The Project Gutenberg eBook of The Composition of Indian Geographical Names, by J. Hammond Trumbull

Transcriber's Note: The oo ligature in the original book is here represented as [oo]. See footnote 4.

ON THE COMPOSITION OF

INDIAN GEOGRAPHICAL NAMES.

A proper name has been defined to be "a mere mark put upon an individual, and of which it is the characteristic property to be destitute of meaning."[1] If we accept this definition, it follows that there are no proper names in the aboriginal languages of America. Every Indian synthesis—names of persons and places not excepted—must "preserve the consciousness of its roots," and must not only have a meaning but be so framed as to convey that meaning with precision, to all who speak the language to which it belongs. Whenever, by phonetic corruption or by change of circumstance, it loses its self-interpreting or self-defining power, it must be discarded from the language. "It requires tradition, society, and literature to maintain forms which can no longer be analyzed at once."[2] In our own language, such forms may hold their places by prescriptive right or force of custom, and names absolutely unmeaning, or applied without regard to their original meaning, are accepted by common consent as the distinguishing marks of persons and places. We call a man William or Charles, Jones or Brown,—or a town, New Lebanon, Cincinnati, Baton Rouge, or Big Bethel—just as we put a number on a policeman's badge or on a post-office box, or a trademark on an article of merchandise; and the number and the mark are as truly and in nearly the same sense proper names as the others are.

Not that personal or proper names, in any language, were originally mere arbitrary sounds, devoid of meaning. The first James or the first Brown could, doubtless, have given as good a reason for his name as the first Abraham. But changes of language and lapse of time made the names independent of the reasons, and took from them all their significance. Patrick is not now, eo nomine, a 'patrician;' Bridget is not necessarily 'strong' or 'bright;' and in the name of Mary, hallowed by its associations, only the etymologist can detect the primitive 'bitterness.' Boston is no longer 'St. Botolph's Town;' there is no 'Castle of the inhabitants of Hwiccia' (Hwic-wara-ceaster) to be seen at Worcester; and Hartford is neither 'the ford of harts,' (which the city seal has made it,) nor 'the red ford,' which its name once indicated.

In the same way, many Indian geographical names, after their adoption by Anglo-American colonists, became unmeaning sounds. Their original character was lost by their transfer to a foreign tongue. Nearly all have suffered some mutilation or change of form. In many instances, hardly a trace of true original can be detected in the modern name. Some have been separated from the localities to which they belonged, and assigned to others to which they are etymologically inappropriate. A mountain receives the name of a river; a bay, that of a cape or a peninsula; a tract of land, that of a rock or a waterfall. And so 'Massachusetts' and 'Connecticut' and 'Narragansett' have come to be proper names, as truly as 'Boston' and 'Hartford' are in their cis-Atlantic appropriation.

The Indian languages tolerated no such 'mere marks.' Every name described the locality to which it was affixed. The description was sometimes topographical; sometimes historical, preserving the memory of a battle, a feast, the dwelling-place of a great sachem, or the like; sometimes it indicated one of the natural products of the place, or the animals which resorted to it; occasionally, its position or direction from a place previously known, or from the territory of the[Pg 5] nation by which the name was given,—as for example, 'the land on the other side of the river,' 'behind the mountain,' 'the east land,' 'the half-way place,' &c. The same name might be, in fact it very often was, given to more places than one; but these must not be so near together that mistakes or doubts could be occasioned by the repetition. With this precaution, there was no reason why there might not be as many 'Great Rivers,' 'Bends,' 'Forks,' and 'Water-fall places' as there are Washingtons, Franklins, Unions, and Fairplays in the list of American post-offices.

With few exceptions, the structure of these names is simple. Nearly all may be referred to one of three classes:

I. Those formed by the union of two elements, which we will call adjectival and substantival;[3] with or without a locative suffix or post-position meaning 'at,' 'in,' 'by,' 'near,' &c.

II. Those which have a single element, the substantival or 'ground-word,' with its locative suffix.

III. Those formed from verbs, as participials or verbal nouns, denoting a place where the action of the verb is performed. To this class belong, for example, such names as Mushauwomuk (Boston), 'where there is going-by-boat,' i.e., a ferry, or canoe-crossing. Most of these names, however, may be shown by rigid analysis to belong to one of the two preceding classes, which comprise at least nine-tenths of all Algonkin local names which have been preserved.

The examples I shall give of these three classes, will be taken from Algonkin languages; chiefly from the Massachu[Pg 6]setts or Natick (which was substantially the same as that spoken by the Narragansetts and Connecticut Indians), the Abnaki, the Lenni-Lenâpe or Delaware, the Chippewa or Ojibway, and the Knisteno or Cree.[4]

Of names of the first class, in central and southern New England, some of the more common substantival components or 'ground-words' are those which denote Land or Country, River, Water, Lake or Pond, Fishing-place, Rock, Mountain, Inclosure, and Island.

1. The Massachusetts ohke (Narr. aûke; Delaware, hacki; Chip. ahke; Abnaki, 'ki;) signifies land, and in local names, place or country. The final vowel is sometimes lost in composition. With the locative suffix, it becomes ohkit (Del. hacking; Chip. ahkiⁿ; Abn. kik;) at or in a place or country.

To the Narragansetts proper, the country east of Narragansett Bay and Providence River was waⁿpan-auke, 'east land;' and its people were called by the Dutch explorers, Wapenokis, and by the English, Wampanoags. The tribes of the upper St. Lawrence taught the French, and tribes south of the Pis[Pg 7]cataqua taught the English, to give the name of East-landers—Abenaquis, or Abinakis—to the Indians of Maine. The country of the Delawares was 'east land,' Wapanachki, to Algonkin nations of the west.

The 'Chawwonock,' or 'Chawonocke,' of Capt. John Smith,—on what is now known as Chowan River, in Virginia and North Carolina,—was, to the Powhattans and other Virginian tribes, the 'south country,' or sowan-ohke, as Eliot wrote it, in Gen. xxiv. 62.

With the adjectival sucki, 'dark-colored,' 'blackish,' we have the aboriginal name of the South Meadow in Hartford,—sucki-ohke, (written Sicaiook, Suckiaug, &c.), 'black earth.'

Wuskowhanan-auk-it, 'at the pigeon country,' was the name (as given by Roger Williams) of a "place where these fowl breed abundantly,"—in the northern part of the Nipmuck country (now in Worcester county, Mass.).

'Kiskatamenakook,' the name of a brook (but originally, of some locality near the brook) in Catskill, N.Y.,[5] is kiskato-minak-auke, 'place of thin-shelled nuts' (or shag-bark hickory nuts).

2. River. Seip or sepu (Del. sipo; Chip. sēpē; Abn. sip[oo];) the Algonkin word for 'river' is derived from a root that means 'stretched out,' 'extended,' 'become long,' and corresponds nearly to the English 'stream.' This word rarely, if ever, enters into the composition of local names, and, so far as I know, it does not make a part of the name of any river in New England. Mississippi is missi-sipu, 'great river;' Kitchi-sipi, 'chief river' or 'greatest river,' was the Montagnais name of the St. Lawrence;[6] and Miste-shipu is their modern name for the Moise or 'Great River' which flows from the lakes of the Labrador peninsula into the Gulf of St. Lawrence.[7]

Near the Atlantic seaboard, the most common substantival components of river names are (1) -tuk and (2) -hanne, -han, or -huan. Neither of these is an independent word. They are inseparable nouns-generic, or generic affixes.

-Tuk (Abn. -teg[oo]é; Del. -ittuk;) denotes a river whose waters are driven in waves, by tides or wind. It is found in names of tidal rivers and estuaries; less frequently, in names of broad and deep streams, not affected by tides. With the adjectival missi, 'great,' it forms missi-tuk,—now written Mystic,—the name of 'the great river' of Boston bay, and of another wide-mouthed tidal river in the Pequot country, which now divides the towns of Stonington and Groton.

Near the eastern boundary of the Pequot country, was the river which the Narragansetts called Paquat-tuk, sometimes written Paquetock, now Pawcatuck, 'Pequot river,'—the present eastern boundary of Connecticut. Another adjectival prefix, pohki or pahke, 'pure,' 'clear,' found in the name of several tidal streams, is hardly distinguishable from the former, in the modern forms of Pacatock, Paucatuck, &c.

Quinni-tuk is the 'long tidal-river.' With the locative affix, Quinni-tuk-ut, 'on long river,'—now Connecticut,—was the name of the valley, or lands both sides of the river. In one early deed (1636), I find the name written Quinetucquet; in another, of the same year, Quenticutt. Roger Williams (1643) has Qunnihticut, and calls the Indians of this region Quintik-óock, i.e. 'the long river people.' The c in the second syllable of the modern name has no business there, and it is difficult to find a reason for its intrusion.

'Lenapewihittuck' was the Delaware name of 'the river of the Lenape,' and 'Mohicannittuck,' of 'the river of the Mohicans' (Hudson River).[8]

Of Pawtucket and Pawtuxet, the composition is less obvious; but we have reliable Indian testimony that these names mean, respectively, 'at the falls' and 'at the little falls.'[Pg 9] Pequot and Narragansett interpreters, in 1679, declared that Blackstone's River, was "called in Indian Pautuck (which signifies, a Fall), because there the fresh water falls into the salt water."[9] So, the upper falls of the Quinebaug river (at Danielsonville, Conn.) were called "Powntuck, which is a general name for all Falls," as Indians of that region testified.[10] There was another Pautucket, 'at the falls' of the Merrimac (now Lowell); and another on Westfield River, Mass. Pawtuxet, i.e. pau't-tuk-es-it, is the regularly formed diminutive of paut-tuk-it. The village of Pawtuxet, four miles south of Providence, R.I., is "at the little falls" of the river to which their name has been transferred. The first settlers of Plymouth were informed by Samoset, that the place which they had chosen for their plantation was called 'Patuxet,'—probably because of some 'little falls' on Town Brook.[11] There was another 'Pautuxet,' or 'Powtuxet,' on the Quinebaug, at the lower falls; and a river 'Patuxet' (Patuxent), in Maryland. The same name is ingeniously disguised by Campanius, as 'Poaetquessing,' which he mentions as one of the principal towns of the Indians on the Delaware, just below the lower falls of that river at Trenton; and 'Poutaxat' was understood by the Swedes to be the Indian name both of the river and bay.[12] The adjectival pawt- or pauat- seems to be derived from a root meaning 'to make a loud noise.' It is found in many, perhaps in all Algonkin languages. 'Pawating,' as Schoolcraft wrote it, was the Chippewa name of the Sault Ste. Marie, or Falls of St. Mary's River,—pronounced poú-at-ing´, or pau-at-uⁿ, the last syllable representing the locative affix,—"at the Falls."[Pg 10] The same name is found in Virginia, under a disguise which has hitherto prevented its recognition. Capt. John Smith informs us that the "place of which their great Emperor taketh his name" of Powhatan, or Pawatan, was near "the Falls" of James River,[13] where is now the city of Richmond. 'Powatan' is pauat-hanne, or 'falls on a rapid stream.'

Acáwmé or Ogkomé (Chip. agami; Abn. agaⁿmi; Del. achgameu;) means 'on the other side,' 'over against,' 'beyond.' As an adjectival, it is found in Acawm-auké, the modern 'Accomac,' a peninsula east of Chesapeake Bay, which was 'other-side land' to the Powhatans of Virginia. The site of Plymouth, Mass., was called 'Accomack' by Capt. John Smith,—a name given not by the Indians who occupied it but by those, probably, who lived farther north, 'on the other side' of Plymouth Bay. The countries of Europe were called 'other-side lands,'—Narr. acawmen-óaki; Abn. agaⁿmen-[oo]ki. With -tuk, it forms acawmen-tuk (Abn. agaⁿmen-teg[oo]), 'other-side river,' or, its diminutive, acawmen-tuk-es (Abn. agaⁿmen-teg[oo]éss[oo]), 'the small other-side river,'—a name first given (as Agamenticus or Accomenticus) to York, Me., from the 'small tidal-river beyond' the Piscataqua, on which that town was planted.

Peske-tuk (Abn. peské-teg[oo]é) denotes a 'divided river,' or a river which another cleaves. It is not generally (if ever) applied to one of the 'forks' which unite to form the main stream, but to some considerable tributary received by the main stream, or to the division of the stream by some obstacle, near its mouth, which makes of it a 'double river.' The primary meaning of the (adjectival) root is 'to divide in two,' and the secondary, 'to split,' 'to divide forcibly, or abruptly.' These shades of meaning are not likely to be detected under the disguises in which river-names come down[Pg 11] to our time. Râle translates ne-peské, "je vas dans le chemin qui en coupe un autre:" peskahak[oo]n, "branche."

Piscataqua, Pascataqua, &c., represent the Abn. peské-teg[oo]é, 'divided tidal-river.' The word for 'place' (ohke, Abn. 'ki,) being added, gives the form Piscataquak or -quog. There is another Piscataway, in New Jersey,—not far below the junction of the north and south branches of the Raritan,—and a Piscataway river in Maryland, which empties into the Potomac; a Piscataquog river, tributary to the Merrimac, in New Hampshire; a Piscataquis (diminutive) in Maine, which empties into the Penobscot. Pasquotank, the name of an arm of Albemarle Sound and of a small river which flows into it, in North Carolina, has probably the same origin.

The adjectival peské, or piské, is found in many other compound names besides those which are formed with -tuk or -hanne: as in Pascoag, for peské-auké, in Burrilville, R.I., 'the dividing place' of two branches of Blackstone's River; and Pesquamscot, in South Kingston, R.I., which (if the name is rightly given) is "at the divided (or cleft) rock,"—peské-ompsk-ut,—perhaps some ancient land-mark, on or near the margin of Worden's Pond.

Nôeu-tuk (Nóahtuk, Eliot), 'in the middle of the river,' may be, as Mr. Judd [14] and others have supposed, the name which has been variously corrupted to Norwottock, Nonotuck, Noatucke, Nawottok, &c. If so, it probably belonged, originally to one of the necks or peninsulas of meadow, near Northampton,—such as that at Hockanum, which, by a change in the course of the river at that point, has now become an island.

Tetiquet or Titicut, which passes for the Indian name of Taunton, and of a fishing place on Taunton River in the north-west part of Middleborough, Mass., shows how effectually such names may be disguised by phonetic corruption and mutilation. Kehte-tuk-ut (or as Eliot wrote it in Genesis xv. 18, Kehteihtukqut) means 'on the great river.' In the[Pg 12] Plymouth Colony Records we find the forms 'Cauteeticutt' and 'Coteticutt,' and elsewhere, Kehtehticut,—the latter, in 1698, as the name of a place on the great river, "between Taunton and Bridgewater." Hence, 'Teghtacutt,' 'Teightaquid,' 'Tetiquet,' &c.[15]

(2). The other substantival component of river-names, -hanne or -han (Abn. -ts[oo]aⁿn or -taⁿn; Mass. -tchuan;) denotes 'a rapid stream' or 'current;' primarily, 'flowing water.' In the Massachusetts and Abnaki, it occurs in such compounds as anu-tchuan (Abn. ari'ts[oo]aⁿn), 'it over-flows:' kussi-tchuan (Abn. kesi'ts[oo]aⁿn), 'it swift flows,' &c.

In Pennsylvania and Virginia, where the streams which rise in the highlands flow down rapidly descending slopes, -hanné is more common than -tuk or sepu in river names. Keht-hanné (kittan, Zeisb.; kithanne, Hkw.) was a name given to the Delaware River as 'the principal or greatest stream' of that region: and by the western Delawares, to the Ohio.[16] With the locative termination, Kittanning (Penn.) is a place 'on the greatest stream.' The Schuylkill was Ganshow-hanné, 'noisy stream;' the Lackawanna, Lechau-hanné, 'forked stream' or 'stream that forks:'[17] with affix, Lechauhannak or Lechauwahannak, 'at the river-fork,'—for which Hendrick Aupamut, a Muhhekan, wrote (with dialectic exchange of n for Delaware l) 'Naukhuwwhnauk,' 'The Forks' of the Miami.[18] The same name is found in New England, disguised as Newichawanock, Nuchawanack, &c., as near Berwick, Me., 'at the fork' or confluence of Cocheco and Salmon Fall rivers,—the 'Neghechewanck' of Wood's Map (1634). Powhatan, for Pauat-hanne, 'at the Falls on a rapid stream,' has been previously noticed.

Alleghany, or as some prefer to write it, Allegheny,—the Algonkin name of the Ohio River, but now restricted to one[Pg 13] of its branches,—is probably (Delaware) welhik-hanné or [oo]lik-hanné, 'the best (or, the fairest) river.' Welhik (as Zeisberger wrote it)[19] is the inanimate form of the adjectival, meaning 'best,' 'most beautiful.' In his Vocabulary, Zeisberger gave this synthesis, with slight change of orthography, as "Wulach'neü" [or [oo]lakhanne[oo], as Eliot would have written it,] with the free translation, "a fine River, without Falls." The name was indeed more likely to belong to rivers 'without falls' or other obstruction to the passage of canoes, but its literal meaning is, as its composition shows, "best rapid-stream," or "finest rapid-stream;" "La Belle Riviere" of the French, and the Oue-yo´ or O hee´ yo Gä-hun´-dä, "good river" or "the beautiful river," of the Senecas.[20] For this translation of the name we have very respectable authority,—that of Christian Frederick Post, a Moravian of Pennsylvania, who lived seventeen years with the Muhhekan Indians and was twice married among them, and whose knowledge of the Indian languages enabled him to render important services to the colony, as a negotiator with the Delawares and Shawanese of the Ohio, in the French war. In his "Journal from Philadelphia to the Ohio" in 1758,[21] after mention of the 'Alleghenny' river, he says: "The Ohio, as it is called by the Sennecas. Alleghenny is the name of the same river in the Delaware language. Both words signify the fine or fair river." La Metairie, the notary of La Salle's expedition, "calls the Ohio, the Olighinsipou, or Aleghin; evidently an Algonkin name,"—as Dr. Shea remarks.[22] Hecke[Pg 14]welder says that the Delawares "still call the Allegany (Ohio) river, Alligéwi Sipu,"—"the river of the Alligewi" as he chooses to translate it. In one form, we have wulik-hannésipu, 'best rapid-stream long-river;' in the other, wuliké-sipu, 'best long-river.' Heckewelder's derivation of the name, on the authority of a Delaware legend, from the mythic 'Alligewi' or 'Talligewi,'—"a race of Indians said to have once inhabited that country," who, after great battles fought in pre-historic times, were driven from it by the all-conquering Delawares,[23]—is of no value, unless supported by other testimony. The identification of Alleghany with the Seneca "De o´ na gä no, cold water" [or, cold spring,[24]] proposed by a writer in the Historical Magazine (vol. iv. p. 184), though not apparent at first sight, might deserve consideration if there were any reason for believing the name of the river to be of Iroquois origin,—if it were probable that an Iroquois name would have been adopted by Algonkin nations,—or, if the word for 'water' or 'spring' could be made, in any American language, the substantival component of a river name.

From the river, the name appears to have been transferred by the English to a range of the "Endless Mountains."

3. Nippe, Nipi (= n'pi; Narr. nip; Muhh. nup; Abn. and Chip. nebi; Del. m'bi;) and its diminutives, nippisse and nips, were employed in compound names to denote Water, generally, without characterizing it as 'swift flowing,' 'wave moved,' 'tidal,' or 'standing:' as, for example, in the name of a part of a river, where the stream widening with diminished current becomes lake-like, or of a stretch[Pg 15] of tide-water inland, forming a bay or cove at a river's mouth. By the northern Algonkins, it appears to have been used for 'lake,' as in the name of Missi-nippi or Missinabe lake ('great water'), and in that of Lake Nippissing, which has the locative affix, nippis-ing, 'at the small lake' north-east of the greater Lake Huron, which gave a name to the nation of 'Nipissings,' or as the French called them, 'Nipissiriniens,'—according to Charlevoix, the true Algonkins.

Quinnipiac, regarded as the Indian name of New Haven,—also written Quinnypiock, Quinopiocke, Quillipiack, &c., and by President Stiles [25] (on the authority of an Indian of East Haven) Quinnepyooghq,—is, probably, 'long water place,' quinni-nippe-ohke, or quin-nipi-ohke. Kennebec would seem to be another form of the same name, from the Abnaki, k[oo]né-be-ki, were it not that Râle wrote,[26] as the name of the river, 'Aghenibékki'—suggesting a different adjectival. But Biard, in the Relation de la Nouvelle-France of 1611, has 'Kinibequi,' Champlain, Quinebequy, and Vimont, in 1640, 'Quinibequi,' so that we are justified in regarding the name as the probable equivalent of Quinni-pi-ohke.

4. -Paug, -pog, -bog, (Abn. -béga or -bégat; Del. -pécat;) an inseparable generic, denoting 'water at rest,' 'standing water,' is the substantival component of names of small lakes and ponds, throughout New England.[27] Some of the most common of these names are,—

Massa-paug, 'great pond,'—which appears in a great variety of modern forms, as Mashapaug, Mashpaug, Massapogue,[Pg 16] Massapog, &c. A pond in Cranston, near Providence, R.I.; another in Warwick, in the same State; 'Alexander's Lake,' in Killingly; 'Gardiner's Lake,' in Salem, Bozrah and Montville; 'Tyler Pond,' in Goshen; ponds in Sharon, Groton, and Lunenburg, Mass., were each of them the 'Massapaug' or 'great pond' of its vicinity.

Quinni-paug, 'long pond.' One in Killingly, gave a name to Quinebaug River and the 'Quinebaug country.' Endicott, in 1651, wrote this name 'Qunnubbágge' (3 Mass. Hist. Coll., iv. 191). "Quinepoxet," the name of a pond and small river in Princeton, Mass., appears to be a corruption of the diminutive with the locative affix; Quinni-paug-es-it, 'at the little long pond.'

Wongun-paug, 'crooked (or bent) pond.' There is one of the name in Coventry, Conn. Written, 'Wangunbog,' 'Wungumbaug,' &c.

Petuhkqui-paug, 'round pond,' now called 'Dumpling Pond,' in Greenwich, Conn., gave a name to a plain and brook in that town, and, occasionally, to the plantation settled there, sometimes written 'Petuckquapock.'

Nunni-paug, 'fresh pond.' One in Edgartown, Martha's Vineyard, gave a name (Nunnepoag) to an Indian village near it. Eliot wrote nunnipog, for 'fresh water,' in James iii. 12.

Sonki-paug or soⁿki-paug, 'cool pond.' (Sonkipog, 'cold water,' Eliot.) Egunk-sonkipaug, or 'the cool pond (spring) of Egunk' hill in Sterling, Conn., is named in Chandler's Survey of the Mohegan country, as one of the east bounds.

Pahke-paug, 'clear pond' or 'pure water pond.' This name occurs in various forms, as 'Pahcupog,' a pond near Westerly, R.I.;[28] 'Pauquepaug,' transferred from a pond to a brook in Kent and New Milford; 'Paquabaug,' near Shepaug River, in Roxbury, &c. 'Pequabuck' river, in Bristol and Farmington, appears to derive its name from some 'clear pond,'—perhaps the one between Bristol and Plymouth.

Another noun-generic that denotes 'lake' or 'fresh water at rest,' is found in many Abnaki, northern Algonkin and Chippewa names, but not, perhaps, in Massachusetts or Connecticut. This is the Algonkin -gămi, -gŏmi, or -gummee. Kitchi-gami or 'Kechegummee,' the Chippewa name of Lake Superior, is 'the greatest, or chief lake.' Caucomgomoc, in Maine, is the Abn. kaäkou-gami-k, 'at Big-Gull lake.' Temi-gami, 'deep lake,' discharges its waters into Ottawa River, in Canada; Kinou-gami, now Kenocami, 'long lake,' into the Saguenay, at Chicoutimi.

There is a Mitchi-gami or (as sometimes written) machi-gummi, 'large lake,' in northern Wisconsin, and the river which flows from it has received the same name, with the locative suffix, 'Machigāmig' (for mitchi-gaming). A branch of this river is now called 'Fence River' from a mitchihikan or mitchikan, a 'wooden fence' constructed near its banks, by the Indians, for catching deer.[29] Father Allouez describes, in the 'Relation' for 1670 (p. 96), a sort of 'fence' or weir which the Indians had built across Fox River, for taking sturgeon &c., and which they called 'Mitihikan;' and shortly after, he mentions the destruction, by the Iroquois, of a village of Outagamis (Fox Indians) near his mission station, called Machihigan-ing, ['at the mitchihikan, or weir?'] on the 'Lake of the Illinois,' now Michigan. Father Dablon, in the next year's Relation, calls this lake 'Mitchiganons.' Perhaps there was some confusion between the names of the 'weir' and the 'great lake,' and 'Michigan' appears to have been adopted as a kind of compromise between the two. If so, this modern form of the name is corrupt in more senses than one.[30]

5. -amaug, denoting 'a fishing place' (Abn. aⁿmaⁿgan, 'on pêche là,') is derived from the root âm or âma, signifying 'to take by the mouth;' whence, âm-aü, 'he fishes with hook and line,' and Del. âman, a fish-hook. Wonkemaug for wongun-amaug, 'crooked fishing-place,' between Warren and New Preston, in Litchfield county, is now 'Raumaug Lake.' Ouschank-amaug, in East Windsor, was perhaps the 'eel fishing-place.' The lake in Worcester, Quansigamaug, Quansigamug, &c., and now Quinsigamond, was 'the pickerel fishing-place,' qunnosuog-amaug.

6. Rock. In composition, -pisk or -psk (Abn. pesk[oo]; Cree, -pisk; Chip. -bik;) denotes hard or flint-like rock;[31] -ompsk or oⁿbsk, and, by phonetic corruption, -msk, (from ompaé, 'upright,' and -pisk,) a 'standing rock.' As a substantival component of local names, -ompsk and, with the locative affix, -ompskut, are found in such names as—

Petukqui-ompskut, corrupted to Pettiquamscut, 'at the round rock.' Such a rock, on the east side of Narrow River, north-east from Tower Hill Church in South Kingston, R.I., was one of the bound marks of, and gave a name to, the "Pettiquamscut purchase" in the Narragansett country.

Wanashqui-ompskut (wanashquompsqut, Ezekiel xxvi. 14), 'at the top of the rock,' or at 'the point of rock.' Wonnesquam, Annis Squam, and Squam, near Cape Ann, are perhaps corrupt forms of the name of some 'rock summit' or 'point of rock' thereabouts. Winnesquamsaukit (for wanashqui-ompsk-ohk-it?) near Exeter Falls, N.H., has been transformed to Swampscoate and Squamscot. The name of Swamscot or Swampscot, formerly part of Lynn, Mass., has a different meaning. It is from m'squi-ompsk, 'Red Rock' (the modern name), near the north end of Long Beach, which[Pg 19] was perhaps "The clifte" mentioned as one of the bounds of Mr. Humfrey's Swampscot farm, laid out in 1638.[32] M'squompskut means 'at the red rock.' The sound of the initial m was easily lost to English ears.[33]

Penobscot, a corruption of the Abnaki paⁿna[oo]aⁿbskek, was originally the name of a locality on the river so called by the English. Mr. Moses Greenleaf, in a letter to Dr. Morse in 1823, wrote 'Pe noom´ ske ook' as the Indian name of Old Town Falls, "whence the English name of the River, which would have been better, Penobscook." He gave, as the meaning of this name, "Rocky Falls." The St. Francis Indians told Thoreau, that it means "Rocky River."[34] 'At the fall of the rock' or 'at the descending rock' is a more nearly exact translation. The first syllable, pen- (Abn. paⁿna) represents a root meaning 'to fall from a height,'—as in paⁿn-tek[oo], 'fall of a river' or 'rapids;' penaⁿ-ki, 'fall of land,' the descent or downward slope of a mountain, &c.

Keht-ompskqut, or 'Ketumpscut' as it was formerly written,[35]—'at the greatest rock,'—is corrupted to Catumb, the name of a reef off the west end of Fisher's Island.

Tomheganomset[36]—corrupted finally to 'Higganum,' the name of a brook and parish in the north-east part of Haddam,—appears to have been, originally, the designation of a locality from which the Indians procured stone suitable for making axes,—tomhegun-ompsk-ut, 'at the tomahawk rock.' In 'Higganompos,' as the name was sometimes written, without the locative affix, we have less difficulty in recognizing the substantival -ompsk.

Qussuk, another word for 'rock' or 'stone,' used by Eliot and Roger Williams, is not often—perhaps never found in local names. Hassun or Assun (Chip. assin´; Del. achsin;)[Pg 20] appears in New England names only as an adjectival (assuné, assini, 'stony'), but farther north, it occasionally occurs as the substantival component of such names as Mistassinni, 'the Great Stone,' which gives its name to a lake in British America, to a tribe of Indians, and to a river that flows into St. John's Lake.[37]

7. Wadchu (in composition, -adchu) means, always, 'mountain' or 'hill.' In Wachuset, we have it, with the locative affix -set, 'near' or 'in the vicinity of the mountain,'—a name which has been transferred to the mountain itself. With the adjectival massa, 'great,' is formed mass-adchu-set, 'near the great mountain,' or 'great hill country,'—now, Massachusetts.

'Kunckquachu' and 'Quunkwattchu,' mentioned in the deeds of Hadley purchase, in 1658,[38] are forms of qunuⁿkqu-adchu, 'high mountain,'—afterwards belittled as 'Mount Toby.'

'Kearsarge,' the modern name of two well-known mountains in New Hampshire, disguises k[oo]wass-adchu, 'pine mountain.' On Holland's Map, published in 1784, the southern Kearsarge (in Merrimack county) is marked "Kyarsarga Mountain; by the Indians, Cowissewaschook."[39] In this form,—which the termination ok (for ohke, auke, 'land,') shows to belong to the region, not exclusively to the mountain itself,—the analysis becomes more easy. The meaning of the adjectival is perhaps not quite certain. K[oo]wa (Abn. k[oo]é) 'a pine tree,' with its diminutive, k[oo]wasse, is a derivative,—from a root which means 'sharp,' 'pointed.' It is possible, that in this synthesis, the root preserves its primary signification, and that 'Kearsarge' is the 'pointed' or 'peaked mountain.'

Mauch Chunk (Penn.) is from Del. machk, 'bear' and wachtschunk, 'at, or on, the mountain,'—according to Heckewelder, who writes 'Machkschúnk,' or the Delaware name of 'the bear's mountain.'[Pg 21]

In the Abnaki and some other Algonkin dialects, the substantival component of mountain names is -ádené,—an inseparable noun-generic. Katahdin (pronounced Ktaadn by the Indians of Maine), Abn. Ket-ádené, 'the greatest (or chief) mountain,' is the equivalent of 'Kittatinny,' the name of a ridge of the Alleghanies, in New Jersey and Pennsylvania.

8. -komuk or komako (Del. -kamik, -kamiké; Abn. -kamighe; Cree, -gómmik; Powhatan, -comaco;) cannot be exactly translated by any one English word. It denotes 'place,' in the sense of enclosed, limited or appropriated space. As a component of local names, it means, generally, 'an enclosure,' natural or artificial; such as a house or other building, a village, a planted field, a thicket or place surrounded by trees, &c. The place of residence of the Sachem, which (says Roger Williams) was "far different from other houses [wigwams], both in capacity, and in the fineness and quality of their mats," was called sachimâ-komuk, or, as Edward Winslow wrote it, 'sachimo comaco,'—the Sachem-house. Werowocomoco, Weramocomoco, &c. in Virginia, was the 'Werowance's house,' and the name appears on Smith's map, at a place "upon the river Pamauncke [now York River], where the great King [Powhatan] was resident."

Kuppi-komuk, 'closed place,' 'secure enclosure,' was the name of a Pequot fastness in a swamp, in Groton, Conn. Roger Williams wrote this name "Cuppacommock," and understood its meaning to be "a refuge, or hiding place." Eliot has kuppóhkomuk for a planted 'grove,' in Deut. xvi. 21, and for a landing-place or safe harbor, Acts xxvii. 40.

Nashaue-komuk, 'half-way house,' was at what is now Chilmark, on Martha's Vineyard, where there was a village of praying Indians [40] in 1698, and earlier.

The Abnaki keta-kamig[oo] means, according to Râle, 'the[Pg 22] main land,'—literally, 'greatest place;' teteba-kamighé, 'level place,' a plain; pépam-kamighek, 'the all land,' 'l'univers.'

Néssa[oo]a-kamíghé, meaning 'double place' or 'second place,' was the name of the Abnaki village of St. Francis de Sales, on the St. Lawrence,[41]—to which the mission was removed about 1700, from its first station established near the Falls of the Chaudière in 1683.[42]

9. Of two words meaning Island, munnohan or, rejecting the formative, munnoh (Abn. menahan; Del. menatey; Chip. minís, a diminutive,) is the more common, but is rarely, if ever, found in composition. The 'Grand Menan,' opposite Passammaquoddy Bay, retains the Abnaki name. Long Island was Menatey or Manati, 'the Island,'—to the Delawares, Minsi and other neighboring tribes. Any smaller island was menatan (Mass. munnohhan), the indefinite form, or menates (Mass. munnises, manisses), the diminutive. Campanius mentions one 'Manathaan,' Coopers' Island (now Cherry Island) near Fort Christina, in the Delaware,[43] and "Manataanung or Manaates, a place settled by the Dutch, who built there a clever little town, which went on increasing every day,"—now called New York. (The termination in -ung is the locative affix.) New York Island was sometimes spoken of as 'the island'—'Manaté,' 'Manhatte;' sometimes as 'an island'—Manathan, Menatan, 'Manhatan;' more accurately, as 'the small island'—Manhaates, Manattes, and 'the Manados' of the Dutch. The Island Indians collectively, were called Manhattans; those of the small island, 'Manhatesen.' "They deeply mistake," as Gov. Stuyvesant's agents declared, in 1659,[44] "who interpret the general name of Manhattans, unto the particular town built upon a little Island; because it signified the whole country and province."

Manisses or Monasses, as Block Island was called, is an[Pg 23]other form of the diminutive,—from munnoh; and Manhasset, otherwise written, Munhansick, a name of Shelter Island, is the same diminutive with the locative affix, munna-es-et. So is 'Manusses' or 'Mennewies,' an island near Rye, N.Y.,—now written (with the southern form of the locative,) Manussing.

Montauk Point, formerly Montauket, Montacut, and by Roger Williams, Munnawtawkit, is probably from manati, auke, and -it locative; 'in the Island country,' or 'country of the Islanders.'

The other name of 'Island,' in Algonkin languages, is ahquedne or ocquidne; with the locative; ahquednet, as in Acts xxvii. 16. (Compare, Cree, ákootin, "it suspends, is sit-uate, e.g. an island in the water," from âkoo, a verbal root "expressive of a state of rest." Howse's Grammar, p. 152. Micmac, agwitk, "it is in the water;" whence, Ep-agwit, "it lies [sits?] in the water,"[45] the Indian name of Prince Edward's Island.) This appears to have been restricted in its application, to islands lying near the main land or spoken of with reference to the main land. Roger Williams learned from the Narragansetts to call Rhode Island, Aquiday, Aquednet, &c., 'the Island' or 'at the Island,' and a "little island in the mouth of the Bay," was Aquedenesick,[46] or Aquidneset, i.e. 'at the small island.'

Chippaquiddick, the modern name of an island divided by a narrow strait from Martha's Vineyard, is from cheppi-aquidne, 'separated island.'

Abnaki names ending in -kaⁿtti, or -kontee (Mass. -kontu; Etchemin or Maliseet, -kodiah, -quoddy; Micmac, -kaⁿdi, or -aikadee;) may be placed with those of the first class, though this termination, representing a substantival component, is really only the locative affix of nouns in the indefinite plural. Exact location was denoted by affixing, to inanimate nouns-[Pg 24]singular, -et, -it or -ut; proximity, or something less than exact location, by -set, (interposing s, the characteristic of diminutives and derogatives) between the noun and affix. Plural nouns, representing a definite number of individuals, or a number which might be regarded as definite, received -ettu, -ittu, or -uttu, in the locative: but if the number was indefinite, or many individuals were spoken of collectively, the affix was -kontu, denoting 'where many are,' or 'place of abundance.' For example, wadchu, mountain; wadchu-ut, to, on, or at the mountain; wadchu-set, near the mountain; wadchuuttu (or -ehtu), in or among certain mountains, known or indicated (as in Eliot's version of Numbers xxxiii. 47, 48); wadchué-kontu, among mountains, where there are a great many mountains, for 'in the hill country,' Joshua xiii. 6. So, nippe-kontu, 'in the waters,' i.e. in many waters, or 'where there is much water,' Deut. iv. 18; v. 8. In Deuteronomy xi. 11, the conversion to a verb of a noun which had previously received this affix, shows that the idea of abundance or of multitude is associated with it: "ohke wadchuuhkontu[oo]," i.e. wadechué-kontu-[oo], "the land is a land of hills," that is, where are many hills, or where hills are plenty.

This form of verb was rarely used by Eliot and is not alluded to in his Grammar. It appears to have been less common in the Massachusetts than in most of the other Algonkin languages. In the Chippewa, an 'abundance verb,' as Baraga [47] calls it, may be formed from any noun, by adding -ka or -ĭka for the indicative present: in the Cree, by adding -skow or -ooskow. In the Abnaki, -ka or -k[oo], or -ik[oo], forms similar verbs, and verbals. The final 'tti of kaⁿtti, represents the impersonal a'tté, eto, 'there belongs to it,' 'there is there,' il y a. (Abn. meskik[oo]i'kaⁿtti, 'where there is abundance of grass,' is the equivalent of the Micmac "m'skeegoo-aicadee, a meadow."[48])

Aⁿmes[oo]k-kaⁿtti, 'where there is plenty of alewives or herrings;' from Abn. aⁿms[oo]ak (Narr. aumsûog; Mass. ômmissuog, cotton;) literally, 'small fishes,' but appropriated to fish of the herring tribe, including alewives and menhaden or bony-fish. Râle gives this as the name of one of the Abnaki villages on or near the river 'Aghenibekki.' It is the same, probably, as the 'Meesee Contee' or 'Meesucontee,' at Farmington Falls, on Sandy River, Me.[49] With the suffix of 'place' or 'land,' it has been written Amessagunticook and Amasaquanteg.

'Amoscoggin,' 'Ammarescoggen,' &c., and the 'Aumoughcawgen' of Capt. John Smith, names given to the Kennebec or its main western branch, the Androscoggin,[50]—appear to have belonged, originally, to 'fishing places' on the river, from Abn. aⁿm's[oo]a-khíge, or aⁿm's[oo]a-kaⁿgan. 'Amoskeag,' at the falls of the Merrimack, has the same meaning, probably; aⁿm's[oo]a-khíge (Mass. ômmissakkeag), a 'fishing-place for alewives.' It certainly does not mean 'beavers,' or 'pond or marsh' of beavers,—as Mr. Schoolcraft supposed it to mean.[51]

Madamiscomtis or Mattammiscontis, the name of a tributary of the Penobscot and of a town in Lincoln county, Me., was translated by Mr. Greenleaf, in 1823, "Young Alewive stream;" but it appears to represent met-aⁿms[oo]ak-kaⁿtti, 'a place where there has been (but is not now) plenty of alewives,' or to which they no longer resort. Compare Râle's[Pg 26] met-aⁿm[oo]ak, "les poissons ont faites leurs œufs; ils s'en sont allés; il n'y en a plus."

Cobbosseecontee river, in the south part of Kennebec county, is named from a place near "the mouth of the stream, where it adjoineth itself to Kennebec river,"[52] and 'where there was plenty of sturgeons,'—kabassak-kaⁿtti.

'Peskadamioukkanti' is given by Charlevoix, as the Indian name of "the river of the Etchemins," that is, the St. Croix,—a name which is now corrupted to Passamaquoddy; but this latter form of the name is probably derived from the Etchemin, while Charlevoix wrote the Abnaki form. The Rev. Elijah Kellogg, in 1828,[53] gave, as the meaning of 'Passamaquoddie,' 'pollock fish,' and the Rev. Mr. Rand translates 'Pestumoo-kwoddy' by 'pollock ground.'[54] Cotton's vocabulary gives 'pâkonnótam' for 'haddock.' Perhaps peskadami[oo]k, like aⁿms[oo]ak, belonged to more than one species of fish.

Of Etchemin and Micmac words having a similar termination, we find among others,—

Shubenacadie (Chebenacardie on Charlevoix' map, and Shebenacadia on Jeffry's map of 1775). One of the principal rivers of Nova Scotia, was so named because 'sipen-ak were plenty there.' Professor Dawson was informed by an "ancient Micmac patriarch," that "Shuben or Sgabun means ground-nuts or Indian potatoes," and by the Rev. Mr. Rand, of Hantsport, N.S., that "segubbun is a ground-nut, and Segubbuna-kaddy is the place or region of ground-nuts," &c.[55] It is not quite certain that shuben and segubbun denote the same esculent root. The Abnaki name of the wild potato or ground-nut was pen, pl. penak (Chip. opin-īg; Del. obben-ak); 'sipen,' which is obviously the equivalent of sheben, Râle describes as "blanches, plus grosses que des penak:" and sheep'n-ak is the modern Abnaki (Penobscot) name for the bulbous roots[Pg 27] of the Yellow Lily (Lilium Canadense). Thoreau's Indian guide in the 'Maine Woods' told him that these bulbs "were good for soup, that is to cook with meat to thicken it,"—and taught him how to prepare them.[56] Josselyn mentions such "a water-lily, with yellow flowers," of which "the Indians eat the roots" boiled.[57]

"Segoonuma-kaddy, place of gaspereaux; Gaspereau or Alewife River," "Boonamoo-kwoddy, Tom Cod ground," and "Kata-kaddy, eel-ground,"—are given by Professor Dawson, on Mr. Rand's authority. Segoonumak is the equivalent of Mass. and Narr. sequanamâuquock, 'spring (or early summer) fish,' by R. Williams translated 'bream.' And boonamoo,—the ponamo of Charlevoix (i. 127), who confounded it with some 'species of dog-fish (chien de mer),'—is the ap[oo]naⁿ-mes[oo] of Rasles and papônaumsu, 'winter fish,' of Roger Williams, 'which some call frost-fish,'—Morrhua pruinosa.

The frequent occurrence of this termination in Micmac, Etchemin and Abnaki local names gives probability to the conjecture, that it came to be regarded as a general name for the region which these tribes inhabited,—'L'arcadia,' 'l'Accadie,' and 'la Cadie,' of early geographers and voyagers. Dr. Kohl has not found this name on any earlier map than that published by Girolamo Ruscelli in 1561.[58] That it is of Indian origin there is hardly room for doubt, and of two or three possible derivations, that from the terminal -kâdi, -kodiah, or -kaⁿtti, is on the whole preferable. But this termination, in the sense of 'place of abundance' or in that of 'ground, land, or place,' cannot be used separately, as an independent word, in any one of the languages which have been mentioned; and it is singular that, in two or three instances, only this termination should have been preserved[Pg 28] after the first and more important component of the name was lost.

There are two Abnaki words which are not unlike -kaⁿtti in sound, one or both of which may perhaps be found in some local names: (1) ka[oo]di, 'where he sleeps,' a lodging place of men or animals; and (2) ak[oo]daï[oo]i, in composition or as a prefix, ak[oo]dé, 'against the current,' up-stream; as in ned-ak[oo]té'hémen, 'I go up stream,' and [oo]derak[oo]daⁿnaⁿ, 'the fish go up stream.' Some such synthesis may have given names to fishing-places on tidal rivers, and I am more inclined to regard the name of 'Tracadie' or 'Tracody' as a corruption of [oo]derak[oo]daⁿ, than to derive it (with Professor Dawson [59] and the Rev. Mr. Rand) from "Tulluk-kaddy; probably, place of residence; dwelling place,"—or rather (for the termination requires this), where residences or dwellings are plenty,—where there is abundance of dwelling place. There is a Tracadie in Nova Scotia, another (Tregaté, of Champlain) on the coast of New Brunswick, a Tracody or Tracady Bay in Prince Edward's Island, and a Tracadigash Point in Chaleur Bay.

Thevet, in La Cosmographie universelle,[60] gives an account of his visit in 1556, to "one of the finest rivers in the whole world which we call Norumbegue, and the aborigines Agoncy,"—now Penobscot Bay. In 'Agoncy' we have, I conjecture, another form of the Abnaki -kaⁿtti, and an equivalent of 'Acadie.'

II. Names formed from a single ground-word or substantival,—with or without a locative or other suffix.

To this class belong some names already noticed in connection with compound names to which they are related; such as, Wachu-set, 'near the mountain;' Menahan (Menan), Manati, Manathaan, 'island;' Manataan-ung, Aquedn-et, 'on the island,' &c. Of the many which might be added to these, the limits of this paper permit me to mention only a few.[Pg 29]

1. Nâïag, 'a corner, angle, or point.' This is a verbal, formed from nâ-i, 'it is angular,' 'it corners.' Eliot wrote "yaue naiyag wetu" for the "four corners of a house," Job i. 19. Sometimes, nâi receives, instead of the formative -ag, the locative affix (nâï-it or nâï-ut); sometimes it is used as an adjectival prefixed to auke, 'land.' One or another of these forms serves as the name of a great number of river and sea-coast 'points.' In Connecticut, we find a 'Nayaug' at the southern extremity of Mason's Island in Mystic Bay, and 'Noank' (formerly written, Naweag, Naiwayonk, Noïank, &c.) at the west point of Mystic River's mouth, in Groton; Noag or Noyaug, in Glastenbury, &c. In Rhode Island, Nayatt or Nayot point in Barrington, on Providence Bay, and Nahiganset or Narragansett, 'the country about the Point.'[61] On Long Island, Nyack on Peconick Bay, Southampton,[62] and another at the west end of the Island, opposite Coney Island. There is also a Nyack on the west side of the Tappan Sea, in New Jersey.

2. Wonkun, 'bended,' 'a bend,' was sometimes used without affix. The Abnaki equivalent is [oo]aⁿghíghen, 'courbe,' 'croché' (Râle). There was a Wongun, on the Connecticut, between Glastenbury and Wethersfield, and another, more considerable, a few miles below, in Middletown. Wonki is found in compound names, as an adjectival; as in Wonki-tuk, 'bent river,' on the Quinebaug, between Plainfield and Canterbury,—written by early recorders, 'Wongattuck,' 'Wanungatuck,' &c., and at last transferred from its proper place to a hill and brook west of the river, where it is disguised as Nunkertunk. The Great Bend between Hadley and Hatfield, Mass., was called Kuppo-wonkun-ohk, 'close bend place,' or 'place shut-in by a bend.' A tract of meadow west of this bend was called, in 1660, 'Cappowonganick,' and 'Capa[Pg 30]wonk,' and still retains, I believe, the latter name.[63] Wnogquetookoke, the Indian name of Stockbridge, Mass., as written by Dr. Edwards in the Muhhecan dialect, describes "a bend-of-the-river place."

Another Abnaki word meaning 'curved,' 'crooked,'—pikaⁿghén—occurs in the name Pikaⁿghenahik, now 'Crooked Island,' in Penobscot River.[64]

3. Hócquaun (uhquôn, Eliot), 'hook-shaped,' 'a hook,'—is the base of Hoccanum, the name of a tract of land and the stream which bounds it, in East Hartford, and of other Hoccanums, in Hadley and in Yarmouth, Mass. Heckewelder [65] wrote "Okhúcquan, Woâkhúcquoan or (short) Húcquan," for the modern 'Occoquan,' the name of a river in Virginia, and remarked: "All these names signify a hook." Campanius has 'hóckung' for 'a hook.'

Hackensack may have had its name from the húcquan-sauk, 'hook mouth,' by which the waters of Newark Bay find their way, around Bergen Point, by the Kill van Cul, to New York Bay.

3. Sóhk or Sauk, a root that denotes 'pouring out,' is the base of many local names for 'the outlet' or 'discharge' of a river or lake. The Abnaki forms, saⁿg[oo]k, 'sortie de la rivière (seu) la source,' and saⁿghede'teg[oo]é [= Mass. saukituk,] gave names to Saco in Maine, to the river which has its outflow at that place, and to Sagadahock (saⁿghede'aki), 'land at the mouth' of Kennebeck river.

Saucon, the name of a creek and township in Northampton county, Penn., "denotes (says Heckewelder [66]) the outlet of a smaller stream into a larger one,"—which restricts the denotation too narrowly. The name means "the outlet,"—and nothing more. Another Soh´coon, or (with the locative)[Pg 31] Saukunk, "at the mouth" of the Big Beaver, on the Ohio,—now in the township of Beaver, Penn.,—was a well known rendezvous of Indian war parties.[67]

Saganaum, Sagana, now Saginaw[68] Bay, on Lake Huron, received its name from the mouth of the river which flows through it to the lake.

The Mississagas were people of the missi-sauk, missi-sague, or (with locative) missi-sak-ing,[69] that is 'great outlet.' In the last half of the seventeenth century they were seated on the banks of a river which is described as flowing into Lake Huron some twenty or thirty leagues south of the Sault Ste. Marie (the same river probably that is now known as the Mississauga, emptying into Manitou Bay,) and nearly opposite the Straits of Mississauga on the South side of the Bay, between Manitoulin and Cockburn Islands. So little is known however of the history and migrations of this people, that it is perhaps impossible now to identify the 'great outlet' from which they first had their name.

The Saguenay (Sagnay, Sagné, Saghuny, etc.), the great tributary of the St. Lawrence, was so called either from the well-known trading-place at its mouth, the annual resort of the Montagnars and all the eastern tribes,[70] or more probably from the 'Grand Discharge'[71] of its main stream from Lake St. John and its strong current to and past the rapids at Chicoutimi, and thence on to the St. Lawrence.[72] Near Lake[Pg 32] St. John and the Grand Discharge was another rendezvous of the scattered tribes. The missionary Saint-Simon in 1671 described this place as one at which "all the nations inhabiting the country between the two seas (towards the east and north) assembled to barter their furs." Hind's Exploration of Labrador, ii. 23.

In composition with -tuk, 'river' or 'tidal stream,' sauki (adjectival) gave names to 'Soakatuck,' now Saugatuck, the mouth of a river in Fairfield county, Conn.; to 'Sawahquatock,' or 'Sawkatuck-et,' at the outlet of Long Pond or mouth of Herring River, in Harwich, Mass.; and perhaps to Massaugatucket, (missi-saukituk-ut?), in Marshfield, Mass., and in South Kingston, R.I.,—a name which, in both places, has been shortened to Saquatucket.

'Winnipiseogee' (pronounced Win´ ni pe sauk´ e,) is compounded of winni, nippe, and sauki, 'good-water discharge,' and the name must have belonged originally to the outlet by which the waters of the lake pass to the Merrimack, rather than to the lake itself. Winnepesauke, Wenepesioco and (with the locative) Winnipesiockett, are among the early forms of the name. The translation of this synthesis by 'the Smile of the Great Spirit' is sheer nonsense. Another, first proposed by the late Judge Potter of New Hampshire, in his History of Manchester (p. 27),[73]—'the beautiful water of the high place,'—is demonstrably wrong. It assumes that is or es represents kees, meaning 'high;' to which assumption there are two objections: first, that there is no evidence that such a word as kees, meaning 'high,' is found in any Algonkin language, and secondly, that if there be such a word, it must retain its significant root, in any synthesis of which it makes part,—in other words, that kees could not drop its initial k and preserve its meaning. I was at first inclined to accept the more probable translation proposed by[Pg 33] 'S.F.S.' [S.F. Streeter?] in the Historical Magazine for August, 1857,[74]—"the land of the placid or beautiful lake;" but, in the dialects of New England, nippisse or nips, a diminutive of nippe, 'water,' is never used for paug, 'lake' or 'standing water;'[75] and if it were sometimes so used, the extent of Lake Winnepiseogee forbids it to be classed with the 'small lakes' or 'ponds,' to which, only, the diminutive is appropriate.

4. Nashaué (Chip. nássawaiï and ashawiwi), 'mid-way,' or 'between,' and with ohke or auk added, 'the land between' or 'the half-way place,'—was the name of several localities. The tract on which Lancaster, in Worcester county (Mass.) was settled, was 'between' the branches of the river, and so it was called 'Nashaway' or 'Nashawake' (nashaué-ohke); and this name was afterwards transferred from the territory to the river itself. There was another Nashaway in Connecticut, between Quinnebaug and Five-Mile Rivers in Windham county, and here, too, the mutilated name of the nashaue-ohke was transferred, as Ashawog or Assawog, to the Five-Mile River. Natchaug in the same county, the name of the eastern branch of Shetucket river, belonged originally to the tract 'between' the eastern and western branches; and the Shetucket itself borrows a name (nashaue-tuk-ut) from its place 'between' Yantic and Quinebaug rivers. A neck of land (now in Griswold, Conn.) "between Pachaug River and a brook that comes into it from the south," one of the Muhhekan east boundaries, was called sometimes, Shawwunk, 'at the place between,'—sometimes Shawwâmug (nashaué-amaug), 'the fishing-place between' the rivers, or the 'half-way fishing-place.'[76]

5. Ashim, is once used by Eliot (Cant. iv. 12) for 'fountain.' It denoted a spring or brook from which water was obtained for drinking. In the Abnaki, asiem nebi, 'il puise de l'eau;' and ned-a'sihibe, 'je puise de l'eau, fonti vel fluvio.' (Rasles.)

Winne-ashim-ut, 'at the good spring,' near Romney Marsh, is now Chelsea, Mass. The name appears in deeds and records as Winnisimmet, Winisemit, Winnet Semet, etc. The author of the 'New English Canaan' informs us (book 2, ch. 8), that "At Weenasemute is a water, the virtue whereof is, to cure barrennesse. The place taketh his name of that fountaine, which signifieth quick spring, or quickning spring. Probatum."

Ashimuit or Shumuit, an Indian village near the line between Sandwich and Falmouth, Mass.,—Shaume, a neck and river in Sandwich (the Chawum of Capt. John Smith?),—Shimmoah, an Indian village on Nantucket,—may all have derived their names from springs resorted to by the natives, as was suggested by the Rev. Samuel Deane in a paper in Mass. Hist. Collections, 2d Series, vol. x. pp. 173, 174.

6. Mattappan, a participle of mattappu (Chip. namátabi), 'he sits down,' denotes a 'sitting-down place,' or, as generally employed in local names, the end of a portage between two rivers or from one arm of the sea to another,—where the canoe was launched again and its bearers re-embarked. Râle translates the Abnaki equivalent, mataⁿbe, by 'il va au bord de l'eau,—a la grève pour s'embarquer,' and metaⁿbéniganik, by 'au bout de delà du portage.'

Mattapan-ock, afterwards shortened to Mattapan, that part of Dorchester Neck (South Boston) where "the west country people were set down" in 1630,[77] may have been so called because it was the end of a carrying place from South Bay to Dorchester Bay, across the narrowest part of the peninsula, or—as seems highly probable—because it was the temporary[Pg 35] 'sitting-down place' of the new comers. Elsewhere, we find the name evidently associated with portage.

On Smith's Map of Virginia, one 'Mattapanient' appears as the name of the northern fork (now the Mattápony) of Pamaunk (York) River; another (Mattpanient) near the head waters of the Pawtuxunt; and a third on the 'Chickahamania' not far above its confluence with Powhatan (James) River.

Mattapoiset, on an inlet of Buzzard's Bay, in Rochester, Mass.,—another Mattapoiset or 'Mattapuyst,' now Gardner's Neck, in Swanzea,—and 'Mattapeaset' or 'Mattabesic,' on the great bend of the Connecticut (now Middletown), derived their names from the same word, probably.

On a map of Lake Superior, made by Jesuit missionaries and published in Paris in 1672, the stream which is marked on modern maps as 'Rivière aux Traines' or 'Train River,' is named 'R. Mataban.' The small lake from which it flows is the 'end of portage' between the waters of Lake Michigan and those of Lake Superior.

7. Chabenuk, 'a bound mark'; literally, 'that which separates or divides.' A hill in Griswold, Conn., which was anciently one of the Muhhekan east bound-marks, was called Chabinuⁿk, 'Atchaubennuck,' and 'Chabunnuck.' The village of praying Indians in Dudley (now Webster?) Mass., was named Chabanakongkomuk (Eliot, 1668,) or -ongkomum, and the Great Pond still retains, it is said, the name of Chaubenagungamaug (chabenukong-amaug?), "the boundary fishing-place." This pond was a bound mark between the Nipmucks and the Muhhekans, and was resorted to by Indians of both nations.

III. Participials and verbals employed as place-names may generally, as was before remarked, be referred to one or the other of the two preceding classes. The distinction between noun and verb is less clearly marked in Indian grammar than in English. The name Mushauwomuk (corrupted to Shaw[Pg 36]mut) may be regarded as a participle from the verb mushau[oo]m (Narr. mishoonhom) 'he goes by boat,'—or as a noun, meaning 'a ferry,'—or as a name of the first class, compounded of the adjectival mush[oo]-n, 'boat or canoe,' and wom[oo]-uk, habitual or customary going, i.e., 'where there is going-by-boat.'

The analysis of names of this class is not easy. In most cases, its results must be regarded as merely provisional. Without some clue supplied by history or tradition and without accurate knowledge of the locality to which the name belongs, or is supposed to belong, one can never be certain of having found the right key to the synthesis, however well it may seem to fit the lock. Experience Mayhew writing from Chilmark on Martha's Vineyard, in 1722, gives the Indian name of the place where he was living as Nimpanickhickanuh. If he had not added the information that the name "signifies in English, The place of thunder clefts," and that it was so called "because there was once a tree there split in pieces by the thunder," it is not likely that any one in this generation would have discovered its precise meaning,—though it might have been conjectured that neimpau, or nimbau, 'thunder,' made a part of it.

Quilútámende was (Heckewelder tells us [78]) the Delaware name of a place on the Susquehanna, in Pennsylvania, where, as the Indians say, "in their wars with the Five Nations, they fell by surprise upon their enemies. The word or name of this place is therefore, Where we came unawares upon them, &c." Without the tradition, the meaning of the name would not have been guessed,—or, if guessed, would not have been confidently accepted.

The difficulty of analyzing such names is greatly increased by the fact that they come to us in corrupt forms. The same name may be found, in early records, written in a dozen different ways, and some three or four of these may admit of as many different translations. Indian grammatical[Pg 37] synthesis was exact. Every consonant and every vowel had its office and its place. Not one could be dropped or transposed, nor could one be added, without change of meaning. Now most of the Indian local names were first written by men who cared nothing for their meaning and knew nothing of the languages to which they belonged. Of the few who had learned to speak one or more of these languages, no two adopted the same way of writing them, and no one—John Eliot excepted—appears to have been at all careful to write the same word twice alike. In the seventeenth century men took considerable liberties with the spelling of their own surnames and very large liberty with English polysyllables—especially with local names. Scribes who contrived to find five or six ways of writing 'Hartford' or 'Wethersfield,' were not likely to preserve uniformity in their dealings with Indian names. A few letters more or less were of no great consequence, but, generally, the writers tried to keep on the safe side, by putting in as many as they could find room for; prefixing a c to every k, doubling every w and g, and tacking on a superfluous final e, for good measure.

In some instances, what is supposed to be an Indian place-name is in fact a personal name, borrowed from some sachem or chief who lived on or claimed to own the territory. Names of this class are likely to give trouble to translators. I was puzzled for a long time by 'Mianus,' the name of a stream between Stamford and Greenwich,—till I remembered that Mayano, an Indian warrior (who was killed by Capt. Patrick in 1643) had lived hereabouts; and on searching the Greenwich records, I found the stream was first mentioned as Moyannoes and Mehanno's creek, and that it bounded 'Moyannoe's neck' of land. Moosup river, which flows westerly through Plainfield into the Quinebaug and which has given names to a post-office and factory village, was formerly Moosup's river,—Moosup or Maussup being one of the aliases of a Narragansett sachem who is better known, in the history of Philip's war, as Pessacus. Heckewelder [79] restores 'Pyma[Pg 38]tuning,' the name of a place in Pennsylvania, to the Del. 'Pihmtónink,' meaning, "the dwelling place of the man with the crooked mouth, or the crooked man's dwelling place," and adds, that he "knew the man perfectly well," who gave this name to the locality.

Some of the examples which have been given,—such as Higganum, Nunkertunk, Shawmut, Swamscot and Titicut,—show how the difficulties of analysis have been increased by phonetic corruption, sometimes to such a degree as hardly to leave a trace of the original. Another and not less striking example is presented by Snipsic, the modern name of a pond between Ellington and Tolland. If we had not access to Chandler's Survey of the Mohegan Country, made in 1705, who would suppose that 'Snipsic' was the surviving representative of Moshenupsuck, 'great-pond brook' or (literally) 'great-pond outlet,' at the south end of Moshenups or Mashenips 'great pond?' The territories of three nations, the Muhhekans, Nipmucks and River Indians, ran together at this point.

'Nameroake,' 'Namareck' or 'Namelake,' in East Windsor, was transformed to May-luck, giving to a brook a name which 'tradition' derives from the 'luck' of a party of emigrants who came in 'May' to the Connecticut.[80] The original name appears to have been the equivalent of 'Nameaug' or 'Nameoke' (New London), and to mean 'the fishing place,'—n'amaug or nama-ohke.

But none of these names exhibits a more curious transformation than that of 'Bagadoose' or 'Bigaduce,' a peninsula on the east side of Penobscot Bay, now Castine, Me. Williamson's History of Maine (ii. 572) states on the authority of Col. J. Wardwell of Penobscot, in 1820, that this point bore the name of a former resident, a Frenchman, one 'Major Biguyduce.' Afterwards, the historian was informed that 'Marche bagyduce' was an Indian word meaning 'no good[Pg 39] cove.' Mr. Joseph Williamson, in a paper in the Maine Historical Society's Collections (vol. vi. p. 107) identifies this name with the Matchebiguatus of Edward Winslow's quitclaim to Massachusetts in 1644,[81] and correctly translates the prefix matche by 'bad,' but adds: "What Biguatus means, I do not know." Purchas mentions 'Chebegnadose,' as an Indian town on the 'Apananawapeske' or Penobscot.[82] Râle gives, as the name of the place on "the river where M. de Gastin [Castine] is," Matsibig[oo]ad[oo]ssek, and on his authority we may accept this form as nearly representing the original. The analysis now becomes more easy. Matsi-aⁿbaga[oo]at-ek, means 'at the bad-shelter place,—bad covert or cove;' and matsi-aⁿbaga[oo]at[oo]s-ek the diminutive, 'at the small bad-shelter place.' About two miles and a half above the mouth of the Kenebec was a place called by the Indians 'Abagadusset' or 'Abequaduset'—the same name without the prefix—meaning 'at the cove, or place of shelter.'

The adjectivals employed in the composition of Algonkin names are very numerous, and hardly admit of classification. Noun, adjective, adverb or even an active verb may, with slight change of form, serve as a prefix. But, as was before remarked, every prefix, strictly considered, is an adverb or must be construed as an adverb,—the synthesis which serves as a name having generally the verb form. Some of the most common of these prefixes have been mentioned on preceding pages. A few others, whose meanings are less obvious and have been sometimes mistaken by translators, may deserve more particular notice.

1. Pohqui, pohquae´; Narr. pâuqui; Abn. p[oo]'k[oo]ié; 'open,' 'clear' (primarily, 'broken'). In composition with ohke, 'land,' or formed as a verbal in -aug, it denotes 'cleared land' or 'an open place:' as in the names variously written 'Pahquioque,' 'Paquiaug;' 'Pyquaag;' 'Poquaig,' 'Payquaoge,' &c., in Danbury and Wethersfield, and in Athol, Mass.[Pg 40]

2. Pahke (Abn. paⁿg[oo]i,) 'clear,' 'pure'. Found with paug, 'standing water' or 'pond,' in such names as 'Pahcupog,' 'Paquabaug,' &c. See page 16.

3. Pâguan-aü, 'he destroys,' 'he slaughters' (Narr. paúquana, 'there is a slaughter') in composition with ohke denotes 'place of slaughter' or 'of destruction,' and commemorates some sanguinary victory or disastrous defeat. This is probably the meaning of nearly all the names written 'Poquannoc,' 'Pequannoc,' 'Pauganuck,' &c., of places in Bridgeport (Stratfield), Windsor and Groton, Conn., and of a town in New Jersey. Some of these, however, may possibly be derived from paukunni and ohke, 'dark place.'

4. Pemi (Abn. pemai-[oo]i; Del. pimé-u; Cree, peemé;) denotes deviation from a straight line; 'sloping,' 'aslant,' 'twisted.' Pummeeche (Cree, pimich; Chip. pemiji; Abn. pemetsi;) 'crosswise; traverse.' Eliot wrote 'pummeeche may' for 'cross-way,' Obad. 14; and pumetshin (literally, 'it crosses') for 'a cross,' as in up-pumetshin-eum, 'his cross,' Luke xiv. 27. Pemiji-gome or Pemiji-guma, 'cross water,' is the Chippewa name for a lake whose longest diameter crosses the general course of the river which flows through it,—which stretches across, not with the stream. There is such a lake in Minnesota, near the sources of the Mississippi, just below the junction of the two primary forks of that river; another ('Pemijigome') in the chain of small lakes which are the northern sources of the Manidowish (and Chippewa) River in Wisconsin, and still another near the Lacs des Flambeaux, the source of Flambeau River, an affluent of the Manidowish.

The same prefix or its equivalent occurs in the name of a lake in Maine, near the source of the Alligash branch of St. John's River. Mr. Greenleaf, in a list of Indian names made in 1823,[83] gave this as "Baam´chenun´gamo or Ahp´moojee`negmook." Thoreau [84] was informed by his Penobscot[Pg 41] guide, that the name "means 'Lake that is crossed;' because the usual course lies across, not along it." There is another "Cross Lake," in Aroostook county, near the head of Fish River. We seem to recognize, and with less difficulty, the same prefix in Pemigewasset, but the full composition of that name is not clear.

Pemi- denotes, not a crossing of but deviation from a straight line, whether vertical or horizontal. In place-names it may generally be translated by 'sloping' or 'aslant;' sometimes by 'awry' or 'tortuous.' Pemadené, which Râle gives as the Abnaki word for 'mountain,' denotes a sloping mountain-side (pemi-adené), in distinction from one that is steep or precipitous. 'Pemetiq,' the Indian name of Mount Desert Island, as written by Father Biard in 1611, is the Abnaki peme'teki, 'sloping land.' Pemaquid appears to be another form of the word which Râle wrote 'Pemaaⁿkke,' meaning (with the locative suffix) 'at the place where the land slopes;' where "le terre penche; est en talus."[85] Pymatuning, in Pennsylvania, is explained by Heckewelder, as "the dwelling place of the man with the crooked mouth; Pihmtónink" (from pimeu and 't[oo]n).

Wanashque, Anasqui, 'at the extremity of,' 'at the end;' Abn. [oo]anask[oo]i[oo]i, 'au bout;' Cree, wánnusk[oo]tch; Chip. ishkuè, eshqua. See (pp. 18, 19,) Wanashqu-ompsk-ut, Wonnesquam,[86] Winnesquamsaukit, Squamscot. Wonasquatucket, a small river which divides North Providence and Johnston, R.I., retains the name which belonged to the point at which it enters an arm of Narragansett Bay (or Providence River), 'at the end of the tidal-river.' A stream in Rochester, Mass., which empties into the head of an inlet from Buzzard's Bay, received the same name. Ishquagoma, on the upper Embarras[Pg 42] River, Minnesota, is the 'end lake,' the extreme point to which canoes go up that stream.

Names of fishes supply the adjectival components of many place-names on the sea-coast of New England, on the lakes, and along river-courses. The difficulty of analyzing such names is the greater because the same species of fish was known by different names to different tribes. The more common substantivals are -amaug, 'fishing place; -tuk or sipu, 'river;' ohke, 'place;' Abn. -kaⁿtti, 'place of abundance;' and -keag, -keke, Abn. -khigé, which appears to denote a peculiar mode of fishing,—perhaps, by a weir;[87] possibly, a spearing-place.

From the generic namaus (namohs, El.; Abn. namés; Del. namees;) 'a fish'—but probably, one of the smaller sort, for the form is a diminutive,—come such names as Nameoke or Nameaug (New London), for namau-ohke, 'fish country;' Namasket or Namasseket (on Taunton River, in Middleborough, Mass.) 'at the fish place,' a favorite resort of the Indians of that region; Namaskeak, now Amoskeag, on the Merrimack, and Nam'skeket or Skeekeet, in Wellfleet, Mass.

M'squammaug (Abn. mesk[oo]amék[oo]), 'red fish,' i.e. salmon, gave names to several localities. Misquamacuck or Squamicut, now Westerly, R.I., was 'a salmon place' of the Narragansetts. The initial m often disappears; and sometimes, so much of the rest of the name goes with it, that we can only guess at the original synthesis. 'Gonic,' a post office and railroad station, near Dover, N.H., on the Cocheco river, was once 'Squammagonic,'—and probably, a salmon-fishing place.

Kaúposh (Abn. kabassé, plu. kabassak), 'sturgeon,' is a component of the name Cobbosseecontee, in Maine (page 26, ante), 'where sturgeons are plenty;' and Cobscook, an arm of Passamaquoddy Bay, Pembroke, Me., perhaps stands for kabassakhigé, 'sturgeon-catching place.'[Pg 43]

Aumsuog or Ommissuog (Abn. aⁿms[oo]ak), 'small fish,'—especially alewives and herrings,—is a component of the name of the Abnaki village on the Kennebec, Aⁿmes[oo]k-kaⁿtti; of Mattammiscontis, a tributary of the Kennebec (see p. 25, ante), and probably, of Amoscoggin and Amoskeag.

Qunnôsu (pl. -suog; Abn. k[oo]n[oo]sé; Old Alg. kinoⁿjé; Chip. keno´zha;) is found in the name of Kenosha, a town and county in Wisconsin; perhaps, in Kenjua or Kenzua creek and township, in Warren county, Pa. Quinshepaug or Quonshapauge, in Mendon, Mass., seems to denote a 'pickerel pond' (qunnosu-paug). Maskinongé, i.e. massa-kinoⁿjé, 'great pike' or maskelunge, names a river and lake in Canada.

Pescatum, said to mean 'pollock,' occurs as an adjectival in Peskadamioukkaⁿtti, the modern Passamaquoddy (p. 26).

Nahaⁿm[oo], the Abnaki name of the 'eel,' is found in "Nehumkeag, the English of which is Eel Land, ... a stream or brook that empties itself into Kennebec River," not far from Cobbissecontee.[88] This brook was sometimes called by the English, Nehumkee. The Indian name of Salem, Mass., was Nehumkeke or Naümkeag, and a place on the Merrimac, near the mouth of Concord River (now in Lowell, I believe,) had the same name,—written, Naamkeak.

In view of the illustrations which have been given, we repeat what was stated in the beginning of this paper, that Indian place-names are not proper names, that is unmeaning marks, but significant appellatives, each conveying a description of the locality to which it belongs. In those parts of the country where Indian languages are still spoken, the analysis of such names is comparatively easy. Chippewa, Cree, or (in another family) Sioux-Dakota geographical names may generally be translated with as little difficulty as other words or syntheses in the same languages. In New England, and especially in our part of New England, the case is different.[Pg 44] We can hardly expect to ascertain the meaning of all the names which have come down to us from dead languages of aboriginal tribes. Some of the obstacles to accurate analysis have been pointed out. Nearly every geographical name has been mutilated or has suffered change. It would indeed be strange if Indian polysyntheses, with their frequent gutturals and nasals, adopted from unwritten languages and by those who were ignorant of their meanings, had been exempted from the phonetic change to which all language is subject, as a result of the universal disposition "to put more facile in the stead of more difficult sounds or combination of sounds, and to get rid altogether of what is unnecessary in the words we use."[89] What Professor Haldeman calls otosis, 'that error of the ear by which words are perverted to a more familiar form,'[90] has effected some curious transformations. Swatara,[91] the name of a stream in Pennsylvania, becomes 'Sweet Arrow;' the Potopaco of John Smith's map (p[oo]tuppâg, a bay or cove; Eliot,) on a bend of the Potomac, is naturalized as 'Port Tobacco.' Nama'auke, 'the place of fish' in East Windsor, passes through Namerack and Namalake to the modern 'May Luck.' Moskitu-auke, 'grass land,' in Scituate, R.I., gives the name of 'Mosquito Hawk' to the brook which crosses it.[92]

In Connecticut and Rhode Island special causes operated to corrupt and transform almost beyond possibility of recognition, many of the Indian place names. Five different dialects at least were spoken between Narragansett Bay and the Housatonic River, at the time of the first coming of the English. In early deeds and conveyances in the colonial and in local records, we find the same river, lake, tract of land or bound-mark named sometimes in the Muhhekan, sometimes in the Narragansett, or Niantic, or Nipmuck, or Connecticut valley, or Quinnipiac (Quiripee) dialect. The adopted name is often extra-limitary to the tribe by which it was given. Often, it is a mixture of, or a sort of compromise between, two dialects; half Muhhekan, half Narragansett or Nipmuck. In the form in which it comes to us, we can only guess from what language or languages it has been corrupted.

The analysis of those names even whose composition appears to be most obvious must be accepted as provisional merely. The recovery of a lost syllable or of a lost guttural or nasal, the correction of a false accent even, may give to the synthesis another and hitherto unsuspected meaning. It would be surprising if some of the translations which have been hazarded in this paper do not prove to be wide of their[Pg 46] mark. Even English etymology is not reckoned among the exact sciences yet,—and in Algonkin, there is the additional disadvantage of having no Sanskrit verbs "to go," to fall back on as a last resort.

Recent manifestations of an increasing interest in Indian onomatology, or at least of awakened curiosity to discover the meanings of Indian names, may perhaps justify the writer in offering, at the close of this paper, a few suggestions, as to the method of analysis which appears most likely to give correct results, and as to the tests by which to judge of the probability that a supposed translation of any name is the true one.

1. The earliest recorded form of the name should be sought for, and every variation from it should be noted. These should be taken so far as possible from original manuscripts, not from printed copies.

2. Where the difference of forms is considerable, knowledge of the character and opportunities of the writer may sometimes determine the preference of one form to others, as probably the most accurate. A Massachusetts or Connecticut name written by John Eliot or Experience Mayhew—or by the famous interpreter, Thomas Stanton—may safely be assumed to represent the original combination of sounds more exactly than the form given it by some town-recorder, ignorant of the Indian language and who perhaps did not always write or spell his own correctly.

3. The name should be considered with some reference to the topographical features of the region to which it belongs. These may sometimes determine the true meaning when the analysis is doubtful, or may suggest the meaning which would otherwise have been unsuspected under the modern form.

4. Remembering that every letter or sound had its value,—if, in the analysis of a name, it becomes necessary to get rid of a troublesome consonant or vowel by assuming it to have been introduced 'for the sake of euphony,'—it is probable that the interpretation so arrived at is not the right one.

5. The components of every place-name—or to speak more generally, the elements of every Indian synthesis are[Pg 47] significant roots, not mere fractions of words arbitrarily selected for new combinations. There has been no more prolific source of error in dealings with the etymology and the grammatical structure of the American languages than that one-sided view of the truth which was given by Duponceau [93] in the statement that "one or more syllables of each simple word are generally chosen and combined together, in one compound locution, often leaving out the harsh consonants for the sake of euphony,"—and repeated by Heckewelder,[94] when he wrote, that "in the Delaware and other American languages, parts or parcels of different words, sometimes a single sound or letter, are compounded together in an artificial manner so as to avoid the meeting of harsh or disagreeable sounds," &c. The "single sound or letter" the "one or more syllables," were chosen not as "part or parcel" of a word but because of their inherent significance. The Delaware "Pilape, a youth," is not—as Heckewelder and Duponceau represented it to be [95]—"formed from pilsit, chaste, innocent, and lenape, a man," but from pil- (Mass. pen-, Abn. pir-,) strange, novel, unused (and hence) pure,—and -aⁿpe (Mass. -omp, Abn. aⁿbé) a male, vir. It is true that the same roots are found in the two words pil-sit (a participle of the verb-adjective pil-esu, 'he is pure,') and len-aⁿpe, 'common man:' but the statement that "one or more syllables" are taken from these words to form Pilape is inaccurate and misleading. It might with as much truth be said that the English word boyhood is formed from selected syllables of boy-ish and man-hood; or that purity 'compounds together in an artificial manner' fractions of purify and quality.

We meet with similar analyses in almost every published list of Indian names. Some examples have been given in[Pg 48] the preceding pages of this paper,—as in the interpretation of 'Winnipisiogee' (p. 32) by 'the beautiful water of the high place,' s or ēs being regarded as the fractional representative of 'kees, high.' Pemigewasset has been translated by 'crooked place of pines' and 'crooked mountain pine place,'—as if k[oo]-a, 'a pine,' or its plural k[oo]-ash, could dispense in composition with its significant base, k[oo], and appear by a grammatical formative only.

6. No interpretation of a place-name is correct which makes bad grammar of the original. The apparatus of Indian synthesis was cumbersome and perhaps inelegant, but it was nicely adjusted to its work. The grammatical relations of words were never lost sight of. The several components of a name had their established order, not dependent upon the will or skill of the composer. When we read modern advertisements of "cheap gentlemen's traveling bags" or "steel-faced carpenters' claw hammers," we may construe such phrases with a latitude which was not permitted to the Algonkins. If 'Connecticut' means—as some have supposed it to mean—'long deer place,' it denotes a place where long deer abounded; if 'Piscataqua' was named 'great deer river,' it was because the deer found in that river were of remarkable size. 'Coaquanock' or, as Heckewelder wrote it, 'Cuwequenaku,' the site of Philadelphia, may mean 'pine long-place' but cannot mean 'long pine-place' or 'grove of long pine trees.' If 'Pemigewasset' is compounded of words signifying 'crooked,' 'pines,' and 'place,' it denotes 'a place of crooked pines,'—not 'crooked place of pines.'

Again—every Indian name is complete within itself. A mere adjectival or qualificative cannot serve independently, leaving the real ground-word to be supplied by the hearer. River names must contain some element which denotes 'river;' names of lakes or ponds something which stands for 'lake' or 'pond.' The Indians had not our fashion of speech which permits Hudson's River to be called 'the Hudson,' drops the word 'lake' from 'Champlain' or 'Erie,' and makes "the Alleghanies" a geographical name. This difference[Pg 49] must not be lost sight of, in analysis or translation. Agawam or Auguan (a name given to several localities in New England where there are low flat meadows or marshes,) cannot be the equivalent of the Abnaki ag[oo]aⁿn, which means 'a smoke-dried fish,'[96]—though ag[oo]aⁿna-ki or something like it (if such a name should be found), might mean 'smoked-fish place.' Chickahominy does not stand for 'great corn,' nor Pawcatuck for 'much or many deer;'[97] because neither 'corn' nor 'deer' designates place or implies fixed location, and therefore neither can be made the ground-word of a place-name. Androscoggin or Amoscoggin is not from the Abnaki 'amaskohegan, fish-spearing,'[98] for a similar reason (and moreover, because the termination -hēgan denotes always an instrument, never an action or a place; it may belong to 'a fish-spear,' but not to 'fish spearing' nor to the locality 'where fish are speared.')

7. The locative post-position, -et, -it or -ut,[99] means in, at or on,—not 'land' or 'place.' It locates, not the object to the name of which it is affixed, but something else as related to that object,—which must be of such a nature that location can be predicated of it. Animate nouns, that is, names of animate objects cannot receive this affix. 'At the rock' (ompsk-ut), 'at the mountain' (wadchu-ut), or 'in the country' (ohk-it, auk-it), is intelligible, in Indian or English; 'at the deer,' 'at the bear,' or 'at the sturgeons,' would be nonsense in any language. When animate nouns occur in place-[Pg 50]names, they receive the formative of verbals, or serve as adjectival prefixes to some localizing ground-word or noun-generic.

8. Finally,—in the analysis of geographical names, differences of language and dialect must not be disregarded. In determining the primary meaning of roots, great assistance may be had by the comparison of derivatives in nearly related languages of the same stock. But in American languages, the diversity of dialects is even more remarkable than the identity and constancy of roots. Every tribe, almost every village had its peculiarities of speech. Names etymologically identical might have widely different meanings in two languages, or even in two nations speaking substantially the same language. The eastern Algonkin generic name for 'fish' (nâma-us, Del. namai-s) is restricted by northern and western tribes to a single species, the sturgeon (Chip. namai´,) as the fish, par excellence. Attuk, in Massachusetts was the common fallow-deer,—in Canada and the north-west the caribou or reindeer. The Abnaki Indian called his dog (atié) by a name which the Chippewa gives his horse (oti-un; n'di, my horse).[100] The most common noun-generic of river names in New England (-tuk, 'tidal river') occurs rarely in those of Pennsylvania and Virginia, where it is replaced by -hanne ('rapid stream'), and is unknown to western Algonkin tribes whose streams are undisturbed by tides. The analysis of a geographical name must be sought in the language spoken by the name-givers. The correct translation of a Connecticut or Narragansett name is not likely to be attained by searching for its several components in a Chippewa vocabulary; or of the name of a locality near Hudson's River, by deriving its prefix from an Abnaki adverb and its ground-word from a Chippewa participle,—as was actually done in a recently published list of Indian names.

INDIAN NAMES.

FOOTNOTES

[1] Mill's Logic, B. I. ch. viii.

[2] Max Müller, Science of Language, (1st Series,) p. 292.

[3] These terms, though not strictly appropriate to Indian synthesis, are sufficiently explicit for the purposes of this paper. They are borrowed from the author of "Words and Places" (the Rev. Isaac Taylor), who has employed them (2d ed., p. 460) as equivalents of Förstemann's "Bestimmungswort" and "Grundwort," (Die deutschen Ortsnamen. Nordhausen, 1863, pp. 26-107, 109-174). In Indian names, the "Bestimmungswort" sometimes corresponds to the English adjective—sometimes to a noun substantive—but is more generally an adverb.

[4] It has not been thought advisable to attempt the reduction of words or names taken from different languages to a uniform orthography. When no authorities are named, it may be understood that the Massachusetts words are taken from Eliot's translation of the Bible, or from his Indian Grammar; the Narragansett, from Roger Williams's Indian Key, and his published letters; the Abnaki, from the Dictionary of Râle (Rasles), edited by Dr. Pickering; the Delaware, from Zeisberger's Vocabulary and his Grammar; the Chippewa, from Schoolcraft (Sch.), Baraga's Dictionary and Grammar (B.), and the Spelling Books published by the American Board of Commissioners of Foreign Missions; and the Cree, from Howse's Grammar of that language.

The character [oo] (oo in 'food;' w in 'Wabash,' 'Wisconsin'), used by Eliot, has been substituted in Abnaki words for the Greek ou of Râle and the Jesuit missionaries, and for the ω of Campanius. A small ⁿ placed above the line, shows that the vowel which it follows is nasal,—and replaces the ñ employed for the same purpose by Râle, and the short line or dash placed under a vowel, in Pickering's alphabet.

In Eliot's notation, oh usually represents the sound of o in order and in form,—that of broad a; but sometimes it stands for short o, as in not.

[5] Doc. Hist. of New York (4to), vol. iii. p. 656.

[6] Jesuit Relations, 1633, 1636, 1640.

[7] Hind's Exploration of Labrador, i. 9, 32.

[8] Heckewelder's Historical account, &c., p. 33. He was mistaken in translating "the word hittuck," by "a rapid stream."

[9] Col. Records of Connecticut, 1677-89, p. 275.

[10] Chandler's Survey of the Mohegan country, 1705.

[11] See Mourt's Relation, Dexter's edition, pp. 84, 91, 99. Misled by a form of this name, Patackosi, given in the Appendix to Savage's Winthrop (ii. 478) and elsewhere, I suggested to Dr. Dexter another derivation. See his note 297, to Mourt, p. 84.

[12] Descrip. of New Sweden, b. ii. ch. 1, 2; Proud's Hist. of Pennsylvania, ii. 252.

[13] "True Relation of Virginia," &c. (Deane's edition, Boston, 1866), p. 7. On Smith's map, 1606, the 'King's house,' at 'Powhatan,' is marked just below "The Fales" on 'Powhatan flu:' or James River.

[14] History of Hadley, pp. 121, 122.

[15] See Hist. Magazine, vol. iii. p. 48.

[16] Heckewelder, on Indian names, in Trans. Am. Phil. Soc. vol. iv.

[18] Narrative, &c., in Mem. Hist. Society of Pennsylvania, vol. ii. p. 97.

[19] Grammar of the Lenni-Lenape, transl. by Duponceau, p. 43. "Wulit, good." "Welsit (masc. and fem.), the best." "Inanimate, Welhik, best."

[20] Morgan's League of the Iroquois, p. 436.

[21] Published in London, 1759, and re-printed in Appendix to Proud's Hist. of Penn., vol. ii. pp. 65-132.

[22] Shea's Early Voyages on the Mississippi, p. 75.

La Metairie's 'Olighinsipou' suggests another possible derivation which may be worth mention. The Indian name of the Alleghanies has been said,—I do not now remember on whose authority,—to mean 'Endless Mountains.' 'Endless' cannot be more exactly expressed in any Algonkin language than by 'very long' or 'longest,'—in the Delaware, Eluwi-guneu. "The very long or longest river" would be Eluwi-guneu sipu, or, if the words were compounded in one, Eluwi-gunesipu.

[23] Paper on Indian names, ut supra, p. 367; Historical Account, &c., pp. 29-32.

[24] Morgan's League of the Iroquois, pp. 466, 468.

[25] Ms. Itinerary. He was careful to preserve the Indian pronunciation of local names, and the form in which he gives this name convinces me that it is not, as I formerly supposed, the quinnuppohke (or quinuppeohke) of Eliot,—meaning 'the surrounding country' or the 'land all about' the site of New Haven.

[26] Dictionary, s.v. 'Noms.'

[27] Paug is regularly formed from pe (Abn. bi), the base of nippe, and may be translated more exactly by 'where water is' or 'place of water.'

[28] A bound of Human Garret's land, one mile north-easterly from Ninigret's old Fort. See Conn. Col. Records, ii. 314.

[29] Foster and Whitney's Report on the Geology of Lake Superior, &c., Pt. II p. 400.

[30] Râle gives Abn. mitsegan, 'fianté.' Thoreau, fishing in a river in Maine, caught several sucker-like fishes, which his Abnaki guide threw away, saying they were 'Michegan fish, i.e., soft and stinking fish, good for nothing.'—Maine Woods, p. 210.

[31] Primarily, that which 'breaks,' 'cleaves,' 'splits:' distinguishing the harder rocks—such as were used for making spear and arrow heads, axes, chisels, corn-mortars, &c., and for striking fire,—from the softer, such as steatite (soap-stone) from which pots and other vessels, pipe-bowls, &c., were fashioned.

[32] Mass. Records, i. 147, 226.

[33] Squantam, the supposed name of an Algonkin deity, is only a corrupt form of the verb m'squantam, = musqui-antam, 'he is angry,' literally, 'he is red (bloody-) minded.'

[34] Maine Woods, pp. 145, 324.

[35] Pres. Stiles's Itinerary, 1761.

[36] Conn. Col. Records, i. 434.

[37] Hind's Exploration of Labrador, vol. ii. pp. 147, 148.

[38] History of Hadley, 21, 22, 114.

[39] W.F. Goodwin, in Historical Magazine, ix. 28.

[40] About half-way from Tisbury to Gay Head.

[41] Râle, s.v. Village.

[42] Shea's Hist. of Catholic Missions, 142, 145.

[43] Description of New Sweden, b. ii. c. 8. (Duponceau's translation.)

[44] N.Y. Hist. Soc. Collections, iii. 375.

[45] Dawson's Acadian Geology, App. p. 673.

[46] 4th Mass. Hist. Collections, vi. 267.

[47] Otchipwe Grammar, pp. 87, 412.

[48] Mr. Rand's Micmac Vocabulary, in Schoolcraft's Collections, vol. v. p. 579.

[49] Coll. Me. Hist. Society, iv. 31, 105.

[50] The statement that the Androscoggin received its present name in compliment to Edmond Andros, about 1684, is erroneous. This form of the name appears as early as 1639, in the release by Thomas Purchase to the Governor of Massachusetts,—correctly printed (from the original draft in the handwriting of Thomas Lechford) in Mass. Records, vol. i. p. 272.

[51] Information respecting the Indian Tribes, &c., vol. iii. p. 526.

[52] Depositions in Coll. Me. Histor. Society, iv. 113.

[53] 3 Mass. Hist. Coll., iii. 181.

[54] Dawson's Acadian Geology, 2d ed., (London, 1868), pp. 3, 8.

[55] Acadian Geology, pp. 1, 3.

[56] Maine Woods, pp. 194, 284, 326.

[57] Voyages, p. 44.

[58] See Coll. Me. Hist. Society, 2d Ser., vol. i. p. 234.

[59] Acadian Geology, l.c.

[60] Cited by Dr. Kohl, in Coll. Me. Hist. Society, N.S., i. 416.

[61] See Narragansett Club Publications, vol. i. p. 22 (note 6).

[62] On Block's Map, 1616, the "Nahicans" are marked on the easternmost point of Long Island.

[63] Judd's History of Hadley, 115, 116, 117.

[64] Mr. Moses Greenleaf, in 1823, wrote this name, Bakungunahik.

[65] On Indian names, in Trans. Am. Phil. Society, N.S., vol. iv., p. 377.

[67] Paper on Indian Names, ut supra, p. 366; and 3 Mass. Historical Collections, vi. 145. [Compare, the Iroquois Swa-deh´ and Oswa´-go (modern Oswego), which has the same meaning as Alg. sauki,—"flowing out."—Morgan's League of the Iroquois.]

[68] Saguinam, Charlevoix, i. 501; iii. 279.

[69] Relations des Jésuites, 1658, p. 22; 1648, p. 62; 1671, pp. 25, 31.

[70] Charlevoix, Nouv. France, iii. 65; Gallatin's Synopsis, p. 24.

[71] This name is still retained.

[72] When first discovered the Saguenay was not regarded as a river, but as a strait or passage by which the waters of some northern sea flowed to the St. Lawrence. But on a French map of 1543, the 'R. de Sagnay' and the country of 'Sagnay' are laid down. See Maine Hist. Soc. Collections, 2d Series, vol. i., pp. 331, 354. Charlevoix gives Pitchitaouichetz, as the Indian name of the River.

[73] And in the Historical Magazine, vol. i. p. 246.

[74] Vol. i. p. 246.

[75] See pp. 14, 15.

[76] Chandler's Survey and Map of the Mohegan country, 1705. Compare the Chip. ashawiwi-sitagon, "a place from which water runs two ways," a dividing ridge or portage between river courses. Owen's Geological Survey of Wisconsin, etc., p. 312.

[77] Blake's Annals of Dorchester, p. 9; Winthrop's Journal, vol. i. p. 28.

[78] On Indian Names, in Trans. Am. Philos. Society, N.S. iv. 361.

[79] On Indian Names (ut supra), p. 365.

[80] Stiles's History of Ancient Windsor, p. 111.

[81] Printed in note to Savage's Winthrop's Journal, ii. 180.

[82] See Thornton's Ancient Pemaquid, in Maine Hist. Collections, v. 156.

[83] Report of American Society for Promoting Civilization of the Indian Tribes, p. 52.

[84] Maine Woods, 232.

[85] Abnaki Dictionary, s.v. Pencher. Compare, p. 545, "bimk[oo]é, il penche naturellement la tête sur un côte."

[86] Wonnesquam (as should have been mentioned on the page referred to) may possibly represent the Abnaki [oo]anask[oo]aⁿaⁿmi[oo]i or -mek 'at the end of the peninsula' ('au bout de la presqu'ile.' Râle).

[87] Schoolcraft derives the name of the Namakagun fork of the St. Croix river, Wisc., from Chip. "namai, sturgeon, and kagun, a yoke or weir."

[88] Col. William Lithgow's deposition, 1767,—in New England Historical and General Register, xxiv. 24.

[89] Whitney's Language and the Study of Language, p. 69.—"Ein natürliches Volksgefühl, oft auch der Volkswitz, den nicht mehr verstandenen Namen neu umprägte und mit anderen lebenden Wörtern in Verbindung setzte." Dr. J. Bender, Die deutschen Ortsnamen (2te Ausg.) p. 2.

[90] Haldeman's Analytic Orthography, §279, and "Etymology as a means of Education," in Pennsylvania School Journal for October, 1868.

[91] "Swatawro," on Sayer and Bennett's Map, 1775.

[92] "Whiskey Jack," the name by which the Canada Jay (Perisoreus Canadensis) is best known to the lumbermen and hunters of Maine and Canada, is the Montagnais Ouishcatchaⁿ (Cree, Ouiskeshauneesh), which has passed perhaps through the transitional forms of 'Ouiske Jean' and 'Whiskey Johnny.' The Shagbark Hickory nuts, in the dialect of the Abnakis called s'k[oo]skada´mennar, literally, 'nuts to be cracked with the teeth,' are the 'Kuskatominies' and 'Kisky Thomas' nuts of descendants of the Dutch colonists of New Jersey and New York. A contraction of the plural form of a Massachusetts noun-generic,—asquash, denoting 'things which are eaten green, or without cooking,' was adopted as the name of a garden vegetable,—with conscious reference, perhaps, to the old English word squash, meaning 'something soft or immature.' Sometimes etymology overreaches itself, by regarding an aboriginal name as the corrupt form of a foreign one. Thus the maskalongé or 'great long-nose' of the St. Lawrence (see p. 43) has been reputed of French extraction,—masque elongé: and sagackomi, the northern name of a plant used as a substitute for or to mix with tobacco,—especially, of the Bearberry, Arctostaphylos uva-ursi,—is resolved into sac-à-commis, "on account of the Hudson's Bay officers carrying it in bags for smoking," as Sir John Richardson believed (Arctic Searching Expedition, ii. 303). It was left for the ingenuity of a Westminster Reviewer to discover that barbecue (denoting, in the language of the Indians of Guiana, a wooden frame or grille on which all kinds of flesh and fish were dry-roasted, or cured in smoke,) might be a corruption of the French barbe à queue, i.e. 'from snout to tail;' a suggestion which appears to have found favor with lexicographers.

THE COMPOSITION OF

INDIAN GEOGRAPHICAL NAMES,

ILLUSTRATED FROM THE

ALGONKIN LANGUAGES.

BY J. HAMMOND TRUMBULL.

ON THE COMPOSITION OF

INDIAN GEOGRAPHICAL NAMES.

INDIAN NAMES.

FOOTNOTES