^' IMAGE EVALUATION TEST TARGET (MT-3) 1.0 I.I |50 '™=== e m M Z2 1.8 1-25 1.4 1,6 ■• 6" ► V] /J >> % Photographic Sciences Corporation 23 WEST MAIN STREET WEBSTER, N.Y. 14580 (716) 872-4503 CIHM/ICMH Microfiche Series. CIHM/ICMH Collection de microfiches. Canadian Institute for Historical Microreproductions / Institut Canadian de microreproductions historiques Technical and Bibliographic Notes/Notas techniques et bfoliooraphiques The Institute has attempted to obtain the best original copy available for filming. Features of this copy which may be bibliographically unique, wiilch may alter any of the images in the reproduction, or which may significantly change the usual method of filming, are checked below. D Coloured covers/ Couverture de couleur I I Covers damaged/ D Couverture endommagie Covers restored and/or laminated/ Couverture restaur^ et/ou pelliculte I I Cover title missing/ Le titre de couverture manque I I Coloured maps/ D Cartes giographiques en couleur Coloured ink (i.e. other than blue or black)/ Ence de couleur (i.e. autre que bloue ou noire) □ Coloured plates and/or illustrations/ Planches et/ou illustrations en couleur Q Bound with other material/ ReiiA avec d'autres documents D D D Tight binding may cause shadows or distortion along interior margin/ La reiiure serrie peut causer de i'ombre ou de la distortion le long de la marge IntArieure Blank leaves added during restoration may appear within the text. Whenever possible, these have been omitted from filming/ ii se peut que certaines pages blanches ejouttes iors d'une restauration apparaissent dans le texte, mais, iorsque cela 6tait possible, ces pages n'ont pas it* filmies. Additional comments:/ Commentaires supplAmentaires; L'Institut a microfilm^ le meiileur exempiaire qu'il iul a 6t6 possible de se procurer. Les details de cet exempiaire qui sont peut-Atre uniques du point de vue bibllographlc.ue, qui peuvent modifier une image reproduite, ou qui peuvent exiger une modification dans la methods normaie de fi'maga sont indiquAs ci-dessous. □ Coloured pages/ Pages de couleur □ Pages damaged/ Pages endommagies I — I Pages restored and/or laminated/ D This item is filmed at the reduction ratio checked below/ Ce document est film* au taux de reduction indiquA ci-dessous. Pages restaurtes et/ou peiiiculAes Pnges discoloured, stained or foxe( Pages dicolories, tacheties ou piqu6es Pages detached/ Pages d6tach6es Showthroughy Transparence Quality of prir Quality inigaie de i'impression includes supplementary materii Comprend du materiel suppiimentaire Only edition available/ Seule Mition disponlble Tl to I — I Pnges discoloured, stained or foxed/ I I Pages detached/ r~~| Showthrough/ I I Quality of print varies/ |~~| includes supplementary material/ I — I Only edition available/ P< of fil Oi bfl th si( ot fir si< or Pages wholly or partially obscured by errata slips, tissues, etc., have been ref'imed to ensure the best possible image/ Les pages totalement ou partieilement obscurcies pur un feuiilet d'errata, une pelure, etc., ont AtA fiimies A nouveau de fa9on A obtenir la meilleure image possible. Th sh Til wl Mi dii enl be rig ret m< 10X 14X 18X 22X 26X 30X \y ^ 12X 16X 20X • 24X ".s( 28X 32X The copy filmad h«r« has bean reproduced thanks to the generosity of: Library Division Provincial Archives of British Columbia The images appearing iiere are the best quality possible considering the condition and legibility of the original copy and in keeping with the filmiug contract specifications. Original copies in printed paper covers are filmed beginning wy heterogeneous tribes, and inhabit a country some 1,500 miles south o( the most sou- thern Denes' liunting grounds differ in physical typ.;, natural dispositions, manners and customs from our Indians. Their legends, myths ami i -ligious observances have no eciuivalents here, and yet their speech stamps them at once as an offshoot of the great D6no family. +Am. Ant. Coll., Vol. II. KS89-9I'. THE DtSE LANGUAGES. 171 *»* ago before the American Association for the Advancement of Science, the fundamental proposition of which is that " the only satisfactory evi- dence of affiliation or direct relationship of two communities, apart from authentic historical records, is their speech."* Well might that multifariously-gifted philosopher Leibnitz deplore the carelessness ot the explorers of his time, who did not think it worth their while to collect vocabularies of the languages spoken by the nations they pretended to make known to the civilized world. " C'est un grand ddfaut," he says, " que ceux qui font des descriptions de pays et des re- lations de voyages oublient d'ajouter des essais des langues des peuples, of car cela servirait pour en faire connaitre les origines."f This reproach the German philosopher, though addressed to travellers in foreign lands, might be construed as applying \yith even a greater degree of force to those who, like the missionaries, are by the nature itself of their avoca- tion bound to reside among the natives of the countries they evangelize. Not to deserve it, I, for one, shall attempt to unfold to the appreciation of the indulgent philologist the beauties of the languages spoken by a family of American aborigines who, if low in the social scale, still possess in thei." native dialects vehicles for thought more expressive, and, in their own way, richer than that of many civilized nations. I mean the lan- guages of the D6n6 Indians, of whom I have given a sociological outline in a late fasciculus of the " Proceedings" of the Canadian Institute. '*^ For the benefit of those who may not have read it, let me state that by D^n^s I mean that large family of Indians more commonly known under the inappropriate§ names of T\nn6, Tinnch, or Athabaskan. It extends west of the Rockies from the 51° latitude north and east of that range of mountains from the Southern Branch of the Saskatchewan to the territory of the Esquimau i. Apart from the Nabajoes of New Mexico, who are ethnologically connected therewith, it is divided into a dozen or more tribes speaking as many dialects. For the sake of briefness, all the abo.i^, lal terms unavoidably used in tht' course of this monograph siiall be, unless otherwise noted, in the dialect of the Carriers, the most important of thf Western tribes. I shall also, to facilitate the intelligence of some of my remarks, occasion- ally point out the relations of these idioms to the principal other linguistic groups, especially the classical tongues. *Am. Antiquarian, IV., Nos. I and 2. •(■Leibnitz, ATonHmenta vatia intditii, ex MiiHto J FcHer, torn. IX., p, 595, Jena 1717. t Procee(liiit;s of tiae Canadian Institute, Octob. 1889, p. 109 and seij. §See "Tlie Western D6n6s " ; Proc. Can. Inst. Oct. 1889, p. I09, note 2. im 172 TRANSACTIONS OF THE CANADIAN INSTITUTE. [Vol. I. CHAPTER I. PHONETICS AND GRAPHIC SIGNS. And first a word concerning the D6u6 Alphabet. To be complete, it should count — apart from reduplicated but phonetically unchanged letters — no less than 60 graphic signs, 13 of which to represent vowels, 39 for simple and 8 for double consonants. The vowels are a, d, ae, e, 6, h, i, i, o, o, u, fl, ii. They are all pro- nounced as in French except is which corresponds to the French e in jc, tc, le ; 6*' which is sounded as the e in the French word " mets" ; c as that of the English "ten "; e and u which have the Italian and u the German sound. The use of the latter is confined to a few insignificant bands of Aborigines who have made the Rocky Mountains thc'r home. The 29 simple consonants are b, d, f, g, J, k, x, k, /, f, m, n, fl, N, p, q, q, r, 4, R, s, s, t, t, V, w,y, z. They are all sounded as in English with the ex- ception of the following : Ji is strongly aspirated ; j is pronounced as in French ; fi as in Spanish* ; N is nasal ; /" is a lingualo-sibilant which is obtained by the emission of a hissing sound on both sides of the tongue curved upward^; previous to its striking the lingual letter ; r is the result of uvular vibrations ; \ and x are respectively k and r pronounced with a very guttural inflection ; R is the common r of the Romance languages, and is proper to a handful of Rocky Mountain Ddnds ; q corresponds to the hard c in the words " ccvitr, am'" such as pronounced by North-western Frenchmen : it can be described as approaching the sound of ty both letters being consonants and sounded simultaneously. The dot in k, t, q, adds to the regular pronunciation of those letters the exploding sound peculiar to most Indian languages, s is pho- netically intermediate between s and s/i. The /"and its co-relative v are found only in the speech of a few Eastern tribes. By double consonants I do not mean the mere succession or groupmg of two or more of the above letters retaining their original value as tl, kw, kfw. I have in mind those consonants which, either are differently pronounced when agglomerated, as ch, or represent sounds which, though phonetically one, cannot be expressed by any of our consonants taken separately. There are eight such double consonants : sh, ch, th, kh, kr, kr, ts and //. The first two are pronounced as in English ; th and kh are equivalent to t-\-h and k-^h but are produced by a single emission of voice- It . proper to a few Northern D^n^s. 1889-90.] THE Dtsi LANGUAGES. 17S Such is the case with kr and ^r, save that the sound of the r in this phonetic compound is perceptible only for the natives and those who may have made a specialty of the study of their language. Ts and t/ have the exploding sound and their value can hardly be taught otherwise than t>iva voce. To the foregoing should be added the hiatu.> (•) which has the effect of abruptly interrupting the pronunciation and slightly elevating the tone of the syllable it follows. There is no accent in D^nd. Besides, the D^n^ dialects possess two genuine diphthongs , au and ai pronounced respectively as the German au and the English long / in such words as " fire, mire," etc. A peculiarity of the language is that it avoids the succession of two vocalic sounds with as much care as the Polynesian idioms do the accumulation of consonants. Thus " Leo " is pronounced " Leyo " and " Noah, Nowah " by our Indians. Among the consonants b and /, C>[>AV -^^one H < > > > A V h >j < > > > A V // R < > > > A V II W < > > > A v Hw < > > > A V T D (1) cDj:i3nu Th a D D D Q O T G D D D Q O PB(i) aDDBfiO (1) KG, Kr E333finuU X, Kh a B B B m DO K, Kr S B B B Cfi tt) N c J j) J) n o M £. 3 j3 ^ n bu (2) WiiA A (E E I O U e 9 © Q l£) Alone G^ ^9 I® ^9 Q S- e^ ^ ^ ® 9 s Y Q Q L Tl f Tf Tf Z Tz Dz S Sh Ch Ts Ts C D G 10 a D c :d G o S 3 a B s 3 a 3 O !3 Q CJ !D !D Q tJ E) © Q O © Q Q a S Q CQ a 3 3 n U ED 3 Q a 3 3 51R Oi 3 3 ffl tb 3 3 fifi ab 3 3 m Oi 53 3 (fi GO Hiatus • — Accessories : o (3) Z 2 s s iTie 1 "* Explanatory Notes. —(1) These letters are not differentiated in D6n4. (2) - is the nasal «. (3) z is the French/. (4) s is phonetically intermediate between s and J. * is prefixed to proper names, and o is suffixed to syllables the vowel of which it is necessary to render long. iMi 176 TRANSACTIONS OP THE CANADIAN INSTITUTE. [Vol. I. CHAPTER II. GENERAL CHARACTERISTICS OF THE DEN^ LANGUAGES. It would be difficult to assign in one word the place which the D6n6 languages occupy among the chief classifications into which modern philo- logists have divided the human speech on the basis of its grammatical structure. Certain writers, and even eminent scholars, too fond of gene- ralizations, have given as characteristics of the American languages traits which really pertain only to some of them. W. von Humboldt pointed to the agglutinative tendency of their verbs as to their chief characteristic* and Wiseman quotes in support of this view Malte Brun's remark to the effect that " this wonderful uniformity in the particular manner of form- ing the conjugation of verbs from one extremity of America to the other favours in a singular manner the supposition of a primitive people which formed the common stock of the American indigenous natives."-f Now, it so happens that the Ddnd verbs are not formed by agglutination, J and are just as inflected as the Latin or Greek verbs. More recently Dr. J. Hammond Trumbull, in a paper, otherwise full of valuable hints concerning the peculiarities of the American languages considered as an independent linguistic group, makes the following re- marks, the first of which he emphasizes by the use of italics : " T/ie Indian noun is not separable as a part of speech from the verb. Every name is not merely descriptive but predicative. . . In short, every Indian name is in fact a verb."§ Yet, with all the respect due to such an autho- rity on American philology as Dr. Trumbull, I must state that there are in Ddn^ many nouns which have no relation whatever to the verb ; nay, the great majority of them is altogether independent therefrom, and they are just as purely nominative as the English " house," " lake," " bear," etc. In a former paper I have referred to the remarkable propensity of the Ddnd nation for the self-appropriation of foreign practices and customs. Its language likewise presents to the investigator features so varied as to suggest a mixed origin for the whole stock, but more especially for the Western tribes. It is at the same time compounding, agglutinative, in- flective, and polysynthetic. || Not, of course that it possesses each and * Apud Wiseman, xii Lectures on the Connection betw. Science and Revel. Lect. IT. p. 82. t Ibid. X At least as this word is now understood and applied by philologists. § Transact. Am. Philol. Assoc. 1869-70. II I apply to these epithets the sense given them in the 2d Edit, of Powell's Introduc. to the Study of Ind. Lang, p. 56. 1889-90.] THB Dl^N^ LANOUAOES. 177 every one of these attributes in each and every one of its component parts But I do affirm that we find them all in the language considered as a whole. In other words the D4n6 has patent affinities with the Aryan, Turanian and Semitic tongues. This I hope to prove to the satisfaction of the reader in the following chapters. Another assertion still more common in philological writings and, in the light of the Ddnd vocabulary, quite as groundless as the preceding ones, is that abstract terms and words expressive of generality and col- lectivity are wanting in the American languages which are represented by some superficial observers as awkward in their syntax and very limited in their vocabulary. I admit that we must not look to the Ddn<5 idioms as to models of phraseological conciseness. That most important pro- perty of a language, the power of expressing without periphrases the subtlest effi)rts of the human mind, is somewhat wanting in the hpeech of our Aborigines, whilst terms expressive of those abstract ideas the accu- rate rendering of which constitutes linguistic perfection are not, I must confess, over numerous. Yet the philologist who wishes to propose for the guidance of students a rule of general import should, it seems, previously examine whether it is not too sweeping in its comprehensiveness. That words expressive of abstractness are not altogether wanting in Ddnd will be proved by the following nouns taken from the Carrier Vocabulary : — ( Mm .Mr in I '"I Tli, cold (as affecting human body) Hokwoez, cold (referring to the temperature) Seal, warmth Tsazcel, heat (of the sun) Hwozoel, heat (of the tempera- ture) Tlo, smile Tso, weeping Ni, cares Sheen, witchcraft Pcef, sleep Shih, French '■' essoufflement" Qan, old age Tai, famine Noeti, effort (physical) Nini, pleasure Tsus, kiss (childish) "jjoef, diurnal revolution of time Nuntsd, bleeding of the nose Hokos, haze Na-k8s, snow-haze (lit. eye-haze) xwoes, cough Unih, jealousy NcBtqoet, fear Tsafkoes, obscurity Yoexaih, daybreak Untd, love Noetai, dance N8yd, play Yathoek, talking Yuyuz, whistling Huzcef, time Neni, human mind Nezoel, human soul (animating the body) Netsin, second self 178 TRANSACTIONS OF THE CANADIAN INSTITUTE. [Vol. I. Nceta, Indian game (Chinook : lahal ) Atlih, Indian game (played with small sticks) Chahwozun, hunger Khu, vomiting Qiltoh, cramp Tata, disease CEItsa'f, bodily pain • tzi, shivering Qitenkoh, work (to be done) Lla, manual labor Thih, frost Mcessai, Latin, " nihilum,' French, " nt'ant " Nezuf, human shade* Hwolna, difficulty Hwold, easiness QEtata, olden time Horvvcenceta, ceremonial ban- quetting ThcjL'f, kick Yaf, locomotion on two feet (hu- man) Kwoef, locomotion or> four feet (and of birds) Krai, loconioLion by running Pih, " " swimming Khe, " in canoe Kwcez, " in sleigh, etc. Besides, the third person singular of several verbs is also used to render our abstract nouns. On the other hand, collectivity may be said to be expressed by such words as ic^tai, ducks (of any species) ; tivtai-ya::, birds ; Mviinai, animals, etc. To these might be added, ara, French, " poils" and derivatives ; ne-na, eyes ; ne-tzo, ears ; ne-lla, hands ; ne-klu\ feet ; ne-kran, arms, which, though singular in form, are plural or collective in meaning. Lastly. iicfI their forms. What I have already said will, I trust, give a fair enough idea of the dififerentiating superiority of the D^nd over the Arj'ari languages. I shall content myself with stating that the single paradigm of the verb " to put " contains in my dictionary (which could be more comple.;e) over 3,000 verbs all of which differ in meaning as well as irk material structure. And this number is repeated in connection with almost all the other objective verbs, which are quite numerous ! And to say that a child four or five years old possesses these innumerable vocables well nigh as perfectly as does his father and knows his extricate language infinitely better than any French academician does his own plain and easy mother tongue ! Who will now vaunt the so much extolled mental superiority of the white race and despise the intellectual capacity of those poor " savages " ? l?ut we must descend from the heights of admiration to the more pro- saic task of rapidly analyzing the different parts which constitute this wonderful speech. CHAPTER III. THE NOUNS ; THEIR VARIETIES AND INFLECTIONS. There are in D6n6 but eight different parts of speech : the noun, the adjective, the pronoun, the adverb, the verb, the postposition, the conjunc- tion and the interjection. Besides the verbs, the immense majority of the adjectives .as well as many nouns and a few adverbs are susceptible of conjugation. Considered in their material structure and etymology, the Ddnd nouns may be divided into four classes. There are the primary roots which are all monosyllabic as in Chinese. Such areja, sky ; i/ai, water ; tsc, stone ; ^(Ts, black bear, etc. They are essentially nominative : they neither define nor describe the object they de..ignate ; they merely differentiate it from another. I consider them a^ the remnants of the primitive D6n6 language, inasmuch as they are to be found with little or no alteration m all the dialects of the family, whatever may be the distance intervening between the Aborigines who speak them. The second category comprises roots of simple import which are genu- ine unsynthetical substantives though polysyllabic, generally dissyllabic, in form. To this category belong words as fa'/ie, man ; ts<^khd, woman ; pivmjrccn, lake, etc. They possess, to a limited extent, the properties of the monosyllabic roots, being likewise merely determinative c.nd often- times varying but little with the change of dialect. 11 ' ■';*% ■i I 182 TRANSACTIONS OP THE CANADIAN INSTITUTE. [Vol. I. ;iil The third class contains composite nouns formed, as a rule, by com- pounding, though sometimes by agglutinating, monosyllabic or dis- syllabic roots. Such are ne-na-pa-ra (literally : man-eyes-edge-hair), eye lashes ; tcepe-tc', wild sheep horns ; inai-re, vegetable oil, instead of mai-M, literally fruit-oii. These nouns being mere compounds of roots belonging to the two former categories have the same degree of relative immutable- ness with regard to the various dialects as the radicals which enter into their composition. The fourth and last class is made up of verbal nouns which, as their name indicates, are nothing else than verbs in the impersonal or personal moods employed to qualify objects of secondary import with the help, sometimes of a radical noun, sometimes of a pronoun, and always of a prepositive particle prefixed to, or incorporated in, the verbal substantive. Of this description are the woxAs pe-ycen-* i^ is assimilated by the vowel of the possessive pronoun or the desinential letter of the words with which it is agglutinated. Thus, " his nest " is said u-to instead of u-a'to, and the prefix a? likewise disappears in such compounds as soh-to, "robin-nest," tcerccs-tan, " aspen leaf," etc. The D4n6 dialects lack declensions of any description. As in the modern analytic tongues of Aryan descent, the office of the cases is, with one single exception, filled by prepositions or rather /^ j/positions ; for in D6n6, as in the Turanian idioms, it is a general rule that the governed word precedes the governing. The exception is the genitiv<; jr posses- sive, which is expressed as in English by first designating the progenitor or possessor and then prefixing the possessive pronoun in the third per- son to the word denoting the offspring or the object possessed. Thus IFi/- yam n-ycen has exactly the same signification as the semi-Saxon " VVil- lelm hys lond," a disintegration of the original " Willelmes lond," which has come down to us under the modern contracted form " William's land." The possessive pronoun affi3cts some nouns to sucn an extent as to im- part to them a genuine inflection, in fact an inflection which, viewed in the light of the Ddnd phonology is even more radical than that of the Greek or Latin cases, since the element thereby inflected is not, as in those languages, the vowel which in Ddnt^ is unimportant, but the conso- nant which constitutes the quintessence of the word For instance Ci is the Carrier monosyllable for " dog," which, when affected by the pos- sessive pronoun becomes s-hvk. Its Chifxohtin equivalent tCin is equally M4|l 184 TRANSACTIONS OF THE CANADIAN INSTITUTE. [Vol. I. changed into sce-llik when in contact with the possessive pronoun of the first person singular, -^e means " grease " the possessive of which is u-rwe at third person singular.* Were there no other cases of conso- nantal inflection in the languages under study, I think these might suffice to confirm the truth of Horatio Hale's assertion that " the opinion which prevails widely among scholars . . . that the languages of the Aryan and Semitic families are the only tongues in which genuine inflections are to be found ... is utterly erroneous."! In common with the Mongolian idioms, the Ddnd has no genuine plu- ral. When absolutely necessary, that number is expressed, not by the adjunction of a pluralizing consonant ag the s of many modern Aryan languages, or by a vocalic inflection as in Latin and Italian, but by the adjunction of such adjectives as Cai, "many," or tsiyauh, "all." This rule applies to all the nouns (except Ci, dog) expressive of non-human beings. Personal names form their plural by suffixing the particles klu\ feet (which is also applied to //), or more frequently ne, root of tcene, man. Only two nouns of the Carrier dialect tskklit, woman, and tekli^, relation, undergo a vocalic mutation when in the plural number, becoming respec- tively ts^klid and tekho. As in Hebp < — but minus any suffix corresponding to the Semitic dualic increment — the names of those objects which are twofold by nature, as the eyes, the hands, etc., are intrinsically dual in meaning though singular in form. To obtain the grammatical singular, one must add to the Ddnd vocable the participle kcez, contraction of oekwz, "half" Therefore ne-lla refers to both (human) hands, and to get the singular we must say ne-lla-kxez, " (human)-hands-half " Grammatical gender is likewise unknown in Ddn(5, and this is again a link of affinity with the Turanian languages. When necessity requires a generic distinction, it is obtained as in Japanese, by the use of the words " man " and " woman " which, for the purpose become adjectives and mean " male " and " female." Here I cannot resist the temptation to point out the remarkable terminological analogy existing between the Japanese word for " male " and that used by the Chippewayans, one of the most important of the Ddnd tribes. This term is yii in both languages. The Chifxohtins replace it by yoss which also reminds the linguist of the Japanese osz which has the same signification. • To fully appreciate the inflectiveness of these and similar words one should bear in mind that there is in Dt5n6 as much difference betwc.u / and CaxnX ^ and ras between any two non- transmutable consonants of the Kiijjlish alphal)et. + The Dtvelopment of Language, Proc. Can. Inst. Oct. 1888, p. 114. 1889-90.] THE Ditii LANOUAOES. 185 The names of some mammals, however, change with the sex, and a few have even a neuter gender, as is the case with the name of the caribou which in Carrier is hwotsih for the neuter, oetceten for the masculine, and cetnnia for the feminine genders. The peculiar differentiating tendency which we have already noticed in the verbs, extends also to the names of a few fur-bearing animals. Among these we find the beaver whose name is tsa. His offspring, when under two years of age is called tsa-tst'l in Chifxohtin. But when the ani- mal has seen two winters, it receives the name of khoq, which after its third winter is exchanged with cetqal'il, which alludes to its being of age to be mated. The D^nd dialects possess diminutive and amplificative forms which are obtained conformably to the Japanese method with this difference that, while the Japanese /r^fixes to the nouns the words ko, " child," and o, " big," the Ddn^ j«/fixes thereto the words yaz and c/io\wh.\ch have exactly the same signification. CHAPTER IV. \^iym THE ADJECTIVES AND THE PRONOUNS. As heretofore stated, the quasi totality of the Ddnd adjectives are genu- ine verbs. Indeed the only real adjectives that do not admit of the pos- sibility of conjugation, are, in the Carrier dialect, cnyii, another, and tsiya, all. Gistel, naked, yuya, ashamed, iannrwois, cylindrical, tsac/u'sk/nvcun, red hot, might pretend to the title, but their native form and use are more that of adverbs than of real adjectives. This being the case, it follows that intrinsic forms of the comparative and of the superlative are no more possible in that dialect than in Hebrew. Their fu-iction is filled, as in the Semitic tongue, by some circumlocution. To obtain the comparative the Carriers use the adverb onnccs, more, before the adjective, and say, for instance: onnces nziiy " more (he is) good " for better. The superlative, when suggestive of no comparison, is rendered as in English by adverbs corresponding to our " very, much." When it implies some comparison, its expression offers in Ddnd greater difficulty. In Carrier, we generally make use of some comparative adverb as the above mentioned onmes, or more appropriately the particle kccs, both of which are coupled with the relative pronoun (e, I'll, tne or ne) in this wise : kms nzu e, " the best" (thing). Sometimes a 31 fT ■ biH.'i:- m 186 TRANSACTIONS OP THE CANADIAN INSTITUTE. [Vol. I. periphrastic circumlocution takes the place of our superlative, and this is particularly the case among the Eastern Ddnds. Thus, according to Petitot, to translate : " my mother is the best of mothers," a Chippcwayan would be inclined to say : " mothers all anyhow my mother alone is good." The only intrinsically formed superlative is that of the locative and ordina- tive adverbs or adverbial-adjectives. Its distinctive element is the prefix asie, applied to the adverb in its natural state as cbtc-no\ the most northern ; cete-yo, the lowest ; ccte-iis, the nearest (in distance) ; oite-tsA the first (in rank or age) ; cete-chii, the first (in the succession of time), etc. This genuine superlative is proper to the Carrier dialect. A peculiar comparative of similitude obtains through the whole lin- guistic group. But as it is essentially incorporated in the verb, I simply mention the fact in this connection. I deem it more relevant to associate with the regular adjectives the demonstrative pronouns which, together with the numerals, have in T>6\\4 the same material features and follow the same rules as the two genuine adjectives o&yu and tsiya. These rules have reference to their place in the sentence, viz : immediately before the noun, and their grammatical accord therewith, to understand which a word of explanation i:. necessary. The broadest division of the substantives grammatically considered and the only one which affects in any way the unconjugatable adjectives and the demonstrative or relative pronouns connected therewith, is that which differentiates the names of human from those of non-human beings. It has for effect to demand the addition to the adjective of an n — from time — for the singular, and of the suffix ne for the plural of adjec- tives relating to human beings. Thus nyu means " another (thing)," (xyun, " another (person)," a.yune, " other (persons)." The same remark applies — barring what has reference to the plural — to the numeral adjectives. These arc quite rich in variety of forms. In Carrier tlia means three (things) ; thane, three (persons) ; that, three times ; t/iatceu, in three places ; thauh, in three ways ; thatltoh, all of the three (things); i/iaha'ltoh,a.\\ of the three (persons) ; thahultoh, all of the three times, (places or things — massive or spacious). Most of the Eastern dia- le'^ts lack all but the three first forms. Neither do they, as a rule, possess the ordinal numbers which are expressed in Carrier by the cardinal ad- jectives preceded by the postpositions (which for the purpose become pre- positions) /lo'/ or hii'oC with, it," in this way: p(vf tkaiuvn, "with him three " (persons) or the third ; InvoCthat, " with it three times," or the third time, etc. 1889-90.] THE d£n6 LANOUAQKS. 187 In common with the ancient Egyptians and most American aborigines the D^nds use their fingers to count upon, and their system is decimal. Holding his left hand with the palm turned toward his face, the Carrier will bend with the index of his right his little finger and all the others ini succession. Then he repeats the same operation on the fingers of his right hand whose palm is this time turned outside, the number six being named on the thumb. That number is fkce-tha, " on both it is three" (things), and eight is in like manner fkce-tcemige, " on both it is four " (things). The expression used for nine recalls to mind the fvo« and he. The first of these pronouns corresponds to our " his, her, its," — the pronouns admit of no gender in Ddnd. The second (Invo) is prefixed to nouns in regimi- nal connection with such words as " house, lake," and those which denote '! 188 TRANSACTIONS OF THK CANADIAN INSTITUTE. [Vol. 1. cxtensiveness or indefiniteness. Vce refers to a third completive person as in this sentence Po/ yos-pa i-kefa-hivoftsi, Paul paid for him. By changing yOB-pa into u-pa, we would give to understand that he paid for Paul. The plural of this pronoun is he. Tee has relation to the person whose action is expressed in the sentence. It is the exact equivalent of the Latin suus, sua, suiim. It will easily be understood that with such a convenient array of pro- nominal terms, amphibology is a mere impossibility. In a late fasciculus of the " Proceedings " of the Canadian Institute, I Avas struck by a quotation from Peschel to the effect that " in the American languages the connected syllables (of composite words) are always curtailed of some sound."* This is another of the many erro- neous statements of philologists, who, because they have obtained some knowledge of a few native tongues, are ' oo prone to apply to those they are unacquainted with the Latin axiom : ab iino disce otnnes. Or shall we exculpate them from the charge of temerity and lay the blame at the door of those who being in position to acquaint the philological world with new languages, did not take the trouble to do so ? Be it as it may, Peschel could hardly find in the Carrier dialect one really composite word to which to apply his own rule. Nay, I think his remark would be more to the point in reference to such a highly inflected language as Latin than with regard to the confessedly polysynthetical D(?n(; idioms, at least if we are to take such a word as cadaver as an abbreviated com- pound formed from caro data mrmibus. Still, the case might have been different in pristine times as the name {Na-ka-3tli) of the village where these lines are written would seem to warrant us to infer. According to a local tradition, a powerful tribe of dwarfs (atna) once attacked and well nigh swept it out of existence. As it is situated close by the outlet of this (Stuart's) Lake, the enemy's arrows, which were diminutive in proportions as the hands that used them, floated down the river in immense numb'^rs. Hence, to give a graphic idea of the importance of the conflict, the ancients used to say to their children : the river was covered with the floating arrows of the enemy, Atna ka poef tistli, which by contraction has become Nakastli. On the other hand, we find in connection with the pronouns remarkable instances of contractions whereby two words, a pronoun and a postposi- tion, primitively independent have combined to form, not a regular com- posite word as those alluded to by Peschel, but a single monosyllable • Proc. Can. Inst. April, 1889, p. 291. 1889-^0.] THE Vis± LANOUAOES. 189 possessing the meaning of the original two words. Here are example* in two different dialects. Carrier Dialect. Soef, with me, for Nyoef, " thee, s-p-oef m-p-oef Poef, yoef, ^^^ p-cep-cef, y-oep-cef tcef, hwof, ' ■' t-oep-oef, hwo-pcE-f ne-p-cel nuh-p-oef opcE-pce-f, hey-oep-oeC Nef Nohwcef Opoef, heyoef, Chifxohtin Dialect. Sa, for me, instead of s-cep-a Na, " thee, etc., n-cemp-a Pa, ya, p-cep-a, y-oep-a njwa, njo-p-a Jwa, Jo-p-a Kupa, ku-pcE-pa Of the adverbs I shall say nothing besides mentioning the fact that many of our adverbs have for equivalents in D<^n6 non-adverbial particles incorporated in, or refixed to, the verb as formative elements. The sequence will explain this peculiarity. As for the conjunctions and the interjections they offer to the phi- lologist nothing worthy of remark beyond this resemblance with the Mon- golian or Tartaric tongues, that conjunctive terms are but seldom used, their number being reduced to a few indispensable words. CHAPTER V. THE SIMPLE OR PRIMARY VERBS. The verb is indeed the word (verbiim) par excellence in D^n6. In fact, out of every twenty words of the language, I doubt whether there are two that are not susceptible of conjugation. Were it not for the verb the D6n6 grammar might be said to be devoid of serious difficulties. As it is, the multiplicity of the verbs, the varieties of their forms, the frequence of their irregularities, the intricacies of their negative elements and the almost unlimited possibilities of their combinations assig'' to the Ddn6 idioms one of the first places among the most difficult languages known to Philology. The student should not anticipate in these pages a com- plete treatise on that most complex of the parts of speech. I am not writing a grammar and must confine myself to general outlines. Yet I hope he will miss in this and the following chapters very little of what is of real value to form a sound estimate of the chief characteristics of the D^n^ verbs. For the sake of perspicuity I shall treat here exclusively of the simple or primary verbs by opposition to the composite or syntheti- cally constructed verbs which shall form the subject of another chapter, 14 i M 1 1 'k 1 ! 190 THANSACT10N8 OF THE CANADIAN INSTITUTE. [Vol. I. i^ A word concerning the modes, tenses, persons and numbers shall serve us as an introduction to the subject. As in the Hebrew grammar, there is, properly speaking, but one mode, the indicative, in the Dt^nd dialects. The imperative can hardly be called a mode, since it is formed entirely of persons taken from the present tense and eventual future. This future furnishes also the equivalents of our subjunctive and optative. The infini- tive does not exist. It is but imperfectly replaced by the impersonal. As for the tenses, they are of two kinds : primary and secondary. The former are four in number, viz.: the present, preterite, proximate future and eventual future. The proximate future refers to the action as being on the point of taking place and corresponds to the English " I am going to." The eventual future is vague and aleatory in meaning and has no strictly exact equivalent in our languages. Each of these four tenses is expressed by a single word as in Latin and Italian. The secondary tenses are simply the primary tenses accompanied by auxiliary verbs or oarticles, as tnle, ta, silt, etc.* The Ddn^ verbs have three numbers as their Greek co-relatives. But while the dialects of the Eastern D^n^s possess, as a rule, three persons for each number, the verbs of the Western Ddnds have generally but one — the first — person for the dual. Par contre, their verbs of locomotion and their verbs of station have no less than sixteen persons for each tense, viz.: three for the definite singular, three for the indefinite singular, three for the plural and seven for the dual. Besides, all those verbs whose radical varies with the number, as is the case with the verbs of cubation, have always — unless they belong to one of the two categories above mentioned — ten persons. Before submitting to the appreciation of the philologist paradigms illustrative of the foregoing, it will not be amiss to give a brief outline of the internal structure of the verb in general. Every D^n6 verb, no matter of what form or tense it is aflfected, is com- posed of at least two distinct integral parts : — ist. a monosyllabic root which is always the desinential syllable of the verb and is generally — not necessarily — invariable ; and, 2nd, a pronominal crement which varies according to the person and the tense. This combination is identical * In the dialects of the Eastern D^n^s the tenses are, according to Rev. E. Petitot, the present, imperfect, preterite and future. In justice to the student of those dialects, I thinlc it necessary for me to state that I am acquainted with missionaries ministering to the Eastern Dt5n6s who spealc in rather disparaging terms of that author's Dictionary and Grammar, and insinuate that, except as regards the Loucheux dialect, which he is recognized to have thoroughly mastered, both works are very faulty. Yet I hardly thinlc that he could have erred in reference to such impor- tant points of Grammar as are the conjugations and tenses. 1889-90.] THE Diai LANQUAOES. 191 with that of the Latin and Greek verbs, with the unimportant difference that the formative elements in the D^n6 and Aryan verbs are disposed in inverse order. Those persons who feel inclined to question the capacity of Indian languages for genuine inflections are respectfully referred to the following paradigms. I have disposed the Latin synonym side by side with the Carrier words, to show that the inflectiveness of the D6n6 verbs is not merely accidental, bnt pervades the whole conjugation. PRIMARY TENSES OF THE VERB CESTEN, I WORK, I ACT. CARRIER. Sing. Plur LATIN. Present. CES-ten in-ten CE-ten cEtscE-ten (jeh-ten cehoi-ten Dual : i-ten ag-o ag-is ag-it -imus ag-itis ag-unt ambo ag-imus Preterite. is-ten Sing^ an-_ten i-ten cEtsi-ten Plur } ih-ten oehi-ten Dual : a-ten eg-i eg-isti eg- it eg-imus eg-istis eg-erunt ambo eg-imus CARRIER. LATIN. Proximate Future, /oethis-tif Sing J cEthan-tif |a;thi-tif ^CEzthi-tif Plur J cethihltif (oehoethi-tif Dual : cetha-tif I am going ♦^o act, (almost acturus sum " es, etc.) Eventual Future. US-ten Sing<|oNn-ten u-ten / cetsu-ten Plur } uh-ten oehu-ten I ( Dual: 6-ten ag-am * ag-es ag-et ag-emus ag-etis ag-ent ambo ag-emus * In a vague and indefinite sense. 192 TRANSACTIONS OF THE CANADIAN INSTITUTE. Imperative. [Vol. I. Sing. In-ten u-ten Plural : oetsu-ten Plur. oeh-ten (Bhu-ten Dual : 6-tcn As illustrative of the formation of the secondary tenses, I shall give the following, with their Latin equivalents : oesten inle Latin ; agebam isten inle egeram usten inle usten sih egissem agam (definite), etc. Of the primary tenses it may be said that the two futures are the most immutable in their pronominal elements, whilst the desinential r )t of the proximate future is of a rather changeable character. As ,or the preterite, instead of being invariably affected by a vocalic mutation of its personal crement, as is the case with the above example, it undergoes quite frequently consonantal variations, either in its radical or in its pro- nominal part, the is, ann, etc., of which are often changed into sees, shi, etc. Now for the sixteen persons of the verbs of locomotion and of the verbs of station : Present Tense. I ncBshya, I walk Definite^Sing. < ninya, thou walkest ' niya, he walks L noetsiya, one walks IndefSing. ] nihya j^^^^^^^^^ f nceheya soesta, I am sitting sinta, thou art, etc. sta Plural ncetscetif, we w"'k ncehtif, you walk noehcetif, they walk tsa?zta soehta hoezta oeztlftsi ta3ftsi hoetiftsi See below ' nres'as, I walk (with one person) soeskrt^ nin'as, thou walkest (do) sinkr6 noe'as, he walks (do) oeskre Dual noetsce'as ) tsoezkr^ ) > See below . > nitas \ ikr^ \ noeh-as, you walk (two together) soehkr^ noehoe-as, they walk " hazkr^ See below i8«i)-yo.] TUK D^N^ LANUUAUea. 1U3 Some of these personal forms have no equivalents in any languafje that I know of, and consequently require a word of explanation. . Nivtsivas and nttas are distinguished practically by a slight difTerencc of meaning only. Nttas is the regular dual of concomitancy which implies that the locomotion is executed by both of us, while metsce as substitutes to the idea of the first person that of the impersonal : one is walking a deux. Niliya and tue/uya, with their co-relatives of the other tenses, also convey the idea of indefipitcncss, but coupled with that of the second and the third person of tht; plural. For instance, a native orator, while giving orders to a group of fellow countrymen and unwilling to designate any one of the , in particular, will say (in the eventual future) : ndhya," you shall walk," meaning : one of you shall walk. Again, referring to an undesignated person in a crowd which he is not directly addressing, he might say in an indefinite manner : me/ioya, " they shall walk," that is, some unknown or purposely unnamed person among them shall walk. The same remarks apply to the verb saesta and derivatives. Should I say to a group of Indians : twftsi, "sit down," all my audience will at once understand mc. But if, instead, I were to say : sielita — the pro- nominal element of which has the h characteristic of the second person plural, though the radical retains its singular form — the natives would immediately understand that I mean only one of them to sit down, and they would be at a loss to know who is to comply with my request. Abstracting the many irregularities of the Dene verbs, all their inflec- tions may be reduced to three conjugations, which are characterized by the consonant of the pronominal crement of the second person singular.* n denotes the first, C the second, and / the third. The verbs above con- jugated — with the exception of the last (scesta), which, in common with some irregular verbs, takes for the singular and dual the characteristics of the first, and for the plural that of the second conjugation — all belong to the first. Here is the present of verbs of the second and of the third. *Petitot in his Grammaire compaive of the Eastern dialects counts four conjugations, three of which lie bases on the pronominal vocalic inflection, the second being in es, the third in as and the fourth in us, while the first he states to consist in the mere juxtaposition of a personal pro- noun to an adjective, a preposition or an adverb. On this side of the Rockies, we have no other equivalents of this uninflected conjugation — if conjugation it can be called — than five or six irre- gular verbs as s-ra-k'volita' {lit. me-on-account-of-difficulty), " I am difficult." I think it prefer- able to treat them as so many unimportant anomalies to making them constitute a separate con- jugation. '' ■ % iiM^lM^iSiii4k^^sM(7han means "woinb," and the literal translation of that verb should read. "I uiomb, thou wotn/wxf , " etc. 1889-90.] THE D^N^ LANGUAGES. 195 suiting therefrom, as its main object is simply to give emphasis and elegance to the sentence. Naturally, the use of the substantive verb is incompatible with that of the D(^n^ adjectives, which are of themselves genuine verbs, combining in their elements our personal pronoun, auxiliary verb, and adjective. I divide them into two classes : primary and secondary adjectives, the first of which, materially considered, correspond to the primary single-radi- caled verbs, and as such fall under the scope of this chapter. They usually express qualities of simple import, as "good, bad, great, small, wide," etc., and while in their primitive form they are merely de- notive. All of them, in addition to the various forms which they are liable to assume as verbs, are also affectible by internal mutations con- noting the nri-ure of the object they qualify without, however, becoming objective vercj. Thus n-cha (third person singular of cescha) means "big," A without pointing to any peculiarity in the subject ; nin-cha is applicable to a sphere or a spheroid exclusively ; din-cha, to an elongated object and to vocal sounds ; hun-c/ia, to vastness or indefiniteness of proportions.* Dce-nin-cha, without being a composite verb in the estim.ation of th,^;^ p^n(^ grammar, yet combines two forms, dtn and nin, and refers to such an article as a ring which is intended for an elongated object, viz., the finger (hence dm, euphonical inflection of dtn)^ whilst it is of itself circular in A form (hence iitn). A similar analysis applies to hwo-din-c/ia. In accor- dance with the law of euphonic sequence of the vowels, /i?iti, qualihcative of indefinite spaces, is changed into /nuo ; dtn has reference to vocal sounds, and the whole compound denotes the peculiar resonance noticeable in the voice of certain persons whose speech seems to emanate from cavernous depths.! CHAPTER VI. THE COMPOSITE VERBS. Polysynthetism has generally been regarded, and not without reason, as the chief characteristic of the American aboriginal verbs. It pervades the languages of the remotest tribes to such an extent as to permit of iheir being classed under one single denomination, despite their many termino- logical and grammatical dissimilarities. Yet that prrticular feature can- not be said to be their exclusive property. Many a verb of purely Indo- 1. 1 m I * Hun and its co-relalive hwo characterize also the Impersonal. t Identical transmuta ions take place in many a non-adjective verb. 196 TRANSACTIONS OP THE CANADIAN INSTITUTE. [Vol. I. European parentage, for instance, can trace to a synthetical process its original formation. For what are such terms as de-ponere, re-ponere, pre- ponere, ex-com-municare, not to speak of cneir synonyms and derivatives in the modern analytic tongues, if not synthetical compounds'? Scholars are also well aware of the role played in the formation of Greek words by such prepositions as Kara, /xtra, wtp, 8ia, etc. Nor sliould we forget the original synthetic verbs dt-standan, be-standati, for-standan, under-standan, etc., of the early Anglo-Saxon, the main root of our own English. But as no otiier linguistic group displays this propensity for word agglomera- tion in such a degree and with such varied results, its claims to be regarded as the special characteristic of the American languages remain unimpaired. The Ddnd dialects v/hich, as we have seen, are distinctly inflective in the pronominal elements of their verbs, are also thoroughly polysynthetic in the formation of composite verbal terms. Indeed, it is to that pecu- liarity, coupled with the multiplicity of their sense-modifying forms, that we must look for the rause of the prodigious number of their verbs. It were tedious, as wel'. .is unprofitable for the purpose in view, to enter into the detc ils of the several processes whereby two or more dependent or independent terms are united to compose a new word. A few representa- tive examples will, I trust, suffice to give us an idea of the whole system. I have in a previous chapter hinted at the existence in the Carrier dialect of a comparative internally connected with the primary verbal adjectives and a few other verbs of simple import. This shall serve as our first illustration of synthetism as applied to the formation of the D^nd com- posite verbs. The primary verbal adjective mssul means " I am small," and its third person singular is n-tsul. With the help of the proper prefixes we obtain : ndcd-tsul, he is as small ; dal-tsul, he is as small as (with a com- plement) ; nga-dcel-tsul, he is as small as that ; pce-ndi-ytel-tsul, he is as small as he; fce-ndcel-tsul, they are both as small, one as the other; sii-tl- tsul, he is small enough ; ta-l-tsul? how small is he ? Cel-tsul! how small he is ! We must not fail to observe that the comparative prefixes have for effect to change the pronominal part of these verbs from that of the first to that of the third conjugation. Moreover, these and all other cognate verbs are susceptible of assuming, conjointly with the above prefixes, all the internal inflections assumable by a verb as verb, not counting those they can be affected by as primary verbal adjectives. For simpler, and p^irhaps more intelligible, examples we will look to the verbs formed with the help of the particle ta, root of tathi, " door." By prefixing it to such verbs as nceshya, " I (human being) walk," we 1889-90.J THE Dfo£ LANGUAGES. I'.T will get, at the third person singular, ta-ntnya, " he (hum. b.) walks in (tne house)." Should we change the verbal stem, while retaining the modifi- cative prefix, we might say : ta-ntnyAt, he drives in (the house), which, by further verbal modifications, may become ta-nceycek, he usually drives in (the h.); ta-ncetgiit, he is susceptible of being driven in (do.); ta-ncetgcek, it can generally be driven in (do.), etc. If we now try of the objective verbs, we may say, while sti'l keeping th-- modificative particle : ta-s'aih I put in (the house a single object); ta-dtstaih, I put in (the house a single object) for my own use ; ta-taih, it (one single obj.) can be put in (the house) ; ta-dttaih, it (do.) can be put in (do.) for one's own use ; ta- stle, I put in (tho house many objects), etc., etc. Let us not, however, allow ourselves to be allured by the multiplicity of forms assumable by this and similar verbs. It would lead us to the enumeration of many hundreds of verbs before the supply is exhausted. For the same reason, we shall avoid ta-sis, I throw in (the house a single obj.), which is the first of a new series of objective verbs. Selecting less prolific forms for our word-building experiences, we come on ta-ssd (lit. " I cry on the threshold "), I beg ; ta-ncss'a, I order in ; ta-hzi'ojyi, I enter while playing ; ta-ncestsas, I enter while throwing sticks on the snow ;* ta-dazni, I introduce my hand in (the house) ; ta'- nceskra, I introduce my finger in, etc., etc. Some verbs of as simple formation connote an even greater number of ideas. For instance, ntsi\ " he is bad, avaricious," while preceded by the aforesaid prefix ta, becomes ta-ntsv, which we cannot well translate in English by less than nine word;^, viz.: " he proves avaricious to those who enter his house." So that if it is true to say that a single idea expressed in one Aryan word frequently requires a periphrase in the American languages, it can be retorted that sometimes short native verbs cannot be translated by less than a whole sentence in our idioms. Nay, the Carrier dialect might even, with some respect, be adduced as a mode! of phraseolo- gical conciseness. As a proof of this, it may suffice to quote the conjugat- able adverb 'a, which means " promptly." To any person requiring of a native courier, for instance, to cover, in one day, a distance of 70 or 80 miles the latter might simply answer : wa gcennih (lit. it-shall-be-promptly let-it-not), the real signification of which is : " one cannot reasonably expect that I should do it in so short a time." Same remark applies to sa\ a long time ; Cat, often ; su, well ; niCtsa', far ; niCtiik, near, etc., which are similarly conjugatable. To return to the composite verbs. A slight analysis of the examples * Indian game much in Tavor during the winter months. - ' I'i' ' ( h H Wt\i\ V 'I ■ n 'if f ii 't^'i' J:l! 198 TRANSACTIONS OF THE CANADIAN INSTITUTE. [Vol. I. given above will result in the detection of two radicals : a primary root, which is always the final syllable of the verbal stem, and contains the key to the main signification of the compound, — and a secondary radical, which precedes the personal element, and alters the sense of the primary root. Thus the j/rt of ta-ncssh-ya indicates that a human being is walking on two feet, and the prefixed particle ia furthermore denotes that such locomotion takes place from the outside to the inside of a habitation. Now if, in lieu of experiencing with a permanent initial particle, we should retain instead an immutable desinential radical, we might succes- sively prefix thereto various initial radicals, whereby new words with new meanings would be formed. Tsaih, for instance, is the desinential root of such objective verbs as have for complement anything of a granulous nature. We shall prefix thereto the particle t/u', whereby— with the incor- poration of the pronominal .$■ — we obtain tlu'-sizaih, (I take out (of the house gran, obj.), which may be further modified into ha-sizaili, I extract (gran, obj.) from ; kcB-stzaih, I put (do.) on ; pe-stzaih^ I put (do,) in (a recipient) ; Inue-stzaih, 1 take (do.) away ; Invosa-stzaili, I bring (do.) to myself; nence-stzaili^ I lay (do.) on the ground; na-stzaUi, I drop (do., do.) ; nintha-stzaih, I put (do.) in the wrong place, that is, I lose, etc. From which we conclude the existence in D^n6 of two chief syn- thetical processes, viz., one whereby the verbal stem is changed while the initial prefix remains invariable, and the other which retains the verbal stem unaltered, while it prefixes to it divers modificative particles. On the other hand, a primary radical may be modified by many a se- condary radical, or formative particles performing the office of radicals. Instead of ta-ncesliya, for instance, we might say ta-tia-ncvsqa* I walk in again ; ta-fia-hiue'-ncesqa, I only begin to walk in again, etc. Nor should a .student, desirous of seeing in all its expansion the won- derful power of composition peculiar to the languages under review, stop at the examination of compounds resulting from the prefixing to a verbal stem of merely one or two sense-modifying particles. To g've him an idea of what it can lead to, I shall introduce him to the verb rc-kce-na-lnve'-s}i(en-da;-tli(e-n(CZ-krok,\\\\\Q}n is a simple word, meaning: " I usually re-commence to walk to and fro on all fours while singing." Shall we analyse it ? Let us try. fe is a prefix expressive of reciprocity, which, when in connection with a verb of locomotion, indicates that the movement is executed between two certain points without giving promi- nence to either ; k(e denotes direction towards those points ; na is the * Q.i is the equivalent of ya, such as inflected by the iterative particle na. 1889-90.] THK D]£nE LANOUAOE8. 190 iterative particle suggesting that the action is repeated ; hwe' refers to the action as being in its incipient stage ; sheen means " song," and when in- corporated in a verb it indicates that singing accompanies the action ex- pressed by the verbal root ; dee is called for by sJieen, said particle always entering into the composition of verbs denoting reference to vocal sounds ; tliee is the secondary radical of the uncomposite verb thtskret, inflected from tht for the sake of euphony with 7ia:s, the pronominal crement of the whole compound, the n of which is demanded by the previous /nve'\ cb characterizes the present tense, and z the first person singular of the third conjugation ; while krok is the main radical altered here by the usitative from the normal form kret, and is expressive of locomotion habitually executed on four feet or on all fours. '<\% To enable the student to penetrate still further into the synthetism of the D^n^ verbs, I give below lists of the principal word-formative par- ticles, together with examples ill istrative of their use. The following are postpositions with regard to their complement — noun or pronoun — and prepositions relatively to the verbal stem.s to which they are prefixed : — AFFIXES. EXPRESSIVE OF EXAMPLES. X^y ;*fwa* desire n-;ira-ncBszoen (lit. thee by desire of I think), I want thee k^, kw6 love u-kw^-ssA (him-through love of-I cry), I pine after him kcBn, kwcen reference ho-kwoen-nahwoelnoek (it-about-he narrates) he relates it kcenne, ) deference s-koenne-inten (me-in deference to-work), obey kwosnne \ me tfa transmission poe-tfa-s-ai (them-in-hands-I-put), I passed to them Ua co-operation ne-Ila-iten (us- hand in hand-he worked), he helped us ^a opposition hwot-qa-soe-niydt (it-opposite to me-he drives), he prohibits it to me •When in connection with the indefinite completive pronoun ho, .ind sometimes the personal pronoun m, the A. 'f'. of this and the following postpositions are inflected into ^oand kiv. 300 TRANSACTIONS OP THE CANADIAN INSTITUTE. [Vol. I. A The following affixes preclude the possibility of union with a comple- ment, and are so intimately connected with the verb as to possess of themselves no separate existence : — AFFIXES EXPRESSIVE OF EXAMPLES. hwosa approach hwosa-shya, I arrive (where the speaker - stands) hwe recess hwe-shya, I depart (from where the speaker stands) nintha wrong nintha-noes-en, I spoil (lit. I treat wrongly) thee, thcenne ill doing (huntsi') thcBnne-zoesten, I have done (evil) ha derivation ha-shya, I come from •cen rejection •oen-tisno, I have rejected (person, compl. j th^ exit Ih^-noeshya, I go out ta ingress ta-noeshya, I go in kCB partial break koe-noeshyoez, I broke (in one place) ya total breakage ya-shyoez, do. (in pieces) tha relation to water tha-stil, I threw (plur. com.) into the water tsoe " fire tsoe-distil do (do) in the fire ne " the soi' ne-noestla, I put (do) on the ground na « I- * na-stil, I threw (do) to the ground Besides the above and other similar affixes, I may refer to those which are instrumental in modifying the form of the verb, and which we shall presently study. CHAPTER VII. VARIETIE? OF VERBS. y While the whole fabric of the Ddn6 conjugations seems to be the almost exact reproduction of the Latin and Greek verbal inflections, the many forms which modify them, and the nature of these modifications, equally remind the student of the ia/, niphal, piel, etc., of the Hebrew gram- mar. I am well aware that some scholars, among them J. W. Powell in his " Introduction to the study of Indian languages,"f are inclined to as- similate these forms to mere grammatical modes. But I am loath to •This particle, moreover, refers to a prompt action, having the soil for its terminus, f Second Edit. p. 53. 1889-90.] THE D±si LANQUAOES. 201 follow thern, because in D6n6 these alterations of the verbal sti ucture affect the whole of both the affirmative and the negative conjugations, while some of them, as the potential, alter quite as much, if not more, the body of the verb as their English equivalent, which in all the dictionaries is regarded, not as a modal variation of the verb, but as a quite different word. Thus from the verb kmmshyh, " I break," we obtain, with the help of the potential, koesqces, " I am breakable." On the other hand, some of these forms consist simply in the change of conjugation, which ' ^n hardly be likened to a mode. The principal forms affecting the verbs, in the Carrier dialect, are the affirmative, negative, usitative, potential, causative, reciprocal, reflective, iterative, initiative, terminative, plural and impersonal. I shall explain, in as few words as possible, the nature of each of them. Of the affirmative nothing needs be said, since it is the normal state of the verb. Yet it may be well to note that a few verbs, the meaning of which is essentially negative, as JmlcM, " there is not ;" huUil, " it disap- peared," lack the affirmative form, and still are destitute of the elements characteristic of the negative. The negative can, in Carrier, affect the material structure of the verb in three different ways, viz.: by the incorporation of a negative par- ticle (/, Ce, fee, ft, fo, according to the tense and the letter following these particles) ; by an inflection of the personal syllable, and oftentimes, though not in every case, by a modification of the mutable part of the desinen- tial radical, i.e., the vowel or final consonant — the initial consonant of a syllable being, as a rule, immutable in D^nd An example will facilitate the intelligence of these remarks. Here are two tenses of the verb nc&s-'a, " I keep in my custody," conjugated under the affirmative and the negative forms. i • i t 3 1 '1m '■\ r t .I'M 1)., r.s Present Tense. Affirmative. D. Negative. Proximate Future. Affirmative. Negative. noe-s-'a n-ce/ffis-es-o?./ n-oethis-'af n-oefthoezis--al n-in-'a n-oifrnzm-oix n-cEthan-'af n-oefthcezan-'al n-oe-"a n-oifois-ou n-oethi-'af n-oefthis--al n-oetsoe--a n-te/tscej-au n-oez-thi-'af n-oeftsoethis-'al n-cEh-"a n-mfoiXfxh-ou n-oethih--af n-oefthoezih--al n-oehoe-'a n-(x,fo^ods-(£X n-oehoethi-af n-oefothis-'al n-i-ta n-osfcezi-tcM n-CBtha-taf n-cefthoeza-tal This form is proper to the Western Den^s, and, when under its three m TRANSACTIONS OP THE CANADIAN INSTITUTE. [Vol. I. <%> inflections, to the Carriers exclusively. The other Western D^n^s elimi- nate the negative infixes f, fm, etc., and if they were to keep the word un- changed by the lexical exigencies of their own dialects they would simply say : n-oezws-mi, n-a!tlmzh-aC. The usitative form consists in a modification of the desinential radical bearing sometimes on the vowel and sometimes on the final consonant, or both. Thus (Bs'aC, ." I eat,'' becomes in the usitative ms'ueC, while ms- cMt, *T take," is transformed into ces-chuek. This is the simplest of all the forms, in that sense that it is the only one that never affects any other part of the verb than the desinence. I should also remark, in this connection, th; i " usitative" is rather inappropriate when applied to the D6n6 verbs, and is here retained for the sake of conforming to the com- mon wording of American philologists. I think " generalizing " would better answer the purpose. The potential varies according to the conjugation of the verb it affects. It merely modifies the radical of the verbs of the first conjugation. So, oear, " he eats," changes for the potential its radical 'nf into taf, and becomes oetaf, "it is edible." But if the verb belongs to the .second con- jugation, the potential transforms it into a verb of the third. The difference between ♦"he affirmative and the causative is equally one of conjugation rather than of form. As a rule, the causative merely changes verbs of the first into verbs of the second conjugation. Intso, for instance, means " thou criest," and by giving its pronominal syllable the /"distinctive of the second conjugation, the result is iltso, " thou causest to cry." This form, if form we must call it, corresponds to the /lyp/it/ of the Hebrew verbs. It is not simply a transformation of an intransitive into a transitive verb, as is shown in the following : natnkas, " thou sharpen- est" (a transitive verb) ; natCkas, "thou causest to sharpen." In a few cases it has for secondary effect to materially alter the final radical of the verb. Thus te-ntnzoen, " thou knowest" (a person), when modified by the causative is te-ntftzoen, " thou acquaintest with." The distinctive elements of the reciprocal form are identical with the particles which, in Northern Carrier, characterize the negative. The dif- ference is only one of place and of results. The reciprocal, unlike the negative, leaves both the pronominal syllable and the desinence unaltered, except when the desinence is immediately preceded by the hiatus, and in a few other cases. Two prefixes may be assigned as the main distinctive element of the reflective form. They are either oedosd, preceding immediately the pro- 1889-90.] THB nkvk LANQUAGES. 203 nominal crement, or the particle uce, prefixed to the modificative syllable na. Moreover, cedced has frequently for effect to transform verbs of the second into verbs of the third conjugation. Ttntht signifies " thou art valuable," and ttCthi, " thou makcst him valuable," i.e., thou treatest him with consideration. Hence, cedcvd-iitht, " thou makcst thyself important," in other words, " thou art proud." In like manner, by prefixing aob to tia/numznmk, " I narrate," we get JOi-nahivoezimk, " I narrate about myself," that is to say, " I make my confession." This form is susceptible of a few irregularities. The particle na characterizes the iterative form. It is prefixed to, or inserted in, the body of the verb, according to the composition of the latter. When the verb is formed of several radicals, modificative parti- cles or completive pronouns, its place is generally immediately after the initial root and before any complement or formative syllables. Either the radical or the conjugation of the verb is liable to be thereby modi- fied, verbs of the second conjugation being ordinarily given the pronomi- nal elements of the third, and such desinential roots as ya, yaC, 'aih, 'aC, 'as, swn, etc., being converted respectively into ia, taC, taih, iaf, tas, tztien, etc. I must remark, in this connection, that such verbs as express an action which, to the Indian mind, is essentially reiterated, take the itera- tive form, even when in their normal state, as na-skas, " I sharpen "; i/u'- na-dws-tli, " I pray." Divested of the characteristic of iteration, tia-skas is reduced to oes-kas, which should be translated, " I sharpen for the first time a new instrument," while t/u'-dces-tli refers either to a passing suppli- cation, as in a moment of distress, or to the first prayer of the child. As for the initiative and the terminative forms, they are obtained by the prefixing or infixing of the particle /nue' for the former and ne' for the latter. These affixes are generally incompatible with the present tense. They have for effect to prefix an n to the personal elements of the verbs whica lack it while under their regular form, and hive' furthermore changes the pronominal inflection this, characteristic of the proximate future, into nthas, the other persons of that tense being also proportion- ately altered. So, from msten^ " I work," we get for the initiative hive'- wxstcn, " I begin to work," and for the terminative ne'umsten, " I have stopped working," while to translate " I will presently commence to work," we vvill have to say : JiivcnthastiC, instead of hive'-thtstiC. Both forms are, as far as I know, proper to the Carrier dialect. By plural I mean a certain form whereby a verb is made to refer to an act exercised repeatedly or in connection with several objects, without, however, having the exact signification or the material '-rd-fe-yt-'al, " he did not give it to him." It will be remarked that thest verbs combine in their formative elements — ist, the personal pronoini inin,yt); 2nd, the direct completive pronoun (e,yt); 3rd, the indi!i''-c completive pronoun (s, ycc), in addition to which we find also incorporated therein : 4th, the negative particle {fe of the last example) Another feature of the D6id syntax is that these completive pronouns are never expressed in connection with any but the third person singular or plural, while some verbs never admit of their incorporation into their elements unless they be of the plural number. Thus a>sc/aU, dws/i i, etc., may mean either " I took, I told," or " I took it, him or her, I told him or her," according to the context. As a compensation, the direct com- pletive pronouns find place in many a Den^ verb, the English equivalent of which is unrelated to any pronominal complement. Thus, for " God made man," we must say, " God man made-him ;" " he took his pipe," must be turned, his (own) pipe he-took-it." These few remarks will, I hope, suffice to give an idea of the Dene syntax. Shall I now say a word about the idiotisms of the language ? The task is rather inviting on account of the abundance of the material 1889-90.] THE D^N^ LANUUAQES. 207 to select from. In fact, I might almost say that the D6n6 dialects are mainly composed ot idiotisms, since, in the words of Dr. Trumbull, " it is nearly impossible to find an Indian name or verb which admits of exact translation by an English name or verb."* Among many phraseological peculiarities, we find, in Carrier, such expressions as s/mn }>aff//^ similar That changes in an American language are not re. . •.fed to the inflexible syllables of a word will be shown by the following <■,. ;ns, most of which are still understood, even by children, but have grown obsolete : — Old tuords. ul'en, lynx tsoNntzif, ice breaker ni-yutse (he barks inland), coyote ,yaih-pa-tsA, red fox sft's-eaoef, brown or cross bear ken-tsi, species of red willow Nexv words. washi I"rect, or both — pronouns, and the various accessory sense-mcdifying affixes. I ITH. — The substantive verb " to be " exists in its independent form. I2TH — The divisions of the verbs into transitive, intransitive, pas.' 'e, etc., are purely theoretical, and have no effect upon the conjugations. 13TH. — There are three conji' -Rations with only one mode and four primary tenses. i4Tn. — The negation is, at least, triple in Carrier, double in Chif/fohtin and single in the Eastern dialects. 15TH. — The Carrier is the most synthetical and inflective of all the Western or Eastern dialects. ;il