans. Theudas’ followers were cut down in a cavalry
raid, and he himself was captured and beheaded. Roman officials
apprehended danger chiefly from this source, and were particularly on
their guard against it.

Such incidents as the riots provoked by individual soldiers cannot have
been frequent. As has been said in one case, the Roman commander
executed a soldier whose outrage had stirred up a revolt. But a garrison
of foreign soldiers in a warlike country furnishes constant incentives
to friction, which may at any time burst out into a general war. In
Samaria and Galilee there were abundant pretexts for mutual attacks, the
net result of which was that the land was full of brigandage, which
indicates that the Roman police here were strikingly ineffective. And in
all cases the suspicion that attached to every armed leader was that his
motives were treasonable as well as criminal. So Dortus of Lydda was
accused by the Samaritans of directly preaching rebellion.

Under Nero, says Josephus, the country went from bad to worse, and was
filled with brigands and impostors.[317] How little it was possible to
distinguish between these two classes appears from the fact that
Josephus continually mentions them in couples. Those whom he calls
Assassins, or Sicarii, can be placed in neither category. One thing is
evident. Their apparently wanton murders must have had other incentives
than pillage, for even Josephus does not charge them with that; they
were obviously animated by a purpose that may be called either
patriotism or fanatic zeal, depending upon one’s bias. That is shown
plainly enough in a casual statement of Josephus that these brigands
were attempting to foment by force a war on Rome, τὸν δῆμον εἰς
τὸν πρὸς Ῥωμαίοις πόλεμον ἠρέθιζον.

The usual “prophet,” in this case an unnamed Egyptian, appears with his
promise to make the walls of Jerusalem fall at his command, and the
usual attack of armed soldiers on a helpless group of unarmed fanatics.
In the Wars, Josephus speaks of a great number of these self-styled
prophets (II. xiii. 4): “Cheats and vagabonds caused rebellion and total
subversion of society, under the pretense of being divinely inspired.
They infected the common people with madness, and led them into the
desert with the promise that God would there show them how to gain
freedom.” The procurator Felix took the customary measures of treating
these expeditions as open sedition and crushing them with all the power
at his command—acts which can only have inflamed the prevailing
disorders.

The picture drawn by Josephus of the Judea of those days represents a
condition nothing short of anarchy. Such a situation could have existed
only under an incompetent Roman governor. Whether the procurator Gessius
Florus was or was not quite the monster he is depicted as being in the
Wars, he can scarcely have been an efficient administrator. It is very
likely that the various acts of cruelty imputed to him by Josephus were
examples of the intemperate violence of a weak man exasperated by his
own failure to control the situation. However this may be, it certainly
was not the excesses of an individual governor that provoked the
rebellion of 68 C.E., even if we accept Josephus’ account of him in
full, and assume him to have been a second and worse Verres. The
outbreak of that year was the result of causes lying far deeper in the
condition of the time and the character of the people.


------------------------------------------------------------------------




                             CHAPTER XVIII

                         THE REVOLT OF 68 C.E.


The Jews were not the only nation that fought with desperate fury
against complete submergence in the floods of Roman dominance. The
spread of the Roman arms had encountered, from the beginning, seemingly
small obstacles that proved more serious checks than the greater ones.
Thus, after the Second Punic War, when Rome was already in the ascendant
in the world, the relatively fresh strength of a conquering people was
all but exhausted in the attempt to subdue and render thoroughly Roman
the mountain tribes of the Ligurians in the northern part of the
peninsula.[318] In later times, after Caesar’s conquest, the subjugation
of Belgium was a weary succession of revolts and massacres and punitive
expeditions that stretched over several generations. Similarly in
Numidia it was found that formal submission of the tribes that filled
this region insured no permanence of control.[319]

In the last cases, however, the danger that was warded off seemed in
Roman eyes to be remote. In the case of Judea the very existence of the
eastern empire was threatened. On the other side of the Syrian desert
there was a watchful and ready enemy, who might appear in force at any
time and with whose arrival there might break out into open
conflagration the smouldering disloyalty that still was present in the
Asiatic provinces.

The Jewish rebellion of 68 C.E. was not an isolated phenomenon. For the
Jews it formed the beginning of a series of insurrections that did not
end till the founding of Aelia Capitolina put a visible seal on the
futility of all such attempts. To us the outcome seems so inevitable
that the heroism of the Zealots has stood for centuries as a striking
example of unrestrained fanaticism. To take a modern instance, if the
single island of Cyprus were to attempt, by its unaided strength, to
cast off the British rule, it would not seem to be engaged in a more
completely forlorn enterprise than were the Jews who undertook to defy
the power of the legions. And yet those who began and conducted the
revolt were neither fools nor madmen, and the hopes that buoyed them
must have been very real when they attempted the impossible.

We must first of all remember that a foreign suzerainty was not
necessarily incompatible with Jewish theocratic ideals. Tradition had
accustomed the Jews to Assyrian and Persian dominance, and their most
sacred recollections contained ample warrant for those who would bear
the rule of Caesar with complete equanimity. But it had been axiomatic
that the rule of a foreign master was a divinely imposed penalty, a
trial, a test of submission. At some time the period of trials would
cease, and the normal condition of complete freedom from outside control
under the sway of God would be restored. The Messianic hope made that
situation more and more vividly present to the hearts of men.

Nor did actual experience of recorded history make this possibility a
vain dream. The vicissitudes of fortune, the sudden rise of obscure
nations to supremacy, and their quick destruction, were rhetorical
commonplaces. The East knew abundant cases of the kind. Empires had
risen and crumbled almost within the recollection of living men. That
was particularly so after Alexander, when sudden glories and eclipses
were too common to be noteworthy.

And we must further reckon with the fact that a potent incentive was the
living faith in an actual God, who could and did hurl the mighty from
their seat. To these men the destruction of Sennacherib or the triumph
of Gideon was no legend, but a real event, which might occur in their
days as in the days of their fathers. The attempt, accordingly, to
secure the independence of a small portion of the empire need not have
seemed to the men that undertook it quite as insensate a proceeding as
it does to us.

Our most complete source for the period is discredited by the _parti
pris_ of the author, the disloyal Josephus. The Roman sources indicate
that in the Jewish revolt there was nothing different from the revolts
in other parts of the world, revolts to which Romans were accustomed.
There was no direct external provocation. There was no one event that
seemed to account adequately for an outburst just then. But we find no
indication that Romans felt it to be a strange or inexplicable fact for
men to rise in order to recover their freedom. The imperial interests
demanded that the hopelessness of such rising should be made apparent.
It was therefore to the leaders of the community, the aristocracy, that
Romans looked to keep in check the ignorant multitude to whom the
superiority of Romans in war or civilization might not at all be
apparent.

The contemptible young rake who, as Agrippa II, continued for some years
the empty title of “king of the Jews,” was no doubt at one with the smug
Josephus in his sincere conviction of the overwhelming might of the
Romans and the folly of attacking it. We cannot sufficiently admire the
successful way in which the king concealed his heartfelt pity for the
sufferings of the Jews, “since he wished to humble the exalted thoughts
they were indulging,” as Josephus naïvely tells us (Wars, II. xvi. 2).
However, not mere truckling to the Romans, but sober conviction, would
sufficiently account for the pro-Roman leanings of men like Agrippa and
Josephus. The long speech put in the king’s mouth (_ibid._ II. xvi. 4)
was perhaps never delivered, but it states the feeling of the pro-Roman
party and of the Romans themselves eminently well.

Both Josephus and Agrippa could hold no other view than that it was some
single act or series of acts of the procurator Florus that animated the
leaders of the revolt. It seemed to them a “small reason” for engaging
in what was conceded even by the most hopeful to be a desperate and
frightful war. The burden of the king’s supposed speech, however, in
which we are justified in seeing the sentiment of the historian, is
this: “Who and what are these Jews that they can refuse to submit to
that nation to which all others have submitted?”[320] We find enumerated
for us the extent and wealth of the Roman possessions with a fervor of
patriotism that might have shamed many a Roman. “Are you richer than the
Gauls, more powerful in body than the Germans, wiser than the Greeks,
more numerous than all the inhabitants of the earth put together?” he
asks, and enforces his question with a detailed account of the enormous
numbers of people who in the several provinces are kept in check by a
handful of legionaries.

As an appeal to common sense, the speech, in spite of its obvious
exaggerations, ought to have been successful. But what the Romans and
the Romanized Jews chose to overlook was that common sense was scarcely
a factor in producing the “exalted opinions” which Agrippa sought to
abase. The glowing assurance of direct divine interposition was of
course lacking to the speaker, and the wilder and more exuberant fancies
that made the present time big with great upheavals and opened vistas of
strange and sweeping changes, could not be answered by a statistical
enumeration of the forces at the disposal of Romans and Jews
respectively.

In the previous chapter one fact has been frequently mentioned which
Josephus states quite casually as an ordinary incident of the events he
is describing. That fact is the readiness with which the Romans took
alarm, not only at the armed “brigands,” who were really at all times in
open revolt, but at anyone who, posing as preacher or prophet, gathered
a crowd about him for thoroughly unwarlike purposes. We do not find
elsewhere in the empire this quickness of animadversion on the part of
the authorities to such acts. The Armenian Peregrinus was quite
unmolested by the Roman officials when he undertook to perform before
the eyes of the assembled crowd the miracle of Hercules on Mount
Oeta.[321] Nor is there any evidence, however large the multitude was
that surrounded the itinerant magician elsewhere, that riot and
subversion were apprehended from that fact. Yet when the Egyptian
promised to divide the walls of Jerusalem (above, p. 285), or Theudas to
pass dry shod over the Jordan, or another man to discover the hidden
treasures on the Gerizim (above, p. 284), a troop was sent at once to
crush with bloody effectiveness an incipient rebellion. Obviously, in
Judea, and not elsewhere, the assertion of divine inspiration carried
with it a claim to certain political rights, or was deemed to do so,
which was incompatible with Roman sovereignty.

It is easy enough to understand what that claim was, and easy enough to
understand why it does not stand forth more clearly in Josephus’
narrative. The coming of the Messianic kingdom had been looked for by
previous generations as well, but in the generation that preceded 68
C.E. it became more and more strongly believed to be immediately at hand
and to demand from those who would share in it a more than passive
reception.

We are not to suppose that every one of these impostors or thaumaturgs
claimed Messianic rank. That it is not expressly stated by Josephus
proves little, since he actively strove to suppress any indication that
there were rebellious incentives among his people other than the brutal
oppressions of Florus. But to claim to be Messiah was a serious matter
both to the people and to the Roman officials, and we assume that these
rather vulgar swindlers hardly dared to go so far. However, whether
individuals did or did not make these pretensions, it is clear that
during the reign of Nero the sense of an impending cataclysm was
growing, and the most fondly held dreams of the Jews, which clustered
about the Messianic idea, seemed to come near to realization.

Besides the cumulative force which the Jewish eschatology and Messianic
hope acquired by the mere tradition from generation to generation, there
was another and more general factor. The constitution established by
Augustus might strive as it would to resemble with only slight
modifications the republican forms it displaced. The East, for its part,
had never been deceived into regarding it otherwise than a monarchy. And
as such it was an unmistakable notch in the course of events. At a
specific moment, whether it was Caesar’s entry into Rome or Augustus’
investiture with the principate, living men had seen and noted a page
turned in the history of the world.

In this new monarchical constitution, the weak point was the succession.
The glamour of acknowledged divinity rested upon Julius Caesar and
Augustus, and in their blood there seemed to be an assurance of title to
the lordship of the world. What would happen if this blood should fail?
No machinery existed that would automatically indicate who the successor
would be. Changes of dynasty, whether regular or violent, were of course
no new thing to the East, but this was not the same. The Roman empire
was unique. The _imperator_, or αὐτοκράτωρ, was as new in
conception as in title. Divinely established, the imperial dignity would
be divinely maintained in those who by their origin could claim an
unbroken chain of divine descent. He whom we know as Nero was on the
monuments “Nero Claudius Caesar, son of the god Claudius and
great-great-grandson of the god Augustus”; and the last was at all times
officially styled _Divi filius_, “son of the God.”[322]

But Nero’s childlessness made it plain that the divine maintenance would
be wanting. With Nero, the line of Augustus would become extinct. For
Rome that presaged confusion and civil war. For the little stretch of
country between the Lebanon and the River of Egypt, it loosed all the
hopes and fears and expectations to which each generation had added a
little, and which were to be realized in the dissolution that was
hurrying on.

Nor must we forget that the reign of Nero had been marked by frequent
rebellions. Armenia had revolted and been subdued. At the other end of
the Roman world, the Britons had risen in a bloody insurrection. And in
the very midst of the Jewish war, the inevitable Gallic rebellion broke
out, ostensibly against Nero personally, but doubtless impelled by
motives of national feeling as well. Perhaps, if we had as detailed a
narrative of the British, Armenian, and Gallic insurrections as we have
of the Jewish, we should find many preliminary conditions the same.
Perhaps in those countries too “brigands” and “impostors” stirred the
people to revolt by playing upon their sacred traditions and appealing
to their hopes of a national restoration.[323]

One very curious circumstance is the association of this last emperor of
the Julian house with the Jews generally and the Messiahship
particularly. How far it is possible to discover the real Nero under the
mass of slanderous gossip and poisonous rhetoric which Suetonius and
Tacitus have heaped upon him, is not easy to determine, nor is it
necessary to do so at this point. One thing may, however, be insisted
upon. He courted and achieved a high degree of popularity. This is
hinted at, not only in the fact noted in Suetonius (Nero, 37), that in a
public prayer he ostentatiously referred only to himself and the people,
and omitted any mention of the senate, but is expressly referred to in
the same writer (_ibid._ 53): _Maxime autem popularitate efferebatur,
omnium aemulus qui quoquo modo animum vulgi moverent_, “Above all, his
chief desire was for popularity, and, to gain this, he imitated all who
in any way had caught the fancy of the mob.” To this may be added the
confirmatory evidence of the lasting veneration felt for his memory by
the populace (_ibid._ 57) and the assumption of his name by Otho when
the latter desired to court popular favor (Suetonius, Otho, 7).[324]

This favor among the masses in the city would of itself indicate a hold
on the Oriental part of his subjects, which Nero’s personal traits make
especially likely. And of these Oriental or half-Oriental Romans a very
considerable fraction were Jews. The all-powerful Poppaea Sabina, Nero’s
mistress and afterwards his wife, is on good grounds believed to have
been a semi-proselyte, a _metuens_.[325] Josephus ascribes Nero’s
interference to her influence when Agrippa II attempted to make a
display of the temple ceremonies. It is also not unlikely that the
change of attitude on the part of Josephus toward Nero was due to the
general feeling of the Roman Jewry toward his memory—a feeling of which
Josephus had no cognizance in writing the Wars, but which had come to
his attention when the Antiquities was composed. In the Wars (IV. ix. 2)
we hear “how he abused his power and intrusted the control of affairs to
unworthy freedmen, those wicked men, Nymphidius and Tigellinus.” In the
Antiquities (XX. viii. 3) we find a temperate paragraph warning readers
that the extant accounts of Nero are thoroughly unreliable, especially
the accounts of those “who have impudently and senselessly lied about
him.”[326]

That among the Roman populace there were some who believed that Nero was
not dead, but still alive, and would return to be avenged upon his foes,
is not strange. But it is particularly strange that in the extreme East
the hereditary rivals of Rome, the Parthians, cherished his memory, so
that their king Vologaesus expressly asked for recognition of that fact
when he strove to renew his alliance with Rome. It was among the
Parthians that the man who claimed to be Nero found enthusiastic support
about 88 C.E. (Suet. Nero, 57). The Parthians seem to have been ready to
invade the Roman empire to re-establish this “Nero” (Tac. Hist. I. ii.
6). That, it is true, happened long afterward; but directly after Nero’s
death, in the very throes of the Jewish war, a similar belief spread
like wildfire over Greece and Asia Minor, and a slave, by calling
himself Nero, secured temporary control of the island of Cythnus (Tac.
Hist. I. ii. 8).

One phrase of Suetonius is especially noteworthy. Long before Nero’s
death it had been prophesied that he would be deposed, and would return
as lord of the East: _Nonnulli_, Suetonius goes on to say, _nominatim
regnum Hierosolymorum_ [_spoponderant_], “Some assured him specifically
that he would be king of Jerusalem.”

There is no direct confirmation in the Jewish sources of this
association of Nero with a restored kingdom at Jerusalem. The very late
Talmudic legend which states that Nero became a convert and was the
ancestor of Rabbi Meïr[327] must, of course, be disregarded. No notable
heathen sovereign escaped conversion in the Jewish legends. To the
Christians, Nero was Belial or Antichrist for reasons obvious enough,
and the Sibylline verses which so represent him are probably of
Christian origin. But since the Messianic idea of the Jews was
well-known throughout the Roman world (Suet. Vespasian, 4), the
prediction made to Nero meant nothing less than that he was the promised
Messiah, a conception startling enough, but perhaps less so to Nero’s
generation than to ours.

It may further be possible to find an association between Nero and the
Jews in the words that Philostratus[328] (Life of Apollonius, v. 33)
puts in the mouth of the Alexandrian Euphrates. The Jews, Euphrates
says, are the enemies of the human race almost as much as Nero, but it
is the latter against whom Vespasian should direct his arms, not the
former.

Whether, however, it was Nero or someone else, the intense force of the
Messianic idea of the time of the revolt is attested explicitly by
Suetonius in the passage alluded to above. _Percrebuerat Oriente toto
vetus et constans opinio esse in fatis ut eo tempore Iudaea profecti
rerum potirentur_, “Throughout the length and breadth of the East there
was current an old and unvarying belief to the effect that it was
decreed by fate that supreme power would fall into the hands of men
coming from Judea.” If to Tacitus the insurrection was merely the
expected outbreak of a turbulent province, repressed with difficulty in
previous generations, and inevitable under all circumstances; if, to
Josephus, the revolt was the foolish attempt of deluded but unfortunate
men, driven mad by the oppressions of officials and led by selfish
rascals, Suetonius, who retailed the gossip of the seven seas, had
clearer insight when he referred the actual outbreak of hostilities to
the general conviction that the result of the war would determine the
fate of the empire. The Law would go out from Zion: _Iudaea profecti
rerum potirentur_.

The war, which resulted in the fall of Jerusalem, was in the eyes of
Josephus (Wars, Preface, § 1) the greatest war in recorded history. The
words he uses are very much like those of Livy when he is about to
describe the Second Punic War (Livy, XXI. i.), where, it must be
admitted, the statement seems somewhat more fitting. The Roman
historians naturally enough do not attach quite the same importance to a
rebellion in a border province, however dangerous or desperate. But no
one regarded it as an insignificant episode in the maintenance of the
imperial frontier. There were many accounts of it, most of them written
“sophistically” (_ibid._ I. i. 1), _i.e._ with a definite purpose that
was quite apart from that of presenting a true version of the facts.
These men, we are told by the author, wrote from hate or for favor. They
desired to flatter the Romans or to vent their spleen on the Jews. The
accurate truth was, of course, to be found only in the austerely
veridical account written by Josephus in Aramaic, and translated by him
into Greek.

It is, accordingly, strange that in the one narrative which we have from
a source independent of Josephus, there should appear details which
suggest that flattery of the Roman conqueror was not wholly absent from
Josephus’ own narrative. In the Roman History of Cassius Dio (known
principally by the Greek form of his name, Dion Cassius), who wrote
about 225 C.E., we find a version of the siege of Jerusalem in which
Titus is something less than a demi-god, and the Jews something
different from the wretched and besotted fanatics Josephus makes of
them. Dio has little sympathy for the Jews in general, and finds their
institutions repellent on the whole, but his account is simpler and
actually more favorable to the Jews than the one presented in the pages
of the Wars.

Such details as the wound received by Titus (Dio, lxvi. 5), which
Josephus omits or modifies (Wars, V. vi. 2), are of minor significance,
although even they indicate the strain Josephus was put to in his
attempt to make Titus move in the midst of dangers like a present
divinity. But there are other matters that Josephus does not mention,
_e.g._ the desertion of Roman soldiers to the Jews in the very midst of
the siege, the awe of the Romans toward the temple, so that they had to
be actually forced to enter upon the forbidden precinct even when the
building was in flames. But especially it is the Asiatic Roman, and not
the Jew, who lays stress upon the heroic pride which the Jews displayed
in the moment of their utmost extremity. “All believed it was not
destruction, but victory, safety, happiness, to die with their temple”
(Dio, lxvi. 6).

That the conquest of the capital seemed no usual triumph is evidenced by
the closing words of Dio (_ibid._ 7) and by the inscription which was
carved on one of the arches erected to Titus. Several such arches were
erected. One on the lower ridges of the Palatine, at the edge of the
forum, contains the famous relief of the triumph of Titus. The other was
in the Circus Maximus, and of this we have only the copy of the
inscription (C. I. L. vi. 944). It runs as follows:[329]

    The Senate and People of Rome have erected this arch to the first of
    their citizens, His Sacred Majesty, Titus Caesar Vespasian, son of
    the God Vespasian, High Priest, invested for the tenth time with
    tribunician power, hailed commander seventeen times, chosen consul
    eight times, Father of his Country, because, led by the guidance,
    wisdom, and divine favor of his father, he subdued the race of the
    Jews, and destroyed their city of Jerusalem, a city which all kings,
    commanders, and nations before him have either attacked in vain, or
    left wholly unassailed.

Dio notes that the title “Judaicus” was not assumed by either Vespasian
or Titus. The inscription just quoted makes it clear that their motive
in doing so was not any desire to minimize the importance of their
victory. Relatively less important triumphs over such people as the
Adiabeni or Carpi resulted in the assumption of the titles of
Adiabenicus or Carpicus. It has been urged with considerable
plausibility[330] that the term “Judaicus” would suggest to the general
public a “convert to Judaism,” and at a moment when the spread of
Judaism was, if anything, greater and more successful than ever, despite
the fall of the temple, that was an impression dangerous to convey,
particularly since Titus was himself under a strong suspicion of Eastern
proclivities (Suet. Titus, 5). As a matter of fact, however, Dio’s
surprise is due to the conditions of his own time, when the emperors
freely assumed these gentile cognomina. So Septimius Severus is
Parthicus, Arabicus, Adiabenicus, Britannicus. In Vespasian’s time that
was distinctly not customary. None of his predecessors assumed these
titles. The name Germanicus, used by Gaius, Claudius, and Nero, is a
hereditary cognomen, and its assumption by Vitellius is due to a desire
on the latter’s part to associate himself with the memory of a name at
all times endeared to the people.

But that the conquest of Judea seemed at the time quite equal to those
which justified the assumption of such honoring titles, may be seen in
the epigram of Martial (ii. 2):

             _Creta dedit magnum, maius dedit Africa nomen
               Scipio quod victor quodque Metellus habet,
             Nobilius domito tribuit Germania Rheno,
               Et puer hoc dignus nomine, Caesar, eras.
             Frater Idumaeos meruit cum patre triumphos,
               Quae datur ex Chattis laurea, tota tua est._

    Crete granted a great name; Africa, a greater; the former to
    Metellus, the latter to Scipio. Even more renowned a title was
    derived from Germany and the conquered Rhine. That title, Caesar,
    your boyhood valor also earned. The Idumean triumph[331] you must
    share with your brother and father. The laurel wreath inscribed with
    the name of the Chatti—that is all your own.

The destruction of the city and temple affected the imaginations of all
men, Jew and non-Jew, very powerfully. A large number of the various
apocryphal books are referred to this period, especially those which are
filled with lamentations over the desolate condition of the former
princess among provinces. But dramatic and affecting as it was, the
destruction of the temple was not at the time the epochal event it seems
to us now. It made only a slight change in the political condition even
of Palestinian Jews, and even in the spiritual condition of the Jews at
large it played seemingly a subordinate part.


------------------------------------------------------------------------




                              CHAPTER XIX

             THE DEVELOPMENT OF THE ROMAN JEWISH COMMUNITY


The Jews in Rome at the time of Cicero formed, we have seen, an
important and numerous class amidst the largely orientalized plebs of
the city. With the other foreigners resident in the city they had a
powerful patron in Caesar, as their grief at his death attested. Under
his successor they found at least an indulgent, if somewhat
contemptuous, toleration, which however was directed not toward them
specially, but toward the other foreigners in the capital as well. And
as we have seen, the religious reformation of Augustus, and his active
disapproval of foreign cults, did not prevent the Jews from spreading
rapidly in all classes of society.

Under Tiberius we hear of a general expulsion of the Jews, as afterward
under Claudius. “Expulsion of Jews” is a term with which later European
history has made us familiar. In the case of such expulsions as the Jews
suffered in England, France, Spain, and Portugal, we know that the term
is literally exact. Practically all Jews were in the instances cited
compelled to leave the country and settle elsewhere. The expulsion
ordered by Tiberius was unquestionably wholly ineffective in practice,
since there were many Jews in Rome shortly after, although we have no
record that the decree was repealed. But it may be questioned whether
even in theory it resembled the expulsions of later times.

The facts are given fully by Suetonius (Tiberius, 36):

    _Externas caerimonias Aegyptios Iudaicosque ritus compescuit,
    coactis qui superstitione ea tenebantur religiosas vestes cum
    instrumento omni comburere. Iudaeorum iuventutem per speciem
    sacramenti in provincias gravioris caeli distribuit: reliquos gentis
    eiusdem vel similia sectantes urbe summovit sub poena perpetuae
    servitutis nisi obtemperassent._

    He checked the spread of foreign rites, particularly the Egyptian
    and Jewish. He compelled those who followed the former superstition
    to burn their ritual vestments and all their religious utensils. The
    younger Jews he transferred to provinces of rigorous climate under
    the pretense of assigning them to military service. All the rest of
    that nation, and all who observed its rites, he ordered out of the
    city under the penalty of being permanently enslaved if they
    disobeyed.

Undoubtedly the same incident is mentioned by Tacitus in the Annals (ii.
85), where we hear that “action was taken about the eradication of
Egyptian and Jewish rites. A senatusconsultum was passed, which
transferred four thousand freedmen of military age who were affected by
this superstition to Sardinia in order to crush brigandage there.... The
rest were to leave Italy unless they abandoned their impious rites
before a certain day.”

Between these two accounts there are discrepancies that cannot be cured
by the simple process of amalgamating the two, as has generally been
done. These divergences will be treated in detail later. For the present
it will be well to compare an independent account, that of Josephus,
with the two.

Josephus (Ant. XVIII. iii. 5) tells us of a Jew, “a thoroughly wicked
man,” who was forced to flee from Judea for some crime, and with three
worthy associates supported himself by swindling in Rome. This man
persuaded Fulvia, a proselyte of high rank, the wife of a certain
Saturninus, to send rich gifts to the temple. The presents so received
were used by the four men for themselves. Upon the complaint of
Saturninus, “Tiberius ordered all the Jews [πᾶν τὸ
Ἰουδαϊκόν] to be driven from Rome. The consuls enrolled four
thousand of them, and sent them to the island of Sardinia. He punished
very many who claimed that their ancestral customs prevented them from
serving.” Apart from the incident which, Josephus says, occasioned the
expulsion, we have a version here which is not quite in accord with the
one either of Tacitus or of Suetonius.

Of these men Josephus is probably the nearest in time to the events he
is describing, but also the most remote in comprehension. Besides the
story just told, Josephus tells another, in which it is a votary of Isis
who is deceived, with the connivance of the priests of the Egyptian
goddess. The two incidents which he relates are placed in juxtaposition
rather than connection by him, but the mere fact that they are told in
this way indicates that a connection did exist in the source, written or
oral, from which he derived them. Josephus does not mention that the
Egyptian worship was attacked as well as the Jewish, and indeed he takes
pains to suggest that the two incidents were not really connected at
all.

From all these statements, and from the reference that Philo makes in
the _Legatio ad Gaium_,[332] there is very little that we can gather
with certainty. This much, however, seems established: an attempt was
made to check the spread both of Judaism and of Isis-worship. In this
attempt a certain number of Jews were expelled from the city or from
Italy. Four thousand soldiers—actual or reputed Jews—were transferred to
Sardinia for the same reason. There are certain difficulties, however,
in the way of supposing that it really was a general expulsion of all
Jews, as Josephus and Suetonius, but not Tacitus, say.

Tacitus’ omission to state it, if such a general expulsion took place,
is itself a difficulty; but like every _argumentum ex silentio_, it
scarcely permits a valid inference. It seems strange, to be sure, that a
severe and deserved punishment of the _taeterrima gens_, “that
disgusting race,” should be represented to be something much milder than
really was the case. But Tacitus is neither here nor in other places
taking pains to cite the decree accurately, and the omission of even a
significant detail may be laid to inadvertence.

But what Tacitus does say cannot be lightly passed over. Four thousand
men, _libertini generis_, “of the freedmen class,” were transferred to
Sardinia for military service. All these four thousand were _ea
superstitione infecti_, “tainted with this superstition.” Now, the Jews
who formed the community at Rome in the time of Cicero may have been
largely freedmen, but their descendants were not classed as _libertini
generis_. The phrase is not used in Latin of those who were of servile
origin, but solely of those who were themselves emancipated slaves.
There is, however, scarcely a possibility that there could have been at
Rome in 19 C.E. so large a body of Jewish freed slaves of military age.
There had been no war in recent times from which these slaves could have
been derived. We may assume therefore that most, if not all, of these
men were freedmen of other nationalities who were converts to Judaism.

This is confirmed by the words _ea superstitione infecti_, “tainted with
this superstition.” These words are meaningless unless they refer to
non-Jewish proselytes.[333] Men who were born Jews could not be so
characterized. If Tacitus had meant those who were Jews by birth, it is
scarcely conceivable that he would have used a phrase that would suggest
just the opposite. The words, further, imply that many of these four
thousand were rather suspected of Jewish leanings than definitely
proselytes. Perhaps they were residents of the districts largely
inhabited by Jews, notably the Transtiberine region.

Again, to suppose that all the Jews were banished by Tiberius involves
an assumption as to that emperor’s methods wholly at variance with what
we know of him. A very large number of Jewish residents in Rome were
Roman citizens (Philo, 569 M), and so far from being a meaningless
distinction in the early empire, that term through the influence of the
rising science of jurisprudence was, in fact, just beginning to have its
meaning and implications defined. A wholesale expulsion of Roman
citizens by either an administrative act or a senatusconsultum is
unthinkable under Tiberius. Exile, in the form of relegation or
expulsion, was a well-known penalty for crime after due trial and
conviction, which in every instance would have to be individual. Even in
the Tacitean caricature[334] we find evidence of the strict legality
with which Tiberius acted on all occasions. No senatusconsultum could
have decreed a general banishment for all Jews, whether Roman citizens
or not, without contravening the fundamental principles of the Roman
law.

How thoroughly confused the transmission of this incident had become in
the accounts we possess, is indicated in the final sentence from
Suetonius: “He ordered them out of the city, under the penalty of being
permanently enslaved if they disobeyed.” The very term _perpetua
servitus_, as though there were a limited slavery in Rome at the time,
is an absurdity. It becomes still more so when we recall that slavery,
except in the later form of compulsory service in the mines and galleys,
was not known as a penalty at Roman law. The state had no machinery for
turning a freeman into a slave, except by his own will, and then it did
so reluctantly. We shall be able to see what lies behind this confusion
when we have considered one or two other matters.

The alleged expulsion is not mentioned by Philo in the extant fragments.
The allusion to some oppressive acts of Sejanus (In Flaccum, § 1. ii. p.
517 M; and Leg. ad Gaium, § 24. ii. p. 569 M) is not clear. But it is
difficult to understand the highly eulogistic references to Tiberius,
then long dead, if a general Jewish expulsion had been ordered by that
emperor.

That the senatusconsultum in question was general, and was directed
indiscriminately at all foreign religions, appears not merely from the
direct statement of Suetonius and Tacitus, and the association of the
two stories by Josephus, but also from a reference of Seneca. In his
philosophic essays, written in the form of letters to his friend
Lucilius (108, 22), he says: “I began [under the teaching of Sotion] to
abstain from animal food.... You ask me when I ceased to abstain. My
youth was passed during the first years of Tiberius Caesar’s rule. At
that time foreign rites were expelled; but one of the proofs of
adherence to such a superstition was held to be the abstinence from the
flesh of certain animals. At the request of my father, who did not fear
malicious prosecution, but hated philosophy, I returned to my former
habits.”

The words of Seneca, _sacra movebantur_, suggest the τῶν ἐν Ἰταλίᾳ
παρακινηθέντων of Philo (_loc. cit._), “when there was a general
agitation [against the Jews?] in Italy.” It is further noticeable that
the _mathematici_, _i.e._ the soothsayers, against whom the Roman laws
were at all times severe, were also included in this decree.[335]


[Illustration: SYMBOLS AND INSCRIPTIONS FROM JEWISH CATACOMBS AND
CEMETERIES IN ROME
(From Garrucci)]


It has been pointed out before (above, p. 242) that the observance of
foreign religious rites was never forbidden as such by Roman laws. From
the first of the instances, the Bacchanalian persecution of 186 B.C.E.,
it was always some definite crime, immorality or imposture, that was
attacked and of which the rites mentioned were alleged to be the
instruments. The “expulsion” of the Isis-worshipers during the republic
meant only that certain foreigners were summarily ordered to leave the
city, something that the Lex Junia Penni in 83 B.C.E. and the Lex Papia
of 65 B.C.E. attempted to enforce, and which the Roman police might do
at any time when they thought the public interest demanded it. Roman
citizens practising these rites could never be proceeded against, unless
they were guilty of one of the crimes these foreign practices were
assumed to involve.

The two stories cited by Josephus, one concerning an Isis-worshiper, the
other a Jew, may not be true. Whether true or not, the incidents they
record surely did not of themselves cause the expulsion of either group.
But these are fair samples of the stories that were probably told and
believed in Rome, and similar incidents no doubt did occur. The
association of the _mathematici_ with the other two makes it probable
that the senatusconsultum was directed against fraud, the getting of
money under false pretenses, and that the Jewish, Isiac, and other
rites, as well as astrology, were mentioned solely as types of devices
to that end.

What actually happened was no doubt that in Rome and in Italy
overzealous officials undertook to treat the observance of foreign rites
as conclusive or at least presumptive evidence of guilt under this act.
Perhaps, as Philo says, it was one of the instances of Sejanus’ tyranny
to do so. But there is no reason to doubt Philo’s express testimony that
Tiberius promptly checked this excess of zeal and enforced the decree as
it was intended (_loc. cit._): ὡς οὐκ ἐπὶ πάντας προβάσης τῆς
ἐπεξελεύσεως, ἀλλ’ ἐπὶ μόνους τοὺς αἰτίους—ὀλίγοι δὲ ῆσαν—κινῆσαι δὲ
μηδὲν ἐξ ἔθους; _i.e._ “since the prosecution was not directed
against all, but only against the guilty, who were very few. Otherwise
there was to be no departure from the customary attitude.”

The transference of the four thousand recruits, _libertini generis_, to
Sardinia undoubtedly took place, and was very likely the expression of
alarm on the part of Sejanus or Tiberius at the spread of Judaism in
Rome. It may well be that the removal of these men was caused rather by
the desire to withdraw them from the range of proselytism than by the
purpose of allowing them to die in the severe climate of Sardinia. There
is as a matter of fact no evidence that Sardinia had a noticeably
different climate from that of Italy. It was one of the granaries of the
empire.[336]

Perhaps we may reconstitute the decree as follows: The penalty imposed
was, for foreigners, expulsion; for Roman citizens, perhaps exile; for
freedmen, forfeiture of their newly acquired liberty in favor of their
former masters or the latter’s heirs. This last fact will explain the
statement of Suetonius. Many of the people affected were no doubt
freedmen, and several instances where such a penalty was actually
inflicted would account quite adequately for the words _perpetua
servitus_ of Suetonius. The “malicious prosecution,” _calumnia_, which
Seneca asserts his father did _not_ fear, would be based, as against
Roman citizens, on the violation of this law against fraudulent
practices, of which, as we have seen, the adoption of foreign rites
would be taken as evidence.

The personal relations between Gaius and the Jewish king Agrippa seemed
to guarantee an era of especial prosperity for the Roman Jews. However,
the entire principate of that indubitable paranoiac was filled with the
agitation that attended his attempt to set up his statue at Jerusalem.
His death, which Josephus describes in gratifyingly minute detail,
brought permanent relief on that point.

It is during the reign of his successor Claudius that we hear of another
expulsion: _Iudaeos impulsore Chresto adsidue tumultuantis Roma expulit_
(Suet. Claud. 25), “The Jews who engaged in constant riots by the
machinations of a certain Chrestus, he expelled from Rome.” It has
constantly been stated that this refers to the agitation in the Roman
Jewry which the preaching of Christianity aroused. For that, however,
there is no sufficient evidence. Jesus, to be sure, is called Chrestus,
Χρηστός, the Upright, in many Christian documents.[337]
This play upon words is practically unavoidable. But Chrestus is a
common name among all classes of society.[338] Jews would be especially
likely to bear it, since it was a fairly good rendering of such a
frequently occurring name as Zadok. The riot in question was no doubt a
real enough event, and the expulsion equally real, even if it did not
quite imply all that seems to be contained in it.

If it were a decree of general expulsion of all Jews, it would be
strikingly at variance with the edicts in favor of the Jews which
Claudius issued, and which are contained in Josephus (Ant. XIX. v.). As
in the case of other documents cited here, there is no reason to
question the substantial accuracy of their contents, although they are
surely not verbatim transcriptions from the records. It is as clearly
impossible in the case of Claudius as in that of Tiberius to suppose an
arbitrary disregard of law on his part, so that a general ejection of
all Jews from the city, including those who were Roman citizens, is not
to be thought of.

Neither Tacitus nor Josephus mentions the expulsion. The silence of
neither is conclusive, but it lends strong probability to the
assumption that the decree cannot have been so radical a measure as a
general expulsion of all Jews from the city would be. The passage from
Suetonius is concerned wholly with acts of Claudius affecting
foreigners—non-Romans, _i.e._ Lycians, Rhodians, Gauls, Germans—and if
we keep in mind Suetonius’ habits of composition, it is highly likely
that he has put together here all that he found together in his
source. We are to understand therefore by the _Iudaei_ of this passage
only foreign Jews, which implies that the majority of the Jews were
not affected by it at all.

But were even all foreign Jews included? Is there anything in the
passage that is not perfectly consistent with the assumption that some
relatively small group of Jews led by a certain Chrestus was ejected
from the city for disorderly conduct? The silence of the other writers,
the total absence of effect on the growth of the Jewish population,
would seem to make this after all the simplest meaning of Suetonius’
words.

The fact of the expulsion is confirmed by that passage in the Acts of
the Apostles in which the meeting of Paul and Aquila at Corinth is
mentioned (Acts xviii. 1, 2): “[Paul] found a certain Jew born in
Pontus, lately come from Italy with his wife Priscilla, (because that
Claudius had commanded all Jews to depart from Rome).” The testimony is
late,[339] but it will be noticed that Aquila is an Asiatic by birth,
and so very likely had no legal right of residence at Rome in any
circumstances.

Finally, expulsion “from Rome” may have meant only exclusion from the
_pomoerium_, the sacral limit of the city that followed an imaginary
line not at all coincident with its real walls. To escape from the
operation of the decree, it would merely have been necessary to cross
the Tiber, where as a matter of fact the Jews generally lived, since the
Transtiberine region was not included in the _pomoerium_. In general,
expulsion from the city specified that the expelled person might not
come within the first milestone, but in view of the difficulties
presented by the assumption of a real expulsion, this supposition may
also be considered.

Mention has already been made of the special association of Claudius’
successor, Nero, with the Jews. The success that attended their efforts
at propaganda during that emperor’s reign is evidenced by the fact that
Poppaea Sabina became a semi-proselyte. And during Nero’s reign occurs
an event of special importance to the Jews of Rome, the first Christian
persecution.

In the reign of Nero, possibly in that of Claudius, there was brought to
the various Jewish congregations of the Roman world, seemingly not
beyond that, the “good news,” εὐαγγέλιον, that a certain
Jesus, of Nazareth in Galilee, was the long-promised Messiah. To most,
perhaps, the facts cited of his life indicated only that he was one of
the “many swindlers,” γόητες ἄνθρωποι, like those whom
Felix captured and put to death (Jos. Ant. XX. viii. 5). But some
believed. If we are to credit the Acts of the Apostles, this belief at
once produced a bitter conflict between those who did so believe,
afterwards called Christians, and those who did not.[340] But the Acts
in the form in which it has come down to us represents a recension of
much later date, made when the enmity between Jew and Christian was real
and indubitable.

It may be that in certain places those Jews who accepted the evangel
almost at once formed congregations of their own, synagogues or
ecclesiae (the terms are practically synonymous),[341] different from
the synagogues of those who rejected it. But there were from the
beginning differences of degree in its acceptance, and even in the
existing recension of the Acts there is good evidence that its
acceptance or rejection did not immediately and everywhere produce a
schism.

In the city of Rome a persecution of Christians, as distinct from Jews,
took place under Nero. That fact is attested by both Suetonius and
Tacitus and by the earliest of the Christian writers. Tertullian quotes
the _commentarii_, the official records, for it.

The record as it appears in Suetonius is characteristically different
from that in Tacitus. In Suetonius we have a brief statement (Nero, 16):
_Afflicti suppliciis Christiani, genus hominum superstitionis novae ac
maleficae_, “Punishment was inflicted upon the Christians, a class of
men that maintained a new and harmful form of superstition.” This
statement is made as one item, apparently of minor importance, in the
list of Nero’s creditable actions, as Suetonius tells us later (_ibid._
19): _Haec partim nulla reprehensione, partim etiam non mediocri laude
digna, in unum contuli_, “These acts, some of which are wholly
blameless, while others deserve even considerable approbation, I have
gathered together.” Whether the punishment of the Christians is in the
former or the latter class does not appear.

In Tacitus, on the other hand, we have the famous account that Nero
sought to divert from himself the suspicion of having set Rome on fire,
by fastening it upon those “whom the people hated for their wickedness,
the so-called Christians” (Ann. xv. 44). These were torn by dogs, or
crucified, or tied to stakes and burned in a coat of pitch to serve as
lanterns to the bestially cruel emperor. The truth of these stories
depends upon the reliability of Tacitus in general. They have been
received with justifiable doubt, ever since the quite conscienceless
methods of Tacitus’ rhetorical style have been made evident. The last
form of punishment, the _tunica molesta_, has made a particular
impression on the ancient and modern world. It is referred to by Seneca,
Juvenal, and Martial, but by none of them associated with the
Christians. From the passage in Seneca (Epist. ad Lucil. xiv. 4) it is
simply a standard form of cruelty, such as the rack, thumbscrew, and
maiden of later times. The very fact that the courtier Seneca dares to
mention it as a form of _saevitia_ would indicate that it was not used
by Seneca’s master, Nero. But what is particularly striking is that
Tertullian[342] in his Apology does not mention any cruelties, in the
sense of savage tortures, inflicted upon the Christians. The context
(Apologeticus, § 5) indicates that the punishment was banishment to some
penal colony, _relegatio_, a punishment considered capital at law, but
still different from the _tunica molesta_.

But a new element was introduced in the case of the Christians, which,
except in the treatment of the Druidic brotherhoods among the Gauls, is
unusual in Roman methods. It is scarcely possible to read the Apology of
Tertullian without being convinced that the profession of Christianity
was in and for itself an indictable offense at Roman law since the time
of Nero, quite apart from the fantastic crimes of which the Christians
were held to be guilty.[343] Tertullian undoubtedly had legal training,
and his exposition of the logical absurdities into which the fact led
Roman officials is convincing enough, but the fact remains. The _nomen
Christianum_, “the profession of Christianity,” was considered a form of
_maiestas_, “treason,” and punished capitally. In effect this was an
attempt to stamp out a religion, just as Claudius had sought to stamp
out the Druids (Suetonius, Claud. 25). (Comp. above, p. 142.)

When Tertullian wrote, perhaps even in the time of Tacitus and
Suetonius, the gulf between Jew and Christian was wide and impassable.
It can hardly have been so in Nero’s time. The statement that Nero’s
measures were instigated by Jews is a later invention for which there is
simply no evidence whatever.[344] The fact that the _nomen Christianum_
was either actually considered treason or partook of the nature of
treason, makes it probable that the Messianic idea, which was the very
essence of the evangel, was the basis of the Roman statute. In Judea the
special and drastic crushing of every “impostor” has been spoken of, and
its significance indicated (above, p. 292). The preaching of
Christianity in Rome itself could only have seemed to Nero, or his
advisers, an attempt at propagating, under the guise of religion, what
had long been considered in the East simple sedition. While therefore
the spread of Judaism, Isis-worship, Mithraism, was offensive, and
attempts were made to check it, the spread of Christianity was an
increase in crime and was treated as such. Perhaps a partial analogy may
be offered in the attitude of conservative Americans to doctrines they
regard as mischievous, like Socialism, and to those which are directly
criminal, like some forms of organized Anarchism.

The elaborate scheme of salvation prepared by the Cilician Jew Paul[345]
gradually gained almost general acceptance among Christians, although in
the mother ecclesia at Jerusalem it found determined and obstinate
resistance long after Paul’s death.[346] The fundamental doctrine, that
the Law was not necessarily the way of salvation for any but born Jews,
and even for them was of doubtful efficacy, was the direct negation of
the Pharisaic doctrine that through the Law there was effected immediate
communion of man with God in this world and the next.

As long as the Christians were merely a heretical Jewish sect, their
fortunes affected the whole Jewish community. When their propaganda
became, not a supplement to that of the Jews, but its rival, and soon
its successful and triumphant rival, its history is wholly separated,
and the measures that dealt with the Christians and those that concerned
the Jews were no longer in danger of being confused. To the Jews the
success of the propaganda of Paul seemed to depend on the fact that he
had abolished the long and severe ritual of initiation; he had increased
his numbers by decreasing the cost of admission. So we find, shortly
after the destruction of the temple, R. Nehemiah ben ha-Kannah asserting
(Ab. iii. 6) that to discard the yoke of the Law was to assume the yoke
of the kingdom and of the world; _i.e._ so far from making the path to
unworldliness easier, it laid insuperable obstacles in the way. The
statement is applicable to Jews of lax observance, but it seems
particularly applicable to the Pauline Christians, who had not merely
lightened the load, but deliberately and _ex professo_ wholly discarded
it.

Outside of the references that give us certain data about the external
history of the Roman Jewish community of the first century, we have
other data of a wholly different sort, data that allow of a more
intimate glimpse into its actual life. They are furnished us by the
Roman satirists, whose literary labors have scarcely an analogue in our
days. Satire itself was assumed to be a Roman genre.[347] Whether or not
it was of Roman invention, the miscellanies that have given us so many
and such vivid pictures of ancient life are known to us wholly in Latin.
It is safe to say that if satirists such as Horace, Persius, Juvenal,
and Martial had not come down to us, ancient history would be a vastly
bleaker province than it is.

Of Horace and his representation of Jewish life we have already spoken.
It will be remembered that the one aspect which earned for the Jews his
none too respectful raillery was their eager proselytism. And it is
excellent evidence of how important this proselytism was in the Jewish
life of the time, that in the two generations that stretched from Nero
to Nerva the same aspect is present to men of such diverse types as
Persius and Juvenal.

With Persius we enter a wholly different stratum of society from that of
Horace and, as we shall later see, of Juvenal. Persius was by birth and
breeding an aristocrat. He was descended from an ancient Etruscan house,
and could boast, accordingly, of a nobility of lineage compared with
which the Roman Valerii and Caecilii were the veriest mushrooms.[348]
But he was almost wholly devoid of the vices that often mark his class.
An austere Stoic, his short life was dedicated to the severe discipline
that his contemporary and fellow-Stoic Seneca found it easier to preach
than to practise.

Persius wrote little, and that little has all come down to us. His
Latin, however, is so crabbed and difficult that he is easily the least
read of Roman poets.[349] His productions are called Satires. They are
less that than homilies, in which, of course, the virtues he inculcates
are best illustrated by the vices he attacks.

One of these vices is superstition. The mental condition that is
terrified by vain and monstrous imaginings of ignorant men is set forth
in the Fifth Satire:[350]

    But when the day of Herod comes and the lamps on the grimy sills,
    garlanded with violets, disgorge their unctuous smoke-clouds; when
    the tail of a tunny-fish fills its red dish and the white jar bursts
    with wine, you move your lips in silent dread and turn pale at the
    Sabbath of the circumcised.

As a picture of Jewish life on the eve of the Sabbath, this passage is
invaluable. We can readily imagine how the activities of a squalid
suburb inhabited by a brawling class of men, mostly of Oriental descent,
must have impressed both the grandee and the Stoic.

But the passage is cited here, not merely as a genre-picture, but more
especially because it is again the phase of Jewish life, so often
neglected in histories, that has brought the Jews to Persius’ attention.
The ordinary Roman, not saved from carnal weakness by Stoicism, is found
to stand in particular dread of the strange and nameless God of the
Jews, to whom he brings a reverence and awe that ought legitimately to
be directed only to the gods of his ancestors.

Persius wrote while the temple was still standing. In 70 the temple was
destroyed. A gaping mob saw the utensils of the inner shrine carried in
triumph through the city, and could feast its eyes, if it chose, on the
admirable portrayal of that procession, on the Arch of Titus near the
Forum. It might be supposed that the God who in Roman eyes could not
save His habitation from the flames, could hope for no adherents among
His conquerors. But after the destruction of the temple, in the lifetime
of the very men who cheered Titus when he returned from Palestine, we
see the propaganda more vigorous, if anything, than before.

It is in the pages of Juvenal that we find evidence of that fact, and
here again we are confronted with a sharply outlined personality.
Decimus Junius Juvenalis was born near Aquinum in Southern Italy, where
the Italic stock had probably suffered less admixture with foreign
elements than was the case at Rome. What his intellectual training was
we can only conjecture from its results, the turgid but sonorous and
often brilliant eloquence of his Satires. Whether they are true pictures
of Roman life and society or not may be doubted. But they indubitably
reflect his own soul. We see there a soured _raté_, a man embittered by
his failure to receive the rewards due to his merits. In the capital of
the world, the city where he, the man of undoubted Roman stock, should
have found a career open before him, he discovered himself to be a
stranger. He was no match for the nimble-witted Greeks that thronged
every profession and crawled into entrances too low to admit the scion
of Cincinnatus and Fabricius. How much of this was the venom of defeated
ambition, and how much was honest indignation at the indescribable
meanness of the lives he depicted, we cannot now determine.

Throughout all his work one note may be heard, the note of rage at a
Rome where everything characteristically Roman was pushed into the
background, a Rome in the hands of Greeks, Egyptians, and Jews. And in
the case of the last it is particularly the danger noted by Strabo and
Seneca,[351] of an actual conquest of Rome by the Jewish faith, that
rouses his savage indignation.

The lines in which he states his feeling are well-known (Juvenal, Sat.
xiv. 96 seq.):

    Some whose lot it is to have a father that reveres the Sabbath,
    worship nothing but the clouds and the sky and think that the flesh
    of swine from which their father abstained is closely related to
    that of man. Soon they become circumcised. Trained to despise the
    laws of Rome they learn, maintain, and revere the Law of the Jews,
    which Moyses has transmitted in a mystic volume;—laws that forbid
    them to show the way to any but members of their cult, and bid them
    guide to a spring none but their circumcised brethren.

We need be at no pains to correct Juvenal’s estimate of Jewish beliefs
or Jewish theology. As in the case of Persius, the interest of the
passage lies in the fact that it gives additional testimony to the
success with which the Jewish synagogues, despite official frowns and
even repressive measures, despite the severe conditions they imposed
upon initiates, were constantly gaining in membership.

Juvenal’s other references to the Jews[352] show us certain unlovely
aspects of their life. The hawkers and fortune-tellers whom he describes
are certainly not the best representatives of the Roman community. It is
no part of his purpose to give a complete picture of the community. But
it is his purpose to denounce the degeneration which made the imperial
city a disagreeable place for real Romans to sojourn in, and the Jewish
peddler at the Grove of Egeria and the swindling hags who sell potent
spells for cash give him the colors he requires.

One other writer must be mentioned, Martial. With him we are in the very
heart of Grub Street. Marcus Valerius Martialis came from Spain to the
capital. He had evidently no definite expectation of any career beyond
that of a man of letters, and such a career involved at that time (as it
continued to do until the nineteenth century) something of the life of a
parasite. He had at least some of the characteristics of a parasite—a
ready tongue, a strong stomach, and an easy conscience. But within his
own field of poetry, the epigram, he was a real master. Subsequent
centuries have rarely equaled the mordancy of his wit or the sting of
his lampoon. At the foot of the banquet tables, jostled by hungry
mountebanks and the very dregs of Roman society, he kept his mocking
eyes open to the foibles of his host no less than to the disgustingly
frank vices of his fellows.

And Martial meets Jews on his way through the teeming city. But if
Horace, Persius, and Juvenal have their eyes upon Romans that were being
Judaized, Martial presents to us the counterpart, Jews that actually
were, or sought to be, as Greek or Roman as possible. In speech it is
likely that most Roman Jews (and Roman Christians as well) were
Greek.[353] But Greek was almost as well understood at Rome as Latin,
and perhaps even better understood among the masses. Two of his Epigrams
(vii. 30, and xi. 94) make it clear enough that the Jew at Rome did not
live aloof from his fellow-citizens, and wealthy Jews did not scruple to
purchase in the market the gratifications they were especially enjoined
by their faith to forego. We can readily believe that Martial is
recounting real experiences, but these cases must have been exceptional.
As we shall see later, the Jewish community was certainly not a
licentious one. That point appears specifically from the controversial
literature. But it is equally well to remember that as individuals they
were subject to human passions, and the excesses found in other classes
of society might also be met with among them.

Grecized in speech and name, and no doubt in dress, the Jews accepted
for their conduct the external forms and standards about them. One very
interesting indication of the completeness with which they identified
themselves with the city in which they lived is the expression
“fatherland” that they used of it; _e.g._ in Akmonia (Ramsay, Cities and
Bishops of Phrygia, no. 561). Again, in Ostia a large and well-carved
slab was recently found in which a decree of the Jews at Ostia was set
forth. The corporation grants to its gerusiarch, Gaius Julius Justus, a
place for a sepulchre. The officers are Livius, Dionysius, Antonius, and
another man whose name is lost (Not. Scav. 1907, p. 479). Surely but for
the unambiguous statement of the inscription itself one would not have
looked for Jews in this assemblage of Julii, Livii, and Antonii.


------------------------------------------------------------------------




                               CHAPTER XX

                     THE FINAL REVOLTS OF THE JEWS


In the generations that followed the fall of the temple, changes of
great moment took place, which we can only partially follow from the
sources at our disposal.

The Mishnah gives in considerable detail the laws that governed the life
of the Jew at this period, and also those that regulated the intercourse
of Jew and non-Jew. But the Mishnah may after all have been the
expression of an ideal as often as it was the record of real
occurrences, and the range of its influence during the time of its
compilation may have been more limited than its necessarily general
phraseology indicates. The Mishnah of Rabbi Judah became the standard
text-book in the Jewish academies of Palestine and Babylonia, although
not to the total exclusion of other sources of Halakah. That did not
occur at once; and even when it was complete, the authority of the
presidents of the schools over the Jews resident throughout the world is
more or less problematic.

For that reason it is especially necessary to note the invaluable
records of actual life that appear in the papyri and inscriptions,
especially where they show that the intercourse between Jews and pagans
was far from being as precisely limited as the Mishnah would compel us
to suppose, and men who are at no pains to conceal their Jewish origin
permitted themselves certain indulgences that would certainly not have
met with the approval of the doctors at Jamnia and Tiberias.

The tractate of the Mishnah which is called Aboda Zara, “Idolatry” or
“Foreign Worship,” lays down the rules under which Jew and heathen may
transact such business as common citizenship or residence made
inevitable. The essential point throughout is that the Jew must not
either directly or indirectly take part, or seem to take part, in the
worship accorded the Abomination. Nor are the seemingly trivial
regulations despicable for their anxious minuteness. In all probability
they are decisions of actual cases, and derive their precision from that
fact.[354]

Certain passages in Aboda Zara (ii. 1) would unquestionably have made
intercourse between Jew and pagan practically impossible except in
public or semi-public places. But in the very same treatise it is
implied that a pagan might be a guest at the Jew’s table (v. 5); and
indeed much of the detail of the entire tractate would be unnecessary if
the provision contained in ii. 1 were literally followed out.

The Epigrams of Martial (above, p. 326), if we believe them, indicate
that so far from fleeing the society of pagans for its sexual vices,
some Jews at least sought it for the sake of these vices, as was the
case with the rival of the Syrian Greek Meleager, more than two
centuries before Martial. But it will be noticed that the subject of the
last Epigram (xi. 94) is a renegade, who swears strange oaths, and is
taunted by Martial with what he is obviously trying to conceal. Besides,
as to the particular vice there mentioned, it rests on the malice of the
satirist alone. The victim of his wit denies his guilt.

Indeed it is just this particular vice, so widely prevalent in the Greek
and Roman world, that the Jewish antagonists of the pagans seized upon
at all times. It unquestionably characterized continental Greece and
Italy much more than the eastern portions of the empire. For the Jews it
seemed to justify the application of the words “Sodom and Gomorrah,”
particularly to the general city life of the Greeks. Some Jew or
Christian scratched those names on a house wall of Pompeii.[355]

It is quite untrue to say that unnatural sexual excesses were so
prevalent as to pass without comment among Greeks and Romans generally.
However large they loom in the writings of extant poets, we may remember
that poets are emotionally privileged people. The sober Roman and Greek
did not find any legal or moral offense in illicit love, but unnatural
lust was generally offensive from both points of view, and, however
widely practised, it was at no time countenanced. Still, Jews and
Christians would be justified in comparing their own unmistakable and
specific condemnations in this matter with the mere disapproval with
which decent heathens regarded it. For the Greek legend that made the
fate of Laius, father of Oedipus,[356] a punishment of his crime in
first bringing pederasty into the world, the Jews had the much more
drastic punishment of Sodom; and, in many passages of the Apocrypha, the
fact of this vice’s prevalence is dwelt upon as a characteristic
difference between Jewish and gentile life.[357]

In many other matters there are evidences that not all the regulations
of Aboda Zara were carried out by all Jews. In the Tosefta[358] we meet
the express prohibition of theatrical representations to the Jews, a
prohibition which, in view of the fact that dramatic performances were
at all times theoretically and actually festivals in honor of Dionysus,
seems perfectly natural. But in spite of that, in the great theater at
Miletus, some extremely desirable seats in the very front rows are
inscribed τόπος τῶν Εἰουδαίων φιλοσεβάστων, “Reserved for
His Imperial Majesty’s most loyal Jews.”[359] It will therefore not be
safe to assume that the Halakic provision which forbade Jews to attend
the theater actually meant that Jews as a class did not do so.

But we find even stronger evidences of the fact that the amenities of
social life in Greek cities seemed to some Jews to override the
decisions of the law schools in Palestine. In Asia Minor a Jew leaves
money not merely for the usual purposes of maintaining his monument, but
also for the astounding purpose of actually assisting a heathen
ceremonial.[360] The instance is a late one, but perhaps more valuable
for that reason, because the spread of the schools’ influence increased
constantly during the third century.

At the fall of the temple the voluntary tax of the shekel or didrachm,
which had formerly been paid to the temple at Jerusalem, and which was a
vital factor in the very first instances of conflict between the Jews
and the Roman authorities (comp. above, p. 226), was converted into an
official tax for the support of the central sanctuary of the Roman state
on the Capitoline Hill. Whether Roman citizens who were Jews paid it,
does not appear. All others however did. The bureau that enforced it was
known as the _fiscus Iudaicus_, the word _fiscus_ indicating here, as
always, that the sums so collected were considered as belonging to the
treasury of the reigning prince during the time of his reign, rather
than to the public treasury.

It does not seem that this tax, except for its destination, was believed
by the Jews to be an act of notable oppression, nor was its enforcement
more inquisitorial than that of other taxes; but it became an especial
weapon of blackmail in Rome and in all Italy, and this blackmail grew
into dimensions so formidable that action had to be taken to suppress
it.

In Rome, we may remember, there was no officer at all resembling our
public prosecutor or district-attorney. The prosecution of criminals was
an individual task, whether of the person aggrieved or of a citizen
acting from patriotic motives. Indeed it had at one time been considered
a duty of the highest insistence, and innumerable Romans had won their
first distinction in this way. The delators of the early empire were in
theory no different, though the reward of their activity was not the
glory or popularity achieved, but the substantial one of a lump sum, or
a share in the fine imposed, a practice still in vogue in our own
jurisdictions. Plainly, under such circumstances, there were temptations
to a form of blackmail which the Greeks knew as συκοφαντία,
and the Romans as _calumnia_; _i.e._ the bringing of suits known to be
unjustified, or with reckless disregard of their justification, for the
purpose of sharing in some reward for doing this quasi-public service.
Private prosecutors at Roman law were required to swear that they were
not proceeding _calumniae causa_, “with blackmailing intent.”[361]

The opportunities presented to delators by the _fiscus Iudaicus_
consisted in the fact that anyone of Jewish origin, with the possible
exception noted above, was liable to the tax, and that there must have
been many who attempted to conceal their Jewish origin in order to
evade it. In view of the wide extent of the spread of the Jewish
propaganda, the delation was plausible from the beginning. Suetonius
tells us at first-hand recollection of a case in which the charge of
evading the tax was made and successfully established.[362] In a very
large number of cases, however, the charge was not established, but in
these cases it was often apparently the policy of prudence to buy off
the accuser rather than risk the uncertainties of a judicial decision.
It is upon people who act in just such a way that blackmailers,
συκοφαντία, _calumniatores_, grew fat. And the charge of
evading the Jewish tax was easily made, and disproved with difficulty,
since all who followed Jewish customs were amenable to it, and many
Jewish customs so closely resembled the practices of certain
philosophic sects that confusion on the subject was perfectly natural.
We have seen this in the case of Seneca some years before this (comp.
above, p. 310).

The emperor Nerva, in 96-98 C.E., removed the occasion of this abuse.
Coins are extant with the legend _Fisci Iudaici calumnia sublata_, “To
commemorate the suppression of blackmail arising from the Jewish tax.”
The _fiscus Iudaicus_ itself continued into much later times, but
blackmail by means of it was ended. How this was done we are not told.
But an obvious and natural method would be to abolish the money reward
which the delator or prosecuting witness received for every conviction.
Plainly there would be no blackmail if there was no incentive thereto.

But this reform of Nerva affected rather those who were not Jews than
those who were, since in the case of actual Jews, whether by birth or
conversion, the tax was enforceable and the accusation of evading it was
not _calumnia_, but patriotic zeal. It is likely enough that the measure
of Nerva discouraged prosecution, even where it was justified, but the
losses which the imperial fiscus sustained by reason of the successful
evasion of the tax on the part of some individuals cannot have been
great, since the Jews not only publicly professed their faith, but
openly and actively spread it.

In the epitome of the sixty-eighth book of Cassius Dio (i. 2), we read
that this measure of Nerva was one of general amnesty for the specific
crime of “impiety,” or ἀσέβεια: “Nerva ordered the
acquittal of those on trial for impiety, and recalled those exiled for
that crime.... He permitted no one to bring charges of impiety or of
Jewish method of living.”

Unfortunately this passage is extant only in the epitome made of this
book by Xiphilinus, a Byzantine monk of the eleventh century. We have no
means of knowing to what extent the epitomator is stating the impression
he received from his reading, largely colored by his time and
personality, and to what extent he is stating the actual substance of
the book. If there really was in Rome an indictable offense which
consisted in adopting Jewish customs as distinguished from the general
charge of impiety, such an offense does not appear elsewhere in our
records. We must remember that there is no indication that the men freed
by Nerva had been suffering under the despotic caprice of Domitian, but
on the contrary there is the specific statement that they were being
duly prosecuted under recognized forms.

It is highly likely that the two accusations which Xiphilinus gives are
really one: that Nerva discouraged prosecutions for impiety, and that
among the instances of men acquitted, which Dio gave, were some who were
converts to Judaism, or believed to be so. In one instance, a constantly
cited one, that is precisely what is the case, and that is the
condemnation, in the last year of Domitian’s reign, of Flavius Clemens
and Flavia Domitilla, both of them kinsmen of the emperor.[363]

In the case of these, we hear that Clemens was executed for “atheism,”
and that under this charge many others who had lapsed into the customs
of the Jews were condemned, some of them to death, others to loss of
their property, Domitilla to exile.

In Suetonius we have a wholly different version (Dom. 15). Flavius
Clemens, we read, was a man _contemptissimae inertiae_, “of thoroughly
contemptible weakness of character,” but enjoying till the very last
year of Domitian’s life the latter’s especial favor. Clemens’ two
children were even designated for the succession. The emperor was,
during this year, a prey to insane suspicions, which amounted to a real
_mania persecutoria_, and on a sudden fit had Clemens executed. The
context and general tone of the passage suggest that the charge, real or
trumped up, against Clemens was one of treason, not impiety.

Clemens’ relationship, his undoubted connection with the palace
conspiracy that ultimately resulted in the assassination of Domitian,
make this account the more likely one, but the “many” mentioned in the
epitome of Xiphilinus require us to assume that at least some of the men
actually prosecuted for “impiety,” or atheism, were so charged upon the
evidence of Jewish practices.

It has been stated, and it must be constantly reiterated, that impiety
was a negative offense, that it implied deliberate refusal to perform a
religious act of legal obligation, rather than the actual doing of some
other religious act. If “impiety” were really the offense here, the
“many” that were charged with it under Domitian and Nerva must have been
so charged because they neglected certain ceremonies which the laws made
obligatory. In Greek communities ἀσέβεια was a relatively
common offense, and indictment for it of frequent occurrence. But it is
doubtful whether there was such an indictment at Roman law. There is no
Latin term for ἀσέβεια. The word _impietas_ is generally
used in a different sense. The Greek Dio or his late Byzantine
epitomator has evidently used that term here to describe in his own
words what seemed to him to be the substance of the accusation rather
than to give a technically exact account of the charge against these
men.

In later law writers certain offenses are discussed under which forms of
impiety or ἀσέβεια might be included. But these offenses
are treated either as sedition or as violations of the Sullan Lex
Cornelia de Sicariis et Veneficis, “Statute of Assassins and Poisoners.”
The latter law seems to have been a general statute containing a varied
assortment of provisions, but all of them relating to acts that tended
to the bodily injury of anyone, whatever the motive or pretext of that
injury.[364]

The “many,” then, who, as Xiphilinus says, were prosecuted for
“impiety,” because they lapsed into Jewish rites, may have been indicted
under the Lex Cornelia—no doubt as a pretext—or charged with treason
upon proof of Jewish proclivities. The Palestinian Jews, we may
remember, were until recently in arms against Rome. In all these cases,
the indictments were probably far-fetched pretexts devised by the morose
and suspicious Domitian during his last year of veritable terror in
order to get rid of men whom he suspected (often justly) of plotting his
assassination. These are the men whom Nerva’s act of amnesty freed.

The famous jurist Paul, who wrote in the first part of the third
century, discusses the restrictions imposed upon the spread of Jewish
rites, under the heading of “sedition” or “treason.” The justification
for that treatment lies in the series of insurrections of the Eastern
Jews of which the rebellion of 68 C.E. was merely the beginning. Our
sources for the events of these rebellions are remote and uncertain, and
the transmission is more than usually troubled; but a chance fragment,
as well as the kernel of the lurid account presented by Xiphilinus’
epitome of Dio, leaves no doubt that the struggle was carried on with
memorable ferocity, and left a lasting impression on the people whom it
concerned.

If Dio is to be believed, the outbreak that took place in the reign of
Trajan (115 C.E.) in Cyprus, Cyrene, and Egypt (Ep. lxviii. 32) was
marked by scenes of indescribable horror. In Cyrene, Dio states, the
Jews devoured the flesh of their victims, clothed themselves in their
skins, threw them to wild beasts, or compelled them to engage in
gladiatorial combats. In Cyrene, two hundred and twenty thousand men
perished; in Cyprus, two hundred and forty thousand. One may say with
Reinach, _Les chiffres et les détails de Dion inspirent la
méfiance_.[365]

It will not be possible to assign the responsibility for these
statements to the epitomator Xiphilinus. Unless they were found in Dio,
he could not have ventured to place them here, since the epitome and the
text were extant together for a long time.

In the Church History of Eusebius (IV. ii.) the revolt is described
somewhat differently. Eusebius mentions the Cyprian revolt in his
Chronicon (ii. 164). Here however he speaks only of the insurrection in
Cyrene and Egypt. The name of the leader is given as Lucua, not Andreas,
as Dio has it, and the whole event is described as an ordinary revolt, a
στάσις, reviving the revolt of 68 C.E. At first the Jews
were generally successful, driving their opponents to take refuge in the
city of Alexandria, while they harried the land. At last the Roman
prefect, Q. Marcius Turbo, crushed them completely.

As far as Egypt is concerned, many papyri mention the revolt. Appian
Arab. Liber (Fg. hist. gr. v. p. 65) gives us a first-hand view of the
situation.

Both the papyri and Appian are in complete accordance with Eusebius’
account, and emphasize the extent of the Jewish insurrection and the
impression it produced upon others.

In Jewish writings the references to what must have been a matter of
prime importance to all Jews are vague and confused. The punishment of
the Mesopotamian Jews by Lusius Quietus[366] is mentioned, but beyond
that we have only much later statements, in which a deal of
legend-making has been imbedded. The “day of Trajan,” which appears as a
festival day, is connected by a persistent tradition with the permission
to rebuild the temple, alleged to have been given by that emperor. The
Roman and Greek writers know nothing of this, and in Jewish tradition
likewise the permission is represented as abortive, and the “day of
Trajan” ceased, according to another story, to be observed when the
martyrs Papius and Lollianus were executed.[367]

However, it must be noted that for Palestine in particular details are
lacking. Indeed we might well believe that Palestine itself took no part
in it whatever. The expedition of Quietus to Mesopotamia may have been
an ordinary military expedition against the Parthians’ territory, with
whom the Romans had been then at war. There is evidence that the Jews of
Parthia were almost autonomous, and a foray into the section which they
happened to control would not be considered as anything more than an
attack on other Parthian dominions. The Mesopotamian provinces of
Parthia were then under the theoretical rule of Rome, but the precarious
character of the conquest was apparent to everyone, so that the first
act of the conqueror’s successor, Hadrian, was to abandon both
Mesopotamia and Armenia. The revolt of the Mesopotamian Jews was, in
consequence, a somewhat different thing from that of the Jews in Cyprus
or Cyrene.

Perhaps the difficulties in Cyprus, Cyrene, and Egypt are to be
considered nothing more than magnified race riots, which, however,
assumed the dimensions of a real war, and demanded systematic military
operations to suppress them. But the friction between the Jews and
Greeks of Salamis or Alexandria could scarcely have resulted in such
serious outbreaks, if the conditions that led to the revolt of 68 C.E.
were not still operative. The fall of the temple did not paralyze the
Jewish propaganda. We find it as vigorous afterward as before. The
Messianic hopes, which were one form of the prevailing spiritual unrest,
had not died out in the East among Jews or non-Jews.[368] The calamity
of the empire, which the death of Nero seemed to bring with it, did not
after all take place.

Our sources represent the era begun by Vespasian, except for a few years
of Domitian’s reign, as one of general and increasing felicity. These
sources, however, are in the highest degree suspect, and while the
period between Vespasian and Marcus Aurelius represents an undoubted
rise in administrative and legal development, they represent a
deterioration in the economic condition due to the gathering pressure of
the huge state machinery itself. The increase of the more degraded forms
of superstition marks the spiritual destitution of the time.

The Jewish communities in Cyprus, Egypt, and Cyrene consisted largely of
craftsmen and small merchants. Perhaps among them were a number of
former Palestinian rebels, sold as slaves in the neighboring markets,
and since ransomed. The conditions, the active Messianic hope, the
presence of former soldiers, were themselves provocative of riot, and
the outbreaks in the places indicated are scarcely surprising. We hear
only of those that became formidable insurrections. It is possible that
slighter ones have failed wholly to be recorded.

But during the reign of Hadrian there broke out an unmistakable
insurrection in Palestine, which more clearly than its predecessors
showed the motive force of these movements. In 131 C.E. a certain Simeon
bar Kosiba led his people again to war on the all-overwhelming power of
the empire. The occasion for the revolt is variously given, but that it
was in the eyes of those that fought in it vastly more than an attempt
to shake off a foreign yoke is shown by the Messiahship to which Simeon
openly laid claim, and for which he had the invaluable support of the
head of the Palestinian schools, the eloquent and passionate Akiba.[369]

Dio[370] states that the immediate instigation of the revolt was the
building on the ruins of Jerusalem the new city and temple that were to
be the official home of the colony of Aelia Capitolina, a community
founded by Hadrian and composed perhaps of native Syrians, since it did
not possess the _ius Italicum_, the full rights of citizenship.[371]
This statement is much more probable than that of Eusebius, which
reverses the order of events, and makes the founding of the Colonia
Aelia Capitolina a consequence and not the cause of the revolt.[372]

The rebellion of 68 had enormously depopulated Judea. Those that were
left had neither the power nor the inclination to try conclusions with
the legionaries again, and, as we have seen, remained passive when
closely related communities rose in arms. But the hopes they nourished,
no doubt systematically fostered by the powerful communities in
Mesopotamia and the Parthian lords of the latter, were none the less
real for their suppression. The erection of Aelia was the signal. Just
as the desecration of the temple by Epiphanes was the last measure of
oppression, which brought upon the king the vengeance of Heaven, so this
second desecration, the dedication of the holy hill to one of the
_elillim_, one of the Abominations of the heathen, roused the frenzy of
the people that witnessed it to such a pitch that the chances of success
could no longer be considered. At the same time, assurances of ultimate
help from Parthia were perhaps not lacking. Among those who streamed to
aid the rebellious Jews were doubtless many of Rome’s hereditary
enemies, since of other rebellions within the empire at that time we
have no evidence.

The Jewish tradition speaks of a systematic and cruel persecution
instituted by Hadrian. The details mentioned are very much like the
remembered incidents of the persecution by Epiphanes. We must keep in
mind that every one of the statements connected with this persecution is
late, and is in so far of dubious historical value.[373] As a matter of
fact the character of Hadrian makes the reality of the persecution in
the highest degree improbable. No doubt the revolt was punished with
ruthless severity, and for the permanent prohibition against the
entrance of a Jew into Aelia Capitolina there is excellent
evidence;[374] but to attempt to root out Judaism as Antiochus had done
is something that simply cannot be credited to Hadrian, if only for the
fact that the overwhelming majority of Jews did not dwell in Palestine
at all, and all the alleged persecutions of Hadrian are localized only
in Palestine. In Hadrian’s letter of 134 C.E., to his brother-in-law
Servianus, the Jews of Egypt are referred to in a manner quite
irreconcilable with the theory that Judaism was then a proscribed
religion.[375]

In this connection we may mention a decree which, according to Jewish
tradition, constituted one of the most deeply resented of Hadrian’s
persecutions—the prohibition of circumcision. Here again the late
biographer of Hadrian, Spartianus, makes this edict precede and not
follow the war; but the reliability of the _Historia Augusta_, of which
Spartianus’ biography is part, is not very high. We have the _Historia
Augusta_, if it is not wholly a fabrication of the fourth century, only
in a recension of that time, so that its testimony on such a detail is
practically valueless.[376]

As a matter of fact, all bodily mutilation had been under the ban of the
Roman law, but that prohibition applied only to Roman citizens. In
practice circumcision had been openly carried on both by Jews who were
Roman citizens and by their converts, in disregard of this provision,
probably under the tacit assumption that the privileges of the Jewish
corporations covered this as well. Primarily the prohibition was
directed against castration, but it was quite general. The only
formulation which the edict against these practices had received was in
the Sullan Lex Cornelia de Sicariis et Veneficis (above, p. 241). This
was a _lex per saturam_, or miscellaneous statute. Under one of its
captions, any act, perhaps any act performed with a weapon or instrument
of any kind, that resulted in bodily injury, was prohibited. A
senatorial decree of the year 83 C.E. specified castration as one of the
mutilations referred to; similarly abortion was punished as a violation
of the Lex Cornelia.[377]

Hadrian’s rescripts seem to have dealt on several occasions with this
law. His obvious intention to extend the statute may have caused him to
use terms of general effect. Perhaps an isolated case of the practice of
circumcision among people outside of those to whom it was an ancient
custom may have been followed by indictment and punishment. If Hadrian
really had attempted to carry out this prohibition generally, he would
have provoked a rebellion in Egypt as well as in Judea, since in Egypt
the priests practised it likewise.[378] The rescript of Antoninus, a few
years later, which expressly exempted Jews from the broad condemnation
of the practice, simply restated established law.[379] Indeed it may
well be that the occasion of Pius’ rescript was rather one that
restricted the Jews than one that enlarged their privileges. Even in the
case of the severest form of mutilation, it is forbidden if it is done
_promercii aut libidinis causa_. A similar insistence on criminal intent
must have been present in the case of the lesser mutilation involved in
the Jewish rite. There could of course never have been any question that
circumcision was not performed _promercii aut libidinis causa_, and
therefore there seems to be little reason for the rescript of Pius,
unless we assume it to have been a direct attempt to check the spread of
Judaism by making the performance of the rite in the case of non-Jews
criminal _per se_, without proof of wrongful intent.

Paul, writing about seventy-five years later, states the limitation on
the performance of the rite even more broadly, by including within it
slaves of non-Jewish origin.[380] In all circumstances there does not
seem to have been any real effort to enforce it. The Jewish propaganda
went on in spite of it, not surreptitiously, as in the case of the
still-proscribed Christians, but quite frankly. The statement of Paul is
the stranger because of the open favor shown by Paul’s master, the
Syrian Severus Alexander, toward all foreign cults, including that of
the Jews. The Sentences of Paul may have been written before the decree
of the emperor which his biographer mentions, by which, he says, Severus
strengthened the privileged position of the Jews, _Iudaeis privilegia
reservavit_.[381] When one contrasts this with the immediately following
statement, _Christianos esse passus est_, “He allowed the Christians to
profess their faith,” it is plain that in the case of the Jews there is
no question of mere toleration, but of the recognition of an established
position, and that is not quite in accord with the statement in Paul’s
Sentences, according to which the spread of Judaism was rigorously
checked, even to the extent of modifying one of the fundamental concepts
of the law—the unlimited character of the master’s dominion over his
slaves.

As has been said, the authenticity of the _Historia Augusta_ is dubious,
but the number of details offered to show the interest of both Alexander
and his predecessor Elagabalus in Judaism and Christianity is too great
to be ignored. The Sentences of Paul, it must be noted, have come down
to us only in the abridged and perhaps interpolated form in which they
are found in the Lex Romana Wisigothorum, a code issued by Alaric II in
506, and called therefore the Breviarium Alaricianum. At that time,
however, proselytizing on the part of the Jews had been expressly
prohibited by a rescript of Theodosius (Cod. Theod. 16, 8, 9, 19) of
415. Even then it was completely ineffective, but at any rate the rite
of circumcision was definitely under a legal ban.[382]

Whether or not a qualified restriction on the spread of Judaism has been
changed in our texts of the Sentences into a general and all-embracing
one, it is impossible to say, but that some such change has taken place
may be called even likely, by reason of the point just raised; viz.,
that it is wholly contrary to the spirit and principles of the Roman law
to impose any restrictions whatever on the master’s authority.

We have examined the decrees that regulated the rite of circumcision,
merely because general inferences have been drawn from it—inferences
that are in no sense justified. The Roman law regarded bodily
mutilation, when practised as part of a religious rite, and especially
for sordid purposes, as against public policy. It was a _privilegium_ of
the Jews, that to the members of their organizations the general rule of
the law did not apply, and the various statements quoted from the
jurists were simply judicial decisions limiting, by a well-known
principle of interpretation, the exercise of the privilege to the
narrowest possible bounds.

The rebellion of Bar-Kosiba was probably the last time that the Jews
confronted the Roman troops on issues that were even partly national. We
hear that between 150 and 161, under Antoninus Pius, another rebellion
broke out, but we have no other record of it than the notices in the
_Historia Augusta_,[383] upon which little reliance can be placed. After
the death of Commodus and Pertinax,[384] the eastern empire, including
Palestine, sided with the local claimant Pescennius Niger, and Palestine
became the scene of battles sufficiently important to justify the
decreeing of a “Jewish triumph” to Caracalla. It is likely that these
various “rebellions” were the more or less serious insurrections of
bandits, who terrorized the countryside until suppressed by the
authorities. This view derives some support from the fact that of one of
these bandits who submitted to Severus we know the name, Claudius (Dio
Cass. Ep. lxxv. 2). There is even no certainty as to whether those who
took part in them were wholly or mainly Jews. At any rate, there were no
national ends which they attempted to serve.

A fact, which may be accidental, and is certainly noteworthy, is that,
of all the struggles of the Jews with their surroundings, after 68, none
are localized in Asia Minor.

It was, however, in Asia Minor that the Jews were especially numerous
and influential. To a certain extent their propaganda had become most
firmly established there, and their position was so intrenched that even
the hostile legislation of the later Byzantine emperors found them in
successful resistance. We find evidences of certain laxity in the
practice of Jewish rites, but neither in 68 nor under Trajan or Hadrian
did the Asiatic Jews take part in the movements that convulsed that
section of the Jews of the empire. And yet it was in the cities of Asia
that the Jews in earlier days did meet hostility and direct attacks, and
needed the assistance of the Roman central government, to be maintained
in the position which they claimed for themselves.[385] However, in that
most ancient and fertile nursery of beliefs and mysteries, the Jewish
mystery evidently found a grateful soil and, as we have seen, sent its
roots deep.[386]


------------------------------------------------------------------------




                              CHAPTER XXI

           THE LEGAL POSITION OF THE JEWS IN THE LATER EMPIRE


The empire established by Augustus was, as has been set forth (above, p.
259), a more or less abstract thing. It was the _imperium_, or supreme
authority, which a single community, the city-state of Rome, exercised
over all the other communities existing within certain not over sharply
defined geographic limits. This _imperium_ was, by Roman statute or
series of statutes, almost completely delegated to a single individual.
The delegation however was not quite complete, and the legal theory that
made it incomplete remained to work no little mischief in a crisis like
the death of Nero or Domitian or Commodus.

When Diocletian reorganized the empire in 286 C.E., the theory was
completely changed. The _imperium_ was now a _dominium_; it was the
authority that a single man possessed over all the inhabitants of a
region greater even than it was under Augustus, and that authority was
in point of law as limitless as that of a master over his slaves.

Between Augustus and Diocletian the reign of the Severan emperors,
particularly the promulgation of the Edict of Caracalla, the Constitutio
Antonina, which extended Roman citizenship to almost all the free
inhabitants of the empire, may be considered the turning-point of the
tendency toward absolutism.[387] It broke finally and completely with
the legal theory that the _populus Romanus_ was a paramount community
within a complex of other similar and inferior communities. From that
time on nearly all those who could possess rights and obligations at
all, whether in regard to one another or to the state, were members of
the paramount community, and the delegation of the _imperium_ to the
_princeps_, which had until then been subject to the remote but still
conceivable possibility of revocation, became irrevocable by the sheer
impossibility of conceiving the _populus_ as acting in the only way the
_populus_ could legally act, by direct vote when assembled in mass in
the Campus Martius.

In the period between Caracalla and Diocletian the vast political
machine snapped at many points. Diocletian’s skill enabled it to go on
for a considerable time, and yet the changes he instituted were
administrative rather than social. Internally the new _populus Romanus_
took its form in the third century.

A calculation of doubtful value makes the population of the empire at
that time about 85,000,000.[388] Of these about half were slaves, _i.e._
at law not participants in the empire at all. The other half were nearly
all _cives Romani_, Roman citizens, and it is the position of these
_cives_ that now concerns us.

Upon the _civis Romanus_ devolved the task of maintaining a frightfully
expensive governmental machinery. The expense consisted in the fact that
a huge army had to be maintained on what was practically a war footing
all the time, because, as a matter of fact, war with the barbarians on
the northern frontier and with the Parthians in the East was always
going on. Compared with that, the expenses of the court itself, although
considerable, were scarcely important; but an important item was the
vast horde of civil employees which the execution of so tremendous a
budget necessitated. Then the local civic centers, generally the remains
of old independent communities, had an organization of their own that
was partly ornamental, but in all circumstances costly. That is to say,
a very large share of the available wealth of the empire was diverted
into unproductive channels, since it was devoted to the purpose of
maintaining a machinery not altogether necessary to guard that wealth.

Many of the nations of modern Europe have a military budget relatively
and absolutely greater than that of the Roman empire of the third
century; but in these nations the economic system has a high degree of
efficiency, compared with that of the older state, and the waste is
incalculably less. The great difference lies in the slave system, which
was the foundation of ancient society. The total absence of individual
incentive wherever the slaves were worked in gangs—and that was,
perhaps, true of the majority of slaves—made the efficiency and
consequent productivity of each laborer much less.

We must further remember that human waste was also much greater, owing
to the absence of all measures to restrict it. Only the most elementary
of sanitary precautions existed, and they were directed against definite
diseases of plainly infectious character. With a great percentage of the
population undernourished, the ravages of any disease with epidemic
tendencies must have been enormous. Even in the absence of any plague,
such a scourge as consumption alone must have been much more generally
destructive than it is now. As has been recently suggested, malaria in
Italy had a heavy account to answer for in producing the physical
debilitation of the _populus Romanus_, and was therefore a real factor
in the gradual decay of the Roman state.[389]

The incidence of the state burdens was not regulated as it is at the
present time. Taxes were imposed within certain districts, and upon each
district devolved the duty of satisfying the impost. For a long time
Italy had been free from such a burden, but even this exceptional
position was abrogated by Constantine in 300 C.E.

How each district accomplished its task was a local matter, and was
determined by its individual development. Until the reorganization
effected by Diocletian, the old national units had in the main been kept
intact. That is to say, Egypt remained what it had been under the
Ptolemies and for thousands of years before—a strongly centralized
kingdom, rigidly bureaucratic, but measurably well organized. Asia,
again, was a group of independent cities and certain larger districts,
principally rural, the kingdoms of Bithynia, Cappadocia, Galatia, etc.
The tax which the particular province had to deliver was apportioned
among the various units according to their apparent capacity. Here and
there a poll-tax existed, levied upon every inhabitant alike, and on the
existence of this poll-tax far-reaching theories have been constructed.

The obligation of the individual toward the state was determined by one
fundamental fact, viz., domicile, or right of residence. Before the
Constitutio Antonina there was only one class of inhabitants that
possessed an almost unlimited right of residence, the _cives Romani_.
But even these could not live indiscriminately in Egypt, for example,
which was at all times an exceptional province, and was considered a
sort of imperial appanage. As a matter of fact, it is in Egypt that we
see the first development of the _colonatus_, destined to be of so
fundamental importance in the creation of the feudal system. It may be
that the _colonatus_ was found practically everywhere in the Hellenistic
states, but its growth in Egypt goes back to Pharaonic times, and its
fullest expansion was found there.

The principle of the _colonatus_ was the permanent obligation of the
agricultural free laborer to remain on the soil he tilled. Originally it
applied only to the state lands, but in the third century these state
lands became largely private property, and the serf-like _coloni_ went
with them. All over the empire there were still, in spite of the
_latifundia_, or agriculture on a big scale, a large number of peasant
proprietors; but with the impossibility of competing with the production
of the _latifundia_, these peasant proprietorships were soon converted
into holdings resembling the _colonatus_, or actually that.

Now, as long as the _civis Romanus_, as a prerogative of his position,
paid no tax, his right of residence was unqualified. When he too had to
submit to a direct tax, the place where he resided became a matter of
prime importance. The tax that was imposed upon any given locality could
be met only if all those subject to tax, living there, paid their dues.
Consequently those who by birth were domiciled there could not remove
themselves without lessening to that extent the power of that district
to meet its state obligations. At first, to be sure, this cannot have
been a matter of first-rate importance. Changes of domicile after all
were rare, and took place principally among the wealthier classes, a
fact that made it easy to insure that no loss would accrue to the
community abandoned. But as conditions of ordinary living deteriorated,
the practice of deserting one’s legal residence became more frequent,
and needed the intervention of the central authorities, since the local
magistrate had no jurisdiction whatever beyond the strictly
circumscribed limits of his commune. As soon as it was possible for a
commune to claim from its members, wherever they happened to be, their
contribution to the communal tax, there arose the corollary that for all
practical purposes the tax-paying member might not leave the place where
his tax was due. The _colonatus_ had been applied to the urban laborer.

But the chaining of the individual to his commune was not sufficient
unless his paying power was maintained. The same motives that impelled
men to evade their fiscal duties by change of domicile, would make them
idle and sullen paupers in the places where they were forced to remain.
It was a part of the state system which the Severan emperors introduced
to make the paying power of the citizen certain by means of the
compulsory guilds.[390] These latter were natural outgrowths of former
voluntary associations. The formation of guilds of laborers, either free
or consisting partly of freemen and slave laborers, was as old as the
state itself. The evident superiority of training which such groups
insured alone justified them. From time to time certain privileges and
exemptions were conferred upon them—always in return for definite state
functions[391] which they took upon themselves as well as the industrial
functions which were their reason for existence. Indeed, in the
municipal towns the _collegiati_, or members of these publicly
sanctioned industrial guilds, formed an order of citizenship second only
to that of the decurions, or municipal senate.

While the various _collegia_ were at first voluntary associations, it is
evident that the sons of members would tend to follow the callings of
their fathers without statutory command to that effect. When, however,
the dues of the corporation to the state became onerous, the voluntary
choice of a calling might leave certain _collegia_ quite deserted. At
what time this danger became so serious that special legislation was
required, we do not know, but there is a vague and textually uncertain
passage in the Life of Alexander Severus, in the _Historia Augusta_,
which indicates that a reorganization of the trade-guilds was undertaken
by that emperor. If it was so, the appearance soon afterwards of the
compulsory guild in full development makes it likely that the compulsory
principle was officially recognized or perhaps extended then.

But it was not merely the artisans of the empire that were included in
any organization or reorganization of the _collegia_. Like all other
corporate bodies the trade-guilds, if not wholly religious in form,
possessed a common cult or ceremony, and this common possession made it
easy to consider them as not essentially different from _collegia_
directly and solely religious—the Greek θίασοι for example.
In these, the voluntary principle remained even after the compulsory
guilds were fully developed, although in point of fact they were
generally rigidly hereditary at all times. Here too, after Alexander
Severus, there must have been a certain legal restriction placed upon
arbitrary withdrawal from such cult-organizations, even if their ritual
was openly and unmistakably foreign, such as that of the Jews, the
orgies of Atthis, or the mysteries of Mithra. Some restriction would be
necessary, because membership in these organizations, as far as they
were tolerated by law, involved the payment of certain dues to the
state, and the state could not see with equanimity the obligation to pay
these dues discarded and no new ones assumed in its place.

The dues to the state did not consist altogether, and soon not even
principally, in the actual taxes levied upon a community, and portioned
among its constituent members, whether individuals or corporations.
Indeed these latter were paid to what seems to us a wholly
disproportionate extent by a small and wealthy class in the community.
The taxes, whether they consisted of ground-rent for state lands,
harbor-dues, or taxes on certain sales, were principally paid by the
large traders and investors, who were in every case the governing body
of the local communes. In provinces where a poll-tax was levied, and
where a tribute was imposed as on conquered territory, which the
province really was, these direct taxes, when brutally executed on the
peasant’s grain, were oppressive enough, but in many parts of the Roman
world they were in effect λειτουργίαι, “liturgies,” _i.e._
the burdens assumed by or imposed upon private persons of making large
contributions in service to the state in proportion to their means. The
principle of the liturgy was common to most Greek states, and was
capable of indefinite extension.

And there was one state burden rapidly increasing in gravity, which was
generally met on the principle of the liturgy, although the state too,
as early as the time of Trajan,[392] was compelled to attempt it in
part. That was the care of incompetents, by which term we may understand
all free individuals who could not support themselves wholly by their
personal efforts, _i.e._ widows and orphans, as well as destitute
freemen. The proletariat of the empire not only had no share in its
burdens, but itself formed the empire’s chief economic burden.

The organization of the system was of very old standing. From time
immemorial the minor children and the women of a family and of a clan
had been under the legal control and care of the family’s head. In the
developed system of law, the technical terms were _tutela_ and _cura_,
the former being the guardianship of a child until fourteen, the latter
the guardianship of a youth until twenty-five, as well as the care of an
adult incompetent. This system of guardianship was further extended, but
always remained the same in principle. It was the duty of the family to
provide for its destitute members, and the legal extension the system
underwent was simply that of widening the family circle. Not merely
close relatives but remoter kinsmen were drawn into it as far as the
obligations of guardianship were concerned; and in default of kinsmen,
the guild, society, or commune assumed the wardship of minors, and was
answerable for their maintenance.

It is easy to understand how important this item of state service
became, when we recall how large a part of the municipal budgets in
England during many centuries was concerned with the care of the poor.
But after the disintegration of the slave system on its economic side,
the number of persons for whose care this provision had to be made must
have been much greater than it was in England at any time. If nothing
else, the minute care with which the burdens of wardship were
apportioned, the precautions against their evasion, the great part its
discussion played in legal literature,[393] will make it evident that
wardship of minors was a vitally important matter, and its
administration one of the chief functions of citizenship in the empire.
Many groups of men were practically exempted from all other state dues,
provided the guardianship of minors within that group was assumed.

The maintenance of the poor is almost a corollary of the compulsory
wardship of women and minors. The artisan whose efforts no longer
sufficed to maintain his family often absconded, or in very many cases
succumbed physically to his tasks, leaving in either case a family for
whose wardship his kinsmen or colleagues had to provide. The state
foundations instituted and maintained by Trajan and his successors were
probably abandoned during the third century, when the _tutela_ was
systematized and minutely regulated.

All in all, every member of the state as such had certain fiscal duties
to the state, _munera_, and his performance of these _munera_ determined
his place in the state. The social cleavage between the _honestiores_,
the “better classes,” and the _humiliores_, “the lower classes,” was of
very great importance in criminal law, since the severity of the penalty
varied according to the class to which the convicted criminal belonged;
but we are not told on what basis the judge determined whether any given
man was _honestior_ or _humilior_, and the whole distinction seems
somewhat un-Roman.[394] For other purposes the various honors and ranks
which multiplied in spite of the sinking significance of the many
constituent communities were much less important than the drastically
enforced classification of citizens by the taxes they paid.

The Jews of the Roman empire were to be found in all the classes that
existed. As long as innumerable forms of local citizenship existed,
distinct from citizenship in the Roman state, Jews might be met in all
those groups. But when the Constitution of Caracalla merged all the
local forms of citizenship in the _civitas Romana_, practically all the
Jews then living in the empire became Roman citizens, although it is
highly likely that the old names did not at once disappear.

Only one exception is known to have been made by Caracalla. A certain
class of inhabitants known as the _dediticii_ were excluded from his
general grant. To analyze the exact position of these _dediticii_ would
demand more detailed argument than can here be offered, especially since
it is a highly controversial matter. Recently it has been urged that all
those who paid a poll-tax, particularly in Egypt and Syria, were classed
as _dediticii_ and consequently excluded from Roman citizenship. For
this, however, there is not the remotest evidence. In the Institutes of
Gaius[395] there is an unfortunate lacuna where the matter is discussed,
but from what is said there, it is likely that as early as the Antonines
the _dediticii_ in Rome were a class of freedmen suffering legal
disabilities for proven offenses, and that there were few others. The
exemption of the _dediticii_ from the benefits of the Edict of Caracalla
was therefore perfectly natural, and did not in the least imply the
exemption of those who paid the poll-tax in Egypt and Syria, among whom
were many Jews.

As Roman citizens domiciled in the various quarters of the empire, the
Jews were subjected to the obligations that went with that domicile. So
in Egypt a great number of Jews paid a poll-tax, although many of them,
especially in Alexandria, were exempt. In Syria and Asia, where many
communities still had tribute to pay, the Jewish members of those
communities were equally assessed.

But besides being legally domiciled in some definite place, the Jews in
every place formed cult-organizations. Apostasy in the case of the Jew
meant no more than the abandoning of this organization, “separating
himself from the congregation.”[396] Those who did so found themselves
at once obliged to perform the rites of the state worship in the many
cities where such rites were legally enforced, or to enter other
cult-associations, since it was only as a member of the Jewish
corporation that he secured the privilege of abstention.

These Jewish corporations were known as “synagogues,” a term more
properly denoting the meetings of the societies. The word was used of
other associations as well as of the Jewish. A word of kindred origin
and meaning, _synodos_, was almost a general term for corporation
everywhere.[397] However “synagogue” became gradually appropriated by
the Jewish _collegia_, and in inscriptions in which the word occurs it
is generally safe to assume a Jewish origin.

Like all other similar corporations or guilds, the Jewish synagogues had
special _munera_. One which was almost unique was the Jewish tax, the
_fiscus_ _Iudaicus_, or didrachm, which, since 70, had been levied on
all the Jews, originally for the support of the Capitoline temple, but
probably long merged into the general fiscus, or imperial treasury. It
is unique, because there does not seem to have been any other tax which,
like this one, was wholly devoid of local basis, and did not depend on
domicile at all. Otherwise membership in the Jewish synagogue conferred
a highly valued and general exemption. The Jews could not be required to
perform any task that violated their religious conviction. This
privilege is formulated in a constitution of Caracalla, but it seems
rather a confirmation of one already existing than a new grant.[398]

According to this privilege, Jews were immediately relieved from all
dues connected with local or state worship or with the temples. As many
Jews were in a financial position that would ordinarily invite the
imposition of just these liturgies, that meant a very great relief. All
other liturgies, including the _tutela_ both of Jews and of non-Jews, we
are expressly told the Jews were subject to.

We know further that the demands upon them did not end there. In
Palestine the organization of the Sanhedrin had maintained itself,
although only in the form of several schools under the general
presidency of the Nasi, whom Romans and Greeks called the Patriarch. The
maintenance of these schools and those who labored in them was a
religious duty which most Jews voluntarily assumed. The money was
collected by _apostoli_, “envoys,” despatched to the various Jewish
synagogues for that purpose.[399] The early Christian emperors believed,
or professed to believe, that the payment of this tax was a grave burden
to the poorer Jews, and that irregularities were committed in its
enforcement. The Jewish sources, all of which are Palestinian, naturally
show no trace of this complaint; nor is it likely that there was much
foundation for it except in certain localities already grievously
burdened by constantly increasing dues.

Besides these various classes into which the tax-paying Jewish citizens
fell, there were also Jews who did not share in the support of the state
at all. Jewish slaves existed in the third and fourth centuries too, but
they can scarcely have been numerous. A Jewish slave belonging to a
Jewish master was practically only a servant bound for a term of
years.[400] Within a relatively short space of time he could demand his
freedom by Biblical law. If his master was a pagan, a religious duty
devolved upon all other Jews, and particularly the local synagogue, to
redeem him.[401] Often, to be sure, that duty could not be carried out.
Not every master would sell, and not every synagogue was financially
able to supply the necessary funds. In general, however, it added
another motive to those already existing that made emancipations
frequent.

The social position and occupations of the Jews throughout the empire
are only slightly known. For Egypt and Rome we have fuller documents
than elsewhere, except for Babylon, which was outside the empire. We
have no means of determining whether the facts found in Egypt and Rome
are in any way typical. One negative statement may however be safely
made. They were only to a very slight extent merchants or money-lenders.
In most cases they seem to have been artisans. The inscriptions in the
Jewish catacombs show us weavers, tent-makers, dyers, butchers,
painters, jewelers, physicians.[402] In Egypt we meet sailors and
handicraftsmen of all description.[403] Vendors, of course, on a small
and large scale were not wholly lacking. Indeed it would be impossible
to understand the individual prosperity of some Jews or of some
communities except on the assumption of commercial occupations and
success. However, in general, commerce was principally in the hands of
Syrians and Greeks, especially the former, whose customs and cults
spread with them over the Mediterranean.

We may say, in conclusion, that the economic and political position of
the Jews in the empire was unique in one sense. There were no other
groups that had exactly the same rights, or were subject to exactly the
same demands as the Jews. But in another sense that position was not at
all unique. Many other groups of men had rights somewhat like those of
the Jewish synagogues, and played a part in the social economy similar
to theirs; and, as individuals, there was probably nothing to mark out
the Jew from his fellows in the community.

We cannot tell how far and how long the Jews would have been able to
maintain their position. There seems however to have been nothing in the
conditions of the Diocletianic empire that threatened the stability of
the synagogues in the form in which they were then found. The religious
basis of the state—the maintenance of a common cult for the whole
empire—had practically been abandoned. At one time, under Aurelian,[404]
the emperor’s devotion to the solar cult had almost made of that the
state religion. But in general it may be said that the absolutism of
Diocletian rendered such bonds unnecessary. Where all men were born
subjects or slaves (“slaves of their duties,” _servi functionum_, the
guild-men are called explicitly[405]) of the same master, it could be
considered indifferent whether they all maintained the same theology.

But whether the Jews might have maintained their position or not, if the
conditions had remained the same, is a purely hypothetical question.
When Christianity became the state religion, under Theodosius,[406] a
step was taken that Jews must perforce regard as retrogressive. In
ancient times participation in the common _sacra_ was of the essence of
membership in a state.[407] That principle was, however, tolerantly
enforced. In the first place the mere existence of private _sacra_ was
not deemed to imperil the public _sacra_. Secondly, exceptions and
exemptions that did not take offensive forms were freely allowed. But
when Theodosius established Christianity, he consciously strove to make
the ecclesia coterminous with the empire. “As well could those be saved
who were not in the ark with Noah,” Cyprian[408] had cried, “as they be
saved who are not in the church.” What was originally a group of elect,
a company of saints (ἅγιοι), “the salt of the earth,”[409]
had been expanded into a world-filling community.

Not only was the ancient theory revived, but it was revived without the
qualifications that had made the ancient theory a livable one. No other
_sacra_ could be permitted to exist. Not to be in the ecclesia, was not
to be in the empire. Only the practical impossibility of really
enforcing that theory restrained the zealous and triumphant leaders. Of
course, the development of law was continuous. The new basis of
citizenship was never actually and formally received as a legal
principle. Yet gradually the limitation of civic rights, which
non-membership in the church involved, operated to work an exclusion
from citizenship itself. In a very short time those who were not within
the church were in a very real sense outside the state, merely tolerated
sojourners, and subject to all the risks of that precarious condition.


------------------------------------------------------------------------

                                SUMMARY


What has been attempted in the foregoing pages is an interpretation of
certain facts of Jewish, Roman, and Greek history within a given period.
For that purpose it has been necessary to analyze fully the terms used,
and in many cases rather to clear away misconceptions than to set forth
new points of view. A brief retrospect is here added.

The Jews, as one of the Mediterranean nations, began to come into close
contact with Greek civilization about the time of Alexander. Greece was
then entering on a new stage in her development. The Macedonian hegemony
produced a greater degree of political unity than had been previously
achieved, but above all a real cultural unity had been created, and was
carried by arms and commerce over the East. To this the Jews, as other
nations did, opposed a vigorous resistance; and this resistance was
successful in so far as it allowed the creation of a practically
independent nation, and particularly it stimulated the independent
development of Jewish institutions, especially religious ones.

In religion the Jews came into further and more extensive conflict with
their Greek environment. For many centuries all the East had known a
great spiritual unrest, from which had grown various religious
movements. Of all these the common goal was the attainment of a personal
immortality, the “salvation of the soul.” Among the Jews too this
movement had been active, and had produced concrete results in sects and
doctrines. The Jewish aspect of this general movement would have
remained a local development, had it not been given a wider field by the
unusual position of the Jews, due to their dispersion.

For this dispersion various causes can be assigned. Perhaps the most
potent single cause was the fact that the Jews, who rigorously opposed
exposure of infants, and encouraged in other ways the growth of their
population, increased too rapidly for the very limited resources of
their small and niggardly territory. At any rate the kingdoms of the
successors of Alexander found Jews as colonists in many of the new
foundations in Asia, Syria, and Egypt, especially the last, where, as a
matter of fact, Jews had lived from pre-Persian times. Within these new
and, in many cases, old communities the doctrines preached in Palestine
became a means of propaganda, and enabled the Jews to do more than
maintain themselves in the exceptional position which their highly
specialized religion necessitated.

The Jews were by no means the only religious group in the Greek
communities with proselytizing tendencies. But they were unique in so
far as they were permanently connected with an existing national group,
with which they maintained relations. This made friction of some sort
inevitable at first, since some community of religious observances for
all citizens of a single state was axiomatic for ancient times. However,
the anomaly of the Jewish position became less glaring in course of
time.

The first stage of Jewish influence is marked by two things, a
constantly increasing dispersion and an equally increasing propaganda
that reached all stages of society.

The advance of the power of Rome at first did not change these
conditions. In fact that advance materially assisted both the dispersion
and its propaganda, since the support of Rome was an invaluable asset
for the Hasmonean kingdom. Even the conquest by Pompey had no other
effect than to accelerate the indicated development, especially within
Italy and Rome itself.

But the relations of the Jews with the Greco-Roman world entered upon a
second stage, the stage of armed conflict, when the national and
religious aspirations of certain classes of Jews, which culminated in
the Messianic hope, came into contact with the denationalizing
tendencies of the imperial system. This conflict was in no sense
inevitable, and might easily have been avoided. In addition to the
internal movements that provoked the series of rebellions between 68 and
135, there was a constant excitation from without. The hereditary
enemies of the Greek East and its successor, the Roman Empire—the
Persians and their kinsmen and successors, the Parthians—maintained not
only their independence but also their hostility, and the fact that the
Jews lived in both empires, and that Parthian Jews communicated freely
with the others, presented a channel for foreign stimulation to revolt.

The third stage of Jewish relations consists of an adjustment of the
Jews to the rapidly centralizing empire, of which the administrative
center was moving eastward. The center of wealth and culture had always
been in the East. The reforms of Hadrian and his successors prepared the
way for the formal recognition of the new state of things in the
Constitutio Antonina, the Edict of Caracalla, which gave Roman
citizenship to almost all the freedmen of the empire. This is the great
period of Roman law, when, in consequence of the enormously extended
application of the civil law, a great impetus was given to the
scientific analysis and application of juristic principles. Out of this
grew the bureaucratic system perfected by Diocletian, and begun perhaps
by Alexander Severus, in which, as told in the last chapter, the attempt
was made to classify every form of human activity in its relation to the
state.

A new stage of Jewish relations begins with the dominance of
Christianity; and that, as was stated at the beginning of this study,
lies outside of its scope.


------------------------------------------------------------------------




                                 NOTES

                              INTRODUCTION


Footnote 1:

  To what extent the Jews of the present day or those of earlier times
  may be considered racially pure, depends upon what criteria of race
  are adopted. At present there is no general agreement among
  ethnologists on this subject. The historical data are very uncertain.
  At all events absolute racial unity of the Jews of the Dispersion
  cannot be maintained. The facts of their vigorous propaganda and their
  extensive slave-property are too well attested. But it is wholly
  impossible to determine how far the admixture went.

Footnote 2:

  The best edition of Philo is the still unfinished one which is being
  prepared by two German scholars, Wendland and Cohen. In this the
  _Apologia_ has not yet appeared. Earlier editions are those of Mangey
  (1742) and Holtze (1851).

  Philo’s works were translated into English by C. D. Yonge (Bohn’s
  Library, London, 1854).

Footnote 3:

  In Greek the two commonest editions of Josephus’ works are those of
  Niese (1887-1895) and of Naber (1896). Neither completely satisfies
  all the demands that may be made for the adequate presentation of the
  text.

  The old English translation of W. Whiston, so widely circulated both
  in England and America, is very inaccurate. The revision of this
  translation by A. R. Shilleto (1889-1890) has only slightly improved
  it.

Footnote 4:

  The references to the Jews in the inscriptions and papyri have not, as
  yet, been collected. Mr. Seymour de Ricci planned a collection of the
  Greek and Latin inscriptions to be called Corpus Inscriptionum
  Judaicarum. This Corpus was, at least partly, in manuscript form in
  1912, but no part has been published. Mr. de Ricci’s article on
  “Inscriptions” in the Jewish Encyclopedia, and Johannes Oehler,
  Epigraphische Beiträge zur Geschichte des Judentums (Monatsschrift f.
  Gesch. u. Wiss. d. Jud. 1909, xvii. 292-302, 443-452, 524-538) give a
  practically complete collection.


                               CHAPTER I

                        GREEK RELIGIOUS CONCEPTS

Footnote 5:

  It is nowhere directly stated that the power of a god did not extend
  beyond a definite locality. But the numerous local epithets applied to
  the various gods indicate it. We need mention only such typical
  references to the θεοὶ ἐγχώριοι as Aesch. Septem. 14,
  Soph. Trach. 183, and Thuc. ii. 74.

Footnote 6:

  Cf. Dionysus in the “Frogs” of Aristophanes, Heracles and Poseidon in
  the “Birds.” The other comic poets, even Epicharmus, the oldest, dealt
  with even greater freedom with the gods. Even the scanty fragments of
  Cratinus and Amphis indicate that fact. In Sicily, an entire dramatic
  genre, that of the Φλύακες, contained practically nothing
  but situations in which the divine personages of the myths were the
  subjects of the coarsest fun.

Footnote 7:

  Such heroic friendships as that of Achilles and Patroclus were
  perverted early in the imagination of Greeks. Cf. Aeschylus, in Athen.
  xiii. 601 A, and Aeschines, i. 142. So also the story of Apollo and
  Admetus became a love story for Alexandria; Callimachus H. ii. 49.

Footnote 8:

  The subject has been discussed in full by de Visser, _De Graecorum
  deis non referentibus speciem humanam_ (Leyden, 1900;) 2d ed. in
  German, 1903. So at Phigaleia, in Arcadia, Demeter had the form of a
  horse; the Brauronian Artemis was a bear; Apollo Lykeios was sometimes
  adored in the form of a wolf.

Footnote 9:

  Aegean and Mycenean are both used to designate the civilization that
  preceded that of historical Greece. Aegean, however, has, to a large
  extent, superseded the older term. For the specifically Cretan form of
  it, Minoan is generally employed.

Footnote 10:

  In spite of the apparently well-defined personalities of the Homeric
  gods and a poetic tradition of many centuries, the sculptors of later
  times found it necessary to indicate the subject of their labors,
  either by some well-known attribute, such as the caduceus, or a sacred
  animal, or a symplegma representing a scene of a known legend. Without
  these accessories, archeologists often find themselves at a loss when
  they are required to name the god intended. Cf. Koepp, Archäologie ii.
  88 seq.

Footnote 11:

  It is not suggested that prayer could not exist without sacrifice. But
  where sacrifice did take place, the act of worship did not lie in the
  sacrifice alone, or in the propitiatory allocution that accompanied
  it, but in the two together.

Footnote 12:

  Cf. Apollo Soter, Soph. O. T. 149, Dionysus Soter, Lycophr. 206, Zeus
  Soter, Aristoph. Plut. 1186, etc.

Footnote 13:

  Max Müller, Lectures on the Science of Language, _passim_. The term is
  rarely used by recent investigators.

Footnote 14:

  For the sacrificial act when addressed to gods, the word was
  θύειν; addressed to heroes, ἐναγίζειν.
  Herod, ii. 44. The color of the sacrificial animal for heroes was
  usually black, and no part of the flesh was eaten. Cf. Sch. Hom. Il.
  i. 459.

Footnote 15:

  For heroes whose position in the state was as high as that of gods, we
  have only to refer to the eponyms of the Cleisthenic tribes at Athens,
  Theseus, Cecrops, Erechtheus, etc.

Footnote 16:

  Local deities, such as Pelops at Olympia (Sch. Pind. Ol. i. 149),
  Archemorus at Nemea (Arg. Pind. Nem. i), Tlepolemus at Rhodes (Sch.
  Pind. Ol. vii. 146).

Footnote 17:

  Cf. Suidas. s. v. Ἀναγυράσιος, Alciphro, iii. 58.

Footnote 18:

  The doctrine of Socrates cited by Xenophon, Memor. iv. 7, represents
  popular Greek feeling on the subject of theological speculation.

Footnote 19:

  Xenophanes of Colophon (sixth cent. B.C.E.) cited in Sex. Emp. adv.
  Math. ix. 193. The lines are frequently quoted, and are to be found in
  any history of philosophy.

Footnote 20:

  A monotheistic or pantheistic tendency showed itself in the attempt on
  the part of poets like Aeschylus and Pindar to absorb the divine world
  into the personality of Zeno. Cf. Aesch. Heliades, 71:

            Ζεύς ἐστιν αὶθήρ, Ζεὺς δὲ γῆ Ζεὺς δ’ οὐρανός,
            Ζεύς τοι τὰ πάντα χ ὥτι τῶνδ’ ὑπέρτερον.

Footnote 21:

  The solar myth theory was especially advocated by Max Müller in his
  various books and articles. Most of the older writers on mythology,
  _e.g._ in the earlier articles of Roscher’s Lexikon, accept it as an
  established dogma. There can be no reasonable doubt that the celestial
  phenomena of sun, moon, and stars exercised a powerful influence on
  popular imagination.

Footnote 22:

  Dionysus came into Greece probably from Thrace and Macedon about the
  tenth century B.C.E. By the sixth century there was no Greek city in
  which he was not worshiped. As far as any center of his worship
  existed, it may be placed in Boeotia. Cf. Farnell, Cults of the Greek
  States, chs. iv. and v.

Footnote 23:

  We find Aphrodite firmly established among Greek gods from the
  earliest times. It may be that the Semitic or Oriental connections
  which have been found for her (cf. Roscher, s. v. Aphrodite, Roscher’s
  Lex. i. 390-406) are due to the readiness with which she was
  associated with Oriental female deities. That fact, however, is itself
  significant.

Footnote 24:

  The merchants of Citium formally introduced into Athens the worship of
  their local Aphrodite; Dittenberger, Syll. no. 551. Sarapis, Isis, and
  Sabazios also early found their way into Athens.

Footnote 25:

  The statement that ἀσέβεια was a negative offense, that
  its gravamen consisted not in introducing new divinities, but in
  neglecting the established ones, is made by Wilamowitz (Antigonus von
  Karyst, p. 277). It is, however, only qualifiedly true. The Greeks
  found purely negative conceptions difficult. Impiety, or
  ἀσέβεια, was not the mere neglect, but such a concrete
  act as would tend to cause the neglect of the established gods. The
  indictment against Socrates charged the introduction of καινὰ
  δαιμόνια, but only because that introduction threatened the
  established form. The merchants of Citium (cf. previous note) might
  introduce their foreign deity with safety. No such danger was deemed
  to lie.

Footnote 26:

  The stories of Lycurgus (Il. vi. 130) and of Pentheus (Euripides,
  Bacchae) are a constant reminder of the difficulties encountered by
  Dionysus in his march through Greece. Then, as has always been the
  case in religious opposition, the opponents of the new forms advanced
  social reasons for their hostility (Eurip. Bacchae, 220-225).

Footnote 27:

  The Egyptian origin of the Eleusinian mysteries is maintained
  especially by Foucart, Les grands mystères d’Eleusis.

Footnote 28:

  The Homeric Hymn to Demeter dates from the close of the seventh
  century B.C.E., perhaps earlier. In it we find the Eleusinian
  mysteries fully developed, and their appeal is Panhellenic.

Footnote 29:

  Homer certainly knows of no general worship of the dead. But the
  accessibility of the dead by means of certain rites is attested not
  only by the Νέκυια (Od. x. 517-520), but by the slaughter
  of the Trojan captives at the funeral of Patroclus (Il. xxiii. 174).
  The poet’s own attitude to the latter is not so important as his
  evidence of the custom’s existence.

Footnote 30:

  In later times any dead man was ἥρος, and his tomb a
  ἡρῷον; C. I. G. 1723, 1781-1783.

Footnote 31:

  The kinship of gods and men was an Orphic dogma, quickly and widely
  accepted. Pindar formulated it in the words ἕν ἀνδρῶν, ἕν θεῶν
  γένος; Nem. vi. i. Cf. Plato, Timaeus, 41 C.

Footnote 32:

  Od. iv. 561.

Footnote 33:

  Hesychius, s. v. Ἁρμοδίου μέλος.


                               CHAPTER II

                        ROMAN RELIGIOUS CONCEPTS

Footnote 34:

  Adolph Bastian presents his theory of _Grundideen_ in his numerous
  writings. It has, however, been found difficult, if not impossible,
  even for anthropologists to present the details of that theory with
  either definiteness or clearness.

Footnote 35:

  Cf. W. Warde Fowler, Roman Religion, in Hasting’s Dictionary of
  Religions (consulted in proof).

Footnote 36:

  The relation, or the contrast, between magic and religion has been a
  constant subject of discussion since the publication of Tylor’s
  Primitive Culture. For the present the contrast stated in the text may
  suffice.

Footnote 37:

  _Sei deo sei deivae sac_ (C. I. L. vi. 110); _sive deo sive deae_
  (_ibid._ iii, 1212); _sei deus sei dea_ (_ibid._ xiv. 3572). _Cf._
  also Not. d. Sc. 1890, p. 218.

Footnote 38:

  Such a story as that of Mars and Nerione may belong to genuine Roman
  mythology. The enormous spread of Latin translations of Greek poems,
  and the wide popularity of Greek plays, rapidly drove out all the
  native myths which had attained no literary form.

Footnote 39:

  Livy V. xxi. 3, 5.

Footnote 40:

  Macrob. Sat. III. ix. 7-8.

Footnote 41:

  The authenticity of this particular application of the formula has
  been questioned; Wissowa, s. v. Evocatio (deorum); Pauly-Wiss. vi.
  1153. The proofs that the formula has been extensively modified are
  not conclusive. The _evocati di_ received a special form of ritual at
  Rome. Festus, p. 237, a, 7. Cf. Verg. Aen. ii. 351-352.

Footnote 42:

  For the Dioscuri, Livy, II. xx. 13. Apollo, Livy, III. lxiii. 7; IV.
  xxv. Both introductions are placed in the fifth century B.C.E. The
  historical account of the reception of Cybele and of Asclepius, Livy,
  Per. ix. and xxix. 10 seq.

Footnote 43:

  The _lectisternium_ is generally conceded to be of Greek origin. The
  ceremony consisted in formally dressing a banquet table and placing
  thereat the images of some gods, who reclined on cushions and were
  assumed to be sharing in the repast.

Footnote 44:

  Cic. De Nat. Deor. i. 119.


                              CHAPTER III

                    GREEK AND ROMAN CONCEPTS OF RACE

Footnote 45:

  The extreme of racial fanaticism will be found in H. S. Chamberlain,
  Grundzüge des neunzehnten Jahrhunderts.

Footnote 46:

  Aristophanes, Acharn. 104, Ιαοναῦ and the Schol. _ad
  loc._: ὂτὶ πάντας τοὺς Ἕλληνας Ἰάονας ἐκάλουν οι
  βάρβαροι.

Footnote 47:

  After the defeat of the Persians, the victors set up a tripod at
  Delphi, about the stem of which a bronze serpent was coiled. About
  this serpent ran an inscription, τοίδε τὸν πόλεμον
  ἐπολέμεον, “The following took part in the war.” Then follows
  the list of the Greeks beginning with the Lacedemonians. Here, if
  anywhere, a collective term denoting the common origin of all these
  nations might have been expected.

Footnote 48:

  Euripides, Iph. Aul. 1400; Aristotle, Pol. I. ii. 4; ὡς ταὐτὸ
  φύσει βάρβαρον καὶ δοῦλον ὄν.

Footnote 49:

  Isocrates, Pan. 181.

Footnote 50:

  Demosthenes, In Mid. 48 (xx. 530).

Footnote 51:

  Daniel xi. 3.

Footnote 52:

  Besides the flings at barbarian descent scattered throughout the
  orators (cf. Dem. In Steph. A. 30), Hellenic origin was required for
  all the competitors in the Olympian games. Herodotus, v. 22.

Footnote 53:

  The secretary of Appius Caecus was a certain Gnaeus Flavius, grandson
  of a slave, who became not merely _curule aedile_, but one of the
  founders of Roman jurisprudence. (Livy, IX. xlvi.). Likewise the
  Gabinius that proposed the Lex Tabellaria of 139 B.C.E. was the son or
  grandson of a slave, _vernae natus_ or _nepos_. (Cf. the newly
  discovered fragment of Livy’s Epitome, Oxyr. Pap. iv. 101 f.) The
  general statement is made by the emperor Claudius (Tac. Ann. xi. 24),
  in a passage unfortunately absent in the fragments of the actual
  speech discovered at Lyons.

Footnote 54:

  Cicero, In Pisonem (Fragments 10-13). Aeschines, In Ctes. 172.

Footnote 55:

  Muttines, a Liby-Phoenician (cf. Livy, XXI. xxii. 3, _Libyphoenices
  mixtum Punicum Afris genus_), becomes a Roman citizen (_ibid._ XXVI.
  v. 11).

Footnote 56:

  Ennius ap. Cic. de. Or. iii. 168.

Footnote 57:

  Mucius defines _gentiles_, _i.e._ true members of Roman _gentes_, as
  follows (ap. Cic. Topica, vi. 29): _Gentiles sunt inter se qui eodem
  nomine sunt, qui ab ingenuis oriundi sunt, quorum maiorum nemo
  servitutem servivit, qui capite non sunt deminuti._ Literally taken,
  that would exclude descendants of former slaves to the thousandth
  generation. But Pliny demands somewhat less even for Roman knights.
  The man is to be _ingenuus ipse, patre, avo paterno_ (H. N. XXXIII.
  ii. 32).

Footnote 58:

  Gallic was still spoken in southern Gaul in the fourth century C.E.,
  Syriac at Antioch in the time of Jerome, and Punic at Carthage for
  centuries after the destruction of the city.

Footnote 59:

  The racial bond upon which modern scientific sectaries lay such stress
  was constantly disregarded in ancient and modern times. The Teutonic
  Burgundians found an alliance with the Mongol Avars against the
  Teutonic Franks a perfectly natural thing.


                               CHAPTER IV

    SKETCH OF JEWISH HISTORY BETWEEN NEBUCHADNEZZAR AND CONSTANTINE

Footnote 60:

  The Carduchi, Taochi, Chalybes, Phasiani (Xenophon, An. IV. iii. 6),
  make friends with the Greek adventurers, or oppose them on their own
  account without any apparent reference to the fact that the army of
  the Ten Thousand was part of a hostile force recently defeated by
  their sovereign.

Footnote 61:

  Herodotus, vii. 89: παρείχοντο δὲ αὐτὰς (sc. τὰς
  τριήρεας) εοἵδε, Φοίνικες μὲν σὺν Σὺροισι τοῖσι ὲν τῆ
  Παλαιστίν ῃ, and he later defines the name specifically
  (_ibid._): τῆς δὲ Συρίας τοῦτο τὸ χωρίον καὶ τὸ μέχρι Αιγύπτου
  πᾶν Παλαιστίνη καλέεται.

Footnote 62:

  Aramaic Papyri Discovered at Assuan, edited by Sayce and Cowley,
  London, 1906. Aramäische Papyri ... zu Elephantine, ed. Sachau,
  Leipzig, 1911.

Footnote 63:

  Josephus, Antiquities, XI. vii. Reference to the same incident in
  Eusebius, Chron. (Ol. 103), Syncellus (486, 10), and Orosius (iii. 7)
  depends upon Eusebius. The general statement of pseudo-Hecataeus (ap.
  Joseph, in Ap. i. 22) is, of course, worthless as evidence.

  Ochus was especially noted for his sacrilege. (Cf. Aelian, N. A. x.
  23).

Footnote 64:

  After the death of Antiochus Sidetes, in 129 B.C.E., the various
  occupants or claimants of the Syrian throne are scarcely to be
  distinguished by nickname or number. They are uniformly imbeciles or
  puppets, and the last of them, Antiochus XIII, dies miserably at the
  hands of a Bedouin sheik.

Footnote 65:

  In the Talmud John Hyrcanus is always יהוחנן כהן הגדל,
  but Alexander is יני המלך. On the coins John styles
  himself High Priest; but Jannai, on both his Hebrew and Greek coins,
  bears the title of King, יהונתן המלך and Αλεξάνδρου
  βασίλεως. Cf. Madden, Coins of the Jews. We have no record that
  the royal title was specifically bestowed upon Jannai, either by the
  Seleucids or by the people. It is therefore likely that it was assumed
  without such authorization. The high-priesthood, on the other hand,
  was duly conferred upon Simon and his descendants.


                               CHAPTER V

       INTERNAL DEVELOPMENT OF THE JEWS DURING THE PERSIAN PERIOD

Footnote 66:

  Cf. especially the Testaments of the Twelve Patriarchs, in the
  editions of Kautzsch or Charles.

Footnote 67:

  That the name is Sira and not Sirach, as it appears in the LXX, is
  generally accepted. It was the practice of Greeks to put a final
  Χ to foreign names to indicate that they were
  indeclinable. Cf. Ἰωσήχ (Luke iii. 26) for José.

Footnote 68:

  Ecclesiasticus xlviii. 24.

Footnote 69:

  Job iv. 7 seq.


                               CHAPTER VI

                THE FIRST CONTACT BETWEEN GREEK AND JEW

Footnote 70:

  Σύριος means scarcely more than “Oriental” in Aeschylus
  (Persae, 81, Σύριον ἅρμα; and Ag. 1312, Σύριον
  ἀγλάϊσμα).

Footnote 71:

  Except Hittite and Amorite, these names have no non-Biblical
  occurrence.

Footnote 72:

  Caphthor is rendered Cappadocia in the LXX (Amos ix. 7), for no better
  reason, it may be, than the similarity between the first syllables.
  The Keftiu ships of the Egyptian monuments are scarcely other than
  Mycenean, and if they came from Crete, Minoan (Breasted, Ancient
  Records of Egypt, ii. 492). That the Philistines are of Cretan origin
  is, in the absence of monumental sources, a pure theory. It fits in
  well, however, with what we do know of them.

Footnote 73:

  The Jews were commanded by Ezra to put away their “strange wives”
  (Ezra x. 10) for the specific reason that the latter incited them to
  idolatry. Instances of intermarriage occur in the papyri from
  Elephantine (see ch. IV., n. 3).

Footnote 74:

  Datis and Artaphernes commanded the Persian troops defeated at
  Marathon, 490 B.C.E. Mardonius was defeated at Plataea in 479.

Footnote 75:

  Joel iii. 6. There is nothing in the extant Book of Joel inconsistent
  with a pre-Exilic date. Such slave raids as the Phoenicians are here
  accused of making, the Greeks made freely in Homeric times, and Greek
  merchants were already in every mart. In the famous picture of a
  golden age in Isaiah, Jewish captives are to be assembled “from
  Assyria, Egypt—and from the islands of the sea” (Isaiah xi. 11), a
  passage indubitably pre-Exilic. The “islands of the sea,” however, are
  obviously Greek.

Footnote 76:

  In the lexicon of Stephen of Byzantium (s. v.) we read Σύροι
  κοινὸν ὄ νομα πολλῶν ἐθνῶν. Strabo, writing in the time of
  Augustus, includes most of the nations of Asia Minor, such as the
  Cappadocians, etc., under that term (xvi. 2).

Footnote 77:

  The famous Harpy-tomb from Xanthus in Lycia, now in the British
  Museum, dates from the sixth century. It is, however, so highly
  developed a work that it presupposes a long history of mutual artistic
  influence between Greece, Ionia, and Lycia.

Footnote 78:

  One of the magnificent sarcophagi found in 1887 at Sidon by Hamdi Bey.
  They are all published in sumptuous form by Hamdi Bey and Reinach, Une
  nécropole royale á Sidon, Paris, 1892. An excellent and convenient
  description may be found in Hans Wachtler, Die Blütezeit der
  griechischen Kunst im Spiegel der Reliefsarcophage, Teubner, 1910 (Aus
  Natur u. Geisteswelt, no. 272).

Footnote 79:

  Strato, king of Sidon in 360 B.C.E. Athen. xii. 531. Cf. Gerostratos
  of Arados at about the same time.

Footnote 80:

  Herodotus, ii. 104 (cf. ii. 37).

Footnote 81:

  Aristotle states the fact in the Meteorologica, II. iii. 39, but does
  not mention the Jews.

Footnote 82:

  Textes, p. 8. n. 3.

Footnote 83:

  In the royal tombs at Sidon excavated by Hamdi Bey (see above, n. 9.),
  one of the monuments bears a long Phoenician inscription of a king of
  Sidon. It begins: “I, Tabnit, priest of Astarte and king of Sidonians,
  son of Eshmunazar, priest of Astarte, and king of the Sidonians.”

Footnote 84:

  Plato, Euthyphro, 3 C., and _passim_.

Footnote 85:

  Aristotle, Rhetoric, III. vii. 6.

Footnote 86:

  Reinach, Textes, pp. 10-12. Müller, Frag. hist. graec. ii. 323, quoted
  in Josephus, In Ap. i. 22.

Footnote 87:

  The untutored philosophers of Voltaire’s stories were quite in the
  mode of the eighteenth century, which had discovered the “noble
  savage,” and were quite convinced that civilization was a
  retrogression from a state of rude and primitive virtue. It was,
  further, a convenient cloak behind which one might criticise an
  autocratic régime. Hence the flood of “Turkish,” “Chinese,”
  “Japanese,” etc. “Letters,” of which Montesquieu’s Lettres Persanes
  are the most famous. Modern instances are “The Traveller from
  Altruria” of Mr. Howells, and Mr. Dickinson’s “Letters of a Chinese
  Official.”

Footnote 88:

  Cited by Diogenes Laertius, i. 9 (Müller, Frag. hist. graec. ii. 328).

Footnote 89:

  Reinach, Textes, p. 13; Müller, Frag. ii. 437; Clemens Alex, i. 15.
  Megasthenes had previously resided at the court of Sibyrtius, satrap
  of Arachosia (southern Afghanistan). Arrian, Anab. V. vi. 1.

Footnote 90:

  Clemens Alex. Str. v. (Sylberg), pp. 607 seq. Justin Coh. ad Graecos,
  25.

Footnote 91:

  Cf. Ecclesiasticus l. 26; Zech. ix. 2.

Footnote 92:

  At Elephantine we learn from the papyri recently from there (Pap. 1,
  Sachau) that the Jews had a shrine consecrated to יהו and
  that in 410 B.C.E. it was destroyed by the priests of a rival Egyptian
  temple.

Footnote 93:

  Reinach, Textes, p. 39. Müller, Frag. iii. 35.


                              CHAPTER VII

                                 EGYPT

Footnote 94:

  This fragment, of the authenticity of which little doubt can be
  entertained, must be distinguished from the books attributed to
  Hecataeus about the Jews and Abraham. Josephus uses both in his
  “Defense” against Apion (i. 22 seq.), but their authenticity was
  questioned even in ancient times (cf. Herennius Philo, cited by
  Origenes, C. Cels. i. 15; Reinach, Textes, p. 157). They are almost
  certainly Jewish works of the first century B.C.E.

  The text of the real Hecataeus (Reinach, Textes, p. 14 seq.) is
  anything but certain. We have it only in a long citation by Diodorus,
  xl. 3. This book of Diodorus, however, has disappeared, and is found
  only in the _Bibliotheca_ made by the Byzantine patriarch Photius in
  the ninth century C.E. (cod. 244).

Footnote 95:

  There were in Egypt a number of colonies of military settlers. They
  are distinguished by certain privileges, and, in legal terminology, by
  the term τῆς ἐπιγονῆς, placed after the words of
  nationality. Just as there are Πέρσαι τῆς ἐπιγονῆς, so
  there are Ἰουδαῖοι τῆς ἐπιγονῆς. In the Hibeh Papyri, i.
  96, of 259 B.C.E., we read an agreement between the Jew Alexander, son
  of Andronicus, decurion in the troop of Zoilus, and Andronicus, a Jew
  τῆς ἐπιγονῆς The groom Daniel (?) in a papyrus of the
  second century B.C.E. (Grenfell, An Alexandrian Erotic Fragment and
  Other Papyri, no. 43.) and the farm laborer Teuphilus (Grenfell-Hunt,
  Fayûm Towns and their Papyri, no. 123) are also humble men, and
  probably in the same stage of cultivation as other men of their
  calling.

Footnote 96:

  Elephantine Pap. (ed. Sachau), no. 6.

Footnote 97:

  Osiris appears as a theophoric element, not only in Egyptian names and
  in those of Grecized Egyptians, but also in purely Phoenician names,
  and joined to Semitic elements. So Osirshamar, from Malta, and
  Osiribdil, from Larnaca (Notice des Mon. Phén du Louvre, nos. 133,
  162).

Footnote 98:

  Reinach, Textes, pp. 20 seq. Müller, Frag. ii. 511-616.

Footnote 99:

  Tac. Hist. V. ii.

Footnote 100:

  Reinach, Textes, p. 362. Photius Bibl. no. 279.


                              CHAPTER VIII

                        JEWS IN PTOLEMAIC EGYPT

Footnote 101:

  Naucratis was founded, on the Canopic mouth of the Nile, about 550
  B.C.E.

Footnote 102:

  However completely oligarchical in practice the government became, the
  sovereignty of the dēmos was recognized in theory. In the ancient doom
  ascribed to Lycurgus (Plutarch, Lyc. 6), which may be said to form the
  constitution of Sparta, occur the words δάμῳ δὲ κὰν κυρίαν ἦμεν
  καὶ κράτος.

Footnote 103:

  Fränkel, Inschriften, v. Perg. no. 5, 18 _et passim_.

Footnote 104:

  Mitteis und Wilcken, Grundzüge und Chrestomathie der Papyruskunde, I.
  v. i, pp. 14 seq.

Footnote 105:

  Mitteis-Wilcken, _op. cit._ p. 15.

Footnote 106:

  Xenophon, De Reditibus, ii. 4-7.

Footnote 107:

  Josephus often refers to the Jews of Alexandria as oἱ ἐν
  Ἀλεξανδρείᾳ Ἰουδαῖοι (Ant. XIII. iii. 4) or oἱ ἐν
  Ἀλεξανδρείᾳ κατοικούντες Ἰουδαῖοι (Ant. XIV. vii. 2), but he
  refers similarly to the Greeks there (Ant. XVIII. viii. i), and
  plainly understands κατοικεῖν simply as “inhabit.” The
  question is fully discussed in Contra Ap. ii. 5, where the general
  statement is made that Jews might and did become Alexandrian citizens,
  but that Egyptians were at first excluded.

Footnote 108:

  Jewish Μακεδόνες, Berliner Griechische Urkunden (B. G.
  U.), iv. 1068 (62). In other classes of citizenship, B. G. U. iv.
  1140; iv. 1151, 7. For humbler classes of Jews cf. ch. VII., n. 2. A
  Jewish house-slave is manumitted in Oxyrhyncus Pap. ix. 1205.

Footnote 109:

  The discussion is fully set forth by Brandis, s. v. Arabarches in the
  Pauly-Wissowa Realenzyklopädie, ii. 342. The word “alabarch” or
  “arabarch” impressed the Romans somewhat as “mogul” impresses the
  English, and was used with the same jocular intent. Cic. ad Att. II.
  xvii. 3. Juvenal, Satires, i. 130.

Footnote 110:

  Apuleius, Met. xi. 30. Drexler in Roscher’s Lexikon Myth., s. v. Isis,
  ii. 409 seq. gives a list of the cities through which the worship of
  Isis spread.

Footnote 111:

  Sarapis was not Osiris-Apis, but a deity of Sinope in Asia Minor, duly
  “evoked” into Alexandria by Ptolemy. The matter is left an open
  question by Cumont, Les religions orientales dans le paganisme romain,
  p. 112, but the general consensus of opinion is in favor of the theory
  just mentioned. The opposition referred to in the text was less an
  aggressive one than it was an assertion of the distinction between
  Greeks and Egyptians. It broke down with the fourth Ptolemy, and
  Sarapis was more or less officially identified with Osiris.

Footnote 112:

  Alexandronesus. Cf. Reinach, in Mélanges Nicolle, p. 451; Pap. of
  Magdola, n. 35.

Footnote 113:

  Greek Pap. of the Brit. Mus. iii. 183, the ἄρχοντες α Ἰουδαίων
  προσευχῆς pay their water tax.

Footnote 114:

  B. G. U. iv. n. 562.

Footnote 115:

  The cartouches representing the Ptolemies contain all the royal titles
  of the Pharaohs.

Footnote 116:

  Mitteis-Wilcken, Grundzüge und Chrestomathie, I. p. 42.


                               CHAPTER IX

            THE STRUGGLE AGAINST GREEK CULTURE IN PALESTINE

Footnote 117:

  Ecclesiasticus xxxi. 12-30; vi. 2-4.

Footnote 118:

  Cf. ch. III., n. 14.

Footnote 119:

  A full bibliography is given in Schürer, Geschichte der Juden 4th ed.,
  iii. 472 seq.

Footnote 120:

  Flinders Petrie Pap. iii. 31, g, 13.

Footnote 121:

  By Mishnic tradition Antigonus was a pupil of Simon the Just (Abot i.
  3). A later legend makes him the founder of the Sadducees (Abot R. N.
  v.). The saying of Antigonus is: “Be not like servants who minister to
  their master for the sake of a reward, but be like servants who
  minister to their master without the expectation of reward, and let
  the fear of Heaven be upon you.”

Footnote 122:

  Andronicus (Hibeh Pap. i. 96), Helenus and Trypho (B. G. U. iv. 1140),
  Dionysius (Dittenberger, Syll. no. 73).

Footnote 123:

  Cf. Oesterley’s edition of Ecclesiasticus, pp. xxiv-xxv.

Footnote 124:

  Josephus, Ant. XII. iv.

Footnote 125:

  Abot i. 4; Shab. 46 a; Eduy. viii. 4; Pes. 15 a.


                               CHAPTER X

                       ANTIOCHUS THE MANIFEST GOD

Footnote 126:

  Polybius, XXVI. i. 1: Ἀντίοχος ὁ ν Ἐπιφανὴς μὲν κληθεὶς Επιμανὴς
  δ’ ἐκ τῶν πράξεων ὀνομασθείς. Cf. also Athenaeus, v. 5 (193),
  and x. 10 (439).

Footnote 127:

  Ptolemy Euergetes II (Athenaeus, x. 10, 438 D).

Footnote 128:

  It is usual to speak of the Seleucid kingdom as Syria. That, however,
  conveys a wholly wrong impression of either the pretensions of the
  house or the actual extent of its dominion. Seleucus himself actually
  maintained his authority within what is now Hindustan and was styled
  “king of Asia,” where he was not called simply “the king” as Alexander
  and the Persians had been before him. Even when Antiochus the Great
  gave up all his Asiatic possessions north of the Taurus, he did not
  renounce his claim to the Persian and Oriental patrimony of Alexander.

Footnote 129:

  Zeitschr. d. deut. morg. Gesell. xxiii. 371; Nöldeke, Die sem. Spr. 41
  f.; Zeitschr. f. Assyr. vi. 26. Cf. also Gardner, Greek and Scythic
  Kings of Bactria and India.

Footnote 130:

  The full title is Θεὸς Επιφανής, as it appears upon
  coins.

Footnote 131:

  The στρατηγός ἐπὶ τὰ ὅ πλα _i.e._ “general of infantry,”
  was at that time practically equivalent to the chief magistracy.
  Athenian coins of the year 175 B.C.E. bear his name and the elephant
  which was the heraldic emblem of his house. Reinach, Rev. d. et. gr.
  1888, 163 f.

Footnote 132:

  Josephus, Ant. XII. v.

Footnote 133:

  The titles ἀγοράνομος and δήμαρχος are
  translations of “aedilis” and “tribunus,” which Antiochus sought to
  transfer to his capital. Polyb. XXVI. i. 5-6. Livy XLI. xx.

Footnote 134:

  Livy (_loc. cit._), Polyb. (_loc. cit._), Athenaeus, x. 438 D and E.

Footnote 135:

  Hybristas is mentioned in Livy XXXVII. xiii. 12.

Footnote 136:

  Polyb. XXXI. xi. 3; Josephus, Ant. XI. ix.

Footnote 137:

  I Macc. i.

Footnote 138:

  Cf. ch. I., n. 22.

Footnote 139:

  Cf. the article Druidae, Pauly-Wissowa, Realenzykl.

Footnote 140:

  Isocrates Nicocles (III), 54. King Nicocles of Salamis in Cyprus, the
  type and exemplar of a benevolent despot, states to his subjects:
  ἑταιρείας μὴ ποιεῖσθε μήτε συνόδους ἄνευ τῆς ἐμῆς γνώμης αἱ γὰρ
  τοιαῦται συστάσεις ἐν μἐν ταῖς ἄλλαις πολιτείαις πλεονεκτοῦσιν, ἐν δὲ
  ταῖς μοναρχίαις κινδυνεύουσιν.

Footnote 141:

  Jerome in Dan. xi. 21 f.

Footnote 142:

  So the Spartans actively assisted the oligarchical party in Megara,
  Argos, Sicyon, and Achaea (Thuc. iv. 74; v. 81; v. 82).


                               CHAPTER XI

                         THE JEWISH PROPAGANDA

Footnote 143:

  Cumont, Les religions orientales dans le paganisme romain, gives the
  best and clearest account of the spread of these foreign cults. The
  Cabiri came from Samothrace. They were generally referred to as
  Θεοὶ μεγάλοι, and are found in many parts of the empire.

Footnote 144:

  Athenian criminal statutes often contain in the penalty clause
  καὶ τὸ γένος αὐτοῦ. Cf. Glotz, La solidarité de la
  famille dans le droit Ath. Cf. for Teos C. I. G. 3044.

Footnote 145:

  Homer, Odys. xi. 489-491.

Footnote 146:

  Frequently pictured relief (Gardner, Greek Sculpt. p. 136) formerly in
  the Sabouroff Coll. Pl. i., Ath. Mitth. 1877. Taf. xx-xxiv.

Footnote 147:

  Il. iii. 243-244; v. 638-651; xviii. 117-119.

Footnote 148:

  Cf. the translation of Menelaus, ch. I, notes 28, 29.

Footnote 149:

  Hymn in Dem. 480-482.

Footnote 150:

  Ben Sira knows of no life after death except Sheol. Perhaps it is
  better to say that he refuses to acknowledge any. His repeated
  affirmations have the air of consciously repudiating a doctrine
  advanced by others. The author of Wisdom (iii. 4) is sure of an
  immortality of the elect. It is in the apocryphal literature
  generally, in Enoch, the Testaments of the Patriarchs—most of them
  written in the first century B.C.E.—that the scattered and
  contradictory references to a future life are to be found.

Footnote 151:

  Josephus, Wars, II. viii. 14. His words are (οἱ
  Σαδδουκαῖοι) ψυχῆς τε τὴν διαμονὴν καὶ τὰς καθ’ Ἄδου
  τιμωρίας καὶ τιμὰς ἀναιροῦσι. The passages in Josephus are our
  only contemporary authority for the sects and their differences; and
  Josephus was a Pharisee. The word αναιροῦσι would in this
  context naturally have the meaning “deny,” but it might also simply
  indicate that the Sadducean belief on the subject was, in his opinion,
  so vague or so qualified as to render their whole transcendental
  scheme ineffectual. It is, however, more natural to give the word its
  dialectic sense (Cf. Plato, Rep. 533 c).

Footnote 152:

  Joseph. Ant. XIII. x. 10. Kid. 43 a.

Footnote 153:

  The vision of a Messianic age in Isaiah ii. 4, and Micah iv. 1,
  expressly includes the gentiles. This is the more important as it is
  highly likely that both Micah and Isaiah are here quoting an ancient
  and widely-accepted prophecy.

Footnote 154:

  There is no direct evidence about the extent of proselytizing in
  pre-Maccabean times. But there are two forms of proselytizing which
  always seemed natural and even inevitable to a man of ancient times.
  The slave, and the stranger actually resident under the roof of a head
  of a household, were, however foreign in blood, practically members of
  that household, and it was a small step when they were brought
  formally into it by appropriate ceremonies. So the first Biblical
  reference to circumcision especially notes that not merely Abraham but
  all his household, the slaves born there and those bought of
  strangers, were circumcised (Gen. xvii. 23, 27).

  The גר, μέτοικος, the sojourning stranger,
  is expressly held to the observance of the religious prohibitions. Ex.
  xii. 43; Lev. xvii. 12. And the relative frequency with which such a
  stranger became a full proselyte is indicated by Ex. xii. 48, and Num.
  ix. 14. It is true that the נכר or “stranger in blood” is
  treated with extreme rigor by Nehemiah, xiii. 30, but it is this same
  נכר who is referred to as a proselyte in Deutero-Isaiah
  (Is. lvi. 3, 6).

Footnote 155:

  Ab. R. Nat. ii. 1.

Footnote 156:

  Josephus, Ant. XV. viii.

Footnote 157:

  Josephus, Wars IV. iv.; VII. viii.

Footnote 158:

  Cf. Catullus, LXIII. The archigallus was not permitted to be chosen
  from Roman citizens till the time of Claudius.

Footnote 159:

  This genre seems to have first taken literary form at the hands of
  Bion of Borysthenes, a pupil of Crates, who was himself a pupil of
  Diogenes.

Footnote 160:

  Wisdom of Solomon xiv. 12-14. Cf. also the entire thirteenth and
  fourteenth chapters of Wisdom.

Footnote 161:

  In Dan. x. 13-20 angels, or “princes,” are the patrons of the various
  nations, as also in the Testaments of the Patr. (Test. Naph. 9). That
  fact of itself indicates a belief in the reality of the divine
  protectors of the heathen nations. And the “devils,” שדים
  (Deut. xxxii. 17), and שעירים (Lev. xvii. 7), are very
  likely the local gods.

Footnote 162:

  Philo, De Specialibus Legibus, ch. 7.

Footnote 163:

  We have already noted the ancient prophecy cited in Is. ii. 4 and
  Micah iv. 1. The fullest statement of this universalist aspiration is
  in Malachi i. 11, and i. 14.


                              CHAPTER XII

                             THE OPPOSITION

Footnote 164:

  The Messenians also expelled the Epicureans (Athen. xii. 547), and
  Antiochus (VI) Dionysius, or rather Tryphon in his name, expelled all
  philosophers from Antioch and all Syria (Athen. _ibid._). The latter
  document has been questioned by Radermacher, Rh. Mus. N. F. lvi.
  (1901), 202, but on insufficient grounds. It is probably genuine, but
  the king referred to is uncertain. It will be remembered that the
  Epicurean Philonides claimed to have converted Epiphanes and to have
  been a favorite of Demetrius (Crönert, Stzb. Berl. (1900), 943, and
  Usener Rh. Mus. N. F. lvi. (1901), 145 seq.) Alexander Balas professed
  Stoicism.

Footnote 165:

  Josephus, Ant. XVIII. ix.

Footnote 166:

  Dio Cassius, lviii. 32; Ens. Chron. ii. 164. The account in its
  details is not free from doubt.

Footnote 167:

  Josephus, Ant. XIV. x.

Footnote 168:

  Senatusconsultum de Bacch. C. I. L. i. 43, n. 196. Bruns Fontes, n.
  35, ll. 14-16.

Footnote 169:

  Cf. the instances cited in Cumont, Les rel. or. dans le pag. rom., p.
  122, and the articles on Isis in the Pauly-Wissowa Realenzykl, the
  Dar.-Saglio Dict., and Roscher’s Lexikon.

Footnote 170:

  In Greek διαβολή. Cf. Aristotle, Rhetoric, II. iii. 30;
  Syrianus, In Hermogenem, ii. (134, 3). Of this διαβολή, a
  favorite form was ἐπηρεασμός, “mockery” (Arist. _op.
  cit._ II. ii. 3), and “Commonplaces,” κοινοὶ τόποι, on
  the subject are cited in Aristotle (_op. cit._ III. xv. 1).

Footnote 171:

  Reinach, Textes, p. 49.

Footnote 172:

  Eratosthenes was head of the Alexandrian Academy.

Footnote 173:

  Apollo is the god named and ascribed to Dora, which, as Josephus
  remarks, is not in Idumaea at all. Nor does Apollo appear as the god
  of Dora on the coins of that city. According to Josephus (Ant. XV.
  vii. 9) the Idumean god was named Koze, who might of course have been
  identified with the Seleucid patron Apollo. It may be a title
  connected with קצין (Josh. x. 24, Micah iii. 1, 9).

Footnote 174:

  An inscription forbidding the approach of gentiles has been found at
  Jerusalem, and is now in Constantinople: μηθένα ἀλλογενῆ
  εἰσπορεύεσθαι ἐντὸς τοῦ περὶ τὸ ιερὸν τρυφάκτου καί περιβόλου · ὅς δ’
  ἂν ληφθῇ ἑαυτῷ αἴτιος ἔσται διὰ τὸ ἐξακολουθεῖν θάνατον.

Footnote 175:

  Reinach, Textes, p. 56. For an estimate of the importance of
  Posidonius for his time, cf. Wendland, Hellenist. Kult. p. 60 seq. and
  134 seq.

Footnote 176:

  Molo in Reinach, Textes, p. 60 seq. Damocritus, _ibid._ p. 121.

Footnote 177:

  Reinach, Textes, p. 131.

Footnote 178:

  Plutarch, Moralia, ii. 813; Reinach, Textes, p. 139.

Footnote 179:

  Pseud-Opp. Cyn. iv. 256. Lact. Inst. i. 21-27.

Footnote 180:

  Cf. also Aelian Var. Hist. xii. 34. Strabo, xv. 1057.

Footnote 181:

  Pseudo-Plut. Sept. Sap. Con. 5. Apul. Met. xi. 6. Ael. Hist. An. x.
  28.

Footnote 182:

  Juvenal, Sat. xv. 1-3. _Quis nescit Volusi Bithynice qualia demens
  Aegyptos portenta colat? crocodilon adorat pars haec, illa pavet
  saturam serpentibus ibïn_; cf. also _latrator Anubis_ (Verg. Aen.
  viii. 698, Prop. iv. 11, 41).

Footnote 183:

  It is not to be inferred that ancient historians as such were
  unreliable. In those times, as in ours, the value of an historical
  narrative must be judged by estimating the character and capacity of
  the writer and the means at his disposal. Many modern historians have
  been special pleaders, some consciously, like Froude and von
  Treitschke, and most have been impelled by personal sympathies and
  antipathies of many kinds.

  It is, however, a fact that the writers of antiquity consciously used
  falsehoods in what they believed to be details, if they supposed that
  they could thereby more forcibly present the essential character of a
  transaction, or better enforce a moral lesson. The extreme danger of
  such a practice need not be insisted on, nor did all writers engage in
  it. But Panaetius and Cicero (Cic. De Orat. ii. 59; De Off. ii. 14),
  Quintillian (ii. 26-39) and the Church Fathers, unhesitatingly defend
  it (Eusebius, Praep. Evan., John Chrysost. De Sac. i. 6-8, Clemens
  Alex. Strom, vii. 9).

Footnote 184:

  Polybius shares the general estimate of Syrians (XVI. lx. 3), but that
  does not prevent him from acknowledging the loyalty and devotion of
  the people of Gaza, whom he classes as Syrians.


                              CHAPTER XIII

                  THE OPPOSITION IN ITS SOCIAL ASPECT

Footnote 185:

  Horace, Sat. I. v. 100.

Footnote 186:

  Apuleius, Florida, i. 6.

Footnote 187:

  Anthol. Pal. v. 160. Reinach, Textes, p. 55.

Footnote 188:

  Fg. hist. gr. iii. 196; Reinach, Textes, p. 42.

Footnote 189:

  Journ. Hell. Stud. xii. 233 seq.

Footnote 190:

  Pausanius, X. xii. 9; Suidas, s. v. Σαμβήθη; Sibyllina,
  iii. 818.

Footnote 191:

  Valerius Maximus, I. iii. 3.

Footnote 192:

  Shab. vi. 2, 4, but cf. Demai iii. 11, and Erub. i. 10.

Footnote 193:

  Cf. above, ch. VII., n. 2.

  The letter of Dolabella to the Ephesians, cited in Josephus, Ant. XIV.
  x. 12, makes it perfectly clear that if the Sabbath restriction had
  actually been enforced in the sense indicated, Jews would have been
  wholly useless for the army. But we have seen that they not merely
  fought their own battles, but engaged freely as mercenaries. We can
  therefore understand the passage in Josephus only in the sense of an
  attempt to escape conscription with the other Ephesians, by alleging
  an extreme application of the Sabbath principle.

  The other passage in Josephus (XVIII. iii.) is in direct contradiction
  with other sources, and will be discussed later.

Footnote 194:

  Saguntum, Livy, XXI. xiv. Abydus, Livy, XXXI. xvii. Cf. also Livy
  XXVIII. xxiii.

Footnote 195:

  Cic. De Nat. Deor. ii. 28, 71, his _fabulis spretis ac repudiatis_....

Footnote 196:

  Reinach, Textes, p. 17. Cf. above, p. 93.

Footnote 197:

  The word itself does not occur in Homer. However, Od. ix. 478, the
  taunt is flung by Odysseus, the blind monster,

           σχέτλι’, ἐπει ξείνους οὐχ ἄζεο σῷ ἐνὶ οἴκῳ
           ἐσθέμεναι τῷ σε Ζεὺς τίσατο καὶ θεοὶ οἄλλοι..

Footnote 198:

  Arrian, Anab. I. ix. 9-10.

Footnote 199:

  Il. iii. 207; Od. iii. 355; vii. 190.

Footnote 200:

  Plutarch, Lycurgus, xxvii.; Ael. V. Hist. xiii. 16; Thuc. i. 144.

Footnote 201:

  Juvenal, Sat. xv. 93-131.

Footnote 202:

  Cf. the undoubted instances of the Gallus-Galla, Graecus-Graeca
  sacrifices at Rome. See article, Gallus et Galla, in Pauly-Wissowa
  Realenzykl, especially the unwilling testimony of Livy, XXII. lvii. 6.

Footnote 203:

  The Tauric Artemis was considered a barbarian goddess, but received
  the veneration of Greeks, and of her we read, Eur. Iph. Taur. 384,
  αὕτη δὲ θυσίαις ἥδεται βροτοκτόνοις. The sacrifices of
  the Trojan captives at the funeral of Patroclus, the sacrifice of
  Polyxena, Astyanax, and Iphigenia are sufficient evidences of the
  familiarity of the practice to Greeks. An historical instance is the
  atonement-sacrifice of Epimenides at Athens. Diog. Laert. i. 111, 112;
  Athen. xiii. 602 C.

Footnote 204:

  For the Gauls, cf. Strabo, iv. 198; the Thracians, vii. 300; the
  Carthaginians, Verg. Aen. i. 525.

Footnote 205:

  The question of the Molech sacrifices in Palestine is too uncertain
  and complicated to be treated here in full. Doubtless some Jews at
  various times sacrificed to Molech; but some Jews in Greek times
  sacrificed to heathen gods, or, at any rate, adored them while still
  professing Judaism, and throughout the Middle Ages individual Jews
  indulged in superstitious practices severely reprobated by the rabbis.
  The passage in Jeremiah (xxxii. 35) does not necessarily imply that
  those who took part in these rites deemed themselves to be worshiping
  Jehovah.

Footnote 206:

  Reinach, Textes, p. 121.

Footnote 207:

  Sat. xv. 78-81 and 93 seq.

Footnote 208:

  Sat. xiv. 103.

Footnote 209:

  It is a curious and instructive fact that Chinese have charged
  Christian missionaries with precisely this same crime, _i.e._ of
  kidnapping and killing children as part of their religious ceremonies.


                              CHAPTER XIV

                       THE PHILOSOPHIC OPPOSITION

Footnote 210:

  Cf. the whole Lucianic dialogue on Images, 459-484, and Zeus
  Tragoedus, 654 seq.

Footnote 211:

  Cicero, De Nat. Deorum, i. 23, 63. Athenag. Supp. xii.

Footnote 212:

  Josephus, Contra Ap. ii. 37.

Footnote 213:

  Euthyphro, viii. 3 (7A).

Footnote 214:

  Sophocles, Oed. Rex, 661.

Footnote 215:

  Cf. ch. XI., n. 19. Also II. Chron. xi. 15. The שדים are
  mentioned in Psalms cvi. 37 as deities to whom human sacrifices are
  made.

Footnote 216:

  Isocr. Pan. 155-156; Lycurgus, In Leocr. 80-81.

Footnote 217:

  For the Boeotians cf. the common ὗς Βοιωτία; Pind. Ol.
  vi. 153; _id._ Fr. iv. 9, and Hor. Epp. II. i. 244; for Egyptian
  _perfidia_, Val. Max. v. 1, 10; for Abdera, Juv. Sat. x. 50; Mart. x.
  25, 4; for the Cretans, the famous Κρῆτες ἀεὶ ψευσταί,
  Call. Hymn in Jov. v. 8., a proverb also quoted from Epimenides by
  Paul, Ep. ad Tit. i. 13. One may also note in this connection the
  Greek proverb, τρία κάππα κάκιστα · Καππαδοκία καὶ Κρήτη καὶ
  Κιλικία.

Footnote 218:

  Livy, XXXIV. xxiv. 4.

Footnote 219:

  Plautus, Rud. v. 50, _scelestus, Agrigentinus, urbis proditor_.

Footnote 220:

  Cicero, Pro Fonteio, 14, 30.

Footnote 221:

  Cicero, Pro Scauro, 17, 38.

Footnote 222:

  Pliny, Hist. Nat. Praef. 25.

Footnote 223:

  Africanus, ap. Eus. Praep. Ev. x. 10, 490 B, Clemens Alex Strom. i.
  22.

Footnote 224:

  Reinach, Textes, p. 122.

Footnote 225:

  Cf. ch. VIII., n. 14.

Footnote 226:

  Cf. ch. XII., n. 12.

Footnote 227:

  Strabo, i. 66; Cic. De Rep. i. 58.

Footnote 228:

  Cicero, Paradoxon, iii.: ὁτι ἴσα τὰ ἁμαρτήματα. _Parva,
  inquit, est res. At magna culpa; nec enim peccata rerum eventis, sed
  vitiis hominum metienda sunt._

Footnote 229:

  Cumont, Les rel. orient. pp. 157 seq.


                               CHAPTER XV

                               THE ROMANS

Footnote 230:

  The first Greek historians to deal with Roman history are Hieronymus
  of Cardia and Timaeus, both of the fourth century B.C.E.

Footnote 231:

  Pliny, Nat. Hist. III. lvii.

Footnote 232:

  Psalms of Solomon, ii.

Footnote 233:

  Livy, XLIX. v.: _Syros omnis esse, haud paulo mancipiorum melius
  propter servilia ingenia quam militum genus_.

Footnote 234:

  Cf. ch. III., n. 9.

Footnote 235:

  Servile origin has been ascribed to such a family as the Sempronian,
  and is assumed for the praenomen Servius, as for the nomen Servilius.

Footnote 236:

  Macrob. Saturn. II. i. 13.

Footnote 237:

  The reading of the last phrase in the mss. is _quod servata_, which is
  scarcely consistent with the rest of the passage. Bernays, Rh. Mus.
  1857, p. 464 seq., conjectured that it was a Jewish or Christian
  marginal gloss which found its way into the text, a supposition by no
  means to be dismissed as cavalierly as Reinach does (Textes, p. 241,
  n. 1). A Christian scribe might easily have been moved by the taunt
  _quam dis cara_, to retort with the triumphant _quod servata!_ It will
  be remembered that the Christians accepted as part of their own all
  the history and literature of the Jews till the birth of Christ, and
  resented as attacks upon themselves any slur against the Jews of
  pre-Christian times. Cf. the very interesting passage in Lactantius,
  Div. inst. iv. 2.

Footnote 238:

  Cic. In Vat. 5, 12.

Footnote 239:

  It may be worth while to indicate briefly the relation between the
  senatorial authority and the executive power at Rome. Unless the
  senate acted at the instance of the magistrate himself, a
  _senatusconsultum_ was an advisory resolution, passed upon motion and
  suggesting to the holder of executive power, or _imperium_, a certain
  course of action. The words were generally: _Placet senatui ut A. A.,
  N. N. consules, alter ambove, si eis videretur, ilia faciant_. In
  practice, it is true, such a resolution was almost mandatory. A strong
  magistrate, however, or a rash one, might and did disregard it. While,
  accordingly, a magistrate might neglect a course of action prescribed
  by the senate, there was nothing to hinder any action on his part
  (whether or not there was senatorial authority for it), except the
  veto power residing in the tribune or in an equal or superior
  magistrate. The only restrictions were made by the laws concerning the
  inviolability of the person of a _civis Romanus_, and of the
  _aerarium_.

Footnote 240:

  The _contio_ was a formal assembly of citizens, called by a magistrate
  holding _imperium_. The purpose was generally to hear projected
  legislation either favorably or unfavorably discussed. No one spoke
  except the magistrate or those whom he designated. The _contio_ took
  no action except to indicate its assent by acclamation, or its dissent
  equally emphatically. At the actual legislative assembly, for which
  the _contiones_ were preparations, no discussion whatever took place.
  The law was presented to be accepted or refused. It will be seen that
  a mass of Orientals who less than two years before had been
  Aramaic-speaking slaves can scarcely have been a power in such
  gatherings as these.

Footnote 241:

  Philo, Leg. ad. Gaium, 23.

Footnote 242:

  The language of the inscriptions in the various Jewish cemeteries at
  Rome is almost always Greek, as is that of most of the monuments in
  the Christian catacombs. Latin is rare and generally later. But these
  monuments belong to Jews who lived several generations after 63 B.C.E.
  As far as Palestine is concerned, both inscriptions and literature
  leave no doubt that the masses spoke only Aramaic or Hebrew.

Footnote 243:

  Caesar, Bell. Gall. II. xxxiii. 7; III. xvi. 4.

Footnote 244:

  Foucart, Mém. sur l’affranchissement des esclaves.

Footnote 245:

  Suet. Div. Iul. 84, 76, 80.

Footnote 246:

  The pretensions of the senatorial party to be the only true Romans
  were not altogether unfounded. The terms _boni_ and _optimates_ which
  they gave themselves were perhaps consciously adapted from the
  καλοὶ κἀγαθοί of Athens. The importance of _nobilitas_ as
  a criterion of true Roman blood lay in the fact that it attested
  lineage in a wholly unmistakable way. We may compare the insistence of
  Nehemiah upon documentary evidence of Israelitish blood (Neh. vii. 61,
  64).

Footnote 247:

  Pro Flacco, 15, 36, compared with 26, 62 seq.

Footnote 248:

  Cf. ch. XIV., notes 11, 12.

Footnote 249:

  The chief political asset of the triumvirs was the orientalized plebs
  of the city, whose origin and poverty would combine to make them
  bitterly detest the organized tax-farmers. Now Crassus, one of the
  triumvirs, was himself the head of a powerful financial group. It may
  be that the tax-farmers persecuted by Gabinius belonged to a rival
  organization, or that Crassus had withdrawn from that form of
  speculation before 60 B.C.E. In the case of Flaccus, the complaint of
  the tax-financier Decianus was a pretext, or else Decianus may have
  been forethoughtful enough to have joined the right syndicate.

Footnote 250:

  Cicero ad Att. ii. 9.

Footnote 251:

  Augustinus, De Civ. Dei, iv. 31, 2.


                              CHAPTER XVI

                  JEWS IN ROME DURING THE EARLY EMPIRE

Footnote 252:

  Myths are understood by modern anthropologists exclusively as a
  “folk-way,” with the effects of single creative imaginations almost
  wholly eliminated. However, the better-known Greek myths are not at
  all folk-devised. As far as the Romans are concerned, it has so far
  been impossible to pick out a definite story which does not appear to
  have been derived from an existing Greek myth by quite sophisticated
  methods.

Footnote 253:

  The phrase referred to is _Ubi bene ibi patria_, although just this
  form of it may not be ancient. However, the idea, that a fatherland
  might brutally ill-use its citizens and still claim their loyalty, was
  something that the average Greek scarcely recognized even in theory.
  When Socrates propounds some such doctrine in Plato’s Crito, 51 B, he
  is consciously advocating a paradox. It was regarded as a noble ideal
  somewhat beyond the reach of ordinary men. Its disregard involved no
  moral turpitude.

  In Cicero, Tusc. v. 37, 108, the phrase runs, _Patria est ubicunque
  est bene_. That is an evident adaptation of a Greek phrase, such as
  the one in Aristoph. Plut. 1151, πατρὶς γάρ ἐστι πᾶσ’ ἵν’ ἂν
  πράττη τις εὖ.

Footnote 254:

  Livy, Epit. lvi. Eunous, the leader, called his followers _Syri_, and
  himself King Antiochus. Cf. Florus, ii. 7 (iii. 9), Diodorus fr.
  xxxiv. 2, 5. Atargatis was the _Dea Syria_ that played so important a
  rôle in the life of the empire.

Footnote 255:

  The philosophic schools had the usual corporate names of θίασος,
  σύνοδος, and the like. Or like other corporations they have a
  cult name in the plural, οἱ Διογενισταί, οἱ Ἀντιπατρισταί, οἱ
  Παναιτιασταί (Athen. v. 186). For the International Athletic
  Union, ἡ περιπολιστικὴ ξυστικὴ σύνοδος, cf. Gk. Pap. in
  Brit. Mus. i. 214 seq.

Footnote 256:

  Cf. ch. III., n. 9.

Footnote 257:

  Cf. Menippus in Lucian’s Icaromenippus, 6 seq. Menippus does not spare
  his fellow Cynics (_ibid._ 16).

Footnote 258:

  Macrobius, Sat. II. i. 13. The jest has unfortunately not come down to
  us.

Footnote 259:

  The book we know as the “Wisdom of Solomon” is unquestionably the
  finest in style and the profoundest in treatment of the Apocrypha.
  Such passages as i.; ii. 1 seq.; ii. 6; iii. 1 seq. can hardly have
  appealed to any but highly cultured men.

Footnote 260:

  Until the time of Claudius, we are told by John Lydus, no Roman
  citizen might actively participate in the rites of Cybele. Cf.
  Dendrophori, Pauly-Wissowa, p. 216. Claudius removed the restriction,
  perhaps to make Cybele a counterfoil to Isis.

Footnote 261:

  The story in Livy, XXXIX., viii. seq. is a case in point. The
  abominable excesses which, as Hispala testifies, took place among the
  Bacchae (_ibid._ 13) are almost certainly gross exaggerations.

  This hostility to new-comers was not a sudden departure from previous
  usage. Sporadic instances are mentioned in Livy’s narrative. As early
  as 429 B.C.E., he tells us, _Datum negotium aedilibus ne qui nisi
  Romani dii neu quo alio more quam patrio colerentur_ (Livy, IV. xxx.
  11). The notice is of value as an indication that the general Roman
  feeling was not always so cordially receptive as is often assumed.

Footnote 262:

  Valerius Max. I. iii. 3.

Footnote 263:

  Cf. Cic. ad Att. iii. 15, 4; Asconius ad Pison. 8.

Footnote 264:

  Suetonius, Div. Iul. 42. Josephus, Ant. XIV. x. 8. Suetonius (_ibid._
  84) states that many _exterae gentes_ enjoyed his favor. The Jews may
  have been only one group among many. However, the statement is
  indirectly made by Suetonius and directly by Josephus, that they
  received his special protection to a striking extent. We have only the
  political support given the triumvirs and Caesar personally to fall
  back upon for a motive.

Footnote 265:

  I undertake with some diffidence to revive a conjecture made before
  without much success, that the 30th Sabbath was the Day of Atonement.
  One remarkable misunderstanding of the Sabbath institution was that it
  was a fast-day. When we consider the number and activity of the Roman
  Jews, it seems scarcely credible that so many otherwise well-informed
  persons supposed that the Jews fasted once a week. Augustus in his
  letter to Tiberius seems to do so (Suet. Aug. 76). Pomp. Trogus
  (Justinus), xxxvi. 2, explicitly states it. Cf. also Petronius
  (Bücheler, Anth. Lat. Frg. 37) and Martial, iv. 4. But at least one
  man, Plutarch, not only knew that it was not so, but was aware that,
  if anything, the Sabbath was a joyous feast-day (Moralia ii., Quaest.
  Con. v. 2). To this testimony must be added that of Persius, Sat. v.
  182 seq. It is in the highest degree surprising that Reinach (p. 265,
  n. 3) could have accepted the theory that the _pallor_ alluded to is
  the faintness brought on by fasting. The tunny fish on the plate
  should have convinced him of his error. It may be remembered that fish
  in all its forms was one of the chief delicacies of the Romans. Tunny,
  however, was a very common fish, and one of the principal food staples
  of the proletariat.

  Persius writes from personal experience. Of the other writers it is
  only Pompeius Trogus who makes the unqualified statement that the
  Sabbath as such was a fast-day. When Strabo writes that Pompey is said
  to have taken Jerusalem τὴν τῆς νηστείας ἡμέραν τηρήσας
  (xvi. 40), he is assumed to have been guilty of the same confusion.
  But it is not easy to see why he should have hesitated to say the
  Sabbath if he meant the Sabbath. Nor is it so certain that Josephus is
  mechanically copying Strabo (Reinach, p. 104. n. 1) when he says (Ant.
  XIV. iv. 3) that Jerusalem was taken περὶ τρίτον μηνα τῇ τῆς
  νηστείας ἡμέρᾳ. The details of Josephus are vastly fuller than
  those of Strabo, and he is not guilty of the latter’s error regarding
  Jewish observance of the Sabbath in times of war (Ant. XIV. iv. 2).
  Besides, the siege lasted several weeks—more than two months—so that
  Pompey’s maneuver, if it depended wholly upon the Sabbath, might have
  been performed at once.

  Hilgenfeld’s supposition (Monatsschrift, 1885, pp. 109-115) that the
  day was the Atonement, is better founded than Reinach would have us
  think. In the mouth of Josephus, ἡ τῆς νηστείας ἡμέρα can
  scarcely have any other sense. And if Josephus believed that Jerusalem
  fell on the Kippur, he believed so from more intimate tradition than
  the writings of Strabo.

  Now, ἡ τῆς νηστείας ἡμέρα, the great fast of the Jews,
  must have been as marked a feature in their life two thousand years
  ago as to-day. While all the other feasts have individual names, it
  does not appear that this one did. יום הכפורים (Lev.
  xxiii. 27; LXX, ἡμέρα ἐξιλασμοῦ) seems rather a
  descriptive term than a proper name. Josephus (Ant. IV. x.) has no
  name for it, although he has for the others. In the Talmud, it is
  ימא “the Day,” יומא רבא “the Great Day,”
  צומא רבא, “the Great Fast.” In Acts xxvii. 9 we meet the
  phrase ἡ νηστεία, “the fast κατ’ ἐξοχήν.”
  Similarly in Philo, _De Septenario_, all the festivals have names
  except this, which is referred to simply as “the Fast.” It must be,
  however, evident that with the institution of other fasts, ἡ
  νηστεία would hardly be adequate. As a distinctive appellation,
  some other name had to be chosen.

  In the Pentateuch the term (שבת שבתון) is used of
  ordinary Sabbaths (Ex. xxxi. 15, xxxv. 2, Lev. xxiii. 3) as well as of
  the Atonement (Lev. xvi. 31, xxiii. 32). But the LXX expressly
  distinguishes the application of it to ordinary Sabbaths from its
  application to the Atonement. The former, it renders σάββατα
  ἀναπαύσεως, the latter σάββατα σαββάτων. This
  latter term may therefore be considered the specific designation of
  the Atonement Day, and it is so used by Philo, De Septen. 23,
  σάββατον σαββάτων, τῶν ἁγίων ἁγιώτεραι (ἑβδόμαδες).

  We may, therefore, assume that in the Greek-speaking Jewish community
  of Rome, σάββατα σαββάτων, “the Great Sabbath,” was the
  common designation—or at least a familiar designation—of the Day of
  Atonement. In that case it could scarcely be otherwise than familiar
  to those who had any dealings whatever with the Jews.

  Fuscus pretends to share a very general observance, and on the
  strength of it to be disinclined to discuss any personal matters with
  his friend. Can that day have been a simple Sabbath? The tone
  indicates a rarer and more solemn occasion. Besides, we are definitely
  told that it is a special Sabbath, the “thirtieth.”

  The Jews at that time seem to have reckoned their festivals by strict
  lunar months (Josephus, Ant. IV. x.) and their civil year by the
  Macedonian calendar. The thirtieth Sabbath, if we reckon by the Roman
  calendar, might conceivably have fallen on the Atonement. By the
  Macedonian or Athenian it could not have done so. However, as the
  Roman calendar was a solar one, the correspondence of the thirtieth
  Sabbath with the Atonement can only have been a fortuitous one in a
  single year. _Tricesima sabbata_ can hardly apply to that.

  It is just possible that the reason for the word “thirtieth” is to be
  found in the widely and devoutly pursued astrology of that time. The
  number thirty had a certain significance in astrology, Firmicus
  Maternus, IV. xvii. 5; xxii. 3. If for one reason or another the
  _mansio_ of the moon, which coincided with the second week of the
  seventh lunar month (cf. Firm. Mat. IV. i. seq. for the importance of
  the moon in astrology), bore the number thirty, then _tricesima
  sabbata_, to initiated and uninitiated, might bear the portentous
  meaning required for the Horatian passage.

  Whether that is so or not, the only Sabbath which we know to have been
  specially singled out from the rest of the year, was this
  σάββατα σαββάτων, the Day of Atonement. Whatever reason
  there was for calling it the thirtieth, the mere fact of its being
  particularly designated makes it likely that Horace referred to that
  day.

  Nearly every one of the festivals in Tishri has already been suggested
  for the phrase, but these results have been reached by elaborate and
  intricate calculations, which bring the thirtieth Sabbath on the
  festival required. The main difficulty with all such calculations has
  been noted. The coincidence can only have been exceptional, and an
  exceptional coincidence will not help us here. Some especially
  rigorous Jews undoubtedly fasted every week like the Pharisee in Luke
  xviii. 11-20, but that was intended as a form of asceticism. The
  custom survived in some Christian communities, notably in Rome, which
  elevated it almost to a dogma, so that Augustine had to combat the
  point with especial vigor. (Ep. xxxvi., and Casulanum, Corp. Scr.
  Eccl. xxxiv. pp. 33 seq.) It may be interesting to remember that from
  a passage of this epistle referring to this Sabbath fast (xiv. 32) is
  derived the famous proverb, “When you are in Rome, do as the Romans
  do.”

Footnote 266:

  Sat. I. iv. 18.

Footnote 267:

  Sat. I. v. 97.

Footnote 268:

  Apellas is a common name for a slave or freedman. Cic. ad Fam. vii.
  25; C. I. L. x. 6114. That a Jew should bear a name derived from that
  of Apollo, is not at all strange. Cf. ch. IX., n. 6.

Footnote 269:

  Cf. Ep. I. vi. 1 seq. The _nil admirari_ of the first line is Horace’s
  equivalent for the ἀταραξία of Epicurus.

Footnote 270:

  As is stated in the text, the _peregrina Sabbata_ and the _septima
  festa_, which is merely a metrical paraphrase for _Sabbata_, are
  treated here as of annual occurrence. The word _redeunt_ itself points
  to that. It has been suggested in Note 264, that the great annual
  Sabbath was the Day of Atonement. If that is referred to here, the
  application is very natural. The season of the Tishri festivals
  coincided in the Mediterranean with rather severe storms. These
  generally began after the Day of Atonement, so that among Jews sailing
  was rarely undertaken after that day. This is strikingly shown by Acts
  xxvii. 9. But the equinoctial storms, while sufficient to make a
  sea-voyage dangerous, do not seem to have caused serious discomfort on
  land. The reference, accordingly, must in each case be understood from
  its context. In the first the courtship is to be begun, _tu licet
  incipias_, at the great Sabbath, to take advantage of the exquisite
  autumn of Italy. In the second, the voyage is not to be deferred even
  for this same Sabbath, which ordinarily marked the danger line of
  navigation.

Footnote 271:

  Vogelstein u. Rieger, Gesch. der Jud. in der Stadt Rom., p. 39 seq.

Footnote 272:

  Reinach, Textes, p. 259.

Footnote 273:

  Pliny, Hist. Nat. XXIX. i. 6. Plaut. Amphitruo, 1013.

Footnote 274:

  Cf. Garrucci, Cimitero ... in Signa Randanini; F. X. Kraus, Roma Sott.
  p. 286 ff.; Garucci, Storia del arte Cristiana, VI. tav. 489-491.


                              CHAPTER XVII

                 THE JEWS OF THE EMPIRE TILL THE REVOLT

Footnote 275:

  Verg. Ecl. i. 6-7; Georg. i. 503; Horace, Odes, I. ii. 43; Ovid, Ex
  Ponto, ii. 8.

Footnote 276:

  Xen. An. IV. i. 2-3.

Footnote 277:

  Cic. ad Att. i. 1.

Footnote 278:

  While notoriously corrupt governors like Cotta (130 B.C. E.), Cic. Pro
  Mur. 58, and Aquilius (126 B.C.E.), Cic. Div. in Caec. 69, were
  acquitted, a rigidly honest man like Rufus was convicted under such a
  charge. Dio Cassius, fr. 97.

Footnote 279:

  Ditt. Or. inscr. no. 456, l. 35; from Mytilene, 457, 659.

Footnote 280:

  The Edict of Caracalla, called the Constitutio Antonina or
  Antoniniana, has been known in substance for a long time. Recently
  fragments of its exact words in Greek were discovered in a papyrus
  (Giessen, Pap. II. (P. Meyer), p. 30 seq): δίδωμι τοῖς συνάπασιν
  ξένοις τοῖς κατὰ τὴν οἰκονμένην πολίτειαν Ρωμαίων μένοντος παντὸς
  γένους πολιτευμάτων χωρὶς τῶν δεδειτικίων. The exact effect of
  the decree is not yet quite clear. It seems evident that the
  _dediticii_ were excluded.

Footnote 281:

  Dio Cassius, xxxvi. 6.

Footnote 282:

  Suet. Aug. 93.

Footnote 283:

  Josephus, Ant. XIV. x.; XII. iii. 2.

Footnote 284:

  The “heterodox Jewish propaganda” is of course Christianity. The
  success of Paul and other missionaries in Asia Minor is best indicated
  by the churches of Asia to which Revelations is addressed.

Footnote 285:

  Horace, Ep. II. ii. 184. The sumptuous present of Aristobulus, which
  formed part of Pompey’s triumphal procession, Josephus, Ant. XIV. iii.
  1. Pliny, Hist. Nat. XXXVII. ii. 12, must have made the Jewish kings
  symbols of enormous wealth. None the less, Herod’s unsparing severity
  toward his own sons was also well known, and it is said to have
  elicited from Augustus the phrase _mallem Herodis porcus esse quam
  filius_—Macrob. Sat. II. iv. 11—a jest which, as Reinach points out
  (Textes, p. 358), is of doubtful authenticity, and certainly not
  original.

Footnote 286:

  Josephus, Ant. XX. iii.

Footnote 287:

  Judea herself was free from tribute, but Herod was responsible for
  certain Arab revenues. Besides, he received from Augustus a number of
  Greek towns (Josephus, Wars, I. xx. seq.), and his kingdom included
  further Batanaea south of Damascus, Galilee, and Peraea, the Greek
  cities across the Jordan and south through Idumaea. All this was held
  by him as the acknowledged beneficiary of Rome (Josephus, Ant. XV. vi.
  7).

Footnote 288:

  Josephus, Ant. XV. i. 2.

Footnote 289:

  Josephus, Ant. XVII. vi. 6.

Footnote 290:

  Cf. ch. XI., n. 15. Cf. also Josephus, Ant. XVII. x.

Footnote 291:

  Not merely composed of Herod’s old soldiers (Josephus, Ant. XVII. x.
  4). Matt. xxii. 16; Mark iii. 6; xii. 13.

Footnote 292:

  Madden, Coins of the Jews. Cf. also Josephus, Ant. XVIII. iii. 1.

Footnote 293:

  Josephus, Ant. XX. viii. 11.

Footnote 294:

  Josephus, Ant. XX. v. 4.

Footnote 295:

  Josephus, Ant. XV. xi. 15.

Footnote 296:

  Josephus, Ant. XVI. vii.-viii. seq. The many children of Herod’s ten
  wives were in almost constant intrigues against him and one another.

Footnote 297:

  Strabo, xvi. 755.

Footnote 298:

  It is necessary at every point to note the uncertain character of our
  evidence. The _Historiae Philippicae_ of Pompeius Trogus written under
  Augustus would have been of inestimable value for us, if we had them
  in full. But we possess them merely in the summary of Justin (third
  century?), which gives us all the substance, but little or none of the
  personality of the writer. And in this case the loss is the more
  serious because Trogus seems to have had a keener feeling for the
  dramatic character of events and a broader sympathy than many other
  ancient historians.

Footnote 299:

  Josephus, Ant. XVII. x. 9.

Footnote 300:

  This is the Varus made famous in the Teutoburg battle. The
  insurrection mentioned in the text is the _polemos shel Varos_ of the
  Seder Olam.

Footnote 301:

  Caesar, Bell. Gall. iii. 10.

Footnote 302:

  Josephus, Ant. XVII. x. 9.

Footnote 303:

  Nicolaus of Damascus, philosopher and historian, was Herod’s principal
  Greek adviser and the advocate of the Jews in many public
  controversies. As far as we can judge from fragments, his History of
  the World, in no less than 114 Books, was a loosely connected
  compilation rather than a work of literary merit.

Footnote 304:

  Josephus, Ant. XVIII. i. 1 and 6.

Footnote 305:

  A complete investigation of this subject is contained in Domaszewski,
  Die Religion des römischen Heeres.

Footnote 306:

  Cagnat. in Dar.-Sagl. Dict. des ant. s. v. legio, p. 1084.

Footnote 307:

  The signa were actually worshiped by the soldiers. They are the
  _propria legionum numina_. Tac. Ann. ii. 17. Cf. Cagnat., _op. cit._
  p. 1065. Domaszewski, _op. cit._ p. 115.

Footnote 308:

  To the sense and tact of this typical Roman official the averting of a
  crisis in the history of Palestinian Jewry is due. The rebellion which
  Gaius would undoubtedly have provoked might have dragged other parts
  of the world with it, and at that time the conditions were less
  favorable for re-establishment of the empire than in 68 C.E.

Footnote 309:

  Josephus, Ant. XVIII. vii. 2.

Footnote 310:

  Josephus, Ant. XIX. vi.

Footnote 311:

  That Tacitus shows a strong antipathy to the Jews can scarcely be
  questioned. It is in these chapters (Hist. v. 2. seq.) more than most
  others, that we are able to see the rhetorical historian of ancient
  times almost in the act of preparing his narrative. The sources of
  Tacitus are open to us. That he used Manetho and Apion instead of
  Josephus and Nicolaus is itself ample indication of the complete lack
  of conscience with which such a writer could select his evidence
  according to the thesis he meant to establish.

Footnote 312:

  Cagnat. Inscr. Gr. ad res Rom. pertin. ii. n. 176.

Footnote 313:

  Cf. for the Jewish feeling toward him, Jos. Ant. VI. i. 2; Ketub. 17a;
  Pes. 88b. He is represented as a rigidly observant and pious Jew.
  However, the boon companion of the young Gaius and the voluptuaries of
  the imperial court must have undergone an overwhelming change of heart
  if he was really worthy of the praise lavished upon him.

Footnote 314:

  Josephus, Ant. XIX. vii.

Footnote 315:

  Josephus, Ant. XX. i. One of the slain rioters is named Hannibal.

Footnote 316:

  Josephus, Ant. XX. v.

Footnote 317:

  Josephus, Ant. XX. viii.


                             CHAPTER XVIII

                         THE REVOLT OF 68 C.E.

Footnote 318:

  Cf. Livy, Books XXXIX and XL.

Footnote 319:

  Tac. Ann. iii. 40 seq.; _ibid._ ii. 52; iv. 23. In 52 C.E., Cilicia
  rose in revolt; _ibid._ xii. 55. The Jewish disturbances of the same
  year are alluded to in Tac. Ann. xii. 54—a passage omitted in Reinach.

Footnote 320:

  Josephus, Wars, II. xvi.

Footnote 321:

  The entire life of this curious impostor, as portrayed by Lucian, is
  of the highest interest. The maddest and most insolent pranks received
  no severer punishment than exclusion from Rome.

Footnote 322:

  C. I. L. vii. 5471.

Footnote 323:

  For the Armenian, British, etc., rebellions, see Suet. Nero, 39, 40.
  In at least one other part of the empire, prophecy and poetry
  maintained the hope of an ultimate supremacy, something like the
  Messianic hope of the Jews. This was in Spain, and upon this fact
  Galba laid great stress. Suet. Galba, 9: _Quorum carminum sententia
  erat, oriturum quandoque ex Hispania principem dominumque rerum_.

Footnote 324:

  Suetonius speaks first of the joy shown at his death, then of the
  grief. It is, however, easy to see that the latter manifestation was
  probably the more genuine and lasting.

Footnote 325:

  Josephus, Ant. XX. viii. 11; Vita, 3.

Footnote 326:

  We learn from the same passage that a great many accounts of Nero
  existed, and many of them were favorable. The implication further is
  that these accounts were written after his death. We have only the
  picture drawn by Tacitus and Suetonius. If we had one written from the
  other side, like Velleius Paterculus’ panegyric of Tiberius (Vell.
  Pat. ii. 129 seq.), we should be better able to judge him.

Footnote 327:

  Gittin 56a.

Footnote 328:

  Reinach, Textes, pp. 176-178.

Footnote 329:

  Neither the arch nor the inscription exists any longer. A copy of the
  inscription was made, before the ninth century, by a monk of the
  monastery of Einsiedeln, to whose observation and antiquarian interest
  we owe more than one valuable record.

Footnote 330:

  The phrase _Iudaica superstitione imbuti_, already quoted, shows what
  the term would be likely to suggest to Roman minds. In Diocletian’s
  time, when the Persians were the arch-enemies of Rome, and Persian
  doctrine in the form of Manicheism was widely spread over the empire,
  the emperors did not hesitate to call themselves _Persicus_. But
  _Persicus_ never meant an adherent of a religious sect.

Footnote 331:

  _Idumaea_ is used for _Iudaea_ in Statius Silvae, iii. 138; v. 2, 138;
  Valerius Flaccus, Argon. 12.


                              CHAPTER XIX

             THE DEVELOPMENT OF THE ROMAN JEWISH COMMUNITY

Footnote 332:

  Philo, Leg. ad Gaium, 24.

Footnote 333:

  We may compare such expressions as _magica arte infecti_, Tac. Ann.
  ii. 2; Cic. Fin. III. ii. 9.

Footnote 334:

  Long before the attempts made in the nineteenth century to
  rehabilitate all the generally acknowledged historical monsters,
  historians had looked askance at the portrait of Tiberius drawn by
  Tacitus. For a recent discussion, cf. Jerome, The Tacitean Tiberius,
  Class. Phil. vii. pp. 265 seq.

Footnote 335:

  Suet. Tib. 36. The _mathematici_ are strictly the astrologers whose
  science was called μάθησις. Cf. the title of Firmicus
  Maternus, _Matheseos libri_. The governmental attempt to suppress the
  _mathematici_ was a total failure, but the law’s attitude toward them
  may be seen from the rescript of Diocletian (294 C.E.): _ars
  mathematica damnabilis interdicta est_ (Cod. Just. IX. xviii. 2).

Footnote 336:

  Nero assigned Sardinia to the senate as ample satisfaction for Achaea,
  which he took under his own jurisdiction.

Footnote 337:

  Acts xi. 26; xxvi. 28. Ιησοῦ χρήστου in the inscription
  quoted in n. 10. In this case the identification of names may be due
  to iotacism.

Footnote 338:

  Cf. the well-known rhetorician Philostr. Vita. Soph. ii. 11, and in
  Rome itself Inscr. gr. Sic. et Ital. 1272; and _ibid._ 2417, 2.

Footnote 339:

  The question of the authenticity and date of the Acts does not belong
  to this study. A thorough discussion will be found in Wendland, Die
  urchristlichen Literaturformen, 3rd p. 314 seq.

Footnote 340:

  Acts xi. 19; xiii. 5, 50.

Footnote 341:

  συναγωγή = ἐκκλησία. Le Bas, 2528 (318
  C.E.), a Marcionite association.

Footnote 342:

  There was a jurist Tertullian of whom some fragments have been
  preserved in the Digest (29, 2, 30; 49, 17, 4). He has on plausible
  grounds been assumed to be the same as the Church Father. There can be
  no question that the latter had legal training. As for the cruelties
  described by Tacitus, it may be said that Eusebius has no word of
  them, even in his denunciation of Nero. (Hist. Eccl. II. xxv.)

Footnote 343:

  All the Church Fathers mention these outrageous charges. Pliny (Ep. x.
  96) refers vaguely to wickednesses charged against them, but the
  _flagitia cohaerentia nomini_ are more likely to be the treasonable
  machinations which the Christian associations were assumed to be
  engaged in than these foul and stupid accusations. It will be
  remembered that Tertullian (_loc. cit._) is more eager to free the
  Christians from the charge of treason than of any other. Treason in
  this case, however, meant not sedition or rebellion, but anarchy,
  _i.e._ attempts at the destruction of the state. The attitude of
  medieval law toward heresy gives a good analogy.

Footnote 344:

  It would scarcely be necessary to refute this slander, if it had not
  recently renewed currency; Harnack, Mission and Ausbreitung.
  Tertullian knows nothing of it, nor Eusebius, although the latter
  refers in the case of Polycarp to Jewish persecution of Christians
  (Hist. Eccl. IV. xv. 29). Tertullian, on the contrary, implies that an
  enemy of the Jews would be likely to be a persecutor of Christians
  (Apol. 5).

Footnote 345:

  Like most men of his time he bore two names, his native name of Saul
  and the name by which he was known among Christians, Paul. This is
  indicated by the phrase Σαῦλος ὁ καὶ Παῦλος (Acts xiii.
  9), which is the usual form in which such a double name was expressed.

Footnote 346:

  The mother church at Jerusalem consisted exclusively of Jews until the
  time of Hadrian (Euseb. Hist. Eccl. IV. v. 2).

Footnote 347:

  Quint. Inst. X. i. 93.

Footnote 348:

  Maecenas, too, was of the highest Etruscan nobility. Horace, Sat. I.
  vi. 1 seq. The antiquity of Etruscan families was proverbial among the
  Romans.

Footnote 349:

  Mommsen seeks to make his crabbed style a racial characteristic. The
  statement is quite gratuitous. His peculiarity of expression is amply
  explained by his youth, his lack of literary practice, and his
  absorption in his philosophical pursuits.

Footnote 350:

  Pers. v. 176. Reinach, Textes, p. 264.

Footnote 351:

  Strabo apud Jos. Ant. XIV. vii. 2: καὶ τόπον οὐκ ἔστι ῥᾳδίως
  εὑρεῖν τῆς οίκουμένης ὅς οὐ παραδέδεκται τοῦτο τὸ φῦλον μηδ’
  ἐπικρατεῖται ὑπ’ αὐτοῦ. Seneca apud Aug. De Civ. Dei, vi. 10:
  _Cum interim usque eo sceleratissimae gentis consuetudo valet ut per
  omnes iam terras recepta sit; victi victoribus leges dederunt_.

Footnote 352:

  Besides the capital passage (Sat. xiv. 96) Juvenal speaks of Jews in
  Sat. iii. 10 seq., 296; vi. 156, 542.

Footnote 353:

  Cf. Garrucci, Cimitero ... in Signa Randanini; Rossi, Roma Sotteranea,
  especially the Indices. As late as 296 C.E. the epitaph of the Bishop
  of the Roman church is given in Greek.


                               CHAPTER XX

                     THE FINAL REVOLTS OF THE JEWS

Footnote 354:

  Perhaps the “egg laid on the Sabbath” would have excited less comment,
  if the fact were kept in mind that a decision in a specific case can
  hardly fail to be particular.

Footnote 355:

  C. I. L. ix. 1. 26.

Footnote 356:

  Laius outraged Chrysippus, son of Pelops, who had been left in his
  care. The Euripidean lost play on Oedipus seems to have adopted that
  version. Pisander, Schol. Eur. Phoen. 1760: πρῶτος δὲ Λάιος τὸν
  ἀθέμιτον ἔρωτα τοῦτον ἔσχεν.

Footnote 357:

  Cf. Philo, De Spec. Leg. 7.

Footnote 358:

  Tosefta Ab. Zar. ii. 6.

Footnote 359:

  Ziebarth, Kulturbilder aus griechischen Städten, p. 73.

Footnote 360:

  In very much earlier times Jews left dedications in the temple of Pan
  Euhodus. Ditt. Inscr. Or. 74: Θεύδοτος Δωρίωνος Ἰουδαῖος σωθεὶς
  ἐκ πελάγου. Cf. 73, Πτολεμαῖος Διονυσίου Ἰουδαῖος.

Footnote 361:

  This became a standing formula and in inscriptions is regularly
  abbreviated N. K. C. (Valerius Probus, 4), _i.e. non kalumniae causa_.
  The use of _k_ for _c_ testifies to the antiquity of the formula.

Footnote 362:

  Suet. Domit. 12.

Footnote 363:

  Dio Cassius (Xiph.), lxvii. 14.

Footnote 364:

  Passed in 81 B.C.E. This law punished offenses as diverse as murder,
  arson, poisoning, perjury, abortion, and abuse of magisterial power.
  In every case it was the effect of the act that was considered.

Footnote 365:

  Reinach, Textes, p. 197, n. 1.

Footnote 366:

  The _polemos shel kitos_ of Mishnah Sota ix. 14 and the Seder Olam.

  Quietus was a Moorish chieftain of great military ability. He seems to
  have hoped for the succession to the throne. After the end of the
  revolt he was transferred to his native province, Mauretania, by
  Hadrian, and was ultimately executed for treason.

Footnote 367:

  Meg. Taan., Adar 12; Grätz, Gesch. der Juden, 3rd iv. 445 seq.

Footnote 368:

  In the case of non-Jews, the Messianic hope was simply the dread of an
  impending cataclysm. As far as this dread was connected with the
  failure of the Julian line, it proved groundless. But the Jewish
  Apocrypha and Pseudepigrapha of this time are full of prophecies of
  the end of the world. It was the general belief that the world was
  very old, and that a fixed cycle, then rapidly coming to its end,
  determined the limits it would reach.

Footnote 369:

  Jerus. Taan. iv. 7, p. 68 d. Ekah Rab. ii. 1.

Footnote 370:

  Dio Cassius (Xiph.), lxix. 12; Reinach, Textes, p. 198.

Footnote 371:

  Dig. 50, 15, 1, 6.

Footnote 372:

  Euseb. Hist. Eccl. IV. vi. 4.

Footnote 373:

  Gen. Rab. lxiii. (xxv. 23) makes Hadrian the typical heathen king, as
  Solomon is the typical Jewish king. His name is followed, as is that
  of Trajan, by a drastic curse. But there are traditions of a kindlier
  feeling toward him. Sibyl. v. 248. In the Meg. Taan. the 29th of Adar.

Footnote 374:

  Eusebius, Hist. Eccl. IV. vi., quoting Aristo of Pella. Jerome in
  Ezek. i. 15. It is here that the famous passage of Jerome occurs,
  which describes the Jews as “buying their tears.” Cf. also Itiner.
  Burdigal. (Hierosolymitanum), I. v. 22.

Footnote 375:

  Vopiscus, Vita Saturn. viii.; Reinach, Textes, p. 326. The
  authenticity of this letter has been questioned, but the transmission,
  although indirect, is better documented than in most such cases.
  Hadrian is known to have written an autobiography, and Phlegon, his
  freedman, who also wrote his life, no doubt used it. Spartianus, Hadr.
  i. 1; xiv. 8.

Footnote 376:

  The writers Spartianus, Capitolinus, etc., dedicate their work to
  Diocletian or Constantine. It was suggested by Dessau, Hermes, 24,
  337, that these writers never existed, and were invented by a forger
  of a century later. Mommsen, Hermes, 25, 298, assumed their existence,
  but regarded the extant works as revised at the time mentioned by
  Dessau. Other investigators, except H. Peter, accept Mommsen’s
  conclusions. Whether they are authentic or not, these biographies are
  alike wretched in style and thought.

Footnote 377:

  Paul, Sent. V. xxiii. 14; Dig. 48, 8, 3, 2; 8, 8. The date is not
  certain; Dig. 48, 8, 3, 4.

Footnote 378:

  B. G. U. 347, 82.

Footnote 379:

  Dig. 48, 8, 11. pr.

Footnote 380:

  Paul, Sent. V. xxii. 3.

Footnote 381:

  Lampridius, Vita Alex. 22.

Footnote 382:

  Jews made converts even after the prohibition of Theodosius (Jerome,
  Migne Patrol, 25, p. 199; 26, p. 311). One further ground for doubting
  the statement of Paul as it appears in the extant texts is the
  following: In the Digest (48, 8, 4, 2) it is only the physician and
  the slave that are capitally punished for castration. The owner of the
  slave (_ibid._ 48, 8, 6) is punished by the loss of half his property.
  Further, the penalty for circumcision is stated to be the same as that
  for castration. That was the case not only in Modestinus’ time, who
  lived after Paul, but as late as Justinian, since it is received into
  the Digest. Yet Paul, according to the extant text, makes the
  circumcision of alien slaves a capital crime (V. xxii. 4). The
  discrepancy can scarcely be reconciled.

Footnote 383:

  Capitol. Antoninus Pius, 5.

Footnote 384:

  193 C.E. It was on this occasion that the Pretorians offered the
  imperial purple to the highest bidder.

Footnote 385:

  Josephus, Ant. XIV. x.

Footnote 386:

  The legend of Polycarp assumes a large and powerful Jewish community.
  In late Byzantine times, the Jews of Asia Minor were still a powerful
  factor. The emperor Michael II, a Phrygian, was suspected of Jewish
  leanings; Theophanes (Contin.), ii. 3 ff.


                              CHAPTER XXI

           THE LEGAL POSITION OF THE JEWS IN THE LATER EMPIRE

Footnote 387:

  The theory advanced by Wilcken-Mitteis (Grundzüge und Chrestomathie
  der Pap. vol. I.) that all who paid a poll-tax were _dediticii_, and
  therefore excluded from the Const. Ant. is wholly gratuitous. There is
  no evidence whatever connecting the _dediticii_ with the poll-tax.

Footnote 388:

  There are few reliable statements in the extant texts for estimating
  the population. Beloch’s work on the subject puts all the data
  together, but nothing except uncertain conjectures can be offered.

Footnote 389:

  Lanciani, Ancient Rome, pp. 50-51; Pelham, Essays on Roman History,
  pp. 268 seq.

Footnote 390:

  Lampridius, Alex. 33: _corpora omnium constituit vinariorum_ ... _et
  omnino omnium artium_.

Footnote 391:

  These are the _collegia, idcirco instituta ut necessariam operam
  publicis utilitatibus exhiberent_ (Dig. 50, 6, 6, 1). They are the
  transportation companies and others engaged in caring for and
  distributing the _annona_, the fire companies and the burial
  associations of the poor. Cf. C. I. L. vi. 85, 29691; x. 1642, xiv.
  2112.

Footnote 392:

  The _institutio alimentaria_ commemorated on the marble slabs
  (_anaglypha_) in the Forum and by the bronze tablets of Veleia and the
  Baebiani (C. I. L. ix. 1147; xi. 1455). It had begun with Nerva:
  _puellas puerosque natos parentibus egestosis sumptu publico per
  Italiae oppida ali iussit_ (Aur. Vict., Nerva, xii.).

Footnote 393:

  An entire article of the Digest (26, 1) is devoted to the _tutela_.
  Another one (27, 1) deals with _excusationes_, which are mainly
  exemptions from the burden of the _tutela_.

Footnote 394:

  The distinction is thoroughgoing in the penal clauses cited in the
  Digest. It was already established in Trajan’s time (Plin. Ep. X.
  lxxix. 3). It is implied in Suetonius, Gaius, 27: _multos honesti
  ordinis_. It is doubtful, however, whether the distinction was already
  recognized in the time of Caligula.

Footnote 395:

  Gaius wrote about 150 C.E., probably in the eastern provinces.

Footnote 396:

  Abot ii. 5. The saying of Hillel has no direct reference to apostasy,
  and concerns rather arrogance or eccentricity of conduct. But it
  literally describes the act by which such a man as Tiberius Julius
  Alexander ceased to be classed as a Jew.

Footnote 397:

  Cf. Plutarch, Numa, 17; Dionys. Hal. iv. 43.

Footnote 398:

  Dig. 50, 2, 3, 3.

Footnote 399:

  Cod. Theod. viii. 14.

Footnote 400:

  Exodus xxi. 2; Josephus, Ant. IV. viii. 28.

Footnote 401:

  Bab. Bat. 3b; Gittin 46b. The duty was regarded as of the highest
  urgency.

Footnote 402:

  Vogelstein and Rieger, Gesch. der Juden, p. 61 seq. Friedländer,
  Darstellungen der Sitt. 7th i. p. 514.

Footnote 403:

  Ox. Pap. ii. no. 276.

Footnote 404:

  Aurelian reigned from 270-275 C.E. The _sol invictus_ whom he adored
  was probably the Baal of Palmyra. Cumont, Les rel. orient, pp. 170,
  367, n. 59.

Footnote 405:

  Cod. Theod. xvi. 4.

Footnote 406:

  In 311 C.E. Galerius, and in 318 C.E. Constantine and Licinius,
  legalized the practice of Christianity. In 380 C.E., by the edict of
  Thessalonica, most of the heathen practices became penal offenses.

Footnote 407:

  Every state as such had its characteristic and legally established
  state ritual. Many centuries later Gladstone, then “the rising hope of
  the stern and unbending Tories,” stated, as a self-evident
  proposition, that a government in its collective capacity must profess
  a religion (The Church in its Relation to the State, 1839).

Footnote 408:

  Cyprian. De catholicae ecclesiae unitate, ch. x.

Footnote 409:

  Matth. v. 13. Cf. generally the Pauline Epistles, _e.g._ II. Corinth.
  xiii. 13.


------------------------------------------------------------------------

                              BIBLIOGRAPHY


                              PERIODICALS


The Jewish Quarterly Review: First Series, London, 1889-1900. Second
Series, Philadelphia, 1910-date.

Revue de études juives, Paris, 1880-date.

Monatsschrift für Geschichte und Wissenschaft des Judentums, Breslau,
1851-date.


                             ENCYCLOPEDIAS


Jewish Encyclopedia: New York, 1901-1906.

Encyclopedia Biblica: London, 1899.

Hastings’ Dictionary of the Bible, 1901-1904.

Hastings’ Encyclopedia of Religion and Ethics, 1908. Not yet completed.

Daremberg-Saglio: Dictionnaire des antiquités grecques et romaines,
1877. Not yet completed.

Pauly-Wissowa: Realenzyklopädie, 1894. Not yet completed.

Schaff-Herzog-Hauck: Realenzyklopädie für protestantische Kirche und
Theologie. 3d ed. Eng. tr. 1908.


                        GENERAL REFERENCE BOOKS


Grätz: Geschichte der Juden (1873-1895). Eng. tr., History of the Jews
(1891).

Schürer: Geschichte des jüdischen Volkes im Zeitalter Jesu Christi (4th
ed.), 1901.

Juster: Les juifs dans l’empire romain, 1914.

Wendland: Die hellenistisch-römische Kultur in ihren Beziehungen zum
Judentum und Christentum, 1912.

Wendland-Poland-Baumgarten: Die hellenistische Kultur.

Friedländer: Darstellungen aus der Sittengeschichte Roms. Leipzig (7th
ed.). Eng. tr. London, 1909.

Cumont: Les religions orientales dans le paganisme romain, 1912.


------------------------------------------------------------------------




                                 INDEX


 Aboda Zara, 329.

 Abstraction, 42.

 Aelia Capitolina, 288, 342.

 Aeolian, 50.

 Africanus, the Younger, 45.

 Agatharchidas, 178.

 Agrippa I, king of the Jews, 283, 313.

 Agrippa II, king of the Jews, 290.

 Akiba, rabbi, 342.

 Akmonia, 327.

 Alexander (Jannai), king of the Jews, 274.

 Alexander Severus, Roman emperor, 346, 356.

 Alexander the Great, 37, 38, 52, 78, 212, 368.

 Alexandria, 91, 107 _seq._, 200, 229, 255, 339, 362.

 Allia, 252.

 Amalekites, 77.

 Antigonus, 128.

 Antigonus, king of the Jews, 267, 270.

 Antigonus of Socho, 128, 133, 386.

 Antinois, 107.

 Antioch, 119, 138, 282.

 Antiochus Cyzicenus, 63, 275.

 Antiochus Epiphanes, 135 _seq._, 205.

 Antiochus Sidetes, 63.

 Anti-Semitism, 97.

 Apamea, 226.

 Apella, 177, 248, 402.

 Aphrodite, 32, 114, 376.

 Apion, 168, 170, 189.

 Apocrypha, 18, 67, 331.

 Apollo, 27, 46, 168.

 Apollonius Molo, 170, 194.

 Appuleius Decianus, 223.

 Aramaic, 118.

 Archelaus, 278.

 Archigallus, 389.

 Aristobulus I, son of John Hyrcanus, 63.

 Aristobulus II, son of Alexander Jannai, 64, 267.

 Aristophanes, 24.

 Aristotle, 81, 84, 212.

 Armenia, 265.

 Army, Roman, 352.

 Artaxerxes Ochus, 61, 181.

 Artemis, 27, 140.

 Asclepius, 46.

 Asebeia, 34, 35, 123, 163 _seq._, 334, 376.

 Asia Minor, 58, 63, 331, 348.

 Asianism, 198.

 Ass, 168 _seq._

 Assideans. See Hasidim.

 Assuan, 60, 96.

 Astrology, 241, 243, 317, 407.

 Atheism, 100, 191 _seq._, 335.

 Athena, 33, 83.

 Athens, 52.

 Atonement, Day of, 399 _seq._

 Atticism, 198.

 Augustus, Roman emperor, 245, 254, 257, 294.

 Aurelian, Roman emperor, 366.

 Avaris, 173.


 Babylon, 56, 266.

 Bacchanalia, 166, 238, 310.

 Bagoas, 62.

 Barbarian, 49, 51.

 Bar Kochba. See Bar Kosiba.

 Bar Kosiba, 65, 342, 348.

 Bastian, 377.

 Ben Sira. See Jesus, son of Sira.

 Bible, 20, 59, 60.

 Byzantine, 411.


 Cabiri, 149.

 Caesar (Gaius Julius Caesar), 222, 244, 294.

 Caligula. See Gaius.

 Calumnia, 332 _seq._

 Camillus, 45.

 Candide, 85.

 Caphthor, 77, 381.

 Caracalla, 348, 351.

 Carthage, 45, 188.

 Cassius, Dio. See Dio Cassius.

 Catiline, 222, 260.

 Celsus, 255.

 Chaeremo, 201.

 Chaldeans, 244, 255.

 Charles, 19.

 Chrestus, 313.

 Christians, 313, 316, 346, 366.

 Cicero, 53, 196, 220 _seq._

 Circumcision, 80, 143, 345, 411.

 Citium, 114, 376.

 City-state, 69, 105 _seq._

 Claudius, Roman emperor, 284, 313.

 Clearchus, 84.

 Clemens. See Flavius Clemens.

 Clemens, of Alexandria, 86, 200.

 Clodius, 244.

 Constantine, 353.

 Constitutio Antoniniana, 262, 350, 361, 371, 403.

 Crassus, 265, 397.

 Credulity, 176, 271.

 Crete, 164, 186.

 Cybele, 46, 158, 161, 238, 242.

 Cynics, 158, 240.

 Cyprian, 366.

 Cyprus, 33, 338.

 Cyrene, 249, 338.


 Damocritus, 170, 189.

 Daniel, Book of, 52, 135.

 David, 73, 179.

 Dead Sea, 81.

 _Dediticii_, 361.

 Deification, 37 _seq._, 261.

 Delphi, 231, 378.

 Deme, 107.

 Demeter, 27.

 Demetrius, Jewish writer, 127.

 Demetrius (the Besieger), 38, 127, 212.

 Demosthenes, 53.

 Diagoras of Melos, 193.

 Diana. See Artemis.

 Diaspora, 60, 208, 369.

 Diatribe, 158.

 Dio Cassius, 300, 334, 338, 342.

 Diocletian, 350, 353, 365.

 Dionysus, 32, 35, 142, 171, 376.

 Dioscuri, 46.

 Domitian, 335.

 Dorian, 50.

 Druids, 142, 319.


 Ecclesia, 366, 408.

 Ecclesiasticus. See Jesus, son of Sira.

 Egypt, 80, 88, 97, 144, 152, 153, 166, 172, 249, 269, 305, 338, 345,
    353, 362.

 Eleazar, 206.

 Elephantine, 60, 96, 97, 180.

 Eleusis, 35, 152, 377.

 Eliphaz, 70.

 Ennius, 53.

 Epicurus, 164.

 Eschatology, 72, 161.

 Essenes, 279.

 Ethiopic, 19.

 Etruscans, 43, 210, 213, 321.

 Euhemerus, 47.

 Euphrates, river, 87.

 Eusebia, 339, 342.

 Euthyphro, 84.

 Exodus, 96, 99.

 Ezra, 57.


 Fast, 399.

 _Fiscus Iudaicus_, 332, 363.

 Flaccus, prefect of Egypt, 200.

 Flaccus, proconsul of Asia, 221.

 Flavius Clemens, 335, 336.

 Florus, Gessius, 283, 286.

 Formula, 43.

 Freedmen, 220, 245, 255, 307, 379.


 Gabinius, 233.

 Gaius, Roman emperor (Caligula), 282, 313.

 Gaius, Roman jurist, 361, 412.

 Galilee, 280.

 Gauls, 187, 232.

 Gerizim, 57, 138.

 Gods, 24, 40.

 Greek names, 123, 128.

 _Grundideen_, 40, 377.


 _Haberim._ See Pharisees.

 Hades, 152, 154.

 Hadrian, 107, 340, 343.

 Hannibal, 174, 188.

 Harmodius, 37.

 Hasidim, 130 _seq._, 147.

 Hasmoneans, 63, 74, 158.

 Hecataeus of Abdera, 92 _seq._, 176, 182.

 Helladius, 102.

 Hellene, 49.

 Hellenization, 79, 116, 133, 145, 207.

 Henotheism, 29.

 Heracles, 46.

 Hermippus, 89.

 Herod (the Great), king of the Jews, 264, 269, 322, 403.

 Herodotus, 80.

 Heroes, 29, 30, 36.

 Hillel, 69.

 Hindu, 85.

 _Historia Augusta_, 344, 357, 411.

 Homer, 23, 25, 49, 150, 184, 200.

 Horace (Quintus Horatius Flaccus), 245 _seq._, 321.

 Human sacrifices, 187.

 Hyksos, 98.

 Hyrcanus II, son of Alexander Jannai, 267.


 Idumaea, 157, 168, 270.

 Images, 273, 280.

 Immortality, 71, 153 _seq._, 237.

 Impiety. See Asebeia.

 Inhospitality, 93, 183 _seq._

 Ionian, 50.

 Isis, 161, 166, 251, 307, 311.

 Isocrates, 78.


 Jerusalem, 178, 224, 233.

 Jesus, founder of Christianity. See Christians.

 Jesus, son of Sira (Ecclesiasticus), 19, 67, 118.

 Joel, Book of, 78, 382.

 John, high priest, son of Simon (Hyrcanus), 63, 155, 274.

 Jonathan, Hasmonean prince, son of Mattathiah, 63.

 Jordan, 73.

 José ben Joëzer, 133.

 Joseph, son of Tobiah, Egyptian tax-farmer, 130.

 Josephus, Titus(?) Flavius, 18, 85, 99, 109, 155, 164, 193, 285, 289,
    296, 306, 373.

 Judah Makkabi, 63, 132, 214.

 Jupiter, 43.

 Juvenal, 54, 189, 323 _seq._


 Kautzsch, 19.


 _Lectisternium_, 46.

 Levant, 72.

 Lex Cornelia de Sicariis, 337, 344.

 Liby-Phoenicians, 53.

 Lucian, 192.

 Lysimachus, 195.


 Ma. See Cybele.

 Maccabeans. See Hasmoneans.

 Maccabees, First Book of, 74, 130, 132, 141, 180, 214.

 Maccabees, Second Book of, 131, 141, 180.

 Maccabees, Fourth Book of, 204, 205.

 Macedonians, 50, 108, 110.

 Macrobius, 45, 223.

 Maecenas, 245, 247, 263.

 Magi, 85.

 Magic, 41, 239.

 Manetho, 99.

 Marathon, 60.

 Martial, 302, 325 _seq._, 329.

 _Mathematici._ See Astrology.

 Mattathiah, 63, 74, 180.

 Megasthenes, 86.

 Megillat Taanit, 20, 410.

 Meïr, rabbi, 297.

 Meleager of Gadara, 177, 329.

 Menelaus, 37.

 Messiah, 72 _seq._, 293, 298, 319, 341, 370, 406, 410.

 Metics, 34, 109, 112.

 Miletus, 331.

 Minoan, 13, 77, 374.

 Misanthropy. See Inhospitality.

 Mishnah, 69, 328.

 Mithra, 241, 357.

 Mithradates, 63.

 Mnaseas, 168.

 Molech, 188, 393.

 Molo. See Apollonius Molo.

 Moloch. See Molech.

 Müller, Max, 375.

 Mysteries, 35, 152.

 Mythology, 25, 26, 44, 236.


 Names, 123, 128.

 Nasi, 265, 363.

 Naucratis, 104.

 Nehemiah, 57, 61.

 Nero, Roman emperor, 285, 294, 315 _seq._

 Nerva, 334.

 Nicarchus, 201.

 Nicocles, 387.

 Nicolaus, of Damascus, 277, 405.

 Nile, 91.


 _Olam ha-bo._ See Immortality.

 Orphism, 153.

 Osarsiph, 100.

 Osiris, 100, 115, 385.

 Ostia, 327.

 Ovid, 250 _seq._



 Pantheism, 31.

 Papyri, 339. See also Elephantine; Assuan.

 Parthians, 265, 297, 340, 370.

 Passover, 97.

 Paul, of Tarsus, 315, 320.

 Paul, Roman jurist, 338, 346.

 Pederasty, 160, 330.

 Pentateuch, 67.

 Pergamon, 107.

 Persians, 52, 108.

 Persius, 321 _seq._

 Petronius, legate of Syria, 282.

 Pharisees, 71, 155, 265, 283.

 Philistia, 72.

 Philo (of Alexandria), 17, 200, 227, 307, 373.

 Phoenicia, 77, 78.

 Pilate (Titus Pontius Pilatus), 280.

 Pirke Abot, 128.

 Plato, 42, 194.

 Pliny, 196.

 Plutarch, 171.

 Polis. See City-state.

 Polybius, 140, 141, 174.

 Polytheism, 160.

 Pompeius Trogus, 274, 404.

 Pompey, 64, 181, 215, 227.

 Poppaea Sabina, 316.

 Porphyrius, 81.

 Poseidon, 33.

 Posidonius, 169, 170, 203.

 Prayer, houses of, 69.

 Propaganda, 148 _seq._, 208, 240, 263, 370.

 Proselyte, 247, 296, 316, 389.

 _Proseucha_, 65.

 Psalms of Solomon, 216.

 Pseudepigrapha, 19.

 Ptolemies, 116, 133, 180.

 Ptolemy Philadelphus, 102.

 Ptolemy Philometor, 175, 178.

 Ptolemy Philopator, 182.

 Ptolemy Soter, 80, 91, 178.

 Pyrrhus, 212.

 Pythagoras, 89.


 Quietus, Lusius, 339.


 Ra, 38, 116.

 Race, 48 _seq._, 379.

 Reinach, Théodore, 17.

 Religion, 21, 22.

 Resurrection, 71, 155.

 Rhetoric, 85, 167, 173, 178, 391.

 Rhodes, 198.

 Ritual murder, 190.

 Rome, 63, 210 _seq._


 Sabazios, 161, 171, 179, 255.

 Sabbath, 143, 177, 181, 246 _seq._, 254, 321.
   See also Thirtieth Sabbath.

 Sabbatistae, 179.

 Sacrifice, 28.

 Sadducees, 155.

 Salamis (Cyprus), 340.

 Salvation, 150.

 Samaritans, 58, 138, 281, 285.

 Sambethe, 179.

 Sanhedrin, 265, 363.

 Sarapis, 114, 385.

 Sardinia, 307, 312.

 Satire, 246, 321.

 Scipionic Circle, 138.

 Scribes, 61.

 Scythians, 186, 190.

 Seder Olam, 20, 404.

 Sejanus, 312.

 Seleucia, 164.

 Seleucid, 63, 146.

 Seleucus, 38.

 Seneca, 310, 318, 324.

 Septuagint, 102.

 Shechem, 57.

 Sheol, 70, 150, 388.

 Sibyl, 179, 298.

 Sidon, 79, 83, 382.

 Simon, high priest, son of Mattathiah, 230.

 Slaves, 219, 237, 309, 352, 364.

 Socrates, 84, 193.

 Sodom, 330.

 Sparta, 51, 151, 186.

 Standards, 280.

 Stoics, 204, 240, 322.

 Strabo, 186, 249.

 Suetonius, 295, 305, 317.

 Suidas, 170.

 Synagogue, 254, 277, 362.

 Syria, 76 _seq._, 215, 264, 277.

 Syrians, 216, 233, 239, 244, 365, 380.

 Tacitus, 102, 170, 189, 283, 307, 317.

 Talmud, 20.

 Tertullian, 318, 408.

 Theodore of Cyrene, 193.

 Theodosius, 347, 366.

 Theophrastus, 81.

 Theudas, 284.

 Thiasi, 244, 357.

 Thirtieth Sabbath, 399 _seq._

 Thracians, 187, 188.

 Tiberius, 200, 254, 304.

 Titus, Roman emperor, 300 _seq._

 Tosefta, 331.

 Trajan, 338.

 Trogus. See Pompeius Trogus.

 Trojan, 49.

 Typhon, 172.

 Tyre, 78.


 Valerius Maximus, 255.

 Varro, Marcus Terentius, 234.

 Varus, Publius Quintilius, 276.

 Veii, 45.

 Vespasian, 341.


 Wisdom of Solomon, 159, 242.


 Xenophanes, 31.

 Xenophon, 194, 258.

 Xiphilinus, 335, 338.


 Yavan, 50, 79.


 Zabidus, 168.

 Zealot, 288.

 Zechariah, 73, 87.

 Zerubbabel, 59.

 Zeus, 28, 138.

 Zion, 268.




                       _The Lord Baltimore Press_

                         BALTIMORE, MD., U.S.A.




------------------------------------------------------------------------




 ● Transcriber’s Notes:
    ○ Missing or obscured punctuation was silently corrected.
    ○ Typographical errors were silently corrected.
    ○ Inconsistent spelling and hyphenation were made consistent only
      when a predominant form was found in this book.
    ○ Text that was in italics is enclosed by underscores (_italics_);
      text that was bold by “equal” signs (=bold=).





End of Project Gutenberg's The Jews among the Greeks and Romans, by Max Radin