The terms “family” and “stock” are here applied interchangeably to a group of languages that are supposed to be cognate.

A single language is called a stock or family when it is not found to be cognate with any other language. Languages are said to be cognate when such relations between them are found that they are supposed to have descended from a common ancestral speech. The evidence of cognation is derived exclusively from the vocabulary. Grammatic similarities are not supposed to furnish evidence of cognation, but to be phenomena, in part relating to stage of culture and in part adventitious. It must be remembered that extreme peculiarities of grammar, like the vocal mutations of the Hebrew or the monosyllabic separation of the Chinese, have not been discovered among Indian tongues. It therefore becomes necessary in the classification of Indian languages into families to neglect grammatic structure, and to consider lexical elements only. But this statement must be clearly understood. It is postulated that in the growth of languages new words are formed by combination, and that these new words change by attrition to secure economy of utterance, and also by assimilation (analogy) for economy of thought. In the comparison of languages for the purposes of systematic philology it often becomes necessary to dismember compounded words for the purpose of comparing the more primitive forms thus obtained. The paradigmatic words considered in grammatic treatises may often be the very words which should be dissected to discover in their elements primary affinities. But the comparison is still lexic, not grammatic.

A lexic comparison is between vocal elements; a grammatic comparison is between grammatic methods, such, for example, as gender systems. The classes into which things are relegated by distinction of gender may be animate and inanimate, and the animate may subsequently be divided into male and female, and these two classes may ultimately absorb, in part at least, inanimate things. The growth of a system of genders may take another course. The animate and inanimate may be subdivided into the standing, the sitting, and the lying, or into the moving, the erect and the reclined; or, still further, the superposed classification may be based upon the supposed constitution of things, as the fleshy, the woody, the rocky, the earthy, the watery. Thus the number of genders may increase, while further on in the history of a language the genders may decrease so as almost to disappear. All of these characteristics are in part adventitious, but to a large extent the gender is a phenomenon of growth, indicating the stage to which the language has attained. A proper case system may not have been established in a language by the fixing of case particles, or, having been established, it may change by the increase or diminution of the number of cases. A tense system also has a beginning, a growth, and a decadence. A mode system is variable in the various stages of the history of a language. In like manner a pronominal system undergoes changes. Particles may be prefixed, infixed, or affixed in compounded words, and which one of these methods will finally prevail can be determined only in the later stage of growth. All of these things are held to belong to the grammar of a language and to be grammatic methods, distinct from lexical elements.

With terms thus defined, languages are supposed to be cognate when fundamental similarities are discovered in their lexical elements. When the members of a family of languages are to be classed in subdivisions and the history of such languages investigated, grammatic characteristics become of primary importance. The words of a language change by the methods described, but the fundamental elements or roots are more enduring. Grammatic methods also change, perhaps even more rapidly than words, and the changes may go on to such an extent that primitive methods are entirely lost, there being no radical grammatic elements to be preserved. Grammatic structure is but a phase or accident of growth, and not a primordial element of language. The roots of a language are its most permanent characteristics, and while the words which are formed from them may change so as to obscure their elements or in some cases even to lose them, it seems that they are never lost from all, but can be recovered in large part. The grammatic structure or plan of a language is forever changing, and in this respect the language may become entirely transformed.

LITERATURE RELATING TO THE CLASSIFICATION OF INDIAN LANGUAGES.

While the literature relating to the languages of North America is very extensive, that which relates to their classification is much less extensive. For the benefit of future students in this line it is thought best to present a concise account of such literature, or at least so much as has been consulted in the preparation of this paper.

The larger part of the volume consists of Gallatin’s paper. A short chapter is devoted to general observations, including certain historical data, and the remainder to the discussion of linguistic material and the affinities of the various tribes mentioned. Vocabularies of many of the families are appended. Twenty-eight linguistic divisions are recognized in the general table of the tribes. Some of these divisions are purely geographic, such as the tribes of Salmon River, Queen Charlotte’s Island, etc. Vocabularies from these localities were at hand, but of their linguistic relations the author was not sufficiently assured. Most of the linguistic families recognized by Gallatin were defined with much precision. Not all of his conclusions are to be accepted in the presence of the data now at hand, but usually they were sound, as is attested by the fact that they have constituted the basis for much classificatory work since his time.

The primary, or at least the ostensible, purpose of the colored map which accompanies Gallatin’s paper was, as indicated by its title, to show the distribution of the tribes, and accordingly their names appear upon it, and not the names of the linguistic families. Nevertheless, it is practically a map of the linguistic families as determined by the author, and it is believed to be the first attempted for the area represented. Only eleven of the twenty-eight families named in this table appear, and these represent the families with which he was best acquainted. As was to be expected from the early period at which the map was constructed, much of the western part of the United States was left uncolored. Altogether the map illustrates well the state of knowledge of the time.

In Chapter XXII of this volume the author gives a brief synopsis of the Indian tribes east of the Mississippi, under a linguistic classification, and adds a brief account of the character and methods of Indian languages. A linguistic map of the region is incorporated, which in general corresponds with the one published by Gallatin in 1836. A notable addition to the Gallatin map is the inclusion of the Uchees in their proper locality. Though considered a distinct family by Gallatin, this tribe does not appear upon his map. Moreover, the Choctaws and Muskogees, which appear as separate families upon Gallatin’s map (though believed by that author to belong to the same family), are united upon Bancroft’s map under the term Mobilian.

The linguistic families treated of are, I. Algonquin, II. Sioux or Dahcota, III. Huron-Iroquois, IV. Catawba, V. Cherokee, VI. Uchee, VII. Natchez, VIII. Mobilian.

The chapter cited is short, but long enough to enable the author to construct a very curious classification of the tribes of which he treats. In his account Scouler is guided chiefly, to use his own words, “by considerations founded on their physical character, manners and customs, and on the affinities of their languages.” As the linguistic considerations are mentioned last, so they appear to be the least weighty of his “considerations.”

Scouler’s definition of a family is very broad indeed, and in his “Northern Family,” which is a branch of his “Insular Group,” he includes such distinct linguistic stocks as “all the Indian tribes in the Russian territory,” the Queen Charlotte Islanders, Koloshes, Ugalentzes, Atnas, Kolchans, Kenáïes, Tun Ghaase, Haidahs, and Chimmesyans. His Nootka-Columbian family is scarcely less incongruous, and it is evident that the classification indicated is only to a comparatively slight extent linguistic.

In addition to a large amount of ethnographic data derived from the Polynesian Islands, Micronesian Islands, Australia, etc., more than one-half of this important volume is devoted to philology, a large share relating to the tribes of northwestern America.

The vocabularies collected by Hale, and the conclusions derived by him from study of them, added much to the previous knowledge of the languages of these tribes. His conclusions and classification were in the main accepted by Gallatin in his linguistic writings of 1848.

In this article, which was read before the Philological Society, January 24, 1845, a large number of North American languages are examined and their affinities discussed in support of the two following postulates made at the beginning of the paper: First, “No American language has an isolated position when compared with the other tongues en masse rather than with the language of any particular class;” second, “The affinities between the language of the New World, as determined by their vocabularies, is not less real than that inferred from the analogies of their grammatical structure.” The author’s conclusions are that both statements are substantiated by the evidence presented. The paper contains no new family names.

It was the purpose of this author, as avowed by himself, to determine whether the races of men are the cooffspring of a single stock or have descended respectively from several original families. Like other authors on this subject, his theory of what should constitute a race was not clearly defined. The scope of the inquiry required the consideration of a great number of subjects and led to the accumulation of a vast body of facts. In volume 5 the author treats of the American Indians, and in connection with the different tribes has something to say of their languages. No attempt at an original classification is made, and in the main the author follows Gallatin’s classification and adopts his conclusions.

The introduction consists of a number of chapters, as follows: First, Geographical notices and Indian means of subsistence; second, Ancient semi-civilization of New Mexico, Rio Gila and its vicinity; third, Philology; fourth, Addenda and miscellaneous. In these are brought together much valuable information, and many important deductions are made which illustrate Mr. Gallatin’s great acumen. The classification given is an amplification of that adopted in 1836, and contains changes and additions. The latter mainly result from a consideration of the material supplied by Mr. Hale, or are simply taken from his work.

The groups additional to those contained in the Archæologia Americana are:

This paper was read before the Ethnological Society on the 11th of December. The languages noticed are those that lie between “Russian America and New California,” of which the author aims to give an exhaustive list. He discusses the value of the groups to which these languages have been assigned, viz, Athabascan and Nootka-Columbian, and finds that they have been given too high value, and that they are only equivalent to the primary subdivisions of stocks, like the Gothic, Celtic, and Classical, rather than to the stocks themselves. He further finds that the Athabascan, the Kolooch, the Nootka-Columbian, and the Cadiak groups are subordinate members of one large and important class—the Eskimo.

No new linguistic groups are presented.

This essay was read before the Ethnological Society February 19, 1845. Brief notices are given of the more important tribes, and the languages are classed in two groups, the Eskimaux and the Kolooch. Each of these groups is found to have affinities—

(1) With the Athabascan tongues, and perhaps equal affinities.

(2) Each has affinities with the Oregon languages, and each perhaps equally.

(3) Each has definite affinities with the languages of New California, and each perhaps equal ones.

(4) Each has miscellaneous affinities with all the other tongues of North and South America.

This, the first edition of this well known atlas, contains, among other maps, an ethnographic map of North America, made in 1845. It is based, as is stated, upon material derived from Gallatin, Humboldt, Clavigero, Hervas, Vater, and others. So far as the eastern part of the United States is concerned it is largely a duplication of Gallatin’s map of 1836, while in the western region a certain amount of new material is incorporated.

1852. In the edition of 1852 the ethnographic map bears date of 1851. Its eastern portion is substantially a copy of the earlier edition, but its western half is materially changed, chiefly in accordance with the knowledge supplied by Hall in 1848.

Map number 72 of the last edition of Berghaus by no means marks an advance upon the edition of 1852. Apparently the number of families is much reduced, but it is very difficult to interpret the meaning of the author, who has attempted on the same map to indicate linguistic divisions and tribal habitats with the result that confusion is made worse confounded.

This short paper by Gallatin consists of a letter addressed to W. Medill, Commissioner of Indian Affairs, requesting his cooperation in an endeavor to obtain vocabularies to assist in a more complete study of the grammar and structure of the languages of the Indians of North America. It is accompanied by a “Synopsis of Indian Tribes,” giving the families and tribes so far as known. In the main the classification is a repetition of that of 1848, but it differs from that in a number of particulars. Two of the families of 1848 do not appear in this paper, viz, Arapaho and Kinai. Queen Charlotte Island, employed as a family name in 1848, is placed under the Wakash family, while the Skittagete language, upon which the name Queen Charlotte Island was based in 1848, is here given as a family designation for the language spoken at “Sitka, bet. 52 and 59 lat.” The following families appear which are not contained in the list of 1848:

The “Observations” are introductory to a series of vocabularies collected in northern California, and treat of the method employed in collecting them and of the difficulties encountered. They also contain notes on the tribes speaking the several languages as well as on the area covered. There is comparatively little of a classificatory nature, though in one instance the name Quoratem is proposed as a proper one for the family “should it be held one.”

Read before the Philological Society, May 13, 1853. A number of languages are examined in this paper for the purpose of determining the stocks to which they belong and the mutual affinities of the latter. Among the languages mentioned are the Saintskla, Umkwa, Lutuami, Paduca, Athabascan, Dieguno, and a number of the Mission languages.

The letter forms half a page of printed matter. The gist of the communication is in effect that the author has heard it said that the Indians of certain pueblos speak three different languages, which he has heard called, respectively, (1) Chu-cha-cas and Kes-whaw-hay; (2) E-nagh-magh; (3) Tay-waugh. This can hardly be called a classification, though the arrangement of the pueblos indicated by Lane is quoted at length by Keane in the Appendix to Stanford’s Compendium.

This paper was read before the Philological Society May 9, 1856, and is stated to be “a supplement to two well known contributions to American philology by the late A. Gallatin.”

So far as classification of North American languages goes, this is perhaps the most important paper of Latham’s, as in it a number of new names are proposed for linguistic groups, such as Copeh for the Sacramento River tribes, Ehnik for the Karok tribes, Mariposa Group and Mendocino Group for the Yokut and Pomo tribes respectively, Moquelumne for the Mutsun, Pujuni for the Meidoo, Weitspek for the Eurocs.

Chapter V of the above report is headed “Vocabularies of North American Languages,” and is by Turner, as is stated in a foot-note. Though the title page of Part III is dated 1855, the chapter by Turner was not issued till 1856, the date of the full volume, as is stated by Turner on page 84. The following are the vocabularies given, with their arrangement in families:

Several of the family names, viz, Keres, Kiowa, Yuma, and Zuñi, have been adopted under the rules formulated above.

This work contains a historic review of early discoveries in New Mexico and of the tribes living therein, with such vocabularies as were available at the time. On pages 315-414 the tribes of British America, from about latitude 54° to 60°, are similarly treated, the various discoveries being reviewed; also those on the North Pacific coast. Much of the material should have been inserted in the volume of 1859 (which was prepared in 1854), to which cross reference is frequently made, and to which it stands in the nature of a supplement.

The above, forming a second supplemental volume of the Transactions for 1854, is an extensive compilation of much previous literature treating of the Indian tribes from the Arctic Ocean southward to Guadalajara, and bears specially upon the Aztec language and its traces in the languages of the numerous tribes scattered along the Pacific Ocean and inland to the high plains. A large number of vocabularies and a vast amount of linguistic material are here brought together and arranged in a comprehensive manner to aid in the study attempted. In his classification of the tribes east of the Rocky Mountains, Buschmann largely followed Gallatin. His treatment of those not included in Gallatin’s paper is in the main original. Many of the results obtained may have been considered bold at the time of publication, but recent philological investigations give evidence of the value of many of the author’s conclusions.

The interesting account of the author’s travels among the Indians, chiefly in the Northwest, and of their habits, is followed by a four page supplement, giving the names, locations, and census of the tribes of the Northwest coast. They are classified by language into Chymseyan, including the Nass, Chymseyans, Skeena and Sabassas Indians, of whom twenty-one tribes are given; Ha-eelb-zuk or Ballabola, including the Milbank Sound Indians, with nine tribes; Klen-ekate, including twenty tribes; Hai-dai, including the Kygargey and Queen Charlotte’s Island Indians, nineteen tribes being enumerated; and Qua-colth, with twenty-nine tribes. No statement of the origin of these tables is given, and they reappear, with no explanation, in Schoolcraft’s Indian Tribes, volume V, pp. 487-489.

In his Queen Charlotte Islands, 1870, Dawson publishes the part of this table relating to the Haida, with the statement that he received it from Dr. W. F. Tolmie. The census was made in 1836-’41 by the late Mr. John Work, who doubtless was the author of the more complete tables published by Kane and Schoolcraft.

The object of this volume is, as the author states in his preface, “to lay before the reader the chief facts and the chief trains of reasoning in Comparative Philology.” Among the great mass of material accumulated for the purpose a share is devoted to the languages of North America. The remarks under these are often taken verbatim from the author’s earlier papers, to which reference has been made above, and the family names and classification set forth in them are substantially repeated.

This is a valuable contribution to our knowledge of the Missouri River tribes, made at a time when the information concerning them was none too precise. The tribes treated of are classified as follows:

I.	Knisteneaux, or Crees.		Algonkin Group, A.
II.	Blackfeet.
III.	Shyennes.
IV.	Arapohos.	Arapoho Group, B.
V.	Atsinas.
VI.	Pawnees.	Pawnee Group, C.
VII.	Arikaras.
VIII.	Dakotas.		Dakota Group, D.
IX.	Assiniboins.
X.	Crows.
XI.	Minnitarees.
XII.	Mandans.
XIII.	Omahas.
XIV.	Iowas.

The work is divided into three parts. (1) Tentative classification of the languages of Mexico; (2) notes on the immigration of the tribes of Mexico; (3) geography of the languages of Mexico.

The author states that he has no knowledge whatever of the languages he treats of. All he attempts to do is to summarize the opinions of others. His authorities were (1) writers on native grammars; (2) missionaries; (3) persons who are reputed to be versed in such matters. He professes to have used his own judgment only when these authorities left him free to do so.

His stated method in compiling the ethnographic map was to place before him the map of a certain department, examine all his authorities bearing on that department, and to mark with a distinctive color all localities said to belong to a particular language. When this was done he drew a boundary line around the area of that language. Examination of the map shows that he has partly expressed on it the classification of languages as given in the first part of his text, and partly limited himself to indicating the geographic boundaries of languages, without, however, giving the boundaries of all the languages mentioned in his lists.

According to the introduction this work is divided into three parts: (1) descriptive; (2) comparative; (3) critical.

The author divides the treatment of each language into (1) its mechanism; (2) its dictionary; (3) its grammar. By “mechanism” he means pronunciation and composition; by “dictionary” he means the commonest or most notable words.

In the case of each language he states the localities where it is spoken, giving a short sketch of its history, the explanation of its etymology, and a list of such writers on that language as he has become acquainted with. Then follows: “mechanism, dictionary, and grammar.” Next he enumerates its dialects if there are any, and compares specimens of them when he is able. He gives the Our Father when he can.

Volume I (1862) contains introduction and twelve languages. Volume II (1865) contains fourteen groups of languages, a vocabulary of the Opata language, and an appendix treating of the Comanche, the Coahuilteco, and various languages of upper California.

Volume III (announced in preface of Volume II) is to contain the “comparative part” (to be treated in the same “mixed” method as the “descriptive part”), and a scientific classification of all the languages spoken in Mexico.

In the “critical part” (apparently dispersed through the other two parts) the author intends to pass judgment on the merits of the languages of Mexico, to point out their good qualities and their defects.

In this important paper is presented much interesting information concerning the inhabitants of Alaska and adjacent territories. The natives are divided into two groups, the Indians of the interior, and the inhabitants of the coast, or Esquimaux. The latter are designated by the term Orarians, which are composed of three lesser groups, Eskimo, Aleutians, and Tuski. The Orarians are distinguished, first, by their language; second, by their distribution; third, by their habits; fourth, by their physical characteristics.

The classification followed is practically the same as is given in the author’s article in the Proceedings of the American Association for the Advancement of Science.

This is an amplification of the paper published in the Proceedings of the American Association, as above cited. The author states that “numerous additions and corrections, as well as personal observations of much before taken at second hand, have placed it in my power to enlarge and improve my original arrangement.”

In this paper the Orarians are divided into “two well marked groups,” the Innuit, comprising all the so-called Eskimo and Tuskis, and the Aleuts. The paper proper is followed by an appendix by Gibbs and Dall, in which are presented a series of vocabularies from the northwest, including dialects of the Tlinkit and Haida nations, T’sim-si-ans, and others.

This is a valuable article, and gives many interesting particulars of the tribes of which it treats. References are here and there made to the languages of the several tribes, with, however, no attempt at their classification. A table follows the report, in which is given by Dall, after Gibbs, a classification of the tribes mentioned by Gibbs. Five families are mentioned, viz: Nūtka, Sahaptin, Tinneh, Selish, and T’sinūk. The comparative vocabularies follow Part II.

The extended paper on the Californian tribes which makes up the bulk of this volume is the most important contribution to the subject ever made. The author’s unusual opportunities for personal observation among these tribes were improved to the utmost and the result is a comparatively full and comprehensive account of their habits and character.

Here and there are allusions to the languages spoken, with reference to the families to which the tribes belong. No formal classification is presented.

This appendix consists of a series of comparative vocabularies collected by Powers, Gibbs and others, classified into linguistic families, as follows:

After some remarks concerning the nature of language and of the special characteristics of Indian languages, the author gives a synopsis of the languages of the Pacific region. The families mentioned are:

This is an important paper, and contains notices of several new stocks, derived from a study of the material furnished by Powers.

The author advocates the plan of using a system of nomenclature similar in nature to that employed in zoology in the case of generic and specific names, adding after the name of the tribe the family to which it belongs; thus: Warm Springs, Sahaptin.

This is a half-column article, the object of which is to assign the several Pueblos to their proper stocks. A paragraph is devoted to contradicting the popular belief that the Pueblos are in some way related to the Aztecs. No vocabularies are given or cited, though the classification is stated to be a linguistic one.

In the appendix are given, first, some of the more general characteristics and peculiarities of Indian languages, followed by a classification of all the tribes of North America, after which is given an alphabetical list of American tribes and languages, with their habitats and the stock to which they belong.

The classification is compiled from many sources, and although it contains many errors and inconsistencies, it affords on the whole a good general idea of prevalent views on the subject.

This is a two-page article in which is set forth a classification of the Pueblo Indians from linguistic considerations. The Pueblos are divided into four families or stocks, viz:

Under the several stocks is given a list of those who have collected vocabularies of these languages and a reference to their publication.

This is a brief article—two and a half pages—on the Twana, Clallam, and Chemakum Indians. The author finds, upon a comparison of vocabularies, that the Chemakum language has little in common with its neighbors.

This paper is a timely contribution to the subject of the Alaska tribes, and carries it from the point at which the author left it in 1869 to date, briefly summarizing the several recent additions to knowledge. It ends with a geographical classification of the Innuit and Indian tribes of Alaska, with estimates of their numbers.