UNIVERSITY OF CALIFORNIA PUBLICATIONS IN AMERICAN ARCHAEOLOGY AND ETHNOLOGY Vol. 9, No. 3, pp. 273-435 April 29, 191 1 THE LANGUAGES OF THE COAST OF CALIFORNIA NORTH OF SAN FRANCISCO BT A. L. KROEBER JSancrof; BERKELEY THE UNIVERSITY PRESS UNIVERSITY OF CALIFORNIA PUBLICATIONS DEPARTMENT OF ANTHROPOLOGY The following publications dealing with archaeological and ethnological subjects issued under the direction of the Department of Anthropology are sent in exchange for the' publi- cations of anthropological departments and museums, and for journals devoted to general anthropology or to archaeology and ethnology. They are for sale at the prices stated, which include postage or express charges. Exchanges should be directed to The Exchange Depart- ment, University Library, Berkeley, California, U. S. A. All orders and remittances should be addressed to the University Press. Price Vol 1 1. Life and Culture of the Hupa, by Pliny Earle Goddard. Pp. 1-88; plates 1-30. September, 1903 ..................................................................... S 1 - 25 2. Hupa Texts, by Pliny Earle Goddard. Pp. 89-368. March, 1904 .......... 3.00 Index, pp. 369-378. Vol 2 1 The Exploration of the Potter Creek Cave, by William J. Sinclair. Pp. 1-27; plates 1-14. April, 1904 .................................... -40 2. The Languages of the Coast of California South of San Francisco, by A L Kroeber. Pp. 29-80, with a map. June, 1904 ............................... GO 3. Types of Indian Culture in California, by A. L. Kroeber. Pp. 81-103. June, 1904 ...................................................... ..... ---" <25 4 Basket Designs of the Indians of Northwestern California, by A. L. Kroeber. Pp. 105-164; plates 15-21. January, 1905 ........... ................ 75 5. The Yokuts Language of South Central California, by A. L. Kroeber. Pp. 165-377. January, 1907 ...................................................................... 2>25 Index, pp. 379-393. Vol 3 The Morphology of the Hupa Language, by Pliny Earle Goddard. 344 pp. June, 1905 ................................................................... 3>5 Vol 4 1. The Earliest Historical Relations between Mexico and Japan, from original documents preserved in Spain and Japan, by Zelia Nuttall. Pp. 1-47. April, 1906 .................................................................. - ........... - >6 2 Contribution to the Physical Anthropology of California, based on col- lections in the Department of Anthropology of the University of California, and in the U. S. National Museum, by Ales Hrdlicka. Pp. 49-64, with 5 tables; plates 1-10, and map. June, 1906 ............ 75 3 The Shoshonean Dialects of California, by A. L. Kroeber. Pp. 65-166. February, 1907 ........................................................... - ..... - ......... --- llBO 4 Indian Myths from South Central California, by A. L. Kroeber. Pp. 167-250. May, 1907 ............................. ._.-_-. .70 5 The Washo Language of East Central California and Nevada, by A. L. Kroeber. Pp. 251-318. September, 1907 ....................................... .----. - 6. The Religion of the Indians of California, by A. L. Kroeber. Pp. 319- 356. September, 1907 ..................................... - ........................................... 50 Index, pp. 357-374. Vol 5 1 The Phonology of the Hupa Language; Part I, The Individual Sounds, by Pliny Earle Goddard. Pp. 1-20, plates 1-8. March, 1907 ........... 35 2. Navaho Myths, Prayers and Songs, with Texts and Translations by Washington Matthews, edited by Pliny Earle Goddard. Pp. 21-63. September, 1907 .................................................................. ~"""~ 3 Kato Texts, by Pliny Earle Goddard. Pp. 65-238, plate 9. December, 4 The Material Culture of the Klamath Lake and Modoc Indians of Northeastern California and Southern Oregon, by S. A. Barrett. Pp. 239-292, plates 10-25. June, 1910 .................................... "" OQ'Q" 5 The Chimariko Indians and Language, by Roland B. Dixon. Pp. 29-i 380. August, 1910 .................. Index, pp. 381-384. Vol 6 1. The Ethno-Geography of the Porno and Neighboring Indians, by Sam- uel Alfred Barrett. Pp. 1-332, maps 1-2. February, 1908 -..--------- S- 25 2 The Geography and Dialects of the Miwok Indians, by Samuel Alfred Barrett. Pp. 333-368, map 3. Tv/r^t 3 On the Evidence of the Occupation of Certain Regions by the Miwok Indians, by A. L. Kroeber. Pp. 369-380. Nos. 2 and 3 in one cover. February, 1908 ............................................................................................... Index, pp. 381-400. ... Vol. 7. 1. The Emeryville Shellmound, by Max Uhle. Pp. 1-106, plates 1-12, with 38 text figures. June, 1907 ............................................................... ; ' 2. Recent Investigations bearing upon the Question of the Occurrence , Neocene Man in the Auriferous Gravels of California, by William J.Sinclair. Pp. 107-130, plates 13-14. February, 1908 ..... ......... -35 3. Porno Indian Basketry, by S. A. Barrett. Pp. 133-306, plates 15-30, 231 text figures. December, 1908 ...... """"^i^- 4. Shellmounds of the San Francisco Bay Region, by N. C. 1, ^ 5 The P EuL 09 LTn%S at Shenmound, b^N.c'. Nelson:"' Pp. 357-426, plates 36-50. April, 1910 .................. Index, pp. 427-441. LINGUISTIC FAMILIES or NORTHERN CALIFORNIA DIALECTS SHOWN ON THE MAP Miivok: 1, 2, 3, Southern, Central, Northern Sierra; 4, Plains; o, 6, Southern, Western Coast; 7, Lake. Porno: 8, Eastern; 9, Northern. Yuki: 10, Wappo; 11, Coast; 12, Huchnom. Yurok: 13, Coast. Karok: 14, Upper. Athabascan: 15, Hupa; 16, Tolowa. Shasta: 17, Atsugewi (Hat Creek); 18, Achomawi (Pit River). UNIVERSITY OF CALIFORNIA PUBLICATIONS IN AMERICAN ARCHAEOLOGY AND ETHNOLOGY Vol. 9, No. 3, pp. 273-435 April 29, 191 1 THE LANGUAGES OF THE COAST OF CALIFORNIA NORTH OF SAN FRANCISCO BY A. L. KROEBEE. CONTENTS. PAGE INTRODUCTION 276 MIWOK 278 I. The Northern Sierra Dialect 278 Phonetics 279 Structure 280 Plural 280 Cases 280 Pronouns 282 Pronominal Affixes 283 Verb 287 Verb Stems 288 Demonstratives 290 Numerals 290 Subordination 290 Order of Words 291 Text 291 II. Miwok Dialects 292 Comparative Phonetics 293 Plural 295 Case Endings 297 Pronominal Forms _ 297 Meaning of the Subjective Forms 300 Use of the Subjective Forms 301 Special Peculiarities 302 Costanoan Connections 306 Demonstratives 308 Verb ... .. 309 274 University of California Publications in Am. Arch, and Ethn. [Vol. 9 PAGE Comparative Etymology 311 Phrases 313 Texts 316 Central Sierra Dialect 316 Southern Sierra Dialect 317 Southern Coast Dialect 318 Summary 319 POMO 320 Phonetics 321 Grammatical Methods 323 Nouns ; 324 Pronouns 325 Demonstratives 327 Interrogatives 330 Adjectives 330 Adverbs 331 Numerals 332 Verbs 332 Composition and Derivation 340 Verb Stems 341 Text 343 Northern Porno 346 YUKI 348 Phonetics 348 Grammatical Processes 350 Derivative Suffixes of Nouns 352 Suffixes of Number and Gender 353 Case Suffixes 355 Derivative Verb Suffixes 357 Syntactical Verb Suffixes 362 Suffixes and Structure 365 Noun 366 Pronoun 366 Adjective 368 Numerals 368 Connectives 369 Verb 370 Sentence 372 Vocabulary 372 Text 375 General Character of the Language 381 Wappo 381 WIYOT 384 Phonetics , 384 Reduplication 387 Composition 387 Derivation 388 Pronoun . - 391 1911] Kroeber: Languages North of San Francisco. 275 PAGE Demonstratives 394 Noun 395 Verb 395 Particles 395 Prefixes 396 Suffixes '. 399 Adjectives 401 Numerals 401 Order of Words 401 Alphabetical List of Affixes 402 Prefixes 402 Suffixes 403 Texts 404 Crow 404 Pelican 405 Skunk 405 Vocabulary 406 Nouns 406 Verbs 411 Adjectives 412 Adverbs _ _. 413 Pronouns 413 Numeral Stems 413 YUROK 414 Phonetics 415 Structure 419 Pronoun 419 Noun 421 Verb 421 Numerals 423 Texts 424 Woxpekumeu 424 Pulekuqwerek 425 Buzzard 425 Summary 426 KAROK 427 Phonetics 427 Structure 428 Noun 430 Pronoun 431 Verb 431 Adjective 433 Numerals 433 Text .. ... 434 276 University of California Publications in Am. Arch, and Ethn. [Vol. 9 INTRODUCTION. To Mrs. Phoebe A. Hearst, whose generosity began ten years ago to secure for the University of California a valuable series of anthropological museum collections, and has since supported an Ethnological and Archaeological Survey of California, the credit is due for the following pages. The paper completes the preliminary studies of a grammatical nature made by the author among the languages of California since 1901. Taken in conjunction with his previous articles in this series and those prepared by other investigators working for the University, together with the studies made of several languages of northeastern California by Dr. R. B. Dixon, and the two or three works published before Mrs. Hearst enabled the University to turn its attention to the field, the present paper brings the knowledge of the subject to a point where at least some information is available on the structure of practically every linguistic family in the state. The territory covered by the present treatise is that lying between the Coast range and the sea from San Francisco to the northern boundary of the state. Two languages in this area have previously been monographically treated in the present series of publications : the Athabascan family as represented by Hupa, by Dr. Goddard, 1 and Chimariko, an isolated stock, by Dr. Dixon. 2 These are accordingly not included here. Those sketched are, in order from south to north, Miwok, Porno, Yuki, Wiyot, Yurok, Karok. Further studies of Yurok are in progress; and the author hopes to continue a more detailed examination of Yuki and Karok. No attempt at an exhaustive treatment of these languages has therefore been made : the descriptions of them are preliminary. The accounts given of the other three languages make use of all the information that has been gathered, and are therefore somewhat fuller. It must be clearly understood that while languages may be spoken of, it is really linguistic families that are dealt with. 1 III, 1905. 2 V, 293-380, 1910. 1911] Eroeber: Languages North of San Francisco. 277 Where several languages are unquestionably kindred, one of the group has been selected for grammatical investigation. Thus the Porno family comprises seven distinct dialects or languages, as evinced by a comparison of vocabularies. As a comparative grammatical study of seven languages is a larger task than was feasible for a preliminary investigation, the Eastern dialect has been chosen to represent the six others, which it undoubtedly does approximately typify in all essentials. In the same way the descriptions of Yuki and Miwok apply each primarily to one language selected from the several constituting the family. So far everything shows that kindred languages in California are very similar in structure, however much they differ in a large proportion of their vocabulary. 278 University of California Publications in Am. Arch, and EtTin. [Vol. 9 MIWOK. As information has been obtained on several Miwok dialects, it seems most practical first to devote a special and more detailed consideration to the best-known of these idioms without refer- ence to its congeners, and then, in a following section, to review and compare all the dialects of the family, both with reference to one another and to their connection with the Costanoan languages, as these now appear to be perhaps genetically related to Miwok. 3 Indians of Miwok family held the territory from the Golden Gate northward to beyond the limits of Marin county, as far as Bodega bay, so that this language was the first to be encountered as the coast is followed northward from San Francisco. A second area of Miwok speech was situated in the Coast range, south of Clear lake. The third, largest, and most important division of the family, and the only one to which the name Miwok in its narrower and original sense properly applies, is neither in the coast region nor north of San Francisco, but occupies the western slope of the Sierra Nevada range between Cosumnes and Fresno rivers. 4 I. THE NORTHERN SIERRA DIALECT. The form of Miwok specially discussed here is the so-called Amador dialect, the northerly of three constituting the Sierra division of the family. The material was obtained in 1908 from the late Eph Cummings of West Point, Calaveras County, and is supplemented by notes, more elaborate on some points, that were secured at Jackson and San Andreas in 1900 by Dr. A. M. Tozzer, and are available through the courtesy of Mr. C. P. Bowditch and Professor R. B. Dixon. Certain inconsistencies in orthography are due to the fact that the material is from two observers. s Present series, IX, 237-271, 1910. * S. A. Barrett, present series, VI, 1-332 and maps, 333-368 and map, 1908. 1911] Kroeber: Languages North of San Francisco. 279 PHONETICS. The phonetics are of the simple Californian type. The vowels are a; i, e, o, u, all open ; and o, u, the quality of which is indis- tinct, and which might be written o, u. Surd and sonant stops are difficult to distinguish, at least as much so as in Costanoan and more than in Yokuts or Porno. The surd stops are rather forcibly articulated. There are interdental or dental and alveolar t and d; the former occur in debe, cut, tolokocu, three, damman, south; the latter apparently in tuina, jump, and doklo, strike. While the existence of two series of dentals is common in Cali- fornian languages, the difference between them is usually greater than in this language, the posterior series being formed more nearly palatally than in Miwok. Dr. Barrett distinguishes three t's t, t, T but in many cases appears to have written t for T where interdental t occurs. The two positions of the dentals have not been distinguished orthographically in these notes. K is formed rather far back. X, spirant of k, is usually heard more nearly like h and is probably the same sound ; n, the nasal of k, occurs, but not initially. There is 1 but no r. Glottal stops occur but are not prominent. Consonants are frequently lengthened or doubled ; n, m, c, k, have been observed thus : kanni, I. kg n t- d- c j t d s z n P b w, y, h, 1, ', tc, dj No words begin or end in two consonants. Juxtapositions of consonants are rare also in words, and in most cases are visibly due to composition, derivation, or affixion. The stress accent of words does not tend so strongly as in some languages to rest on the penult, but may be found on any syllable between the first and the last. The numerals to sixteen illustrate: lu"ti, odi'ko, tolo'kocu, oyi"sa, ma'coka, te'mmoko, ke'nnekakw, ka'winta, wo"e, na'a'tca, lu"sagena, o'diksagena, tolo"teakw, kolo'kaku, yu"ali, o'yodo. It seems that long vowels and vowels followed by a glottal stop or doubled consonant generally bear the accent, which is well marked. 280 University of California Publications in Am. Arch, and Ethn. [Vol. 9 STRUCTURE. The majority of noun and verb stems are disyllabic. Neither etymological duplication nor grammatical reduplication is con- spicuous. There seems to be little vocalic mutation. Position plays an unimportant part syntactically. There is apparently no prefix in the language, even preposed pronouns such as those of Yuki and Yokuts being lacking. Grammatical form is there- fore expressed almost wholly by suffixes. PLURAL. The plural of animate nouns is expressed by -k, sometimes -ko. Thus nafia-k, men, occa-k, women, ole'tcu-k, coyotes, tcummeto-k or tcummeto-ko, southerners. Numerals referring to animate nouns also take the ending : oyica-k tune-ko-nti, four daughter-s- my. It is also further found on miko, ye, from singular mi, and in the subjective and possessive suffixes of the same person, -tok and -moko. It appears also on demonstrative and interrogative stems, as ne-kko-n, their, of these, and mana-ko-n, somebody's. The term gotcayakko, town, from gotca, house, evidently con- tains the suffix. Nouns ending in the diminutive -ti show some irregularity: nafia-ti-koko, boys; uya-guta-k, old men, and ona-guta-k, old women, from uya-ti and ona-ti. Inanimate nouns lack indication of plurality. Efforts made to determine a modification in verbs according to plurality of either subject or object were fruitless. CASES. There are two purely syntactical cases, an objective -i and a possessive -n, which have an extensive use. The objective is not only regularly employed on the object noun, animate or in- animate, but on numerals and verbs used objectively, as masi yinanakama tolokocu-i, we killed three, and gudjikcuangum muli-a-i, I do not wish to sing. It is also used on nouns con- nected with a prepositional adverb, as in lilamadoyi gotca-i, on top of the house. The ending may perhaps also be sought in umedj-i, yesterday, kauleba-i, tomorrow, and willa-i, constantly. The possessive case-suffix is used not only in the noun, but 1911] Kroeber: Languages North of San Francisco. 281 also in the independent pronoun and demonstrative : kannw-n, my, mi-nw-ii, your, ne-cw-n, his, this one's, itci-n, our, mana-ko-ii, somebody 's. When two nouns are possessively related, the posses- sive pronoun as well as the possessive case may be, or is usually, employed: palaia-n hake-cu, close to the ocean, ocean's its edge; kannu-ii sake-nti-n occa-cu, my my-friend's his-wife, the latter construction recalling Yokuts yiwin an limk-in, wife his prairie- falcon's. It will be seen that the possessive case-suffix is added both to the plural and the pronominal suffixes. The same is true of the objective: uye'ayi-ko-i, white men; sake-t, my friend, objective sake-nt-i. Added to a verb with attached subjective suffix, the possessive case renders it subordinate : utcux-ce-te-n, when I had stayed, stay-did-I-when ; tolyok-cuke-te-n wnu-ce-nti, after listening I returned. Yokuts uses the locative case -u in exactly parallel constructions. The possessive is also frequently used on the noun or pronoun subject of a verb, apparently when this is in some way dependent : Kelsi-ri unu-kuke-te-co, Kelsey his-bringing-me ; tolyok-cuke-nti hayapo-ko-n liwakcoko, I-heard captains' speaking; kannw-n tuyan-at, I jumped; itci-n yulu-tcu umedji, we bit yesterday; sake-nti-n huwata-co, my friend ran, my friend's running. Verbs with the potential suffix -uni also may have their subjects in the possessive: mina-n tuyan-uni-na, can you jump ; kannu-n tuyan-uni-t, I can jump. Like almost all languages of California, Miwok possesses locative and instrumental suffixes. Those determined by the author are a general locative -to, an ablative -mo, a terminalis -m, and an instrumental -su. The only other forms obtained are separate postposed words, such as unuk, from, wbuk, for, on account of, or preposed prepositional adverbs governing the noun in the objective case, like the above mentioned lilamadoyi, on top of. sawalo-to, on Saturday gotca-mrao, from the house lelotu-to, on the railroad mokelumne-m, to Mokelumne isako-to, there sanhose-im, to San Jose ne-to, here no'-m, there imaga-to, indoors sawa-m, on the rock min-to, where gudji-su, with a knife gotcayakko-to, to town cawa-su, with a stone mokelumne-mo, from Mokelumne leka-su, with a stick, imaka-ma, from there 282 University of California Publications in Am. Arch, and Ethn. [Vol. 9 Dr. Tozzer found the following suffixes : -to, superessive -ko'-ta, ko-ta, comitative, with -mo, ablative or at -m, -am, inessive -pa, terminalis -pa-zo, instrumental -ta, for It seems that the meaning of the suffixes is not precise, the locative being used to denote the ablative and terminative rela- tions and vice versa. Dr. Tozzer also gives a number of pronominal forms. These consist of the full form of the pronoun, followed by the case- ending, to which in turn a suffix form of the pronoun is added. kani-to-te, on me (I-on-my) ikazo-mo-ko, from him mi-ta-ni, for you mi-ko-ni, at you kani-am-te, in me itci-ko-me, with us It is not certain whether each of these expressions forms one or two words. Possibly kani to-te should be read for kani-to-te. The suffix ko-ta or ko-ta loses its second syllable -ta in these pronominal forms. PRONOUNS. The pronominal forms of Miwok have been most fully deter- mined by Dr. Tozzer, without whose full paradigms their nature would have remained obscure at many points. As in other American languages, the independent personal pronouns and the affixed pronominal elements, or as we might say, the pronouns and the inflections for person, are quite dis- tinct in Miwok. As in most languages that possess both classes of elements, the independent pronouns are used chiefly for emphasis, when they are actually tautological, or in elliptical and unsyntactical constructions. In some languages the longer independent words are clearly expansions of the affix or "inflectional" forms, which must be regarded as primary. In other languages the affixed elements are probably reductions of the originally independent and separate pronouns. In Miwok the two classes of forms are evi- dently of unrelated origin. They show, at least in the singular, no similarity whatever. The independent pronouns, which are throughout treated and declined like nouns, are : Subjective Objective S 1 kanni S 2 mi ' mini S 3 [ikazo ikazo-i P 1 itci, maci P 2 miko miko-i P 3 [ikako ikako-i 1911] Kroeber: Languages North of San Francisco. 283 Possessive kannw-n miim-ii ikazit-n] itci-n, maci-fi miko-n ikakit-n] The forms for the third person are demonstrative. While Dr. Tozzer gives maci, us, as the objective of itci, we, the difference between the two forms is apparently one of duality and plurality respectively, or possibly of inclusion and exclusion of the second person. The first person subjective together with the object of the second, is expressed by the enclitics mu-cu, I thee, and mu-tok-cu. I you. yina mucu', I kill you huwate-ne mucu', I make you run kutcikcu mutokcu, I like you PRONOMINAL AFFIXES. The ' ' inflectional ' ' forms, contrary to the prevailing tendency of American languages, are suffixed. Their most remarkable feature is that the subjective suffixes of the verb show three distinct forms, each used only with cer- tain modes and tenses. The three tense-forms of one person are often entirely dissimilar. One set of forms is employed only for the present and perfect tenses. Another is used with two preterite tenses. Still another, the most common, is used after all other temporal and modal suffixes. This, called hereafter the first form, is perhaps primary, as the objective suffixes of the verb, and in part the possessive suffixes added to nouns, are almost identical. Several of the possessive suffixes, however, resemble the preterite subjective suffixes more closely. Possessive Objective Subjective 1 Subjective 2 Subjective S Future, Passive, etc. Preterite Present and Perfect SI -t, -nti -t, -te -t, -te -nti -ma, -m 82 -no -n, -ni -n, -ni -no -8 S3 -co -k, -ko, -wo -k, -ko, -co -WO PI -ted, -ma (si) -m, -me -me, -m -tco, -ma(f) -ti P2 -moko, -miko -tok, -tokni -tok, -tokni -muko -toksu P3 -ko r -kon -ko, -k -ko -ko -pu 284 University of California Publications in Am. Arch, and Ethn. [Vol. 9 Contrasting with the independent pronoun, the suffixes almost throughout possess forms for the third person. "When both subject and object are expressed in the verb, the objective suffix precedes. Examples of the possessive suffixes : gotea-t, my house occa-t, my wife hana-t, my hair oyaji-nw, your name anei-no, your son leka-sit, his stick occa-cii, his wife hana-tcit, our hair gotca-moko, your house hana-kon, their hair, somebody's hair The possessive suffixes follow the plural ending; case-endings usually but not always follow the possessive suffixes. sake-nt-i, my friend (objective) sake-nti-n, my friend's occa-i-nw, your wife (objective) tune-ko-t, my daughters The first or primary form of the subjective suffixes is em- ployed after the future suffix -i, the passive -si, the usitative -imi, the potential -uni, and at least certain combinations of past suffixes, such as -ke-ce or -kco, and -ce-k. The second form is either attached directly to the stem to express a recent past tense ; or it is added to the preterite suffixes -ce or -ke, which appear to indicate a more remote past. The third form, when immediate to the stem, indicates present time. It also follows the past suffix -naka, which Dr. Tozzer interprets as a perfect. First form of subjective suffixes: huwat-imi-t, I run constantly wokec-i-t, I shall go dobomi-n, you are crazy yulu-in a, will you bite? muli-i-tok a, will you sing? muli-i-me, we shall sing yulu-yi-m, we shall bite hakaine-cakdco-t, I was hungry itei top-i-me, we shall hit itci a hakaine-cak-me, were we hungry? miko a hakaine-i-tokni, will ye be hungry? 1911] Kroeber: Languages North of San Francisco. 285 haline-i-ko, they will be sick haline-imi-su-n, you used to be sick katce-ca-zo liwa-ni-ko, he said he would talk haline-i-tok ane, ye might be sick tokla-bosa-i-te, I shall hit myself itci osati ete-ksoi-m, we had a girl itci osati ete-ma-yi-m, we shall have a girl tcuku yak-te, or yako-zo-te, I had a dog tiwa-i-ko sumnenu-i, they will bring a hat (sombrero) wentete-i-me, we shall sell (Sp. vender) owo-i-ko, they will eat kani mata-si-te, I am shot mini mata-si-yi-ni, you will be shot kani mata-si-zo-te, I was shot tokala-si-zo-te, I was hit tokala-si-te, I am hit itci yiloa-si-me, we are bitten miko yiloa-si-zo-tokni, ye were bitten kalto-i-te, I shall dance hakaine-imi-so-te, I used to be hungry hakaine-pa 6wo-i-te, if I am hungry, I will eat masi hakaine-pa-k, owo-i-me, if we are hungry we will eat hakaine-nit owo-ni-no, if you were hungry, you would eat ndka-ni-zo, wokoe-i-te, if it rains ("its raining"), I shall go Second form of subjective suffixes: huwata-nti, I ran hedea-no, did you see? yitna-nu, did you kill? yulu-tcu, we bit yulu-ce-tco, we bit goyoka-te-no, you looked at me hwla-te-nu, you cut me yulu-te-co, he bit me ika-zo tope-zo, he hit miko tope-muko, ye hit tokla-te-zo, he hit me mini tokla-ni-zo, he hit you ika-zo-i tokla-ko-zo, he hit him masi nana etea-me-zo, the man saw us toloye-nti liwa-zo, I heard her talking moa-se-nti wona-zo, I met him walking moa-tokni-zo wona-muko, he met you walking moa-te-no wona-nti, you met me walking haline-so-tco, we were sick tiwa-nti or tiwa-se-nti, I bought wentete-no or wentete-ka-no, you sold mi owo-no, you ate owo-tco, we ate minii-n a haline-ke-no, were you sick? 286 University of California Publications in Am. Arch, and Ethn. [Vol. 9 haline-ke-tco a itci-fi, were we sick? kalto-zo, he danced eteya-ko-nti, I saw him. muli-ni-no tuyana-nti, when you sang ("your singing"), I jumped moa-in-te mega wone-no, I will meet you walking kani ane topu-pa-nti, I think I was hit Third form of subjective suffixes : goyoku-m, I look hiila-mu, I cut hedeyi-m, I see nrukcu-ma, I go huwate-ma, I run yina-naka-ma, I killed huwate-ti, let us run uhu-ti, let us drink min-to yina-naka-tok, where did ye kill? muli-saino-ma, I wish to sing muli-saino-ano-ma, I do not wish to sing hoyako-wo, he is laughing tokla-bosa-s, you hit yourself mi a hakaine-s, are you hungry? hakaine-ti, we are hungry ika-ko hakaine-pu, they are hungry mi tope-s tcuku-i, you are hitting the dog kani a hakaine-naka-ma, have I been hungry? katco-wo haline-wo, he says he is sick haline-toksu, ye are sick ika-ko woko-saino-pu, they wish to go kani hoyak-saino-ma, I want to laugh muli-saino-wo, he wishes to sing tiwa-vro somnenu-i, he buys a hat tiwa-naka-pu, they bought wentete-ma pulaka-i, I am selling the basket 6wo-s, you are eating mata-pu, they are killing mata-naka-wo, he killed kalto-pu, they are dancing eteya-te-wo, he sees me eteya-ni-ma, I see you kani ane topu-pa-ma, I think I am hit Examples of objective forms, additional to those already given : goyoka-ni-t, I saw you hiila-ni-t, I cut you kutcikce-waco-ni-t, I did not like you goyoka-te-no, you look at me hwla-c-te-ko, he stabbed me wiku-te-cu, his taking me dobe-tii, tcupta-nto, throw it at me! goyoke-to, look at me! 1911] Kroeber: Languages North of San Francisco. 287 VERB. The "inflection" of the verb for person consists of the addition of the pronominal affixes just discussed. The following derivational, modal, and temporal elements, all suffixes, have been found: -ne, causative -ce, -kce, -caino, desiderative -imi, continuative -uni, -ani, potential -anu, -cewa, negative -bo, -bo-sa, reflexive -ce, -cu, -ke, -kco, -cak, past -naka, past, perhaps perfect -i, future -si, -pa, passive Dr. Tozzer sometimes writes the potential or dubitative ani as a separate particle before or after the verb. The subject of the verb in the potential usually has the possessive case-suffix. The order of suffixes is: derivative, modal, temporal. The desiderative and negative precede those that express mode and tense. The potential, the passive, and the usitative come before the preterite and future suffixes. Last of all in the verb are the objective and then the subjective designations of person. huwate-ne-i-t, I will make him run goyok-cu-m, I want to see mi' a tuina-kco, do you wish to jump? tuina-kce-anu-m, I do not wish to jump uhu-kca-nu mi', you do not wish to drink uhuk-imi, he drinks constantly tuyan-imi-t, I jump constantly yina-an-uni-t kannw-n, I cannot kill him tuyan-cewa-t, I do not jump kutei-kce-anu-m, I do not like him (good-wish-not-I) yina-ciwa-cd-n, you did not kill it kaune-naka-ma, I shouted liwa-ni-nd a, can you talk? howato-ni-kd a, can they run? wdke-bo-sa-nti, I burned myself heka-bo, to wash one's self sakizd-bo, to comb one's self 288 University of California Publications in Am. Arch, and Ethn. [Vol. 9 The suffix -ne, to be distinguished from causative -ne, has verbal force on adjective or intransitive stems. kutci-ne-ma, I am good hakai-ne-ma, I am hungry hali-ne-ma, I am sick The interrogative is indicated by the particle a. This is regularly the second word in the sentence; but far from being enclitic, usually carries the heaviest accent in the phrase. Instances occur among examples previously given. In certain verbs the stem in the future appears to end in a consonant, while in the past and present a final vowel appears. In some instances this is brought about by a shift of the second stem-vowel to a place after the final consonant. Present and Past Future wukcu- wokec-i kaune- kauin-i huwate- huwat-i tuyane- tuyan-i yila- yil-i tope- top-i eteya etey-i In Northern Sierra Miwok the verb is certainly as truly conjugated or inflected as in any Indo-European language. The existence of three forms of personal endings whose employment depends on ideas of tense, and the differentiation of all of these from the independent pronouns, make it impossible to describe the language as "agglutinating." VERB STEMS. Verb stems are generally disyllabic, unless those so far determined should ultimately prove to contain affixes of motion, shape, direction, or instrument, of which possibility there is no present indication whatever. ame, give birth doklo, tokla, strike with fist, ameto, beg knock down dekma. tekme, kick duka, dttka, pierce depa, cut ete, etea, eteyo, hete, hideye, dobe, throw see, look at dobome, crazy etepo, lie on stomach 1911] Krocber : Languages North of San Francisco. 289 hakai, hungry ball, sick haye, touch heka, wash henne, ask for hili, pinch hinuwo, gamble grass game hdge, bet hotse, hiccough hdya, laugh hukaye, smell hupa, roll huwa-epo, hasten huwa-te, run huya, start, leave, arrive huyaku, strike hwla, stab kalte, dance kata, shut kauin, kaune, shout kelpe, swallow kole-nak, cough kona, bark kope, open koyok, goyok, see, look kopa, pull kusu, sit with stretched leg kute, kuta, gute, push, knock with hand kuyage, whistle late, suck latci, chop (Spanish la hacha, the ax) lepa, bury liwa, talk lokta, sneeze lometa, fall lutsu, ascend mata, shoot, kill (Spanish matarf) moa, meet mole, spill motea, hide muli, moli, sing mulagu, wash face nawu, dress nepye, swallow nete, count nipito, sit with folded leg noted, notcu, cry, whine nuzu, mizu(t), undress okye, make basket ole, dig d'wd, eat, bite pakal, pay (Spanish pagar) petafie, throw away pilapa, pinch puu, squat sakizd, comb sdtcaya, shine sd'tceld, lie on side sutwa, break a string takya, hit with stick, whip taswa, break temanu, cross tiwa, buy tizd'ye, scratch toloye, hear td'tci, believe, wish tuka, spit tupi, press tuyan, tuina, jump tcamza, die tcime, climb tcunuza, slide tcupta, throw endwise uhu, drink uku, enter uktcu, dream unu, come, return utcu, stay weli, catch welza, hunt for wentete, sell (Spanish vender) wilano, steal wokec, wukuc, wokcu, go woke, burn wokle, swallow wona, walk yana, sleep, lie on back yild, yila, yulu, bite yina, yunu, kill yiya, shake ydtki, hang yuhu, swing yutme, claw 290 University of California Publications in Am. Arch, and Ethn. [Vol. 9 DEMONSTRATIVES. The stems corresponding to this and that are ne and no. Ne and no have been found, both as substantives and adjectives, only with the ending -i; as adverb, here, ne occurs with the ending -to, -kkato. From no is derived no'-m, there. The pos- sessive case of both stems is formed by the ending -cw-fi com- pare mi-nw-n, from mi, you. The possessive plural is ne-ko-n and no-kko-n. Another demonstrative stem denotive either of greater dis- tance than no, or of reference rather than position, appears to be i-. iga-c-i naiia-i, that man imaka-ma, there, from there isako-to, there To these forms are related Dr. Tozzer's ika-zo and ika-ko, usually given in translation for "he" and "they." "He" also appears several times as igas or iga. The interrogatives are mana, who, ti'nw, what, mini, where, mitan, when. Min-to is used for mini when the sentence contains a verb. Somebody's is mana-ko-n, somewhere mini-mta. How large, is miniwitci ; how, is mitciksu. NUMERALS. The numerals, when accompanying animate nouns, take the plural suffix: oyica-k. They also receive case suffixes: tolokocu-i. They also enter into composition: toloko-ma-i, oyica-ma-i, three times, four days; toloko'-me, Ave three, three persons. "Each" is -ameni : otiko-ameni, two each. SUBORDINATION. Dependent clauses have been mentioned as being indicated by the possessive case-suffix. Either this is added to the subject, the verb receiving a possessive instead of a subjective pro- nominal ending, so that the construction is really nominal- possessive; or, to express a temporal clause, the case-suffix is added to the verb, pronominal ending and all. mina-n yulu-no, (I saw) your your-biting sake-nti-n huwata-co, my friend ran, literally, my friend's running tolyok-cu-ke-te-n, after I had listened, literally, of my listening 1911] Eroeber: Languages North of San Francisco. 291 ORDER OF WORDS. The order of words in the sentence is not rigid. The verb sometimes is first, sometimes last. Local modifier and object both precede and follow the verb. Connective words have not been observed. Sawalo-to On Saturday mokelumne-m To Mokelumne Hill wolucprinu-m to Valley Spring. huya-ke-nti I arrived wiku-ke-te-cti his taking me polaia-fi hake TEXT. hoya-na-ke-nti stedji-to tcume-nti I started. On the stage I rode. huya-ke-nti mokelumne-mo wukuc-it I arrived. From Mokelumne Hill I went wolucprinu-mo tcume-nti lelotu-to From Valley Spring I rode on the railroad. sanhose-im wolucprinu-mo sanhose-mo at San Jose from Valley Spring. From San Jose Kelsi-n tcummatc wukucu imaka-ma Mr. Kelsey's south went. From there -cii wiikuc-e-nti imaka-ma huya-yi-ke-nti its close I went. From there I went maunthomoni-mo tolokocu oyisa-i tanalo-i uke-nti to Mt. Hermon three four tunnels I went through huya-ke-t isako-to maunthomon-mo toloko-mai utcu-se-nti I arrived there at Mt. Hermon. Three days I stayed. tolyok-cu-ke-nti hayapo-ko-n liwa-kco-ko tolyok-cu-ke-te-n I listened chiefs' their speaking. After listening wnu-ce-nti sanhose-m hwya-ke-nti Kelsi-n unu-ku-ke-te-co I returned. At San Jose I arrived Mr. Kelsey's his bringing me sanfransisko-mo imaka-ma toloko-mai oyica-mai utcux-se-nti to San Francisco. There three days four days I stayed. heteyi-yi-ke-nti coke-i lapicayu-i wcwmati heteye-nti I saw anything, wafia-ko-i many uye 'ayi-ko-i co 1 Americans, heteye-nti I saw. wukuc-e-nti hii I went < haiapo-to at chief ututi kotca-i large house. fish, bear I saw, cowu-ko-i heteye-nti sakacti-ko-i shows I saw, circuses m-m toloko-mai utcuk-ce-te-n east three days after staying 292 University of California Publications in Am. Arch, and Ethn. [Vol. 9 II. MIWOK DIALECTS. Dr. Barrett, in determining the three areas formerly occupied by the Miwok, 5 has established also the principal dialects. In the coast region immediately north of the Golden Gate, two closely related dialects are spoken, called by Dr. Barrett the Marin or Southern Coast and the Bodega or Western Coast dialects. The speech of the Lake county area is not very divergent from these, and constitutes the Lake or Northern Coast dialect. All the Coast dialects form a unit as compared with the Interior division. This group is divisible first of all into a Plains or Northwestern dialect on the one hand, and a group of foothill or Sierra dialects on the other. The Sierra group consists of three dialects, a Northern or Northeastern, a Central, and a Southern ; or respectively Amador, Tuolumne, and Mariposa. Of these the Southern is the most divergent though less so from the two others than the Plains language. The Northern and Central dialects are similar, though evidently not to the same degree as the Marin and Bodega on the Coast. Minor subdialectic differences, as within the Mariposa dialect, are slight. So much is apparent from a comparison of the vocabularies collected for the purpose. Grouping together the Marin and Bodega dialects on account of their close affiliation, the main linguistic divisions of the Miwok, with their designations by Dr. Barrett and Dr. C. Hart Merriam, 6 are the following: Lake or Northern Coast ; Dr. Merriam 's Tuleamme. Marin-Bodega or Southern and Western Coast ; Dr. Merriam 's Hookooeko, including the Lekahtewutko and Olamentko. Plains or Northwestern Sierra; Dr. Merriam 's Mewko, comprising the Mokozumne, Mokalumne, Ochehak, and others. Amador or Northeastern or Northern Sierra; Dr. Merriam 's Northern Mewuk. Tuolumne or Central Sierra; Dr. Merriam 's Middle Mewuk. Mariposa or Southern Sierra; Dr. Merriam 's Southern Mewuk. s Miwok is the name applied to themselves by the people of the interior only. It was so used by Stephen Powers. Powell, in his appendix to Powers, called both Miwok and Costanoan people Mutsun. Subsequently, in "Indian Linguistic Families," he abandoned Mutsun and called the Miwok family Moquelumnan. Am. Anthr., n.s. IX, 338-357 and map, 1907. 1911] Kroeoer: Languages North of San Francisco. 293 So far as is possible in simple series, this order represents both the geographical position of the dialects, in order from northwest to southeast, and their linguistic similarities. Thus the Lake seems more removed than the other Coast dialects from the interior division. The Plains dialect is the nearest of the interior division, lexically as well as geographically, to the Coast division. Within the interior group the contiguous Northern Sierra is the most similar, and the remote Southern Sierra the most dissimilar, to the Plains dialect. Grammatical material for comparison of the various Miwok dialects is available in Dr. Barrett's vocabularies and the miscel- laneous earlier ones; in notes made by the author on five of the dialects; in information collected by Professor E. B. Dixon on the Central Sierra dialect and kindly placed at the writer's disposal; and in Gatschet's article on the grammar of the Southern Sierra dialect. 7 COMPARATIVE PHONETICS. All four of the interior dialects possess and u. The Coast dialects lack these sounds, 8 which are also wanting in the Costanoan languages. This difference is a reflection of linguistic environment. Besides the interior Miwok, the Maidu, Yokuts, and Chumash, all in contact with Shoshonean languages, have these characteristically Shoshonean vowels. Nearer the coast, Yuki, Porno, and Wintun lack them. At least e and o are open in quality. Sonant stops are difficult to distinguish from surds in all Miwok dialects, as in Costanoan, the two classes being less differ- entiated than in Maidu, Washo, and Porno, in which their relation is more nearly as in English. In the interior dialects the two positions of t are close together, and the sounds difficult to distinguish. In the Coast i Specimen of the Chumeto Language. American Antiquarian, V, 72, 173, 1883. s As written by Dr. Barrett. The writer has recorded kanni or kannit, I, kawul or kawtil, night, huma, huma, or homa, no, untti-ko, people. In the Plains dialect 6 and u are perhaps also of less distinct quality than in the Sierras, as an obscure o and u were generally recorded instead. 294 University of California Publications in Am. Arch, and Ethn. [Vol. 9 idiom, one t is dental, the other, represented by t-, post-alveolar or palatal, almost like the affricative tc, so that the difference between the two sounds is more readily perceptible, as in Cos- tanoan. In the Coast dialect the impression was received that s and c may be but one sound. Doubling or lengthening of consonants, already mentioned as found in Northern Sierra, occurs frequently in all dialects, though but irregularly recorded. Compare Central Sierra mulli, sing, Plains uccu, drink, Coast elli, see, luppu, stone. The posterior palatal nasal n occurs medially in the three Sierra dialects, but is replaced by n in Coast and Plains. Com- pare Sierra and Plums: one kene kenatii (Coast kene) knee hofioyu honoi leg tufiu tuna earthhouse hafii hanepu The dialects actually on the Coast, that is Marin and Bodega, at times show a y corresponding to 1 of Lake, Plains, and Sierra. koya, girl, L kola koyo, leg, L, P, S kolo meye, bird, L mele oye, coyote, L, P, S ole, ole- Another, more irregular, correspondence is of n, 1, t, y, s. two: ot-, oy-, os- large: utu-, unu- oya- white man: uten-, alen-, utel- tongue: nepit, letip (with metathesis) As Dr. Barrett has pointed out, s frequently changes to h in the Southern Sierra dialect. No dialect shows words either beginning or ending in more than one consonant. In the Sierra dialects almost all words that may be supposed to be stems, such as simple nouns, end vocali- cally. On the plains and coast the corresponding words often end in consonants. 1911] Kroeber: Languages North of San Francisco. 295 nose, Sierra huku, Coast and Plains huk tooth, Sierra kutu, Coast and Plains kut night, Sierra kawulu, Coast and Plains kawul water. Sierra kiku, Coast and Plains kik louse, Sierra ketu, Coast ket, Plains ken tongue, Sierra nepitu, Coast letip, Plains nepit It is probable that the Sierra final vowel -u is secondary, and that the original form of the stem of such words ended con- sonantal! y. In Southern Sierra some consonantally-ending words do not show the final -u of the two northern Sierra dialects, but this may be the effect of slurred pronunciation. The northern dialects present -u even on such stems as honoi, knee, melnai, wasp, lapisai, fish. While their stems thus end vocalically, the three Sierra dialects however possess grammatical suffixes that are conson- antal, such as -t, my, -n, the possessive case, -m, the terminalis, and -k, the plural. In all dialects simple nouns and verb stems are rarely mono- syllabic. PLURAL. The usual plural of animate nouns 9 is -k in Central as in Northern Sierra; Plains, Coast, and Lake show -ko, of which occasional instances have also been given in Northern Sierra. The Southern Sierra idiom alone possesses a totally different formation for the plural, of which no trace has appeared in any other dialect, just as no instance of -k has been found in Southern Sierra. 10 The plural of nouns is formed in this dialect most frequently by -ya, sometimes by -ti or -yati ; the plural of verbs and adjectives usually by -ti. In the formation of the plural in nouns, there is probably usually a change of accent, resulting at times in the dropping or moving of vowels. Occasionally there are irregularities. 9 Inanimate nouns show the suffix only occasionally. 10 That is, in the noun, except hiso, hair, plural hiso-k. The pro- nominal and demonstrative forms miko, neiko, iniko, no doubt contain the suffix. 296 University of California Publications in Am. Arch, and Ethn. [Vol. 9 ami-ya-nti, my mothers oha-ya-nti, my wives huggo-ya, heads nito-ya, noses yane-ya, bows tisso-ya, tisso-ti, hands noaha-ti, knives (Spanish) utcu-ti, houses yiita-yati, evenings lokka-yati, summers hika-yati, deer, nan-taya, men (naiia) ox-taya, women (oxa) tuhuhi-ti, black oyani-ti, large huate-ti, run (plural subject) ua-ti, eat (plural object) uhhu-ti, drink (plural object) Numerous other instances are given in Gatschet's vocabu- lary. 11 The following illustrate the shift of the accent : your nail, ha'la-no, pi. hala'-ya-no your belly, otce'-no, pi. otee-ya'-no bow, ya'we, pi. yawe'-ya ax, la'tca, pi. latca'-ya (Sp. la hacha) sun, wa'tu, pi. watu'-ya star, tcala'to, pi. tcaltu'-yate fly, u'tcum, pi. utcmu'-ya salmon, ko'sum, pi. kosmu'-ti Southern Sierra nouns take the plural whether animate or inanimate ; the -k, -ko suffix of the other dialects has been found chiefly on stems denoting animate beings. The radical difference of the Southern Sierra dialect from all others in the formation of the plural is remarkable. The suffixion of -ya with shift of accent and occasional modification of the word, recalls the Yokuts plural in -i or -a with accom- paniment of similar though more developed changes. The suffix -ti suggests the Salinan plural, which while irregular is most frequently produced by the suffixion, or apparent infixion near the end of the word, of 1 or t. Yokuts is adjacent to the Southern Sierra dialect, and Salinan is not far distant. A 11 Powers, Contrib. N. A. Ethnol., Ill, 539, 1877. 1911] Kroeber: Languages North of San Francisco. 297 plural of verbs such as Southern Sierra possesses is found in Salinan and in Costanoan, 12 but no trace of anything of the kind appears in the material collected by Drs. Tozzer, Dixon, and the author in the other Miwok dialects. CASE ENDINGS. The principal case-suffixes are: S. Sierra C. Sierra N. Sierra Plains S. Coatt Costanoan Objective -i -i -i -tc -tc,-t- -c, -se, -e, -ne Possessive -ii -n -n -nu, -xno -n Locative -to, -t -t -to -to -tak, -tka, -ta Ablative -mu -mo -mo -mu -mo ) ,. > -m, -me. -mo 13 Termmahs -m, -ma -m, -am -m -m' e -m 17 J Instrumental -s -s -su -cu -cu -sum, -um, -yum Comitative -li -ko'ta The significance of several of the suffixes does not appear to be rigidly limited even within any one dialect. An ending -wit or -win, sometimes replacing or replaced by -m, has been found on terms of direction in almost all dialects. Southern and Central Sierra olo-win, "west," Northern Sierra olo-wit, Plains etca-wit. Coast olo-m, Lake olo-m-wali, "south." Southern and Central Sierra hisu-m, hihu-m, "east," Northern Sierra hisu-m, hisu-wit, Plains huke-wit. Southern Sierra ne-win, "east, upstream." PRONOMINAL FORMS. The known pronominal forms in the several dialects have been brought together in the appended table. It appears that the four interior dialects agree in expressing all indications of person, except in the independent or emphatic 12 By the suffixion or infixion of -s, in the Mutsun or San Juan Bautista dialect. is General locative i* Given as superessive. IB Given as inessive. i Locative and comitative also. IT On olo-m, south. 298 University of California Publications in Am. Arch, and Ethn. [Vol. 9 pronoun, by suffixes. On the coast this synthetic trait largely disappears. It is true that the subjective and possessive pro- nouns are prefixed and the objective suffixed in the Southern Coast dialect, but the affixion of all is quite loose, and perhaps better described as accentless collocation. A second distinctive feature of the interior dialects is the existence of three series of subjective suffixes, the choice between which is dependent on tense and mode. In the Northern Sierra dialect that series which has been designated as the first is practically identical with the objective suffixes, the second with the possessive, while the third is distinct from both. In Central Sierra and Plains, the first series is also essentially a duplicate of the objective forms. An equivalent of the Northern second series has not yet been found in Plains, while the Central second series is entirely different from the Northern, and but partially similar to its own possessive suffixes. The third series, which is everywhere unconnected with either possessive or objective forms, is characterized by -m in the first person singular, -s in the second singular, and -p in the third plural. It is found with but little variation in Plains, Northern, and Central Sierra, and seems to be represented also in Southern Sierra. On the coast, there is no indication of three subjective series, as indeed might be expected from the analytical or at most loosely synthetic character of the pronominal forms. In the table, the subjective prefixes of the Southern Coast dialect have been coordinated with the second subjective series only because of their identity with the possessive prefixes; it is not certain that they correspond genetically to the interior second series more than to the first or third. 4 S. Sierra C. Sierra N. Sierra Plains S. Coast Independent I kanni kanni kanni kanni kanni thou mi mi' mi' mi' mi' we mahi masi 18 maci maci 1 ' mako /><; 4- 4 f\ 4- 1 -f / i Ir 1 +/! i + ni ye miko miko miko moko 20 miko 18 Or otim. 19 Maci-mi-tc, objective plural. 20 Or moko-tokni. 1911] Kroeber: Languages North of San Francisco. 299 S. Sierra C. Sierra N. Sierra Plains S. Coast Possessive my -nti -kan -nti, -t -ka ka- thy -nu -n -no -nu, -in un- his -hu -s -co -cu, -ic ue- our -ti -ti -ma -mac ma- f ni tilr . -ici your -LlK -mok -moko -mok mikon their -hu -k -ko -ik Subjective 1 I -t -t, -te -t thou -ni -n -n, -ni -n he -k a L- VA -K, -KO, we -m -m -m, -me -m ye -toxni -ton -tok, -tokni they -k 22 -ko -k Subjective 2 I -k 28 -nti* ka-, k> thou -s -no un- he -co UC-, C- we -kti -ma ma- -ktitc -teo ye -ktos -muko mikon- they -P -ko Subjective 3 I -ma, -m -m -m, -ma -m thou -s -s he -wo we -ti -ti -mac -titc ye -tos -toksu -tokun they -pu P -pu -P Objective me -t -t, -te -t, -ti -kanni 21 thee -n -n, -ni -n, -ni -mi him -k, -ko, -wo -t, us -m -m, -me -muk -mako you -ton -tok, -tokni them -k, -ko I-tJiee musu mos mucu cima I-you mutos mutokcu cimatoku 21 Found only after the future suffix -i. 22 Takes the form -ko after the future suffix -i. as The second series subjective do not correspond in Central and Northern Sierra. 2* The single subjective series, in the Coast dialect, is identical with the possessive prefixes. 26 Evidently the objective forms of the Coast dialect are merely the independent pronouns suffixed or postposed to the verb. 300 University of California Publications in Am. Arch, and Ethn. [Vol. 9 MEANING OF THE SUBJECTIVE FORMS. The three distinct series of subjective affixes suggest a feature which is found in other languages and has sometimes been erroneously interpreted. In some American tongues the sub- jective personal elements used with the verb are partly or wholly the same as the possessive forms employed with nouns. In other idioms certain of the subjective elements, especially the in- transitive ones, resemble the objective forms or are identical with them. An explanation that is plausible rather than sub- stantial has repeatedly been made for such phenomena. It is said that in such languages the verb is essentially a noun, or the intransitive verb in reality impersonal and transitive, the form "I eat" being literally "my eating," and "I am sick" more correctly "it is sick to me." This view must be vigorously opposed on general grounds, except where it is capable of proof by specific evidence. It would be just as reasonable to interpret "my food" as "I food." Our own Aryan languages of course do not permit the phrase "I food" as they do allow "my eat- ing," but this circumstance is of no moment in an American tongue. It is saner, if less enticing, to look upon the several series of pronominal forms that many languages possess, as essentially significant only of person, and as primarily undiffer- entiated as to subjective, possessive, and objective relation. In certain types of language such differentiation of form may serve no purpose and may therefore not occur. In other cases the presence of two or more pronominal elements in the same word, or perhaps other circumstances, may cause the existence of several series of forms to be necessary or advantageous. When- ever, in such languages, a word contains only one pronominal element, as in the possessed noun or the intransitive verb, it must then be a matter of indifference to intelligibility which form is drawn upon. "I," "my," and "me" in the native words for "I eat," "my food," and "bites me" would in such case not mean the same as their Aryan equivalents, but would only be indicative of the first person, their grammatical force being a function rather of their position in the word or phrase, the part of speech to which they were joined, its logical meaning 3911] Kroeber: Languages North of San Francisco. 301 or some similar circumstance. As a matter of fact there are abundant instances of identity respectively of possessive and objective, possessive and transitive subjective, and intransitive and transitive subjective pronominal forms, which no one would think of explaining by Indo-European analogy. It may be objected that even in synthetic languages the co- existence of several series of pronominal elements would be impossible except as they originated from a difference in function such as that provided by the subjective, objective, and posses- sive relations. But the ultimate origin of the several sets of pronominal forms can probably be ascertained in but few languages, and is therefore generally a matter of pure specula- tion. Even if the original meaning of a form translatable by "I eat" had been, in any particular idiom, "my eating," it would be a grave error to assume, in the absence of direct and positive evidence, that it still possessed that meaning. There is normally little connection, in any language, between the present force of a word or affix and its "original" function and meaning. Miwok bears on this general question through the fact that at least in certain dialects one of its series of subjective pro- nominal elements resembles the objective, another the possessive, while the third, differing from both, might be interpreted as distinctively subjective. The form employed depends, however, on the tense and mode of the verb. As it would be absurd to assert, because an essentially objective form is used with the future suffix, a possessive with the preterite, and a subjective only with the present, that therefore the future is impersonal, the past a noun, and the present a verb, it can only be said that there are three sets of pronominal elements which have no differ- entiation of meaning to accord with their divergence in form. USE OF THE SUBJECTIVE FOEMS. "While at least two of the three series of subjective pronominal suffixes agree closely in several dialects, there is variation in the affixes of tense and mode by which their choice is determined. 302 University of California Publications in Am. Arch, and Ethn. [Vol. 9 First Series. Second Series: Third Series. Northern Sierra -i, future -si, passive -imi, usitative -uni, potential -ke-ce, kco, past -ce-k, past stem, past -ce, past -ke, past stem, present -na-ka, past Central Sierra -i, future -ke-so, past -fie, past or present -teo-sa, tco-ka, recent past -se, past -ke, past stem, past -na, -fie-na, past or present stem, present -ted', present ew-an-ko, nega- tive present Plains. -ca, future -ma, past -tu, past The second series has not been found in this dialect . stem, present -ka, past Perhaps the most conspicuous difference is that the preterites -ce and -ke require the second form in Northern and the first in Central Sierra. It must be recalled, however, that the forms of the second series are not homologous in these two dialects. 8 l S 2 S 3 SPECIAL PECULIAEITIES. Central Sierra. The independent pronouns in full are : Subjective kani mi' [neal masi, otimei D 1 masi, otim P 1 P 2 P 3 otitcik miko [nekoal, inikal Objective kanii minii neial mikoi nekoial, isakoi Possessive kanu-n minu-fi nesu-n, nosu-n, inisu-n] otime-n otitciku-n miku-n noku-n] The forms given as of the third person are demonstrative, and are derived from the stems ne, no, and ini or isa. The ending 1911] Kroeber: Languages North of San Francisco, 303 -al has not been found in Northern Sierra, and appears to be animate. Professor Dixon's material in this dialect brings out the difference between the dual and plural of the first person, which in Northern Sierra could only be suspected from the presence of the two forms masi and itci. Itci does not occur in Professor Dixon's notes, but its characteristic element tc appears in oti-tci-k, as compared with oti-m, we two, in which -m evidently represents masi. The final -k of otitcik is the suffix of plurality. It is curious that a language should possess a dual in only one person of the pronoun. It seems reasonable to believe that the true distinction between the two forms is rather one of inclusion and exclusion of the second person, which has happened to coincide in the informant's mind, or in some of the examples given, with the difference between a dual and plural. The strange forms oti-m and oti-tci-k 26 confirm this supposition, as they are evidently both derived from oti-ko, two. Further, in many languages that possess a separate inclusive form of the first person plural, this is phonetically more similar to the second person plural than to the first person exclusive. So the Central "dual" masi, essential element m, parallels mi, you, and miko, ye. On the other hand, in Dakota, according to Riggs, 27 an in- clusive form of the first person is restricted to a dual significance, while the exclusive is plural. Four of the linguistic families adjacent to Miwok Wintun, Maidu, Washo, and Yokuts also show a pronominal dual, and besides these, so far as known, no others in California, 28 except Chumash which is in contact with Yokuts. In the possessive the form for the first person is -kan, instead of Northern -nti or -t. It is certain that a mistake has not been made because Professor Dixon's informant knew the form -nti, but gave it as characteristic of the Southern Sierra dialect, which is correct. Evidently the Central dialect, like the Plains idiom, 2 Confirmed by Southern Sierra otit-i. ZT Contrib. N. Am. Ethn., IX, 10, 1893. 28 The Shasta dialects nearest to Maidu show dual forms of the pro- nominal verb affixes only. Dixon, The Pronominal Dual in the Languages of California, Boas Anniversary Volume, 80, 1906. 304 University of California Publications in Am. Arch, and Ethn. [Vol. 9 replaces the suffix form -t or -nti, which has parallels in the per- sonal endings of verbs, by a suffixed abbreviation of the inde- pendent pronoun kanni. It is curious that adjacent dialects differ, and separated ones agree, in this point. Usually the order of objective and subjective suffixes is the same as in Northern Sierra: -ni-t, I-thee; -tokni-t, I-you; -t-an, thou-me. There is however one set of forms that seem to place the objective suffix last : -pu-t, -pu-n, -pu-m, -pu-ton, they-me, -thee, -us, -you. The possessive suffixes are added to verbs which are the object of another verb, the suffix denoting the logical subject of the subordinate verb. teloteo-ke-t mulli-s, I heard him singing, literally, I-heard his-singing soye-ne-ni-t howata-n, I saw you running, literally, I-saw-you your- running ne-al soye-teo-t howata-kan, he sees me running, literally, he sees-me my-running The Central subjective suffixes of the first series added to a substantive or interrogative imply the verb substantive. hayapo-t, chief- 1, I am chief mana-n, who-you, who are you? mini-n, where-you, where are you? Plains. An unexplained form -ikam, meaning they-me, occurs in kanni heat-ikam, they will hit me. The imperative plural shows an ending -tc : uce-tc, drink ! A suffix -k may denote the object of the third person in the imperative : pata-k, strike him ! 29 Elements similar to the subjective suffixes of the first series are added to the independent pronouns before case-suffixes : maci-mi-tc, us; itcu-me-m, with us; kanni-ti-m, with me. This contrasts with the Northern Sierra idiom, in which the locative element follows the pronoun, while the affixed pronominal element is added to this. Coast. An ending -t-, -it-, occurs on transitive forms with pronominal object of the third person : oke-mmi-t-, strike him ! ; k-tcamate-pu-it-, I have it ; k-oke-ka-t-, I struck him. 29 Compare Costanoan : San Juan Bautista, imperative plural, -yuts ; Mon- terey, imperative with object of third person singular, -nk. 1911] Kroeber: Languages North of San Francisco. 305 It has been said before that the subjective and possessive elements in this dialect are probably not true prefixes, but accentless particles or proclitics. They are at times audible as final sounds of the preceding word with which they have no grammatical connection. kenum-unye-s oke-kanni, constantly he hits me eke-m-ka-n tcama-no-ni-n cumuki, from where do you take your pipe? nit-u-n cumuki, this is your pipe eke-to-ni-n huyena-c, where were you born? eke-cu-k lanta, where is my bow? homa-k iolum-oti nome, no, I eat rabbits That these forms are particles is made more probable from the circumstance that they precede the word with which they stand in syntactical relation, while the Interior equivalents are suffixes. It is difficult to understand how a truly incorporated element could change from suffix to prefix, whereas it is readily intelligible that if once a suffix becomes sufficiently detached from the stem to be properly a particle or enclitic, an alteration of idiom without grammatical change may cause it to be proposed. General. The m which in Miwok is the base of the inde- pendent pronouns of the second person is the usual designation of this person in California. K for the first person is common in south central California, appearing in Miwok and Costanoan, Salinan, Chumash, the Tiibatulabal branch of Shoshonean, and the southernmost or Buena Vista dialectic division of Yokuts. The elements of the Miwok suffixes are on the other hand without parallel : m and t for the first person, n and s for the second, and s and k for the third, are quite unique in California. 30 The coexistence of and yet thorough difference between the independent pronouns and the pronominal affixes in the interior dialects of Miwok, is almost without parallel among the languages of California, except Wiyot and Yurok; and even in these the possessive affixes resemble the independent pronoun. The restric- tion of pronominal affix forms to a suffixed position is also unusual. Most the languages that resemble Miwok in the pos- session of case-suffixes and in general phonetic character, treat so Athabascan n of the second person is the only similarity. It is prob- ably necessary to travel as far as the Dakota before encountering m for the first and n for the second person, and then as prefixes. 306 University of California Publications in Am. Arch, and Ethn. [Vol. 9 their pronominal elements as independent words, or at most proclitics or enclitics. Even the affixing languages of California, such as Athabascan, Washo, Chumash, Salinan, and Shoshonean, prefix. The only suffixing languages, besides Yurok and Wiyot in northernmost California, which are of quite a different type of structure and phonetics, are Yana and Wintun, which latter further resembles Miwok in that its suffixed series has nothing in common, in form, with the independent pronoun. 31 The Wintun suffixes are, however, much more restricted than those of most Miwok dialects, being only subjective, and alike in singular and plural. COSTANOAN CONNECTIONS. In view of the fact that the Miwok and Costanoan languages now seem more probably than not to be genetically related as one stock, 32 a comparison of their pronominal forms is desirable. It is clear that as the Miwok interior dialects are compared with the Costanoan idioms, there is a fundamental difference in their employment of pronominal elements, Miwok being elabor- ately synthetic and Costanoan almost purely analytical. As long as the interior Miwok languages were the only ones known, the divergence on this point seemed so insuperable as to render more than doubtful any lexical evidence as to relationship. It appeared that the suffix forms constituted the original pro- nominal apparatus of the Miwok language, and that its inde- pendent pronouns, which are obviously identical with those of Costanoan, had been borrowed from that family. The acquisition of material in the Coast dialect has however altered the situation, in revealing a language which is un- doubtedly Miwok, but more nearly analytic than synthetic in its employment of pronouns. There can thus no longer be objection to a recognition of Costanoan as a branch of Miwok on the score of divergence of pronominal usage, especially as even Costanoan is not absolutely free from traces of synthetic forms. 33 As the two groups of dialects agree in all other revealed points of essen- tial structure, and as they hold a certain lexical element in 31 E. B. Dixon, in Putnam Anniversay Volume, 468, 1909. 32 Present series, IX, 237-271, 1910. ss Present series, II, 72, 73, 1904. 1911] Kroeber: Languages North of San Francisco. 307 common, their relationship can not be questioned except on the ground that the etymological identities may appear insufficient in number. In this case the structural resemblances might be explained as due to contact and influence, and common words as borrowed. Both groups of tongues are however still imperfectly known, and as they agree in perhaps one fourth of the words which may reasonably be assumed to have been determined as stems, it is likely that only fuller data are needed to increase this proportion. At any rate there no longer exist any gram- matical obstacles to a belief in the genetic unity of the two divisions. The conditions existing within the larger family as to pro- nominal matters, may be described thus. The Sierra Miwok dialects have a full array of synthetic forms, and analytic or separate pronouns are superfluous and emphatic. The Plains dialect is also synthetic, but may possess a somewhat reduced apparatus. Coast Miwok is between a synthetic and analytic stage. It lacks most of the Sierra forms, and it is difficult to decide whether its elements are prefixes or particles. Costanoan is distinctly analytic, employing ordinarily only the independent pronominal stems, but it retains vestiges of synthetic forms in the imperative. It seems most reasonable to consider the Sierra synthetic structure as more original, and to regard this as having been lost in successively increasing degrees, but never entirely, in Plains, Coast Miwok, and Costanoan. Theoretically the opposite explanation is equally logical, but does not fit the facts as well. If the synthetic machinery were a late growth, it would hardly be so uniform. Compare the Coast Miwok second and third persons in n and c, which are quite isolated in that dialect but recur in the interior. Lexical divergence proves the Coast dialect to have been detached from the interior for a considerable period. Identical forms could hardly have originated independently in two separate regions. That Coast Miwok k of the first person on the other hand has its source in the independent pronoun kanni and not in the Miwok affix forms t and m, proves nothing, for once a loosening of the synthetic system has set in, it is not unnatural that 308 University of California Publications in Am. Arch, and Ethn. [Vol. 9 abbreviated forms of the independent stems should begin to replace the affixual stems even before these have become entirely analytical. Assuming then, a Miwok-Costanoan family, and the chain of relationship, as established on structural and on the whole con- firmed on lexical grounds, to be : Sierra, Plains, Coast, Costanoan, it is evident that the interior Miwok dialects, which have gener- ally been thought to be geographically in closest contact with the Costanoan languages, 34 are linguistically the farthest removed from them. The gap is bridged by the Coast Miwok dialects, which lie to the northw r est of both. This fact is of primary im- portance in any speculation concerning the origin or movements of the stock. It is curious that an analogous situation is encountered among the Yokuts. The Indians of that family lived in parallel and contiguous stretches of mountain and plain, each division possessing its own group of dialects. In the south the foothill and valley dialects are markedly different. As one proceeds north, the divergence becomes less marked, and finally is almost effaced. In this family also, therefore, the chain of relationship revealed by language runs from south to north and back to the south, so that the most divergent dialects are in geographical contact. The Yokuts division into mountain and plains groups, which are linked only at the northern end, can be paralleled by a geographical separation of the Miwok-Costanoan family into an interior and coast division, also united at their northern extremity. DEMONSTRATIVES. Demonstratives are developed from three stems : ne and no, this and that, and i-, appearing as ini, iti, ika, isa, which appar- ently is a radical of reference rather than of direction or dis- tance, and therefore possesses somewhat the character of a per- sonal pronoun of the third person. Interrogatives and relatives are fairly constant, the principal stems being man, who, min, where, and hi or ti, what. 3* They may really have been separated from them by a strip of Yokuts territory embracing the west side of the lower San Joaquin valley. 1911] Kroeber: Languages North of San Francisco. 309 SS CS NS P C L this ne ne, ne-al nei ne ni no no iti that no noi no no that ini ini it-i i- isa- isa- i- ika- imo ima- \\lio mana mana mana manti manti where mini mini mini- eke 85 what tinu ti'nit hiti hinti In Central and Northern Sierra all these stems take the plural suffix -ko, the objective -i, and the possessive -n, though the latter is added to ne, no, and ini only with the addition of -cu : ini-cu-n. Central -al seems to be a separate particle denotive of animateness ; it follows case and number endings : ne-su-n al, ini-k al, ne-ko-i al. In the Plains dialect the suffix -cu appears in both subject and object: no-cu, no-cu-tc, i-cu, i-cu-tc. Other forms are no-ko, objective no-ko-tc, those; ne'-im, here, no-m, there, mini-m, where; hiti-tc, what, objective. Coast Miwok ni-t-u, this, is perhaps the equivalent of Plains ne-cu. Other forms are: ni-to, here; no-to, it-i-to, there; it-i-ko, they; hinti-tc, what, objective; eke-m, from where; eke-to, at where. VERB. The verb formatives, which are all suffixes, show a similar range in all Miwok dialects and in Costanoan, but the individual suffixes used differ in the several dialects. Many of the blanks in the accompanying list can no doubt be filled. The Central material contains no data on causative, desiderative, continuative, or passive, the Northern no noun-agent forms, and the Plains and Coast forms are still more incomplete. Meaning C. Sierra N. Sierra Plains Coast causative -ne -nuku -ne desiderative -ce, -kce, -caino -caiku desiderative (welak) continuative -imi purposive ( f ) -ukna, mt-o- 85 Compare Wintun heke-, who, where. 310 University of California Publications in Am. Arch, and Ethn. [Vol. 9 potential -ni -uni, -ani negative -ew-an-ko -aim negative -ewa -cewa reflexive -po-sa -bo, -bo-sa past -se -ce, -cu past -ke -ke past -naka past -keso -kco, cak present -ted recent past -tco-sa recent past -tco-ke past past past or present -fie, -na, -ne-na present future -i -i future passive -si, -pa intransitive verbifying -fie -ne noun agent -pe excessive noun agent while -mu i after -se i -ka -ne-po(?) -ka -ma -tu -ca -up -api -ak Southern Sierra. Suffixes of mode and tense are almost un- known. Gatschet gives -t as indicating a preterite, but it is not certain that this is not the suffix expressing plurality of the object. For the future he gives -iku or -eku. His text contains the Central Sierra past suffix -ne. Central Sierra. The ''potential" has conditional force: noka-ni-s wo-ewa-yi-t, if it rains I will not go tokla-ni-t-an howat-it, if you hit me I shall run takya-ni-ma-s yona-im, if he strikes us we will kill him The interrogative is a, identical in use with the Northern form. The forms -mu i, while, and -se i, after, contain the pro- nominal suffixes. As these otherwise follow all modal and tem- poral affixes, the final -i is likely to be the objective case-suffix, used to denote clause subordination. The use of the possessive case for a similar purpose, as in the Northern dialect, is illus- trated in the Central text below. 1911] Eroeber: Languages North of San Francisco. 311 Plains. The preterite suffix -wa of other dialects has in this language a past usitative force. The negative is expressed by the independent preposed particle ket, similar to Southern Sierra ken, and Costanoan ekwe, kue. Coast. The desiderative welak is itself a verb, which pre- cedes or follows the verb which it accompanies. The negative is a preposed particle ela or opu. The suffix -ukna, and the pre- posed particle mt-o, may be purposive, future, or andative, having been obtained with the meaning "to go to" do something. Interrogation is indicated by the particle op, usually op-un with the pronoun of the second person. A common suffix, which has not been included in the preced- ing list, is -pu, used only when an object accompanies the tran- sitive verb. It may be pronominal compare -pu of the third person plural in the interior, or allied to the reflexive -po, -po-sa of other dialects. The Coast reflexive -ne-po seems to contain the same element. The analytic character of the Coast dialect is reflected in the tense-suffixes as well as in the affixual pronouns. The past -ka and present -up are commonly suffixed or postposed, but occa- sionally precede the verb. nitetei ka-k-etc, yesterday I slept ka-n-iolum, you ate manti-ka waya-mi, who gave it to youf kenum-up-c-yolum, constantly you eat As -ka or -ke occurs in the interior dialects, it is clearly an affix which has become an enclitic on the Coast, or an originally separate particle which has been degraded into a suffix in all other dialects, according as the several Miwok languages may have developed in an analytic or synthetic direction. COMPARATIVE ETYMOLOGY. Names of groups of people end in -tci in the Southern Sierra dialect, in -umni in the Plains. 36 In the three Sierra dialects derivatives signifying persons are formed from terms of direc- a This ending, which appears also in Yokuts and Maidu, has been dis- cussed in the present series of publications, VI, 340, 379, 1908, and in the American Anthropologist, n.s., VIII, 662, 1906. 312 University of California Publications in Am. Arch, and Ethn. [Vol. 9 tion by the addition of a vowel, usually -o, accompanied some- times by modification of the final consonant. In the plural, with the ending -ko or -k in the north, and -ya in the south, such terms become designations of groups of people. tcummate, south; tcuinmet-o, plural tcummet-o-ko, or tcumet-o, plural tcumte-ya, southerner, damman or dammalin, north; dammul-e, plural dammul-e-k or damlu-ya, northerner, hisum or hisuwit, hihum, east; hicut-o, plural hicut-o-ko or hi't-o-ya, easterner, olowit or olowin, west; olowit-o, plural olowit-o-ko or olwi-ya, westerner. On the Coast the vowel is omitted: kan-win-ko, northerners; compare tamal-ko, bay people. A diminutive found on such words as boy, girl, old man, old woman, coyote, is -tcki, -tcu in the South, -ti in Central and Northern Sierra, -tci in the Plains, -mbula occurs with diminu- tive meaning on several inanimate nouns in Northern Sierra. In the Southern Coast dialect many nouns in the absolute form or subjective end in -s, which is ordinarily lost in the objective and is lacking from Western Coast and Lake. man, tai-c, objective tai-t-u woman, kulei-c, objective kulei-t- (Maidu kule) old man, oyi-s, objective oyi-t- old woman, potei-s (Barrett), objective potci-t- basket, ewi-c, objective ewi-t- A noun-ending -s or -c is frequent in several Costanoan dialects. Adjectives of color consist, in all dialects of the interior division, of monosyllabic stems which are finally reduplicated. To the reduplicated syllable is appended an -i in the Sierra" dialects, the vowel of the stem in Plains. Southern Sierra: yodj-odj-i, red tuh-uh-i, black Central Sierra: gul-ul-i, black gel-el-i, white Northern Sierra: wet-et-i, red Plains: pud-ud-u, white kul-ul-u, black wet-et-e, red 1911] Kroeber: Languages North of San Francisco. 313 Coast adjectives of color end in -ta: ulu-ta, red, mulu-ta or loko-ta, black, poto-ta, white. Duplication occurs in color adjectives in the Lake dialect : mul-u-mul, black, awa-awa, red. Other adjectival stems appear in duplicated form in all dialects : Plains, wel-wel, good, tcu-tcu-i, sweet; Northern Sierra, tcu-tcu-yu, sweet; Central Sierra, ut-ut-i, large; Lake, tsum-tsum, sweet, koi-koi, bitter; Coast, un-un-i, large, kai-kai, bitter. PHRASES. Southern Sierra: yohe, kill! yohu-ma, I kill yohu-musu, I will kill you kani yoha-k oxxa-i, I kill a woman huine-mu, hina-mu, do you see me? huine-mesu, hina-musu, I see you hilai-fii-ma, do you fear met hilai-ak-ak, I fear hakai-ak-ak, hulwa-k, hungry oxxa-n utcu-hu, woman's her house tcuku-n utcu-hu, dog's his house ne, this, ne-to, here ini, that, ini-m, ini-to, there Central Sierra:" teuku-kan yolla issako-i nana-i, my dog bit that man ne'i nafia yona-na tcuku-i-kan, this man killed my dog tcuku-u suki-s, dog's tail-Ms tceak tcuku-i-kan sawa-s, I hit my dog with a stone sawa-i hata-na-k kiko-m, I threw a stone in the water wani utcu-t, in the house mana-li onotu-ton, with whom did you comet 6'po-ti-li onotu-m, I came with my father ne al haloine, he is sick haloine-ke-so-t, I was sick mil osemo haloine-i-n, you will be sick oti-m haloine-ewan-ko-ti, we two shall not be sick oti-tci-k tunna-na-ktitc, we are cold mii tunna-se-n, you were cold osemo tunna-i-ko, they will be cold haqai-fie-ke-m oti-m, we two were hungry osemo haqai-ne-i-k, he will be hungry nawato-s-a mii, are you tired f nawato-m, I am tired T Dr. K. B. DLxon. 314 University of California Publications in Am. Arch, and Ethn. [Vol. 9 so'yo-sa-k mos, I see you issako soyo-tco-t, he sees me hani miko-i tokla-motos, I hit you tokla-n, he hits you masi tokla-ni-mas, we are hitting you masi soyo-ne-na-k-ni-mas, we see you tekmo-po-sa-s, you are kicking yourself tekmo-msi-tos, ye kick each other mana-i tokla-na-s, whom did you hit? tinu ne, what is this? tinu-s takya-na-n, with what did you hit? toye-mu-i wosa-k, while he slept, I went away owo-se-te-i luk-sa-s, after I ate, you came mulli-se-te-i toye-na, after I sang, he slept toye-ku-m, I am sleepy toye-ti, let us sleep mii a howatu-n, did you run? kani soi-ne-na-k ne-su-n howata-s, I saw him running ne-al soiye-ne-te howata-kan, he saw me running mulli-pe, singer. kalan-pe, dancer Plains : tcama-caiku-m, I wish to eat tcama-ma-t, I ate tcama-ca, he will eat icu tcama-mu unu, he eating-from comes no-ku uccu-p, they drink unu-m uccu-mu, I come from drinking ucce-tc, drink, ye! uccu-ca-tokun, ye will drink pata-t, strike me! pata-muk, strike us kanni pata-cima, I strike you heat-nuku-cima, I make you strike him heat-nuku-caiku-ma, I want you to make him strike icu pata-n, he strikes you ket heta-cima, I do not strike you tcica-tu-n, you saw ket kiwai-m tcico-tc, not can-I see hiti-tc ono-nu, what are you doing? umiy-a tcama-tc, do you like food? ket umiy-im, I do not like it manti anit-u-ni, who gave it to you? mini-m okicca-tu-n, where were you born? lema-ka-p, they used to dance hana-ka-p hanepu-tc, they used to have a sweat-house wtiim-tci, old man ole-tci, ole-na, coyote haye-m cewole-xnw, near the ocean (at-proximity ocean's) 1911] Krocber: Languages North of San Francisco. 315 ocoo-xno kotca-ic, woman 's her house cawe-no haulo-co, friend 's his arrow teloko-te epali-tc, three jackrabbits (obj.) ocooc-itc, woman (obj.) ocoox-mo, with the woman mini-mu, from where? mi '-m kanni, you-with I, you and I ne-im kanni, he and I ata-tci-m-ka, brother-with-my Southern Coast : ele'u-api-ko, fishermen mit-a-ko-n kotca, Indians' house kotca-n wea, inside the house kotca-n lile, on the house kik hawin-to, near the water kik-to, in the water hewai-to k-uyeno-ke, I was born at the beach k-oni-ni it-i kotca-mo, I came from that house luppu-c, with a stone kono-su, with a bow tumai-to, with a stick k-ute-ka-mi, I saw you k-oke-ka-t-, I struck him us-koya-ka, he sang akal-utc kon kawul-o-ke, long-ago they used-to-dance uc-yolum-up, he eats k-welak-op-iolum, I want to eat un-hinti-po, what have you? k-ucu-pu kik, I drink water k-ute-po no-t. kulei-t., I see the woman uc-elli-po-mako, he sees us oke-mmi kanni-tc, hit me! elli-mmi kannu, look at me miko koya-mmi, sing, yel ule-mi-kan, free me! hinti-tc-op-un elli, what do you see? op-un elli-mako, are you looking at us? op-un oke ute-s kawul, can you see at night? yolum-api, comedor yolum-ak, comelon mat-aw-ak, hablador kenum-utc-opu-k cukum-welak, constantly I wish to smoke (i)kon kawul-ukna awe, van a bailar mafiana k-unya ok-nepo, I struck myself c-yolum-ne-t-, he made him eat c-kawul-ne-we-ukna-t-, he made him dance nako-mt-o koya, vamos a cantar 316 University of California Publications in Am. Arch, and Ethn. [Vol. 9 TEXTS. Dr. Dixon secured a short text in the Central Sierra dialect, which is here given, followed by Gatschet's published specimen of the Southern idiom, and a Lord's Prayer in Coast Miwok. The latter occurs in two versions, 38 which however are clearly derived from a single source. They have been combined and translated. Gatschet's text has been slightly systematized and conformed in orthography to the present work; the internal structure of words has been indicated so far as possible by hyphens; and the translation has been rendered somewhat more literally than in the original. Peculiarly, all the existing Miwok texts are very similar in brevity and narrowness of range, and, excepting the Lord's Prayer, even in subject matter. CENTEAL SIEERA DIALECT.a wunuto-so-t I went hunting tuku-se-t I shot kene-mei Once wakal-mo to the river. kaulupa-i In morning itanok kosumai-so-m Then we fished. ew T a-ne-pa-k enatosu when we got none, and wele-so-m temoka-i we caught six owoya-i deer. owoya-i deer. woe-so-m we went, toyese-so-m We slept imo-ok from there olo-win loklo-m below to plains itanok hakai-ne-pa-k Then I got hungry, kosumoy i -ke-so-m we fished motam kene-i half way one woe-so-m isak-mo we went to that. ewa-ka-so-m 40 kosumo-i kene-i One woule-t I came home. tamal-in north kawole-i night. hoya-ke-so-m We got there. hakai-ke-so-m We got none fish. We were hungry hoiyenon kosumoiyi-ke-so-m itanok next day we went fishing. Then kosumo-i enatosu wooltu-me-n 41 fish and returned hinsaiemes wuntoyi-ke-so-m owoya-i kene-i toko-tu-me-n at noon. We went hunting, deer one our-killing 6'toti oyise-nepai sumito-i itanok wilu-se-so-m large four-prong fat. Then we were filled. ssDuflot de Mofras, II, 391. 39 By Dr. E. B. Dixon. 40 Compare the negative suffix -ewa. *i "Of our returning. ' ' 1911] Kroeber: Languages North of San Francisco. 317 SOUTHERN SIERRA DIALECT." tcumte-ya Chumteya maise-i 43 maize, hunema fishing. warai-ko 43 sheep utu Thick tuhuhi black, tcinepitki short, oha-ya women huato run, hawa-t on rock. utcu-pu live aiye-to on flat hale-ya-t in mountains. ken not ne-ok These oni have. hiso-k hair nito-hu their nose laut skin upha swim, weyanu plant natu accurately tuhuhi black. papas-i potatoes, tugo shoot. halgi hunting i-ok They weyanu Plant uo-po they-eat waka 43 cattle tcinipitki short hugu-to-hu on their had. oyani large, yutotci brown. upha Swim esele-te children tcumte-ya Chumteya aitu keiie 44 all. Some kene 4 some ken not hapka climb suku paint lama-i tree. laut-M-i ne-ok These utu skins. kene 44 some humna bead, sekea-hu their- (?) aitu All kotan distant huhu-i wood hame-pu they cover, kefie-t In one kutcotc bone ulato long awuha needle humna-ho their-bead, tisso-no finger. kula coal. Many kene 44 some suku Paint poxau wear hunto-ya-hi Their eyes hupeto-ho their neck wakalmato-t in river, onadju fast ken suku not paint hasanu-i abalone-shell, ewuya Without pama smoke kumsol ulato shell long oki-ta-ho troxot on their chin three nawasu-hu wu theirdressas go.} kahu-i kanni hui-ne-ma miwi ken tobacco. I I have seen people not utcu-yu tolle-m aitu tuye-nu oisa oyani live, on earth all sleep, four large huyu-t oyani utcu-t utcu-pu tolle-s in fire. Large house-in they live, with earth huyu kaweni-m tulu-ma haksi wuksa fire in-middle, through hole smoke goes. utcu aitu tamu lu-pu house all. Differently they speak. 42 Gatschet, work cited. 43 Spanish. 4* Literally: "one." 45 Compare Central Sierra dialect negative suffix -ewa. 318 University of California Publications in Am. Arch, and Ethn. [Vol. 9 SOUTHERN COAST DIALECT. Chocouyem, Rio del Sacramento : Api maeo su lileeo ma n6nas Joukiousm, San Rafael: Api maco sa lile"to man^nas mi au6s omai macono mi taucuchs oyopa mi tauco chaquenit mi aues onia macono michauka oiopa mitauka chakenit opii neyatto chaquenit opu liletto. Tu maco muye genum ji naya opu negate chakenit opu lilSto tumako muye quenunje naya macono sucuji sulia macono mas6cte chague mat opu ma suli macono sucuji sulia mac6no masojte chake mat opu ma suli mayaco. Macoi yangia ume omutto, u!6mi macono omu incapo. mayaco maco yangia ume omut ulemi macono omu in capo. Nette esa Jesus. Netenti Jesus. api father TRANSCRIPTION AND TRANSLATION. ma-ko su lile-to our sky-in, oni-a 1 mako-no 2 mi taulmtc 3 oyo-pa 4 come our thy (rule), (be done) tcake-nit 6 opu wea-to 7 tcake-nit opu like mako us suli-a 9 forgive suli forgive ule-mi free-thou muye all mako-no 2 our ma-ya-ko our (debtors), mako-no 2 our mi awes thy name, mi tau-ko 5 thy (wills) lile-to tu earth-on like sky-in kenum hi 8 waya mako-no suku-hi constantly day give our , ma-sokte tcake-nit opu ma our (sins) like we mako-i yani-a huma omu-to us (lead) not bad-to. omu evil inkapo 10 nete ese Jesus NOTES. 1 The suffix of oni-a reappears below in suli-a and yani-a, possibly also in oyop-a. 2 Perhaps the possessive mako-n, our ; the context seems to demand " us. " 3 Readings taucuchs and chauka ; perhaps confusion with following tauko. 4 Stem perhaps oyo ; a passive suffix -pa occurs in Northern Sierra Miwok. 5 Probably plural in -ko. 6 Tcake mako, asi como nosotros; tcake-tu taic, like a man; tcake-tc hayuca, like a dog; opu recalls the present-tense suffix -op. ? Readings neyatto and negato for ueyato; wea is earth, s Hi, sun, day. s Suli is literally to pity. For the suffix -a see note 1. 10 Hinkapo, cinkapo, lo hizo. 1911] Kroebcr: Languages North of San Francisco. 319 SUMMARY. In most respects Mi wok is a language of the analytic and uncomplex Central Californian type. This appears in its simple phonetics, simple verb and sentence-structure, the presence and nature of case-suffixes, and the character of the plural in the noun. Adverbial, spatial, and instrumental ideas have not been found to be expressed by affixes to verbs. Complex derivational and etymological processes so-called polysynthesis, redupli- cation, and vowel change, are little developed; the principal structural device being suffixation. These characteristics are common to all Miwok dialects and are shared by the Costanoan languages, which, on lexical con- siderations, are probably to be considered as genetically related to Miwok. In pronominal elements of both nouns and verbs, and to a certain extent in the modo-temporal affixes of verbs, there is however a wide difference between the interior Miwok languages on the one hand and the Coast Miwok and Costanoan dialects on the other, the former being suffixing and synthetic, the latter proposing and analytic. The synthetic dialects seem to be more representative of the original structure of the language. Their most pronounced peculiarity is the possession of three distinct series of subjective pronominal suffixes of verbs, each restricted to use in connection with certain suffixes of mode and tense. All Miwok and Costanoan languages are entirely without prefixes. 320 University of California Publications in Am. Arch, and Ethn. [Vol. 9 POMO. The territory of the Porno Indians is south of that of the Yuki, and centers about the present Mendocino, Sonoma, and Lake counties, of which it comprises the greater part. It has recently been fully described in a paper published in this series by Dr. S. A. Barrett, who has in this connection also made an exhaustive determination of the dialectic divisions of the family. These are seven in number, with one additional slight sub- dialect. A comparison of vocabularies of the seven dialects shows the following degrees of affinity among them. The Northern and Central dialects, which are but little differentiated, seem to represent most nearly the original form of the language, as the other dialects resemble these two more nearly than one another. Least divergent from the Northern- Central form of speech are the Eastern dialect and a group consisting of the Southwestern and the Southern dialects, with the additional sub-dialect of the latter. The two most divergent dialects, the Southeastern and Northeastern, are spoken in territories most remote from the heart of the Porno area, and by people in contact with foreign languages. The Northeastern is probably a direct development from the Northern-Central group, and in a different direction from the Southeastern, for the two dialects have less in common with one another than with any others. The material here presented was obtained in 1902 and 1903 from Raymond Brown and Thomas Mitchell, and is from the Eastern dialect, spoken on the shores of the greater part of Clear lake. Dr. Barrett's vocabularies show that this dialect differs from the Northern and Central almost as markedly in verbal endings as in vocabulary. In fact it seems that we have really to reckon with seven Porno languages rather than mere dialects. Porno belongs to the Central Californian morphological type of languages characterized by structural transparency and failure to use pronouns as grammatical machinery. It resembles Yuki in this regard. The differences between the two linguistic 1911] Kroeber: Languages North of San Francisco. 321 stocks are however great. Yuki uses only suffixes, Porno employs prefixes and suffixes. Porno not only possesses whole classes of affixes that have no parallels in Yuki, such as verbal instru- mental, but uses its affixes with wider and at the same time more concrete meaning than the majority of comparatively vague Yuki suffixes possess. Special features of Porno, such as a sex gender, are lacking in Yuki. The nature and employment of demonstratives are very different. On the whole the two languages have but few points of structure in common, other than such as are of a general Central Californian character. Porno often shows an unexpected richness of structural development. Thus the noun, where we should perhaps first look for it, is ordinarily without any designation of plurality. But a few substantives denoting persons show different stems for the plural. Several others possess a plural formed by a suffix -a. A considerable number of verbs have different stems in singular and plural. Several pairs of very frequently used suffixes of verbs express respectively the singular and the plural of the subject. Relatives by marriage are addressed in the plural as a sign of respect a trait found also in the nearby southern Athabascan languages by Dr. Goddard. 46 The plural is also expressed in animate nouns by the use of certain demonstratives. The pronouns show plural forms. Altogether the expression of plurality is much more developed than might at first sight appear, or than is the case in Yuki, where nouns are better provided with suffixes of plurality. A similar condition exists in regard to the expression of other grammatical ideas. PHONETICS. The phonetics of Porno are simple in that the language con- tains no sounds that are obscure or that do not occur in a con- siderable proportion of the languages of mankind, and in that the sounds are little modified by such processes as composition and suffixation. There are no combinations of more than two consonants, and even these quite clearly do not appear in stems. 48 P. E. Goddard, Kato Texts, present series, V, 143, 1909. See also E. Sapir, Yana Texts, ibid., IX, 101, note 150, 1910. 322 University of California Publications in Am. Arch, and Ethn. [Vol. 9 Dr. Barrett has discussed the phonetics of all the Porno dialects. Certain differences between his statements and those made here are explained by the fact that only one dialect is here presented. Other differences are individual, the result of two independent inquiries, neither of them final, and of slightly divergent orthographies. While Dr. Barrett's renderings are probably more correct, they could have been substituted for the author's only in part; for the sake of consistency it has there- fore been necessary in the present account to retain the forms originally written by the author. The vowels of the Eastern dialect are a, e, i, o, u ; e and o being open. The vowels are sometimes short and obscure ; never nasalized, strongly aspirated, or organically of doubtful quality. The most frequent vowel is a, next i. The proportionate occurrences are a 40, i 25, e 15, u 10, o 5, ai 5 times. It will be seen that front vowels are more common than back, and i and u than e and o. The consonants include series corresponding to p, t, tc (English ch), k, and q (velar). The p, t, and tc series include surd, sonant, nasal except of course for tc, and the stressed or fortis surd represented by a following apostrophe. The k series comprises surd, sonant, and fortis, lacks the nasal, but includes both surd and sonant fricative, x and g', the latter an ortho- graphy that has but little justification but which it has seemed best to retain for the sake of consistency and because no more appropriate character is included in the available facilities for printing. The same sounds were written in the q series: q, G, q', X, G' ; but it is doubtful whether all actually occur. A t- like Yuki palatal t-, almost intermediate in sound between t and tc, also d- and t-', were occasionally heard and written, but must as yet be regarded as doubtful. Ts and dz were also heard, but they may be only tc and dj. S and c (sh) bear the relation usual in Indian languages. Their sonants j and z were written, but are denied by Mr. Barrett. R is trilled and sometimes related to t. Ordinary 1 calls for no comment; I is surd 1, not affricative, and is Dr. Barrett's L. Y, w, and h occur; and two sounds written hy and hr seem to represent simple sounds, either 1911] Kroeber: Languages North of San Francisco. 323 distinct or only occasional modifications of initial y and r; com- pare I, originally written hi and xl. Sonant stops were not heard finally; surds between vowels were frequently heard as sonants: mib-al for mip-al, beg-ibax for bek-ibax. These phenomena may partly be due to Indo- European apperception, but they seem to have some Porno foundation, as mit becomes mir-al, evidently for mid-al. Dr. Barrett also writes no final g, d, or b. Words most frequently begin with consonants. Three words out of four end in a vowel, which may be taken also as the proportion in stems. This fact alone accounts for much trans- parence of word structure. At least k, tc, t, p, n, m, x, 1, r occur finally. There is no system of vowel mutations, but the vowel of a suffix is sometimes assimilated to that of the stem, especially in verbs. Thus -hiba, preterite, becomes diko-hoba, garma-haba, kuhu-huba, o'ane-heba. Ui, eye, becomes yu-xa, eye- water, tears The accent of words is rather well marked. In nouns it fre- quently falls on the last syllable : balai', blood, qala'l, liver, cima', ear, xawa's, chin, masa'n, terrible, musu', hair. This would be impossible in Yuki. In verbs the accent is generally on the stem syllable, irrespective of the number of affixes. GRAMMATICAL METHODS. Internal modifications of stems do not occur as an expression of grammatical form. Reduplication is either etymological or confined to a few parts of speech, such as verbs, in which it expresses repetition or continuation. Position is pretty well fixed, the verb coming last, the object after the subject, a pos- sessive noun, adverb, or subordinate clause before its grammatical regent. The principal means of grammatical expression is, as in most languages, affixation. Infixes have not been found; suffixes are more numerous than prefixes, but in the verb the latter are both frequent and important. About seventy-five affixes have been determined. Of these a third are verbal prefixes and a third verbal suffixes. 324 University of California Publications in Am. Arch, and Ethn. [Vol. 9 NOUNS. The noun is formally as undeveloped in Porno as in most American languages. Its principal affixes are a long series of adverbial postpositions or locative case-endings. -u, in -xam, in, into -bai, in -kate, at, to, by, near -a, -ya, at -n, to -mil, to -1, -alal, -nalal, to, toward -awa, from, in -w-ina, on -yu, under -xo-wa, before -na-uwa, behind -ki, for -imak, in company with -i, -ya, -iyai, with, by means of Examples : me-awa, from here ba-y-awa, from there me-a, here, at this xale-na, on tree bihyatsuxai-yai, with fingers bo-1, westward gayu-1, upward dano-nalal, up-hill-ward mo-bai, in a hole gai-na, on the ground xo-xam, in the fire xabe-wina, on the rock xabe-yu, under rock xabe-i, with a stone bihya-i, with the hand musu-i, with hair xai-yai, with a stick tce-una, on a chair ca-u, in the house oa-kate, next to the house oa-xowa, before the house ca-nauwa, behind the house ga-u-wawa, inside, indoors ga-u-waki, from in the house hegibax napo-mil, to their own town wi-wina Zok-a, fell on me gayu-lal galina-lal, up to the sky 1911] Kroeber: Languages North of San Francisco. 325 As will be seen, these endings are suffixed also to pronouns, demonstratives and adverbs. When an enclitic demonstrative follows the noun, this particle and not the noun receives the case-ending. mo-he-bai, in the hole xa-he-bai, in the water la-he-mak, together with the sun xo-he-mil, into the fire masan-ek-himak,<' with a white man Such locative endings as these are common in Central Cali- fornian languages, but are usually accompanied by two or three syntactical case-suffixes, as in Yuki, Maidu, Wintun, Yokuts, Mi wok, and Shasta. In Porno, syntactical case-suffixes are absent, except on names of persons and terms of relationship, which share a possessive -ibax and perhaps an objective -al with pro- nouns and demonstratives. wimaca-ibax oa, my father-in-law's house sulig' am-ibax oa, Sulig'am's house A vocative of terms of relationship is formed by -a. mex-a, older brother! tsets-a, mother's brother! Expression of plurality in the noun by a suffix is confined to a few nouns denoting persons. The plural of animate nouns is usually expressed through accompanying demonstrative elements. person, cauk, pi. hiba-ya woman, da, pi. o'ara-ya; ya-o'ara, deserted wife old woman, da-G'ara, pi. mac'atcur-a. old man, butsi-gi, pi. butsi-a, butsi-yaya young man, cela, pi. cela-ya relatives, o< ametcgi-a PRONOUNS. The Porno pronoun is typically Central Californian. It is never incorporated, has no abbreviated or affix form, is syntacti- cally a noun, and shows a regular development for person, number, and case. As in Yuki, there is no real third person, demonstratives being used. * The h of -himak probably represents an aspiration of the preceding k. Compare notes 49 and 50. 326 University of California Publications in Am. Arch, and Ethn. [Vol. 9 Subject Object Possessive 1 ha wi, u wa-x 2 ma mi mi-bax 3 m. mi-p mi-p-al mi-p-ibax 3 f. mi-t mi-r-al mi-r-ibax 1 pi. wa wa-1 wa-ibax 2 pi. ma-1 ma-1 ma-ibax 3 pi. bek bek-al bek-ibax Mip and mit are he and she, if distant, invisible, or of indefinite reference ; mep and met are used for proximity or visibility. The pronouns are evidently derived from a stem wa for the first person and ma for the second; diversified by -al and -ibax for objective and possessive. 47 The singular objective formed by change of a to i is interesting. It may be due to a lack of separate stems for singular and plural and an unconscious desire to express the distinction. Such a feeling seems to have led to the modification of the first person to ha in the singular, and the adoption of the objective mal to indicate the subjective of the second person plural. As evidenced by the examples already given, pronouns can appear with the locative case-suffixes of the noun. These are added to the objective forms of the pronouns. As in Yuki, terms of relationship furnish the only exception to the rule that the pronouns are not abbreviated or affixed. The possessive pronouns used with such words are, again as in Yuki, prefixed and apparently objective in form. With terms of relationship only : my, wi-, wi-ma- thy, mi- his, mip-i ha-, mip-i ha-mi- our, wa-i ma-, wa-i- your, ma-i mi-, ma-i ha-mi- their, bek-i ha-, bek-i ha-mi- 47 The forms given by Dr. Barrett in volume VI, page 64, of this series show that the Eastern dialect agrees with the others in having ma for the pronoun of the second person. For I and we the normal Porno forms are a and ya, in place of Eastern ha and wa. Southeastern, the Lower Clear Lake dialect, alone has wi for we. This dialect is also the only one that seems to share with the Eastern the possessive suffix -ibax; South- eastern wibax explains the origin of Eastern wax. The other dialects show a possessive suffix -ke: au-ke, ke, my; mi-ke, m-ke, thy; ya-ke, our. Instead of mi-p and mi-t the usual Porno forms for he and she seem to be mu and man. Compare the Northern Porno forms given below. 1911] Kroeber: Languages North of San Francisco. 327 For instance, wima-mex, my older-brother. The ordinary possessives are used only when possession is to be emphasized : two persons, each denying the relationship of the other to a third person, might say, wax mex ba e, my older-brother he-is. DEMONSTRATIVES. Demonstratives are numerous, frequently used, and important in Porno. Me is this, u that. Ba, that, is more common and more indefinite than u, being often best translated by the English article, though almost always somewhat more specific of refer- ence. It is used alone or added to u as u-ba. The apparently contradictory form me-ba has also been found, perhaps for me-baya, this there. An unexplained demonstrative ku-ba also occurs. Me, this, is the base of the pronouns of the third person singular, formed by the addition of -p for males and -t for females. For "it," me, this, or ba, that, is used, if expression is necessary. The original demonstrative form, and probably meaning, of the sex-differentiated forms, are better preserved in mep, met, indicating nearness; mip and mit appear to be modi- fications of these, with corresponding modification of meaning from demonstrative to pronominal significance, as is indicated not only by their expressing distance as opposed to the proximity of mep and met, but by their being less specifically endowed with deiktic reference of any kind. This development of quasi-pro- nominal forms from the demonstrative expressing proximity is interesting because unusual. In Yuki and Yokuts the indefinite demonstrative of distance fulfills the function of the pronoun of the third person ; just as Latin ille, not hie, grow into Romance the and he. The demonstratives not only take the numerous locative and instrumental case-suffixes, but, together with the personal pro- nouns, are the chief recipients of the objective suffix -al and the possessive -ibax. This restriction of use of these two syntactical case-suffixes differentiates them in character from the correspond- ing case-endings of for instance Yuki, Maidu, and Yokuts. The length of -ibax makes it look little like a genuine syntactical case. The objective -al may be related to the directive-termina- 328 University of California Publications in Am. Arch, and Ethn. [Vol. 9 tive -1 or -alal. The possessive has also a benef active meaning: wa-x is my or for me. me hee, this one u gai he, that land u-ba gai he, that land mep ba hikiba, he-is-the-one-who it did u-ba-ya, there u-mip, he, distant o-he-mip, he, more distant o-ya ba ihiba, there it was O-G< ai, that-one too me-G'ai, this-one also u-ba-c'a balai ba e, there blood that is me ba balai he e G'ida-o'ida-k, that blood is red dakir u-ba Gauk e, great that man is u-ba mi-kate hee, that one by you u-ba-ibax, his ku-ba sandia hee dabekmakme, those melons divide ye! The ordinary demonstratives precede the noun; but he and ek follow and are more or less enclitic. He, or hee, means the or this; its omission makes the noun translatable by its English equivalent with the indefinite article a. When an animate noun is subject of a transitive verb, he alone is not used with it. Either the personal pronoun must be added, forming a tautological he-mip; or ek, yek, is postposed. In the plural ek is replaced by the quasi-pronominal bek, to which it is no doubt related either in origin or by analogy. Bek is not confined to the subjective, and shows the form bek-al and bek-ibax. o-ba gai he, that land o-he-mip, he, distant kui hee, not that one! tee hee, which onef uba hee, that one mibax noaha hee da mi mara, your knife do you like? winhawa yek wi kadakhiba, my friend me cut masa'n ek G'ai hi'ntil ek G'ai wa guma kilmawakhiba, white-man and Indian we fought-together ha bekal kilwahiba, I them fought ba masan he mip wi kadakhiba, that white-man he me cut me xa hee dakir xa-mo'-e, this water very deep is ba' ha masa'n hee gutcia-i pako'hoba, that I white-man the knife- with stabbed mex bek wi marakiaika, my-elder-brothers they me like mex milbax ek wi marakiale, my-elder-brothers one-of-them me likes 1911] Eroeber: Languages North of San Francisco. 329 The enclitic he is no doubt related to hi, hi-bax, he-k-ibax, his own, their own. Perhaps the reflexive k'e-hei, self, is also related to he. Ba-bax, his, bek-ibax, their, contrast with hi-bax, he-k-ibax. mi-p hi-bax cawi di-ko-y-a, he struck his own boy mi-p ba-bax oawi di-ko-y-a, he struck his (ejus) boy mi-p mip-ibax cawi di-ko-y-a, he struck his (ejus) boy bek heg-ibax Gawi di-ko-yak-i, they struck their own boy Hi has sometimes relative force: ie'anxa gunula hemip hi hola kidi-1-ba xabe-na pidiakele, then Coyote he which sack he- carried rock-on broke. The element hi also occurs in combination with bek when relatives by marriage (except brothers-in-law or sisters-in-law) are addressed or spoken of. As already mentioned, such rela- tives address each other throughout with plural forms as a sign of respect. Bek would therefore take the place of ek; but the form usually found is hi-bek, or hi-bek-al. Use of singular ek or yek implies that the marriage has terminated in a separation. If however the wife has died, her husband continues to address her relatives with the respectful hi-bek. Such forms seem even to take the place of personal pronouns: bekal, them, meaning you, much like German Sie. wimaca yekal, my ex-parent-in-law wimaca hibekal, my parent-in-law, or parent of my dead wife bekal hane, parent-in-law, say! kanhimi mal butsigi hibekal, say, ye old-man them (say, father-in- law!) Besides hi, u is sometimes combined with bek : kanhimi ubek, say, my daughter-in-law! kanhimi wimaca hubek, say, my mother-in-law! The use of the plural in this connection does not however go beyond demonstrative and inflectional elements. Only singular noun stems and verb stems are used. Thus, mal da-o' ara hibekal, ye old- woman them, not maa'atcura, old women; gok-mi, "stehen Sie," a singular stem with plural ending, as compared with gok-im, "stehe du," and pilik-mi, "stehet ihr!" Enclitics or suffixes expressing indefiniteness are -o'a, -lal, -ula, -laye. Informants translate these by "kind" or "sort." 330 University of California Publications in Am. Arch, and Ethn. [Vol. 9 Kuyula, others, seems to be kui, not, and -ula, kind, -laye seems to be most specifically indefinite. haiu-laye wi G'anehiba, some dog me bit daiawal-ula, young women, young-woman-kind masa'n-ulu, masan o'omtsa-la, whites, plural of masan, white-man kaiu-la mipal G< aneheba, dog-kind them bit ha masan-ula-1 ganud-ed-elhiba, I a-white-man (or, to-the-whites) was telling something. Other indefinites are mil-bax, one of, selected from; kumu, all; o'omtsa, many; dolema, several, literally four-five, dol-lema. Xon-al is one another, objective. eau G'omtsa lia, people many died oau nudal-dal-a G'omtsa e, persons dead many are (lying about) wal gumu lia-y-eg'a, we all shall-die dolema pit 'ahaba G< araya, several there-were women wax haiu mil-bax ek wi G'ania, one of my dogs bit me buragal G'ai ha guta'haba bice' G'ai xotc, bice he mil-bax ha gali go'xhiba, bear I saw deer also two, deer of -them I one shot INTERROGATIVE S. Interrogatives, also used as indefinites and relatives, are am, ki, and tee. Am is who?, what?, something, which, someone, he who. Ki-a is who? Tee-he is which one? tce-a, somewhere, where ? ha am boolhiba, I something was hunting kia da ba, who is that? kia da mi kadake, who you cut? tee hee, which one? me hee, this one am ba da ma marakiala, what-for that you like? masan ki am yeheka, he has done something wonderful tee xabe hee da baten-e, which rock is the largest? ha am xadim milbax ha sididioi, I what biting of I swallow am cauk da ma gar-k-a, what-kind-of-a person do you see? ADJECTIVES. The adjective is more frequently found in duplicated form than any other part of speech. An attributive ending is -k, which is no doubt the same suffix as a -k making nouns of verbs. Adjective stems used as predicates seem to end in -k-i or -e, which is possibly a slurring of the verbal present tense-suffix -a, sometimes heard as -e. Many adjectives show neither ending. 1911] Kroebcr: Languages North of San Francisco. 331 e, ha haiu c'ida-c'ida-k gar-hiba, yes, I saw a red dog wax haiu hee c'ida-G'ida-ki, my dog is red haiu pit 'au, a white dog ha pit'auw-e, I am white Ga-hiem-k, one who watches a house (compare ga-hiem-xale, house- watcher, as a permanent occupation) xo batln ha garhiba, fire large I saw pit'au-k bawe e, it is the white one gili-gili-k, black ADVERBS. Adverbs and conjunctions, so far as expressed by separate words, call for little comment. Interrogation is expressed by the particle da, at the head of the sentence, or preceding the pro- nouns. The negative is similarly expressed by kui, and not by any affix of the verb. Le occurs in optative phrases translated by "let us." Conjunctions answering to English "and" and "or" are o'ai, postposed, and he, preposed. Both often occur in pairs, like English both . . . and, either ... or. There are no connectives or introductory particles corresponding to those which are so important in Yuki and Washo. This lack of intro- ductory words is so complete as to make it difficult to separate one sentence from the next in texts. da, interrogation kui, no, not &, yes le, optative min, thus (cf. me, this) isa, thus maco, like dakere, very (dakir, great, extreme, adj.) si, very, strongly, fast, hard ikwita, perhaps bitemayau, always namk ', always, constantly tibalc' amak, often co, now to, now, immediate future co-a-da-mal, to-day (da, sun) to-a-da-wal, to-day da-ka, yesterday duwe-m, to-morrow (duw-ina, at night) dok, long ago, sometime, ever yu, already yu-1, a while ago ya-yu, for a time yu-pa, again 332 University of California Publications in Am. Arch, and Ethn. [Vol. 9 NUMERALS. The numerals in all the Porno dialects have been given in Dr. Barrett's paper in this series, and analyzed in a comparative paper on the numeral systems of California. 48 The Porno numeral system seems to be quinary-vigesimal, and the Porno territory to be the center of an area in northern California over which this system has to some extent spread; but in the higher numbers, on the other hand, the Porno vigesimal method has in some dialects yielded before the influence of the decimal counting of neighboring people. The adverbial numeral is formed by -nai, "times." xotc-i-nai, twice lema-nai, five times VERBS. The Porno verb is completely non-pronominal, that is, non- incorporative. It seems that in most American languages using both prefixes and suffixes, the affixes to the verb, barring pronominal elements, tend to precede the stem when they are comparatively concrete or non-formal in significance, and to follow it as they are more purely grammatical or modal. Porno is no exception. Adverbial and instrumental ideas are embodied in prefixes, conceptions of tense, mode, and number in suffixes. Many affixes are readily distinguished as such ; but there are some elements whose nature, whether affix or stem, is uncertain. Di and be have the appearance of being affixes denoting singu- larity or plurality of the object of the verb ; but it is more likely that they are indefinite stems, meaning to move one and to move several, which are used in verbal complexes whose specific force is due more to their affixes than to these stems. A similar con- trasting pair are -p and -m, appearing to refer respectively to position, pointing, or intransitiveness, and to motion, putting, or transitiveness. 48 E. B. Dixon and A. L. Kroeber, Numeral Systems of the Languages of California, American Anthropologist, n.s., IX, 676, 686, 1907. 1911] Kroeber: Languages North of San Francisco. 333 ha cau-du-di-kil-hiba, I drove-(him)-in-repeatedly ha oau-du-be-kil-hiba, I drove-(them)-in-repeatedly wa oau-du-di-kima, we drove-(him)-in wa cau-du-be-kima, we drove- (them) -in mi ha cau-du-di-kil-hiba, thee I drove-in-repeatedly mal ha oau-du-be-kil-hiba, you I drove-in-repeatedly wi bihya-xam ba ca-di-m-im, me hand-in that shove-through! wi bihya-xam ba ca-be-m-im, me hands-in that shove-through! bo-1 ca-be-p-ki-mi, west point-them! ca-di-p-ki-m, move it, point it! ca-di-m-ki-m, move it, cause it to travel! Some affixes occur in contrasting pairs : kul-, gur ; di-, co- ; di-, ke- ; -1, -k ; -ma, -ki. Adverbial prefixes are the most loosely connected with the verb, and sometimes are heard as separate words. Thus oa-u-, or ga-u-, in, into, and oa-1 or ga-1, into, to or in the house, are evidently formed from oa, house, and the case-suffixes -u, in, and -1, to. The terminative or directive -1 appears also in xo-1-, out ; compare the adverb xo, out. Ku-1- and li-1-, away, off, probably show the same element. This is the more probable from the fact that, excepting these -1- forms and two prefixes gur- and mar- showing -r-, all ascertained prefixes, whether adverbial or instru- mental, end in vowels. Ga-u-, ga-u-, in, into Ga-1-, ga-1-, in house, to house, into xo-1-, out li-1, away ku-1, away ku-r, gu-r, toward mar-, down gayu-, up xamal-, back tsa-, away ca-, through ma-, across, opposite Examples : gur-uhu-m, come here! le xamal-pil-iba, let us go back! gal-uhu-i ha, I went home ca-t'ol-im, punch it through! lil-ani-a, threw it (away) kul-ani-m, throw it away! me-a gur-ma-hu-m, here come-across! o-ya lil-ma-hu-m, there go-across! 334 University of California Publications in Am. Arch, and Ethn. [Vol. 9 dadai xol-pili-mi, on-the-path go-ye-off! gur-ma-biki-m, face here! kul-ma-gadi-ba, let us run across to there mar-pile-li, they went down gayu-gap-ki, jumped up Gal-oho-le, come home me-a xol-pili-a, here they-came bek tsa-pili-a, they went-away The terms of direction co, east, bo, west, yo, south, kuhu-, north, may be included among prefixes, as they are at times fused with the verb-stem. ha co-hu-bae, I am going east (uhu, go) This is however not always the case. ha bo am boo-1-hiba, I in-the-west something was-hunting Gayu, up, is also both independent adverb and prefix. Instrumental prefixes are numerous. There seem to be several that have not yet been positively determined. They specify actions performed with parts of the body, with spatial parts of objects or objects of certain shape, and with objects of certain properties, such as softness. They are of course used almost entirely with stems denoting motion or dynamic actions. di-, with something round co-, with something long pi, with an edge moving broadside (ax) ga-, with an edge drawn lengthwise Gvotr looked. And gradually walked-toward and si-i'i ki-ma'c-nam-i'1-k ei ha n ca' a"-ii su'up-is ii and what-he-did, again slope-on throwing, 376 University of California Publications in Am. Arch, and Ethn. [Vol. 9 yii-k-i'-mil played. se-e i im-pis And somewhere-from ki-ma'c-k ii doing-that, ei se-e i an And always ki hul-k 'o'i ei that Coyote yii-k-i-nam-i'1-kon wbat-he-was-playing ki'w-is-mil i'iyi asked: "What k 'a'm-1-ite i'i-yi Wild-eat: "What ka cu'uh-uk here sit, ' ' kai n t a n p na n ' w-i'-wi long I ki hul-k 'o'i that Coyote yi'i-k-i-mil played. se-e i And ha n 'ye t 'a n -sa-tl-mil si-i'i now made-himself-be-found. And na'u n ' -mil looked. ha n 'ye now k 'am-l-itc Wild-eat hul-k'o'i Coyote se-ei And a n p I 11 i'iyi something suup-is-u kept-throwing." su'up-is-ha ii was-throwing?" ei a"-ii ii slope-on ta n l-k kup not-so, kii saw si-ei k 'a'm-1-itc smooth i'iyi "What a n p i 11 that ku'up-a sister's-son ya'u n -ha regards?" a n p mis I yon, k 'a'm-1-ite Wild-cat huu'u-t-mil se-e'i stopped. And cinki-mi kup yii-k-i-cilo'o-wi ? -you sister's son play-appeared-to?" a n p yu'u-y-am-ha tanha n 'l-k I was-doing ? Nothing-it-is ; i'mei-mil ta n l-k ta n lk kup said. "Not-it-is, not-it-is, sister's son, mi tat-e'itc neyu' ka'i you pretty ei i'mei-mil And Wild-cat said : i'mei-mil sa n -e'i po'-hot- ha '-nam-il-ki'i said. And oak-gall had-whieh su'up-ak-mil se-e'i hul-k'o'i ta n l-k threw (once). And Coyote: "Not-so, ka'i n t a n p na n w-i'-wi i'mi-ye-kit- I watched," said-when, cu'uh-sil-mil se-ei sat-down. And: sak a n p mi'hi Baby I am, kup kii yi'i-k-a sister's son, that play! long 61 sister's-son, k 'a'm-1-itc Wild-cat iit haik 'a'i-ha me think-foolish? sa n -e'i li's And: "Come, nyi "What yu' makes ku'up-a sister's son si-ka n ' Then kup na'u n -ni sister's-son, will watch." ii ha n ye' kii now that hul-k'o'i Coyote ei si-e i And mil-t'u' deer-heart na'u n -mil watched. ki-pa'u n -kot back a -ei se-e i slope-on. And suup-a'k-mil a' '-ei se-e'i threw (once) slope-on. And hul-k'o'a muk-la'w-e-t-mil sop-ei Coyote involuntarily-moved-to-snap-it. For-that k 'in-i'tc-tl-mil sa n -e'i ha n 'ye ha'a-t-il-mil almost-cried. And now took. ha n ye' kii now that suup-a'k-mil- threw (once) se-e'i ha n ca' And again a n -i'c-t-i ei rolling, k 'a'm-1-itc Wild-cat kai n t ii miit 1 Long I you 1911] Krocber: Languages North of San Francisco. 377 na n na'k-u si-e'i hul-k Vi knew." And Coyote: kup k'on a n p kup sister's-son; only I, sister's-son, sai-ki'1-u se-e'i lis kup almost-did " And: "Come, sister's-son, i'mei-mil hul-k 'o'i sak a n p said Coyote. "Baby I ya'u n -ha ii i'mei-mil se-e'i thinks?" said. And su'up-ak-mil se-e'i a n -e'-t-mil threw (once). And rolled. mu'k-tl-mil sa n -e'i ha n 'tc-tl-mil snapped-it. And split-it. k 'in-i-a'k-mil se-e'i ki-pa'u n ii cried. And together sa n -e'i ta n l-k ta n l-k " It-is-not-so, it-is-not-so, yi'-miwa'a-tl play-help kup ei sister's-son," ku'up-a sister's son and: sa n -e'i And tco' kup tco' " Here, sister's-son, here, tcan-e'-mil se-e'i gave. And se-ei ki-pa'u n -kil a' n -lam-i ei And back-to ward roll-beginning, ha n 'ye lik-i'-t-mil . si-e'i now swallowed. And k'i'n-t-ii-mil si-kit-i'i k'o'l-k'il cried. And then back-toward mis you lis hurry, mi'hi am, ha n ca' a' '-ei again slope-on se-e'i a' n -lam-i ei And, roll-beginning, sop-ei k 'a'm-1-itc Wild-cat tcan-e'-mil gave, i'mei-mil said. su'up-ak-mil threw (once). muk-tl-ki ei snapping-it ha'n-k'il house-to ko'o-t-mil For-that motc-sa ei pressing kup sister's son," a n 'ta n again ha n ye' now k 'a'm-1-itc Wild-cat hul-k Vi Coyote se-e'i k 'a'm-1-itc And Wild-cat nam-nam-li-ki'i ii lay-which ki-mac-i mil-hut 'o'op-in-nam-li-kit- se-e'i those deer-hunt-go-whilc. ha n 'l-t-mil sa n -e'i a n ta n ' heard. And again han-ku'ut-iy-it- ha'-mil se-e'i house-back-near listened. And se-e'i hamlatc-ki ya'-i-t-i ei And smoke-hole-at se-e'i k 'a'm-1-itc And Wild-cat p 'ii-mil-a'k-mil played (once). went ki'm-la'n ki-ma'c-at mil-p 'i'i-mo '1 his-own-brothers of-those deer-flute ha"-ma ei p 'ii-mi'1-mil kim-la'n taking, whistled his-own-brothers a n ta n ' hul-k 'o'-a And again Coyote wo'o '-ma-mil sa n -e'i walked-toward. And kan-u'uk pii-mi'1-mil talk- water whistled. ha n 'ye ta n 'sa-tl-mil now sbowed-himself. ei ha '-ma'-n-kil ei picking-up climbing ki'w-i-pi'i-mo '1 elder-flute 378 University of California Publications in Am. Arch, and Ethn. [Vol. 9 PARTIAL ANALYSIS. k 'am-l-ite, Wild eat, from k 'am-o '1, panther, and the diminutive suffix -ite. ii, particle used in myths to indicate that the narrative does not rest on the personal experience of the narrator. mil-t'u, deer-heart, composite of two independent stems as in English. yii-k-i-k-am, playing, from stem yi, play, appearing usually with an un- known suffix -k; -i-, euphonic, to separate the two -k- suffixes; -k, suffix indicating action, not a condition; -am, suffix expressing on verbs a more or less definite continuative. The word is the predicate of k'am-1-itc, wild-cat, but lacking a tense or mode suffix is participial and subordinate to nau"' -mil, saw. ii, ' ' dubitative " particle described above. im-pis, from somewhere, from im, interrogative where, here used in- definitely, and ablative suffix -pis. hul-k'oi, Coyote. The composition is not clear: hul is eye, k'oi is gopher, regarded as the characteristic food of the coyote. nau"' -mil, looked, from stem na n < w, to see, here used intransitively, and suffix mil, used in narrative in place of tense-suffixes to indicate the finite verb of the sentence, but lacking specific temporal indica- tion. sa D -ei, introductory particle, composed of sa", opposed to si, and indicat- ing that the subject of the verb introduced is the same as that of the finite verb in the last sentence, and the ' ' dubitative ' ' particle ii. ai"-it, gradually, apparently derived from verbal stem a n y, to glide, by a suffix -it or -t of unknown meaning, perhaps found also in kai n t, long ago. woo '-ma, walked toward, from verb stem wo ', to walk, go, and suffix -ma expressing motion toward. As frequently, there are no tense or mode suffixes following this suffix, so that the verb is to be regarded as participial, connected by na n , "and," with the participial form suup-is, throwing; and with it subordinate to the finite verb yii-k-i-mil, played. na", and, connecting the participial clauses containing woo 'ma and suup-is; usually only a connective of words. si-ii, introductory particle, composed of si, opposed to sa n , and indicating that the subject of the verb introduced is different from the subject of the preceding verb; and the particle ii. Sa n -ii and si-ii are ordinarily used only at the head of complete sentences, so that they can be translated by introductory "and"; here they head two parts of the same sentence and are therefore connected by the conjunction na n , and. ki-mac-nam-il-k, what he did, relative verbal form; derived by the final tense suffix -k, indicating an indefinite present and having verbi- fying force; and by the relative suffix nam-li or nam-il, of unknown origin; from a stem ki-mae, primarily demonstrative, from ki, that, and mac, probably meaning in such manner; the combined form having the meaning thus (?) and being used with the plural suffix -i as the regular plural, kimaci, those, of ki, that; but here em- ployed as a verb-stem, as in the form ki-mac-k below. ei = ii. haca, again. a'-ii, on the slope, from a', slope, and locative suffix -i, probably a form of -ki, at. 1911] Kroeber: Languages North of San Francisco. 379 suup-is, throwing, from sup, throw, and suffix -is indicating repeated action; without tense or mode suffix, and therefore participial, coordinated with woo '-ma by na n and subordinated to yii-k-i-mil. ii. yii-k-i-mil, played, from stem yi with its usual derivative suffix -k, euphonic -i-, and finite suffix -mil. se-ei, introductory particle indicative of change of subject, im-pia, somewhere-from, as above ki, that, ordinary demonstrative, equivalent to "the," "he, she or it," and (when contrasted with ka, this) "that"; here attributive to hul-k 'oi, Coyote, and equivalent to ' ' the " or " the before-mentioned. ' ' hul-k 'oi, Coyote, subject of nau n '-mil. nau D < -mil, looked, from stem na n ' w, as before. se-ei, introductory particle indicative of another change of subject. The subject of the last sentence having been Coyote, it is Wild-cat in the present sentence. There is no noun or pronoun or any repre- sentative of the subject in this sentence. an, always, constantly. ki-mac-k, doing that, verbal participle formed by the verbalizing present- suffix -k from the stem ki-mac, from the common demonstrative ki, that; perhaps literally "thus-ing. " Compare ki-mac-nam-il-k above. ii. yii-k-i-mil, played, as above. se-ei, indicates change of subject, a return to Coyote. ha"ye, now; possibly from one root with ha n ca, again, occurring above. ki, that, the, as above with hul-k 'oi. hul-k 'oi, Coyote. ei = ii. ha"ye, now, tautological. t 'a n -sa-tl-mil, showed himself, literally, made himself be found. Stem t'a", find; -sa-, evidently related to causative suffix -si-, a similar form appearing in motc-sa- below; -tl-, frequent transitive-intentional- causative suffix; -mil, suffix of finite verb. si-ii, indicates change of subject to Wild-cat again. k'am-1-itc, Wild-cat. yii-k-i-nam-il-kon, that which he was playing, objective relative parti- ciple. Yii-k-i as before, -nam-il, relative suffix of verbs, -kon, final relative suffix apparently in the objective case. ei = ii. huuu-t-mil, stopped, from stem huu'u, to stop, apparently related to hu', begin, by an unknown process of derivation; -t-, intransitive or involuntary suffix opposed to -tl-; -mil, suffix of finite verb. se-ei, indicating change of subject. hul-k 'oi, Coyote. kiw-is-mil, asked, from kiw, ask; -is, frequentative suffix, often of distinct iterative meaning, as in suup-is above, but here apparently merely habitually used with the stem kiw; -mil, suffix indicating finite verb. iiyi, what, interrogative. cinki, meaning unknown; iiyi-cinki-mi is a frequent interrogative phrase. mi, you, personal pronoun, second person singular, here somewhat enclitic to cinki, but as a matter of accent, not construction; grammatically an independent word, subject of the sentence. kup, sister's son; here an appelation, appositional to mi, you. 380 University of California Publications in Am. Arch, and Ethn. [Vol. 9 yii-k-i-eiloo-wi, appeared to play; from yii-k-i, from usual form yii-k of stem yi, as above; ciloo, to seem, resemble, be like, used as an inde- pendent verb stem and more frequently as a suffix with the force of an auxiliary verb, like lau% be able, and im, try; -wi, ordinary past tense-suffix used in direct discourse, its place being taken in narrative by -mil. se-ei, announcing another change of subject. ' k'am-1-itc, Wild-cat, subject of imeimil, said, at the end of the quotation beginning with the following word. iiyi, what, interrogative, object of the verb yuu-y-am-ha, did. a n p, I, independent pronoun occurring both in this form and as a n p-el; subject of yuu-y-am-ha. yuu-y-am-ha, was doing; from stem yu, to do; -y- phonetic glide frequent before the suffixes -am and -ak after unaspirated vowels; -am, suffix usually having a continuative or iterative force, as here; -ha final interrogative suffix, displacing the tense-suffix. tanha n l-k, it is nothing or it is not so, verbal derivative by the present- tense suffix -k from an enlargement by unknown derivation from the stem tan or ta n l, no, not. Ta n l-k occurs with the same meaning in the next sentence. a n p, I, subject of following cuuh-uk, sit. ka, here, strictly "this," demonstrative pronoun of proximity. Ka may in this sentence form a compound with the following cuuhuk, ka-cuuhuk, I sit here; one should expect the usual locative form kat-a, at this, here. euuh-uk, am sitting, from stem cu<, and present suffix -ik, the u of the suffix being obscure. ii, dubitative particle, here indicating the cessation of the direct dis- course in which it is not used, and the resumption of the narrative. imeimil, said, predicate of k'am-1-itc above, the entire intervening quota- tion with three finite verbs being logically the object. Formed by the suffix -mil indicating the finite verb in narrative, from imei, probably composed of stem im- and euphonic -i- to separate final m of the stem and initial m of the suffix. ta n l-k, it is not so, from ta n l, no, not, and present or verbilizing suffix -k. ta n l-k, it is not so. kup, sister's son, appelation. kai n t, for a long time. a n p, I, subject of the following verb. na nt w-i-wi, looked, was looking, as above. Probably intransitive here, as the following clause, which logically is its object, is not sub- ordinated to it but also contains a finite verb. iiyi, something, interrogative what? here used indefinitely, object of mi- suup-is-u, you threw; possibly equivalent to a relative connecting na"w-i-wi and suup-is-u, though such a construction would not be characteristic. mi, you, subject of suup-is-u. tat-eitc, pretty, from tat, good, and diminutive suffix -ite; also tat-am; attributive to iiyi, something. neyu, smooth, in the same construction as tat-eitc. ka'i, I suup-is-u, kept throwing, predicate of mi, from sup, throw, suffix -is expressive of iteration or continuation, and final past-tense suffix -wi, as in na n 'w-i-wi, here as often contracted to -u. si-ei, particle introducing a new sentence which will have a different subject from the last. 1911] Kroeber: Languages North of San Francisco. 381 GENERAL CHARACTER OF THE LANGUAGE." The most important characteristics of the Yuki language may be said to be the following : phonetic simplicity, both as regards the limited number of sounds and the absence of accumula- tion of either consonants or vowels ; phonetic rigidity or immuta- bility, evident particularly in the lack of modification of stem or affix elements as these are brought together; a strongly mono- syllabic character of the elements of the language, which does not, however, prevent the formation of words of some length; the absence of reduplication; the use of suffixation as the only structural or grammatical means employed; the presence of a moderate number of suffixes but the rigid restriction of these, apart from a few noun-forming endings, to designations of number in nouns, of case and locative relations in nouns, of auxiliary, derivative, and general adverbial ideas in the verb, and of the relations of mode and tense in the verb ; the absence of affixes, particles, or classifiers denoting shape, kind of motion, substance, or instrument; complete absence of any form of incorporation or pronominal affixation; a clear distinction of all words into either nouns, verbs, or adverbs, pronouns and demonstratives being nouns, adjectives verbs; a well developed subordinating sentence structure; the expression of delicate shades of relation between sentences by a series of connectives; and the lack of an absolute differentiation of stem and suffix, as shown by the use of certain stems also in the latter capacity. WAPPO. A few phonetic and grammatical notes obtained on the Wappo dialect reveal certain identities with Yuki proper and a number of interesting divergences. Dr. S. A. Barrett has shown that the Wappo language is spoken in five dialects, or more properly sub-dialects, as the differences are scarcely per- ceptible. The notes here given are on the Northwestern or Russian river sub-dialect. 5* Compare Boas Memorial Volume, 64-79, 1906. 382 University of California Publications in Am. Arch, and Ethn. [Vol. 9 Phonetically Wappo is very similar to Yuki, both the con- stituent sounds and their characteristic combinations being almost identical. The principal differences are the absence of nasalized vowels from Wappo and the presence of a series of sonant stops corresponding to the surds. These have not been recognized by Dr. Barrett. As they had been written in a Wappo vocabulary obtained by the author previous to Dr. Barrett's studies, par- ticular attention was therefore subsequently given to the ques- tion of their occurrence, at the time when the information was secured on which the present notes are based. It is however true that the sonants are less different from the surds than is the case in English. As in Yuki, both ordinary and fortis surd stops occur, k', t', p', t-' tc', and ts' having been heard. Glottal stops are marked and frequent. E and o are open, but i and u seem to be close. The pronouns differ from those of Yuki chiefly in that the possessive, objective, and independent or unsyntactical forms are identical, and somewhat different from the subjective forms. The possessive pronouns are prefixed or preposed. A possessive and objective of the third person singular is de. The pronouns so far as obtained are : Subj. Obj. Poss. Indep. 1 s. a I i I 2 s. mi mi ml mi 3 s. de de 1 p. isi isa isa 2 p. imsa misa The demonstratives are he' and we' and thus unrelated to Yuki ka and ki. Another apparent demonstrative is tse. Self is rendered by mai. Nouns show a plural, of persons only, in -te. The locative and instrumental suffixes determined are : -ti, -i, instrumental -mi, -me, at, in, referring to time -tu, -ta, at, in, referring to time -bi, from, of -deimu, on -ompi, under The tense suffixes for the present, past, and future are -wi, -ta, 1911J Kroeber: Languages North of San Francisco. 383 and -si. Of these the present suffix has a past meaning in Yuki ; the future in Yuki is -pa, not -si ; the Wappo past, -ta, is not known from Yuki. The imperative is formed by a suffix -e, either with or without a prefixed or preposed ina. The optative of the first person plural is indicated by the suffix -si, which is probably the future tense sign, with or without the imperative prefix ma. Interrogation is expressed by the final suffix -ha, as in Yuki. The negative is -laxki, la'k. The use of this form is interesting, as it appears to be the Yuki potential suffix or auxiliary verb lau', lauh. A suffix -la, with or without ho pre- posed to the verb, may be a continuative. Other verbal forms of undetermined significance are -uk and a preposed or prefixed o-. The dubitative or quotative particle ii of Yuki, and the intro- ductory particles or connective words that are so conspicuous in that language, have not been observed. There is no trace of them in the recorded text of a myth. The general appearance of the verb, place of the pronouns before it, and the relative order of words in the sentence, closely parallel Yuki. Specimen phrases : k 'u-ye, k 'uw-e, run ! a ho-k 'uw-ala, I am running, I ran mi I nau-e-ta-ha, did you see me? a' mi nau-ta-la'k, I did not see you. a de hakce-laxk, I do not like him. I okani I hakce, my friend likes me de a k'ewi naw-i-si, I shall see him to-morrow bata a mi nau-wi, I see you now. isi ma-k 'uw-e-si, let us run a ma k 'uw-e-hakce-laxki, I do not want to run ma-ba-e, eat! isi-ba-e-si, let us eat. u a' o-ba-ta, already I ate o-ba-uk, eat he' utci a' o-ba-e-si, this night I will eat i okani o-ba-ta-laxki sumi, my friend did not wish to eat yesterday a lel-i ml o-tcap-i-si, I will strike you with a rock 384 University of California Publications in Am. Arch, and Ethn. [Vol. 9 WIYOT. The Wiyot occupied the Coast from the Bear River mountains north as far as to Little river. Inland they held only to the first watershed. Humboldt bay and the lowest course of Eel river were the most important points in their territory, which was one of the most restricted held by any linguistic family in America. The Wiyot call their language Sulatelak. Those about Humboldt bay call themselves Wiki-daredaliL, from Wiki, the name of the Humboldt bay district. The Mad river portion of their territory they call Patawat, and the country about lower Eel river Wiyat or Wiyot. Viard, a name that has been applied to them, is a rendering of Wiyot. The Yurok call them Weyet and the Karok use a similar term. They have been erroneously called Wishosk, Wic'ack having been interpreted as the designation given them by their Athabascan neighbors. It is the term which they apply to these Athabascan neighbors, wici meaning interior. The Wiyot language, which does not appear to be dialectically differentiated, is rather difficult phonetically and grammatically. Material was recorded from six or eight individuals, none of whom proved a satisfactory linguistic informant. Several texts that were secured throw little light on the structure of the language because they are loosely translated, because the in- formants were unable to render adequate assistance in the analysis of the expressions contained in them, and because the sentences in the texts consist chiefly of independent verbs, so that they present little context. PHONETICS. Wiyot is spoken indistinctly and lacks the phonetic clarity characteristic of the three languages previously considered in this paper. The vowels have but little saturation of quality, and are frequently slurred so as to be obscure, a feature which has been indicated by small capital letters. E and o are open. Nasalized vowels and a, o, ii do not occur. 1911] Kroeber: Languages North of San Francisco. 3S.1 The consonants are : k k' g g ' t t' d s n P p' b m 1, L, r, c, to, is, y, w, h,' There are no velars. Surd and sonant stops are easily con- fused. There are no fricatives outside of the s class, except g', which appears to be a final and medial modification of g. A nasal in the k position does not occur. S and c (sh), as well as ts and tc, do not seem to have sonant equivalents. Surd or lateral L, often affricative, is perhaps at other times only a con- tinuant. The r is rather soft, though less dorso-palatal than Yurok r. It is less trilled than Porno or Karok r. The fortes surds are not very strongly exploded. Glottal stops, ', are fre- quent. Initial w is frequently heard as m, v, or b. It is there- fore probable either that w is habitually spoken with much less rounding of the lips than in English, or that besides w there exists a sound kindred to bilabial v. Owing to uncertainty on this point, only w has been written, though v was often recorded. Yurok w approaches v in quality, and Karok possesses v but no w. Wiyot is of the small number of California linguistic families that permit combinations of consonants both initially and finally and in stems. The variety of such initial and final combinations is however quite restricted, nor are medial combinations con- spicuously frequent. Initial collocations are : kw, which is pos- sibly a development of a simple sound; cw and sw; pi and perhaps bl; tk, tck, and tsk. LW, tew, tw, kc, and kL have also been recorded, but so rarely that their occurrence cannot be looked upon as positively determined. Final combinations are kw, tk, and tck. Besides, tw, ks, and sk have been heard once each ; pc on two names of places in Yurok territory ; and pL in several instances. The rarer uncorroborated occurrences, both initially and finally, must be considered doubtful on account of the habit of many speakers frequently to slur unaccented vowels. It is however clear that kw, tk, and tck are found both at the beginning and end of words; and that k, t, p, tc, c, s, and L, in other words surds, comprise the consonants that occur in first position, and that k, c, s, w, 1, and L occur in second place in 386 University of California Publications in Am. Arch, and Ethn. [Vol. 9 combinations. In the interior of words, where composition and derivation bring other consonants into juxtaposition, there is less restriction on combinations, but the component elements are frequently heard separately. All sounds in the language occur initially and finally, with the following exceptions: Vowels and n are not found at the beginning of words, and the sonant stops g, d, b, besides of course h, y, w, not at the end. It should be added that final vowels are not common. There are a few instances of initial vowels, all in words which appear to be of other than Wiyot origin : ikti 'n, also recorded as hiktl'n and kti'n, Klamath river; eckapc, Gold Bluff, Yurok ecpeu, also in Yurok territory ; itesi, small shells, perhaps obtained by the Wiyot in trade; iewetck, silver. That n does not occur initially is connected with a certain relation betwen it and r and 1. The absence of the sonant stops d and b from final position is possibly due to imperfect apperception rather than to actual non-occurrence. G however seems to become continuant, g' , when final. This may be due to influence of Yurok, in which g is always a continuant. E, 1, and n are closely related. Each has been recorded in place of the others. haluwi, haruwi, boat -helel, -heler, on numerals kac-werar, small house, rat-welar, large house gu-dalew-iL, gu-danow-en, stand won-e'l, wor-e'r, his arm meledal, hi-meredal, walk ritwe-lakwil, ritwe-wacuk-rakwil, crescent-shaped laliL, rariL, stream ' What seems to be original n changes to r or 1 when initial, reappearing after a prefix ending in a vowel. The same process - seems sometimes to occur finally, but may be a conversion of r to n before an added vowel. rawili, right, ka-nawili, not right, left riewom-ot, kwi-niewom, kill -tiar, -tian-ik, suffix of possessive pronouns yi-dar, my father, dan-e'l, his father k-elir, your eye, w-elin-e'l, his eye It is rather striking that vowels are rarely final and never initial. 1911] Kroeber: Languages North of San Francisco. 387 Phonetically Wiyot agrees with Yurok and contrasts con- spicuously with Yuki, Porno, Karok, Chimariko, Yana, Maidu, Wintun, Miwok, Costanoan, Esselen, Washo, and Yokuts, whose enunciation is distinct and whose stems are free from combina- tions of consonants. Nevertheless the indistinctness of speech, the slurring of vowels, and the accumulation of consonants, are not excessive in Wiyot, and are more moderate than in certain more northern languages of the Pacific Coast and than in English. REDUPLICATION. Reduplication is not abundant. It occurs in certain onomato- poetic verbs, at times in iterative verbs and in adjectival stems, and is occasionally used to indicate rhetorical emphasis. It is thus word-forming rather than grammatical. da '-da 'kwa, snore tsi-tsir, sneeze da-dakak, thunder cak-cakw-iL, he is sick, eakw-irak, sickness dak-dakw-iL, it is crossed daru-dalu-i, all gabitcirakw tci-wera-wera-wera-kw, it is too bad! It is apparent that in onomatopoetic verbs the initial syllable is reduplicated as far as the vowel; in other cases an entire syllable or stem is duplicated. There appear to be a few nouns that are normally reduplicated, such as tcatcitckiri, mud-hen, but there is no trace of reduplication expressing a plural or col- lective-distributive. Iterative reduplication in verbs is uncommon. COMPOSITION. In composition of two nouns, the determining precedes the determined; but any other part of speech determining a noun follows it in composition. p'let-kacamale, rocks-small, Little River p 'leta-caweti, rocks-white, Glendale wits-karerer, canine-wild, coyote wopL-akatkera, redwood-branches dikwa-motwiL, white-man-woman ritsowel-ailokwe, night-moon wene-welir, sky-eyes, stars gatsire-weliL, crow-foot 388 University of California Publications in Am. Arch, and Ethn. [Vol. 9 If however a term of direction is united with a noun, it pre- cedes. It seems that such terms are prefixes, and that the process of combination is one of derivation rather than of com- position when these elements are added to nouns. wici-dikwa, inland-spirit gudatri-gakwiL, above-old-man In a number of compounds only part of the elements can be positively determined. haluwi-tulaliyutxu, Medilding village at Hupa,55 (haluwi, boat) da-Lak-dale-waiyits, come-in-ship dog, native dog (Lak, ship; waiyits, dog) cawet-oc-iL, bald-eagle (cawet, white; -iL, suffix of third person) cawet-colig-iL, brant dela-bel-iL, killer-whale (bel, to fish) mes-wululel, fire-place (mes, fire) swaptil-haluwi-laliL, Vance Mill (haluwi, boat; laliL, stream) ta-boderuc-datige-raliL, Lindsey Creek (boderuc, brodiaea roots) we-tapiis, thumb (we's, hand, fingers) guts-er-ol, fresh water (guts, good, ho'l, water) gotso-wen, day (go 'ts-, one, or guts, good, wen, sky) weni-crenim-iL, mole (wen, sky, which according to myth the mole supported) Formations similar to bahuvrihis or possessive compounds occur. baLe-ranaLa, Chinaman (baL, hair; raLa, long) twanagit-erowak, sheep (blanket-make) DERIVATION. Word-forming derivatives that have been determined are suffixes, except terms of direction, which are prefixed. -ate, -hats, -wdts, diminutive : bac-ats, small flat basket hutcwatc-hate, small cooking basket haluwi-wats, small boat p 'let-wits-hats, p'lets-wats, small rocks rariL-wats, small stream wetc-atc, buds wlt-c-atc, small alder-trees watcewaiawatc-hats, small clam shells used as ornaments ss Medilding means ' ' boat-place ' ' in Hupa, as it is accessible only by boats. 1911] Kroeber: Languages North of San Francisco. 389 -tar, on terms of color : mes-iar-etk, red (mes, fire) siswa-ial-ewe-Lak, brown (siswa, black) -Ldk, on terms of color : dukap-Lak, yellow kika-i.ak, red mes-iera-Lak-er, red siswe-Lak-an, black dukap-Layak-an, green baduduwi-Lak, dust-color Limaiusele re-Lak-er, blue-jay color-it-is, blue -gaLet, on adjectives of appearance : coyuwo-geLet, striped tcwetc-gaLet, plaid detcatc-gaLet, spotted -lak, language : sulate-lak, Wiyot language wicl-lak, Athabascan language denakwate-lak, Yurok language guradaliLrakwe-lak, Karok language -welel, -helel, on numerals above four : we 's-ag-eleL, 5 (we's, hand) dakLiluk-elel, 6 halu-welel, 7 hiowita-welel, 8 meceruk-welel, 9 rulok-helel, 10 ritawa-helel, 20 rikwa-helel, 30 rama-helel, 40 we 'sohele-welel, 50 dakLilukhele-welel, 60 haluhele-welel, 70 hlwitahele-welel, 80 mecerukhele-welel, 90 gutseswani-helel, 100 From 50 to 90 this suffix appears twice on each stem. There are several noun-endings : -wil. gu-wil, person (cf. guwi, man) di-wel-a, di-wil-e, somebody, another ir, probably for -in. wel-ir, eye (wel, see) 390 University of California Publications in Am. Arch, and Ethn. [Vol. 9 -&. maL-ak, salmon, food (maL, waL, pL, eat) -S. we '-s, hand (cf. we-tapiis) p'let-s, p'let-k, rock -t. cwa-t, bow (swala, shoot) -VL, the ending of the third person on verbs, forms numerous nouns. The stems from which these are derived are in many cases undetermined, but seem to be verbal. tiger-iL, unmarried man tserar-iL, unmarried woman kakeraw-iL, woman gakw-iL, old man coor-eL, index finger ("pointer") radapir-iL, glutton rakwulir-iii, wolf kanapel-iL, grizzly-bear ("biter") cawet-oe-iL, bald-eagle dakaks-iL, gun dawiLar-iL, glass dawiL-wiw-iL, mirror kagotsikc-iL, lamp As the last examples as well as several previous ones can show, Wiyot is not at all averse to the formation of new words to denote new objects. The majority of the languages of Cali- fornia tend to adopt Spanish or English words. Several endings have been found on verbs of related mean- ing, but it is uncertain whether they are deriving suffixes added to stems or themselves verbs. -tskarer. rari-tskarer, shave, plane bitcewe-tskarer, peel Leriwoke-tskarer, peel ci-tskarer, flay -uiyer. twe-Layer, cut, notch ka-Layer, whittle hawe-Layer, mash gutwaiap-Layer, brush Kroeber: Languages North of San Francisco. 391 Terms of direction precede other word-elements, as before stated. To the examples already given can be added : dat-kasiL, top of head curi-laka, west, ocean Stems of terms of direction are : wur, north at, teatc, south cur, west, across the ocean tinie, wici, east, interior, upland dat, up, above When forming independent words, these are employed with either of the prefixes rak- and ivik-. rak-wur, north wik-tcatc, wik-at, south wik-cur, west rak-tinie, wik-tinie, rak-dat, east PRONOUN. The Wiyot pronoun is incorporated, to use the customary terminology. In other words, it is not a pronoun at all but a pronominal element which is normally affixed to other stems. The independent pronoun occurs only unsyntactically, as in answer to questions, or emphatically, when it is used in addition to the pronominal affixes and is syntactically superfluous. The possessive elements are chiefly prefixed, and show some similarity to the independent pronouns. The subjective and objective elements are always suffixed, and differ entirely both from the possessive and independent forms and from each other. The objective element precedes the subjective, thus standing nearer the stem of the verb. Subjective Objective Possessive Independent 1. 11 ATT , -u, -ow rn- VI -56 ^ j J yil 2. -it" -as ku- kil 3. -iL -a (hu-) e'l 1. -itak-" , -u, -ow (hu-) ik hinar, winar 2. -itawa" -aswa, -wa kiluwa ku- kiluwa 3. -iL -a (hu-) e'l r,6 yi- only before terms of relationship. " -at, -atak, -atawa are also found. 392 University of California Publications in Am. Arch, and Ethn. [Vol. 9 The independent pronouns of the first and second person, yil and kil, have evidently been influenced by mutual analogy. Comparison with the possessive prefixes ru- and ku- makes it appear that 1 or r, representing n, is the original element of yil, and k of kil. In this case the pronominal stems would agree with those of Yurok, where nek and qel are I and you but the possessives ne- and qe- shown n and q to be fundamental. A form yil-il, me, has been found. There is no independent pronoun of the third person, and the possessive is indicated by the suffix -e'l. The first person subjective and objective is often indicated by absence of suffix. 58 An -u or -ow also occurs. The second person is -it or -at, subjective, and -as objective. The third person is respectively -iL and -a. The plural is variously formed. In the third person it does not differ from the singular. The second person uniformly adds a suffix -wa; thus, kil-u-wa, independent; kil-u-wa ku-, posses- sive; -it-a-wa or -at-a-wa, subjective; -as-wa, sometimes merely -wa, objective. The independent form of the first person is hinar or winar. The possessive is a suffix -ik, which reappears in the subjective -itak or -atak, of which the first element resembles the second person -it-. The plural of the first person objective is the same as the singular. Neither the subjective, objective, nor possessive series possess any common elements which may be interpreted as indicative of these relations. It is probably analogizing that has led subjective -it and -iL, objective -as and -a, and possessive ru- and ku- to share respectively the vowels i, a, and u. kilwa-ya wul-al-itwa, were you (pi.) walking? gul-ag' -it-ya, are you going back? hi-gelaw-a, I beat him cu waL-itak, let us eat! do-pL-iL, he is eating hi-wel-a, I saw him hi-wel-aswa, I saw you (pi.) hinar hi-wel-u-L, he saw us hi-wel-as-iL, he saw you Le hi-wel-uw-ituwa, you (pi.) have seen me 58 In Mohave the first person is also denoted by absence of pronominal affixes. 1911] Kroeber: Languages North of San Francisco. 393 bokin-ow-it, you hit me wu-bokin-as, I will hit you wi-kanap-is, I will bite you wi-t-as, I will feed you garewack-iL, he is bad winar wa-kale-waL-u, we will not eat LC hi-kanap-a-it, did you bite it? The suffix -a-, him, plus -it, you, seems to be usually con- tracted to -et : hi-wel-et, you saw him. A suffix -il is sometimes found before subjective suffixes of the first and second person. It occurs also between the objective of the first person -ow and the subjective -iL of the third person. raL-el-at, you are large hi-wel-ow-il-iL, he saw me makL-erakw-il-atawa, you (pi.) are large The first person possessive shows an exceptional yi- instead of regular ru- before certain terms of relationship. This yi- appears to be an abbreviation of the independent pronoun yil. It will be recalled that Porno and Yuki have been found to use a separate class of possessives with terms of relationship. In the case of body-parts, there are also certain peculiar- ities. Many such words begin with an m or w, which seems to be a pronominal element denoting indefiniteness of possessive reference. In some of these terms the first and second persons are expressed by the addition of the prefixes ru- and ku- before the m or w. Such are weliL, foot, we's, hand, wee, vagina, magoks, brain. In other terms initial m- disappears before r- and k- of the first and second persons: m-a'n, pubic hair, r-a'n; m-elak, testicle, k-elak; m-elir, eyes, k-elir, w-elin-e'l. Other body-part terms show an initial element wat- : wat-hel, tail, wat-hewet, head, wat-wi, heart, ra-wat-kai, skin, wat-kerat, bone, wat-melok, ear. This element is also retained with the personal prefixes: ra- wat-kerat, my bones, hu-wat-kerat-i '1, his bones. Still other words denoting parts of the body show neither initial m- nor wat- : sakwer, lungs, dgat, penis, baaL, hair. A few words show unexpected forms : m-etere, nose in general, detere, my nose, kil detere, your nose, detere- '1, his nose; kawik, wat-kawik, blood, r-atkawik, k-atkawik, my, your blood; haluwi, boat, ru-d-aluwi, ku-d-aluwi, my, your boat. 394 University of California Publications in Am. Arch, and Ethn. [Vol. 9 DEMONSTRATIVES. Demonstratives are gic, this, and guru, gur, or gu, that. As in Yuki, the term for that has also some use as a demonstrative of reference, without idea of distance. Gic is used only to express specific proximity. When attributive the demonstratives are proclitic to the noun. A third form, gwilel, occurs with the meaning of he. gun, he, she gic, this one gu-tem-iL, the one sitting there gu-dalew-iL, that one standing there guru waiyits, that dog guru gudatri-gakwiL, that above-old-man gic-garewackiL, this bad one gwilel hu-waiyits-e '1, her dog gwilel kanap-el-iL, he was biting Tciwa means thus, so, that, he who. A related form tcigon or tcigur, him, them, that, appears to be objective. There always appears to be implication of previous reference. tciwi anel-iL, that is what he said tciwi dalow-iL, she lives there kil-ia tciwira ha-tsitsir-iL, you-was-it who sneezed? tciwa daretw, I think so yil-il tciwa-hakil, to me he did it tciwa-wiL delaker, always did thus kiluwa-ya dicgaam-et tcigon, do-you (pi.) like him! tcigur hi-wel-a, them I-saw Interrogatives are ciwa, duwa or dawa, and kwaLwa or WULC, meaning what and where, how and why. The ending -wa of these interrogatives occurs also in the demonstrative tciwa; the stems are therefore probably ci, da, and kwaL or WUL. ciwa, what? duwa, what? duwa kil ka-wol, where is your house? duwa wulal-iL motwiL, where went the woman? duwa dekLelaliL, where is the chief? ciwa kac-welan ka-wol, how small is your house? kwaLwa riewom-ot-ogam, with what did you kill him? kwaLwa wuLe miL kale-waL-e, why do you not eat? "Another" is hikeLe'l, or diwile, somebody. HikeL-e'l per- haps bears the possessive suffix of the third person. All is daru, preposed to the verb. More frequently however ru- enters as a prefix into the verb-complex. 1911] Krocber: Languages North of San Francisco. 395 NOUN. The Wiyot noun is scarcely affected by grammatical con- siderations. It is not reduplicated, and is free from the expres- sion of number, gender, or case, excepting only one general locative suffix -akw. p'let-wakw, on the rock pak-akw, on the ocean mes-akw, in the fire halowi-wakw, in a boat datheri plet-wakw, on top of the rock peL-wakw, place in a house beside the* door wal-akw, in the morning wiril-akw, to-day VERB. Pronominal elements are always suffixed to the verb. On the other hand temporal, modal, and adverbial relations are expressed by prefixes. Similarly the independent adverb precedes the stem. Adverbial prefixes generally follow temporal ones. The scheme of the verb is thus: prefix of tense, prefix of manner, stem, objective pronominal element, subjective pronominal element. Prefixes express several past tenses, a future, and a con- tinuative; a conditional, a subordinate mode, and one form of imperative; the negative; the idea of all; and probably several designations of motion. Suffixes, which predominate in word- formation and in indications of person, are less important in verb structure. Those determined express a causative, an impera- tive, an instrumental, and a reflexive. Several other suffixes occur on intransitive and adjective stems, and seem to serve to render these respectively predicative or attributive. There appear to be several prefixes of motion that have not yet been fully determined; but in general there is little specific expres- sion, by means of affixes to verbs, of kind of action, location, or instrumentality. PARTICLES. gitga is an adverbial particle indicative of futurity and prob- ably of intention. It usually follows the verb. wa-keL-am gitga, I will look for him kanap gitga, I will bite rogal-ia gitga gul-ow-at, soon will you come back! 396 University of California Publications in Am. Arch, and Ethn. [Vol. 9 The interrogative is an enclitic particle, ia or ya, always attached to the first word of the sentence. 59 kil-ia dicgam-at guru guwi, you, do you like that man? gur-ia dekLelaliL, is he a chief? kuna-ya hi-les-at, yesterday did you go by boat? dicgaw-it-ia, do you like me? yil-ya bokin-ew-It, me, did you hit me? kiluwa-ya dicgam-et tcigon, ye do you like him? Le 1-it-waL-et-hia, are you sleepy? co or cu is an optative particle, always at the head of the sentence. co gaw-ak-o, let us start cu wala, I wish I had it cu wirate, let me drink cu gawitwadak, let us go to sleep cu, or cuku-, has negative optative force without the employ- ment of the usual negative prefix. giLa, cuku-laliswu, enough, let us stop singing! cu pugakwiLini, do not touch it! cu ratse-tsaw-inik, kLet, do not touch it, it is hot! cuku-kawi giLa, stop working! cuku-rerir, you must not do it any more! Le, sometimes heard as La, expresses the completed past. It always precedes the verb, and is sometimes heard as a separate syllable, sometimes as a prefix. It is therefore probably a pro- clitic particle. La-wit, I have slept ciwa La-gira-gerak, what have you done? winar Le-ru-ge-da-pL-o, we have all finished eating Le hi-kanap-et, did you bite it? La-gera-le-wel-as, I did not see you Le-rewaLa, it is day Le-kilowa Le-waL-e, have ye eaten? 4 PEEFIXES. hi- is the prefix of ordinary past time. hi-ow-iL, he came hi-wel-as, I saw you hi-esatal, I met hi-rakcem-iL, she pursued him kuna hi-les, yesterday I went by boat 69 In Miwok and Ute-Chemehuevi the interrogative is -a and -ia, and is also the second word in the sentence. Yuki has a verbal suffix -ha. 1911] Kroeber: Languages North of San Francisco. 397 na- seems also to denote past time. na-do-pL-et, you have been eating na-yu-wel-as, I saw you na-Le-weratc, I have drunk already wi-, wa-, indicates the future, wi-kanap-is, I shall bite you wa-detigeliswiw-iL, he -will swim wi-gera-dilegana, 1 shall not become angry wi-letkalegal-iL, he will fall kil-ia wo-bel-at, shall you fish? do- is a continuative. do-bel-iL, he fished do-pL-a, I am eating do-low-iL, they are hanging daru do-pL-iL, they are all eating to- or da- occurs often. Its significance is uncertain. In many cases it seems to be frequentative, iterative, or usitative. da-laLal-iL, he jumps about da-lakwet yil, I was coughing da-digwidiwiri, I am sitting hinar da-ridipu, we live together ta-hokawoweluL, whenever a whale stranded da-kul-ow-iL, always returned ta-weldelaker, always whipped him da-kictawil-iL, constantly asked for food da-wuwokwiwi, have you been training yourself! kul- means back, again, return. It is placed between the tense- prefix and the verb stem. hi-kul-ow-iL, he came back kul-ag-iL, went again La-ka-kul-aker, did not do any more dal- is of undetermined meaning. It occurs in several verbs implying repeated motion. dal-an-iL, dal-anew-iL, buzz, hum dal-adeler, ring da-dal-ak-wer, works hil-, perhaps 1-, is undetermined. hil-ag-iL, went hil-uluwu, takes gawel-, undetermined. gawel-alak, I will move away gawel-ag-iL, they went hi-gawel-uw-iL, they came 398 University of California Publications in Am, Arch, and Ethn. [Vol. 9 let- apparently defines motion in some way. let-kaleg< al-iL, roll dak-let-athanagat, boil violently let-kalegal, fall ru-, occurring also in the independent word daru, all, has the same significance when a prefix. It usually refers to the subject, but also to the object. As a prefix it follows the tense prefixes. winar Le-ru-ge-da-pL-6, we are all done eating hinar ru-raL-el, we are all large ru-rat-er gowil, all the people are large ru-ga-pL-u, will eat wi-ru-bokina-wa, I will eat you all gawe- is inchoative. gawe-rowetger-iL, it is becoming dry gawu-betser, it is becoming dry gawe-rewaLar, it begins to be day gera-, g'ra-, ga-, or ka- makes the negative. It also follows the tense-prefixes. gera-lit-waLa, I am not sleepy g 'ra-la-waL-i, I do not want to eat gwa-gets-Lak, it is not cold g 'ra-dicga-gem yulewe-tsk, I do not like the white ones wi-gera-dilegana, will not be angry La-gera-le-wel-as, I did not see you gera-lu-de-dicga-ge, I do not like him ka-dicgaw-ir-ia miL yil, don't you like me? wi-gera-t-as-a, I cannot feed you ka-, prefix, combined with -ii,ya, suffix, forms an occasional imperative. ka-wa-detigeliswiw-iLya, swim! ka-Lal-iLya, jump! ka-lakwet-ere-iLya, cough! ka-, neither negative nor imperative, is a frequent prefix of entirely undetermined force. It does not seem to be temporal. ya-, yaya- makes the protasis of conditional sentences. ya-kau-kanap-il, if you do not bite me yaya-kwa-dawikw-il, if you do not visit me The demonstrative gu, and probably kic also, are used as prefixes to a verb that is subordinated to another, yu-wel-as gu-bokin-ew-it, I saw you hitting me La-gera-le-wel-as kic-1-ow-et, I did not see you coming gu-dalew-iL wi-bokin-awa, him standing there I will strike g 'ra-wilrekwa-wi gu-kanap-it, I feel sorry that you hit me Kroeber: Languages North of San Francisco. 399 The imperative does not seem to be regularly formed. Some verbs show the ka iLya mentioned, a few -i, a few -ig' or -ag' , and others the stem, as kanap, bite! ka-le-waL-i, eat! gul-ag'-ig', go home! swala-g' -ag' , shoot! waL-ag', eat! SUFFIXES. -dtho seems to be causative. hi-da-tem-atho, made him sit tem-athu, made them sit du-tem-atho, name of the isosceles triangle element in basket pat- terns; the Yurok is wereq 'en, sitting. -wi, -wiwi is reflexive. A more or less pronominal character may be the cause of its final position. iwowok-wiw-iL, trained himself dicgaw-iw-iL, he likes himself kil-ia wi-dukL-wiwi, did you look at yourself? dawiL-wiw-iL, looking-glass What is probably the same suffix occurs normally on a number of verb stems. On some of these its force is clear, on others less intelligible. hie-wi, eat (transitive) gos-wiwi, swim de-tigelis-wiw, swim weta-wi, satisfied rakwa-wi, pity tawik-wi, visit gidid-iw, digwidi-wi, sit dale-wi, dano-wi, stand -ut denotes that the action of the verb is performed with an instrument. It is added to the verb, but has the force of an instrumental case on the noun denoting the instrument. hi-swale-wut cwat, I shot him with a bow kwaLwa riewom-ot-agem, with what did you kill himf dagakciL riewom-ot, a gun I killed him with bumipel da-haka-wut hi-niewu, a knife with I killed him ~er, -erer occurs frequently on numerals, adjectives, intransi- tive verbs, transitive verbs without an object, and nouns. It appears to have something of the force of a verb substantive. 400 University of California Publications in Am. Arch, and Ethn. [Vol. 9 guts-er, rit-er, rikw-er, one, two, three ra't-er, he is large, they are large detel-er, stab gawu-bets-er, it is becoming dry kacam-er, small Lelewit-er, round botcgawin-er, scratched ka-wiluw-er, is hollow gatslag-erer, end wits-kar-erer, coyote -erakw, -rakw, -nakw has similar force. makL-erakw-iL, he is small gots-herakw gu-tigeriL, he is a good-looking young man ru-magoks gots-herakw, my brain is good guts-erakw-iL, good ga-bitc-irakw-iL, he is bad guts-hi-nakw, is good guts-ka-nakw, is not right -pt has been found on color adjectives, apparently when they are attributive. hi-yu-wel-a yulewa-pti holakw', I saw a white deer siswa-pt waiyits, black dog holakw' riewom siswa-pt-ile, deer I killed a black one -tk, or tsk, seems to make adjectives predicative or substantive, herowedi-tk, the moon is shining kike-tk, kike-tsk, red mes-iare-tk, mes-iare-tsk, red p 'letk bele-tk, rock is flat hiwana-tk, square yuwetke-tk, five-cornered siswa-tk, small black seeds dicgaame yulewe-tsk, I like the white ones -u, of uncertain force, also occurs on adjectives, siswa-u, black ra'L-a-u, long, high (ra't, ra'L, large) kacew-a-u, short (kacam-er, small) -nim, meaning unknown. tawakwiLi-nim-iL, pushed him hi-tiekwa-nim-iL, broke him -ikwal, unknown. da-tem-ikwal, sat down gul-ag' -ekwal, went home akome-tal-ikwal, go back ha-maL-ekwel gitga, we shall eat 1911] Kroeber: Languages North of San Francisco. 401 ADJECTIVES. Stems translatable by English adjectives offer more com- plexity than is usual in Californian languages. They appear quite regularly with suffixes, among which have been mentioned -er, -erer, -erakw, -rakw, -nakw, verbifying or equivalent to the verb substantive, -tk and -tsk, predicative or substantive, -pt attribu- tive, and -u. There are also a number of derivative suffixes, chiefly found on adjectives of color and appearance, such as -iar, -Lak, gaLet. There are other complications. Thus the stem ra't, large, appears under the following forms : ra 't-er, or ra 't-ekw-iL, he is large, it is large, they are large; ra'L-el, ra'L-el-at, I am large, you are large; p'letk ra't-etk, large rock; ra't-eter, large (redwood tree) ; ra'tse, largely, much, very; ra'iau, long; ra'iaw-iL, it is long; ra'i/e-, much, on verbs. Adjective stems are not usually reduplicated, as is often the case in Porno, Miwok, and other Californian and American languages. NUMERALS. As already stated, the numerals from five to ten and from twenty to forty bear the suffix -helel or -welel. On fifty, sixty, seventy, eighty, and ninety, the suffix is repeated. The numerals from one to four usually end in -er, whether used in non- syntactical counting or qualifying animate or inanimate nouns. One occurrence without this suffix is of go't, ritwe, rikwo, one, two, three, denoting persons, instead of the usual go'ts-er or gu'tser, rit-er or ritw-er, rik-er or rikw-er. When years are referred to, the numerals have a suffix -eu or -ayu; for days, -dk or -wdk -. gu 'ts-ayu, ritw-eu ; rikw-eu, ramak, we 's-agele-wak. These suffixes are of significance as evidence of the existence of numeral classifiers, such as are abundant in Yurok. ORDER OF WORDS. The order of words as regards noun and verb is not fixed. Both subject and object at times precede and at times follow the verb. Adverbs, interrogatives, and pronouns precede the verb and usually open the sentence. The interrogative particle ia is always attached to the initial word. 402 University of California Publications in Am. Arch, and Ethn. [Vol. 9 ALPHABETICAL LIST OF AFFIXES. PEEFIXES. cu-, on verbs, prohibitive, negative optative euku-, see cu- cur-, west, across ocean da-, see ta- dal-, on verbs, undetermined dat-, up, above do-, continuative ga-, see gera- gawe-, on verbs, inchoative gawel-, on verbs, undetermined gera-, on verbs, negative g'ra, see gera- gul-, see kul- gu-, that, the, demonstrative; on verbs, subordinating hi-, on verbs, past hi!-, on verbs, undetermined hu-, sometimes accompanies the possessive suffixes of the first person plural or third person singular and plural ka-, see gera- ka-, on verbs, undetermined ka-, with suffix -iLya, on verbs, imperative ku-, possessive of second person kul-, on verbs, back, again let-, on verbs, undetermined La-, see Le- Le-, particle, with verbs, completed action m-, indefinite possession, on words denoting body parts na-, on verbs, past rak-, on terms of direction ru-, on verbs, all ru-, possessive of first person singular ta-, on verbs, perhaps usitative or iterative tinie-, east tcatc-, south wa-, see wi- wat-, etymological, on some body-terms wi-, on verbs, future wici-, east, inland wik-, on terms of direction wur-, north ya-, on verbs, if yaya-, see ya- yi-, possessive of first person singular on terms of relationship 1911] Kroeber: Languages North of San Francisco. 403 SUFFIXES. -a, on verbs, third person objective -ag<, see -ig< -ak, on numerals, denotes days -akw, on nouns, general locative -as, on verbs, second person singular objective -aswa, on verbs, second person plural objective -at, see -it -atak, see -itak -atawa, see -itawa -atho, on verbs, causative -ate, diminutive -ayu, see -eu -e'l, possessive of third person -er, probably equivalent to verb substantive -erakw, see -er -erer, see -er -eu, on numerals, denotes years -gaLet, derivative, on terms of appearance -hats, see -ate -helel, see -welel -i, on verbs, imperative -ia, enclitic particle, interrogative -iar, derivative on terms of color -ik, possessive of first person plural -ikwal, on verbs, undetermined -il, before subjective suffixes of first and second person -il, on independent pronoun of first person, perhaps objective -it, on verbs, third person subjective; also agent, and noun formative -iLya, with prefix ka-, imperative -ir, derivative noun-ending -is, see -as -it, on verbs, second person singular subjective -itak, on verbs, first person plural subjective -itawa, on verbs, second person plural subjective k, derivative noun-ending -lak, derivative, denoting language -Lak, derivative on terms of color -nakw, see -er -nim, on verbs, undetermined -ow, on verbs, first person singular subjective, singular and plural objective -pt, on adjective stems, perhaps attributive -rakw, see -er -s, derivative noun-ending -t, derivative noun-ending -tk, on adjective stems, perhaps predicative or substantive -tsk, see -tk 404 University of California Publications in Am. Arch, and Ethn. [Vol. 9 -u, see -ow -u, on adjective stems, undetermined -ut, on verbs, instrumental -wa, ending of demonstratives and interrogatives -wa, denotes the plural on suffixes of the second person; also itself used for -aswa -wak, see -ak -wakw, see -akw -wats, see -ate -welel, on numerals from five up -wi, on verbs, reflexive and medial -wil, derivative noun-ending -wiwi, see -wi -ya, see -ia TEXTS. CROW. gatsir Crow hi-wu'1-ag' -!L he went to. dol-61-o'w-iL he took kerawaga'tkari Porpoises diwe-ru'lakame "What did you do with daTsw-daru'dakw they are with." gul-u'w-iL tsek came back a child. gu-ra'tcetck that boy. wule-ba'iakriL relative-in-law curi-la'kau (Across)ocean curi-la'kau (across.) ocean hi-le's-iL he went by boat. ri'kar woperaga'tck 'c-iL two. He put them wuperaga'tskc-iL he put. tsek the children?" gatsir Crow gatsiri Crow gwa'tc-el His mother dewu-tem-a'Lel ' hi-la'g-iL he put." Hetold(?) hi-wo'kura-Le'kanem-iL gatsiri She caught him. Crow wule-ba'iakriL relative-in-law tsa'ki Children p 'le't-wakw on the rocks. hi-kol-6'w-iL He came back. go'tsker-e '1 Their grandmother gu-ku'nan That night da'wim-iL asked p 'let-wak on the rocks- dil hi-ra'kce'm-iL Eagle. She pursued him. hi-kol-6'twui dil she brought back. Eagle katsir-ie'1-iL Crow said: dil wi'wa-1 eagle his wife (?) godam-i'L di'le ran off. Eagle ya'gitemo't-iL yil-il he told: "Me d 'ane'r-iL did it. mesi-da'lidaks she put mes-akw In the fire mes-a'kwi in the fire. hi-da-tem-a'tho she put him. hi-ka'-tawa'l-iL He died. da't-kasiL The top of his head 1911] Kroeber: Languages North of San Francisco. 405 PELICAN. do-be'1-iL fished tcawera'tci Pelican tcawera'tciL Pelican du'-bel-iL fished. dakanewo'wi he stole hi-6'w-iL came. wa-keLa'm " I will look for him yitawa'ne " Why do you do it ? Do it no more!" dagl'weg' i with dip-net. kawu'kamer-iL stole. dlwi'l Others dlwi'le The others' di'le Eagle ina'g'-iL ina'g'-iL He thought: iy-u'w-iL came SO dil hi-La'k-iL Eagle went to him. yo'ckan da-g' a'tge 'negeL he tore, he tore him to pieces. so ka'-pel-iL Fished gu'ts-hi'nakw He was good. hi-kawe't-6 He gave them food. guts-hi'nakw He was good. gu'ts-ayuta'yeg' er-ak w so for one year. guts-ka'-nakw He thought: '' It is not good." gl'tka da'kiyiwoi dil ' I will, fishing with a dip-net." Eagle cuku-re'rir tcawera'tci hanew ' to Pelican he told. tcawera'tci-ika'n hi-t'ki'n Pelican he seized. He pulled, takwlya'kw 'ter hi-nl'ewa-k He made fire. He killed him. di'wilA hi-ka-kuweye'1-iL the others, were not afraid any longer. hama'-pe'l-i dile rawera'miL Fished Eagle's relatives. ga'-pel-iL tcaweratci ka-guwa' '-pe'l-iL They fished. Pelican no longer fished. SKUNK. bo'tcwi Skunk piLwa'tkotii Flies ra' 'ter large. reg' i'L anus (?) hi'-lew-iL danced, bo'tcwi Skunk. tcewa'-rakw thus did, wa-we'tom-iL went to get dara'kw sick mi'l-iL tciwa medicine-man that bo'tcwi da'herakw sak-sa'kw-iL Skunk was sick, sick. o-si'lakw-e '1 tcewa'-dakw-reg' -iL his-pain there was in. dakdi'skew-iL kana' 'p-iL went to the other side of him, bit (sucked). Wished to kill him hi-nieyaw-er me' 'lakw hi-ewi'wela'kw-iL He killed Elk. Rubbed their hands in joy sak-sa'kw-iL sick. me' 'lakw Elk bo'tcwi Skunk me' 'lakw Elk bo't-iL 406 University of California Publications in Am. Arch, and Ethn. [Vol. 9 piLwa'tkotii Flies. ha-ma/L-ekwel " We eat gitka shall me' 'lakw Elk, hi-mi'w-akwel wi-ri'ewaw-er me' 'lakw bo'tcwi eat the killed Elk." Skunk rer-I'L bo'tcwi kuwe'notw-iL did it. Skunk was well. piLwa'tkotii Flies wita' V-IL rejoiced. hi-ka'waw-iL They cut it up bu'mi 'pel with a knife, piLwa'tkotii Flies, bo'tcwi Skunk. yi'-waL-iL They ate. hi-ku'1-ag' -!L They went home. VOCABULARY. No Wiyot vocabulary has been published since the three printed or reprinted in Powers, 60 which suffer from faulty ortho- graphy and imperfect acquaintance with the language. As no other studies of Wiyot have since that time been undertaken, the author's material is here given, in spite of its no doubt fre- quent inaccuracies, in order to render available for comparative purposes, at least until some more thorough study of the language shall have been made, a more extended body of words than are now accessible. NOUNS. Persons : kowil, guwil, personal kowi, guwi, man motwiL, kakerawiwiL, kawotc, woman2 gakwiL, old man3 ceruki, old woman tigeriL, young man, bachelor tserariL, young woman, ratcetck, boy watcer, girl tsk, tcik, tsak, child hetca, baby wise-pelei, married man wisiL, married woman gwatw, widower, widow wakawe, divorced, separated woman dekLelaliL, chief miliL, danelatwiL, medicine man cokwirak, cirawakw, ghost, dead ancestor* dikwa, tikwa, white man, poison, spirit keldmiL, weramiL, relative eo Tribes of California, Contrib. N. Am. Ethn., Ill, 478, 1877, Appendix, Linguistics, by J. W. Powell. 61 Cf. diwile, another. 2 Of. motw, woman 's front dress. Kawotc perhaps means wife. 83 Probably from gakw, to know. 4 Cakw, sick, die. 1911J Kroeber: Languages North of San Francisco. 407 dar, dan-, father; son gwatc, mother reka, daughter d&k, darewerekere, brother, sister bitcotcker, grandfather gdtcker, grandmother ag6Lek, grandchild wetserakw, son-in-law maiakriL, relative by mar- riage Body-parts : metere, detere, nose melir, welin-, eye 65 melul, mouth niept, teeth wat-melok, ear mit, tongue djipLiL, beard mclokai.. throat wat-wet, head baL, paL, hair we's, hand mokec, fingers we 'tapis, thumb cor-eL, index finger tsewawiL, little finger metkan, nail wo'r, wo'n, arm daletokeru, elbow taLcokra, shoulder weliL, foot tckatc, leg lawel, knee mel, body, flesh meweriL, flesh, fat watswetsaa, breast weser, woman 's breast, milk dau, tau, belly doguganakw, navel hodiLere, umbilical cord duwerec, buttocks dgat, penis melak, testicles bee, vagina cak, clitoris hatageriL, womb ma'n, pubic hair watw, heart tcegeL, liver heL, intestines magoks, brain sakwer, lungs wat-kerat, bone kawik, wat-kawik, blood wat-kai, skin hapLakw, sinew b6kaweriL, tendon hil, urine me'l, excrement betsakw, semen wetsaL, saliva walept, fur, feathers wat-hel, tail wat-6tk, fin wat-iLat, shell merar, horn wat-udatkawi, egg t.-i i k.-uini. breath halokic, -taldkic, shadow tciwarin, name silak, pain Mammals : me 'lakw, elko haLakw, h61akw, deer ta 'wila, buck but-caweti, white deer7 tsetsgeruLigerer, bear makw, kanapeliL, grizzly rakwuLiriL, wolf witskererar, witkaL, coyote' 8 waiyits, waiyets, dog halikwiliL, fox sekseswiL, otter dikwagawi, fisher^ s Wei, to see. The ending -lakw is common to these two terms. T Cawet is white. Kanap-el-iL is biter. c Witskererar is wild dog. Cf. wild cat below. TO Dikwa is poison, white man. 408 University of California Publications in Am. Arch, and Ethn. [Vol. 9 ra 'raweic, tcweLig' atcatci, raccoon gd'miri, mink teigereLariL, civet-cat botcwi, butciwi, skunk datgaeaniL, datkaLaniL, panther datsgagererar, wild-cat tsugatLaiugoner, weasel wit 'hot, gray squirrel seles, becduliL, chipmunk Letc, wood-rat tseretshigarer, wood-mouse yacucagatck, gopher weni-crenimiL, mole 71 da'kere, sea-otter gumayoliL, sea-lion matswaptsire, seal kerawagatkari, porpoise delabeliL, killer-whale 72 kimak, dayugele, whale Birds : tsutskie, bird di'l, eagle cawetociL, bald eagle 73 cataoc, condor butsera, buzzard guletsol, tcanitc, gokwera, bletsul, hawks gatsir, crow ranatwuloiyokit, raven tcakakeLhitcatc, blackbird tcera, Limayusele, bluejay pltanatinu, metsig'e, robin 7 * tsigwatsharawi, kingfisher tseweLiksi, swan tcaiuwetcg, goose katgeragiL, cawetcoligiL, brant 7 s tcatcitckiri, mud-hen pane'r, crane gugitcetck, gull ma 'g' es, shag tcaweratei, pelican Other Animals: gatcu, rattlesnake haretc, garter-snake halunasi, red snake tcitcgiwetcg, turtle matakwiL, lizard kwakw, frog maLak, salmon go 'taw, lamprey-eel hut, surf -fish tcaptcuc, halibut tcgerits, flounder tau'gel, rock cod witiwlnuwi, herring wi'welil, gawui, small fish mo'er, shark cagitsrer, dogfish rit, mekar, gatewac, tsayunuwatcke, clams wuletat, razor clam hiwaklegak, cockle hiwat, haliotis tsar, mussels in bay witcac, mussels in ocean bituwecanagiL, salt water snail butcatc, land snai! 7 tcomack, large slug piLwatkoti, fly gats, bee bie, mosquito tcirawaukw, butterfly swalen, dragon-fly tckLare, grasshopper spina 'g< aralu, larva of locust dakLa'lin, flea heikw, louse botkanawiyuc, spider giLeswaL, spider-web yotu, maggot kwekipLakarer, centipede mireL, angleworm Lwuregat'i, crab gl'bas, small red crab 7 1 Wen, sky, which according to myth the mole supported. 7 2 Bel, to catch fish. 73 Cawet, white. 7 * Cf. mes, fire, red. 75 From maL, waL, pL, eat. MaLak also means food. 7 -ate, diminutive. 1911] Kroeber: Languages North of San Francisco. 409 maLakeL, sand-worm wutwuciL, squid daegalwagigatckarer, jelly- fish wuduyuwetk, sea-urchin" tkayukis, star-fish miplatk, cuwatpiyag' apkwi, holothurian or sea-anemone Plants: wanakw, talewiL, taleg' iL, tree mati, wood hawig' erak, grass, herbs, medicine gutcweratc, pletkapleiwun, leaves? 8 wetcatc, buds*8 dakw, pitch mukweti, pine dap, dak, spruce mopel, wopL, redwood wit, alder tigeL, willow legoLes-weL, hazel himene-weL, Xerophyllum tenax grasses tigwametsha-weL, Wood- war dia fernTB sdpitk, tule we 'taw, salmon-berry mip, blackberry md'kel, huckleberry mikwel, salal-berry kiwatchokwere, thimble- berry bdderuc, Brodiaea roots weL, bldkat, bokitchere, rapcaue, edible roots katsera, soap-root mdt, acorn ga'mak, acorn-soup rakwiyidag' eral, wild oats Ldkai, ecerawen, mokerits, raladethen, edible seeds Nature : wen, wirudala, sky kek, clear sky gotso-wen, day^o tarn, gitcai-ailokwe, sun ritsowel-ailokwe, moon ritsowal, night wene-welir, gumeratck, daruitwl, stars'i gutcetcguciL, Pleiades2 wai'were'iL, morning star liptau, cloud dalaLwala, rainbow daliLak, lightning dadakak, delalater, thunder tamutcikere '1 LakuluwiL, sun-dog 83 hekw, snow bd'ware, rain ho'l, weratci, gutser-ol, waters* pak, salt water, ocean waLa, hot water hiegawi, cold water ha'Lak, steam laliL, rariL, stream, river rariL-wats, small streamss betaw, spring batwar, freshet mes, wes, fire bi'wur, smoke lag' erak, 16 'erak, land patut, earth, soil tetwuka, mud 77 Said to mean round. TS -ate, diminutive: for pletkapleiwun cf. p'letk, rock, bel, flat, blaiatck, wedge, mi-platk, holothurian. TO -weL may refer to use as basket material. so One-sky, or good-sky f si Wene-welir, sky-eyes; gumeratek, cf. gomera, soft, weak. sz Cf. ratcetck, boy; the Pleiades are thought to be girls. 8 Sun his boy holds. 8* Guts-er-ol, good water; weratc, drink. 85 -ate, diminutive. 410 University of California Publications in Am. Arch, and Ethn. [Vol. 9 letkak, sand pLetk, p'letk, rock rakdat, tanatgak, ralitgat, mountains Objects : mol, house kac-werar, small house 87 hikawa, sweat-house mes-wululel, fireplace g 8 haldwi, ha'luwi, boat daL, ship men, paddle hutcwate, cooking basket hutcwatc-hatc, small cook- ing basket 89 gi, woman's basketry cap rael, bitweliL, open-work basketry plate bas, large flat close-woven basket bac-ats, same, smallerss kaluwo, conical open-work carrying basket kiwelauL, basketry dipper bitu, basket mortar dilul, storage basket for acorns dali'Len, small storage bas- ket with cover hitwokwakerawiL, flat sift- ing basket cwat, bow 9 tsapi, arrow kuluwu, quiver bumi'pel, knife meL, ax blaiatck, elkhorn wedge 91 betgl, stone maul tul, stone pestle waLawinewok, slender stone pestle wetsecraweL, metsecakerawiL, slab mortar gamak-watkar, cooking stone 9 2 gawelotgalewiL, digging- stick watk, tule mat dewi'pen, dewi'peliL, string, twine matop, netting shuttle kas-weL, mesh-measure 9 3 da-giweg'iL, a dip-net rathe-giweriL, a dip-net for surf-fish 9 * teaweratc, do'iw, dip-net for salmon in streams gut-wera, dip-net for lam- prey-eels 9 ^ gucager, gill-net for herring cagatagere, gill-net for sal- mon hephagwar, gill-net for trout dalosun, gill-net for sturgeon ha 'ker, hakere, woman 's back dress motw, rewunakwiL, woman's front dress 9 " rulen, clothes 97 twanagit, woven blanket keswakt,, steatite gwageretna, black obsidian 86 Cf . rak, prefix of terms of direction, dat, up. 87 Kac-, small. ssMes, fire. 89 -ate, diminutive, so Cf . swala, shoot. i Cf . bel, wide, flat. 2Ga'mak, acorn-soup. a Perhaps kac-, small. 94 Evidently large-giweriL; ra't, large. 95 Cf. go 'taw, lamprey-eel. 96 Cf. motw-iL, woman. 97 Cf. rulen, nulen, undress. 1911] Kroeber: Languages North of San Francisco. 411 tsagawila, red obsidian b6tcu-caweti, white flint 98 kral, blue rock reni, dentalium shell cur- rency gutserakw, small dentalia used as beads kag, shell disk beads itesi, Lum, small univalve shells used as beads dikwa-lenewiL, glass beads siswatk, yew-seed beads gutcicakwi, pine-nut beads VERBS. Human Relations and Occupa- tions : tsew, give tawik-wi, visit gameratc, play rulen, -nulen, undress wunakwa, steal dabor, lie himan, mark, write bel, catch fish niewom, kill swa-la, shoot t, give food l.-i I in. receive food kictawil, beg da-wim, ask anitw, pay we 'la, buy halewu, olewu, dolewu, danee 101 Mental Action: inag', think daretw, twa, think, remem- ber gakw, know dicgam, like wet, satisfied rag, want, desire wipac, gatsepi, gambling sticks d&pcer, gambling bones, of Southern type maLeL, pipe kakwesiw, medicine-man 's feather head-dress wat-welat, medicine-man 's swallowing feather** dlkwa, dikwa-g'eL, poison Latsik, myth waLel, path, road, trail tceg'ak, corner guts-ewan,- one fathomioo dilegana, angry rakwa, sorry, pity wil, fear kiLat, hurt, pain Senses : athera, smell Lephai, taste tsaw, pugakw, touch kwace, hear wil, wel, see dukL, look at keL, look for, seek Performed with Organs: hanew, iel, anel, delani, atel, say, tell tsowes, shout bawerats, whisper lalisw, sing waL, maL, pL, eat hie-wi, eat somethingioz weratc, drink beLokel, spit kanap, bite tsitsir, sneeze da'kwa, snore lakwet, cough likw, rikw, cry, weep gakwiLet, sweat s Cawet, white. Cf . wat- on body-part terms. 100 Cf. guts-es-wani-helel, one hundred. 101 Cf. dale-wi, stand. 102 - w i reflexive-medial suffix. 412 University of California Publications in Am. Arch, and Ethn. [Vol. 9 Bodily Condition: datw, alive cakw, sick, die tawal, dakw, dead, die it, wit, nit, sleep, dream watap, resemble kawi, make aki, do, affect Bodily Position and Motion: dikweL, lie dale-wi, dano-wi, stand 1 " 2 tern, gididwi, digwidiwi, ak, ag<, a'l, go, move ow, owi, yowi, ohwi, come les, travel in boat hiLak, enter water gos-wi, tigelis-wi, swimioz rakcem, pursue gudam, flee Lai, jump atkag'an, creep unas, crawl takerawae, kneel Bodily Action: yock, ack-ar, tkin, tear, pull, rip ti'n, wakw, push tiar, meet, strike bokin, da-kwicile, hit, strike da-tele, stab da-kwage, slap kiedal, take, pick otw, bringios wolew, get!3 kanew, catches olowo, uluwu, catch, holdios tawi, wawi, butcher cits, flay Dynamic and Spatial: musaw, gadawal, stick pelal, cawat, tiekwa, break, open, cut pawal, split dokap, dokaw, crack welu, wilu, hollow butc, scratch lawil, scrape kaleg'-al, roll, turn wayit, bend cwik, move low, dakwes, komal, hang komat, blow picar, swell tweL, twerie, letka-, fall, descend nole, rise, lift lip, extinguish ADJECTIVES. ra 't, ra 'L, large, long kac, kacam, small, short guts, good gare-wack, ga-bite, bad, rough, strongio* leg, heavy ca'p, light badag'a, hard gomera, soft, weak kLet, hot gets, cold bel, flat, wide io3 Perhaps contain suffix or stem -ew. io* Gera-, ga-, negative prefix. 105 From mes, fire. gatseLak, sharp capo, straight Le'pi, rotten pitag', bitter wukagiw, rich cawanakw, ga-gitgakw, poor mes-iar, kika, red 105 cawet, yulewa, white siswa, black dukaL, dukapL, blue, green, yellow 1911] Kroeber: Languages North of San Francisco. 413 ADVERBS. he, yes kin. no wai, is that sof tawiL, always swawi, very, extremely rogaL, soon wigiL, now wa, far dat, up tcwi, behind wur, north at, tcate, south cur, west, across the ocean wic, tini, east, interior, up- land dalil, indoors geru, gat'gaLil, outdoors kuna, yesterday gowai, to-morrow wiril-akw, to-day 1 "' wal-akw, in the morning gau-kuna, in the eveningioi kil, you hinar, winar, we gic, this gu, guru, that PRONOUNS. tci-wa, that, so, thus ci-wa, du-wa, what, where kwaL-wa, WULO, how, why diwile, hikeL-, another NUMERAL STEMS. go't-, gu'ts-, one rit-, ritw-, two rik-, rikw-, three riaw-, ram-, four we's-ag'-, dekLi-luk, six halu, seven hiowita, eight 10 o mece-rok, nine 11( > ru-lok, ten io Perhaps related to wen, wiru-dala, sky, day. io7 Cf. kuna, yesterday. Jos From we's, hand. io Perhaps related to four. no Cf. -rok of mece-rok, -lok of ru-lok, and -luk of dekLi-luk. 414 University of California Publications in Am. Arch, and Ethn. [Vol. 9 YUROK. The Yurok inhabited the coast from the mouth of Little river to Wilson creek, six miles north of the mouth of the Klamath river; and the Klamath from its mouth to about six miles above the inflow of the Trinity. There are sometimes said to have been four Yurok dialects, that talked along the Klamath being numerically the most important, while three coast dialects, differ- ing in increasing degree with the distance from the river, were spoken in the vicinity of Gold Bluff, Orick, and Trinidad. The coast Yurok have suffered great diminution of numbers and are now much mixed with river Yurok. A final determination of the coast dialects has therefore not been made, but such material as is available seems to show only one divergent dialect on the coast from Stone Lagoon south to Trinidad, and this not very different from the speech along the river. The present account is based on the river dialect. The Yurok have been visited and investigated by the author at various times between 1900 and 1908. A considerable body of texts has been recorded, but there has not yet been opportunity to elucidate from these the grammatical principles governing the language. The account here given therefore consists only of such grammatical notes as were made incidentally to the record- ing of the texts, and is introduced principally to enable a com- parison of Yurok with Wiyot. To the ear the two languages are more similar than any others adjacent, and the suggestion was long ago made 111 that they might be genetically related. The preparation some years ago of a comparative paper on the Native Languages of California, 112 revealed a morphological resemblance between Yurok and Wiyot in most of the points then considered. This structural similarity is extended by the examination made here, and is undeniably close. Whether the two languages are related is however another question. A running acquaintance with both reveals but few in Latham, Trans. Philol. Soe. London, 1856, 84. Opuscula, 343. 112 E. B. Dixon and A. L. Kroeber, Am. Anthr., n.s. V., 1, 1903. 1911] Eroeber: Languages North of San Francisco. 415 words that are similar. Such are mes, mets, fire; welir, welin, eye; go'ts-er, qo't- or qo'ts-, one. 113 This number is so small that unless it is materially increased by further comparison, the resemblances must be regarded as due either to accident or to borrowing. A systematic comparison cannot be made until both languages are farther analyzed and the stems and elements of words, which in most cases are complex, are determined. For instance Wiyot welir, eye, is undoubtedly connected with the stem wel, to see; but in Yurok welin fails to correspond with ne'g'wo, see; so that borrowing is suggested in this instance. That two languages belong to the same morphological type or group, does not by any means prove them genetically related in America. A common origin can be asserted only on the basis of lexical correspondence. Loose unification of languages that may be entirely distinct, based only on general or partial grammatical similarities, is unwarranted. The structural resemblances between Yurok and Wiyot are however so close and often so detailed, as will be seen, as to create a presumption that lexical and genetic relationship may ultimately be established ; and if not, to make it certain that morphological interinfluences between the two languages have greatly modified one or both. Yurok, more properly yuruk, is a Karok word meaning down- stream. The designation Weitspekan is derived from we'tspekw, more properly we'tspuc, now Weitchpec, one of the numerous villages of the Yurok. The Yurok have no name for themselves other than OL, people. PHONETICS. In the phonetic determinations, assistance was rendered by Professor P. E. Goddard with mechanical experiments, by Mr. T. T. Waterman, and by Dr. E. Sapir. Yurok vowel qualities are very shifting and often indeter- minate. There seem to be six vowels, all of peculiarly broad or open quality. I is so open as to be often heard as e. E is also very open, so much so as to be at times the aural equivalent of a in English us It is tempting also to compare Wiyot rit-, two, and rik-, three, with Yurok ni- or na-, two and naxkc, three, as initial n- becomes r- in Wiyot. 416 University of California Publications in Am. Arch, and Ethn. [Vol. 9 bad. A, often written a, is usually between the two sounds as pronounced in father and bad by Americans, but may be heard as either. O is near English aw, sometimes with even more of an a quality. U is also very open. Close vowels do not occur. The sixth vowel is a vocalic r, here represented by er. In its formation the tip of the tongue is bent upward; the tongue meets the palate farther back than in English. Like the related consonantal r of the language, this sound is not trilled and pos- sesses the peculiar quality of English r, as compared with the various forms of continental r, in an exaggerated degree. Unaccented e and o are often hard to distinguish and one is frequently replaced by the other in Coast Yurok as compared with the river dialect. E is often followed by a glide, e 1 . Analogous is a u , in rapid speech o. Glottal stops are abundant. In many stems they are organic. In addition they frequently appear after vowels which close syllables, whenever such syllables are stressed or articulation is forcible. In rapid continuous speech these stops disappear. Thus tinica, what is it, may often be heard as ti'ni'ca' when emphati- cally uttered. An organic stop gives the impression of dis- tinctly doubling the vowel which it follows : pa ', water, is heard as pa' a or pa 'a. A complete cessation of sound formation in the middle of words is not infrequent: wec,ona, world, ololekwic,o 'I, person, we ! n,tsa u kc, woman. It seems likely that such pauses mark etymological divisions. There are two series of stops, both surd, the ordinary and the fortis. English surd and sonant stops are pronounced alike by the Yurok. The ordinary surd stops are more aspirated when final than when followed by a vowel: nepui, but almost nep'. The fortes, or stops accompanied by glottal stop, are of only moderate strength. K and q, palatal and velar k, both occur, but it is uncertain whether they represent organically different sounds or are divergent formations of the same sound influenced by adjacent vowels. In any case k is most frequently audible before i and u, q before e, a, o. K u or qw is frequent, and is felt as a simple sound. 1911] Kroeber: Languages North of San Francisco. 417 T is formed against the lower edge of the teeth and sometimes is clearly interdental. There is one s sound, written c, intermediate between s and sh, perhaps nearer the latter. In ts, which is the equivalent of a simple sound, the s element appears to be more pointed than in the usual c. Yurok g is always a continuant, but not harsh. X, the surd continuant corresponding to g', is limited to combina- tions with following surds, such as k, p, c, and has not been observed initially or finally. It is distinctly harder than a mere ordinary aspiration, h or ', but appears to be only a parasitic development before certain surds: naxkceiL, Woxpekumeu. There are three 1 sounds: 1, I, and L. The first, 1, is sonant and audibly similar to English 1, but a mechanical analysis by Dr. Goddard indicates a suddenness of approach in the formation of the sound, which is found also in Hupa I. 114 . The second, I, is a not very common surd continuant found after glottal stops: qo 'Zqolatswin, ohonecqwetso 7. The third, L, is a surd affricative. In some instances the t-approach is weak, and the sound is almost wholly continuant. Wiyot L was also often heard with this quality. W is sometimes produced with less protrusion or rounding of the lips than English w. It therefore has something of the quality of bilabial v, and initial unaccented we and wo are often hard to distinguish from o or u. Surd m, n, and r occur finally after glottal stops. 118 Con- sonantal sonant r has the peculiar quality already ascribed to vocalic er. N, m, y, and h require no comment. The sounds of the language may be represented as follows : i, e, a, o, u, er 1, e, a, 6, 11, er q q' U) g' qw k k' k" t t' c n n ts ts! p p' mm r, r, 1, I, L w > y " ( it pause (,) 4 Present series, V, 9, 12, 1907. us Compare ibid., 10. 418 University of California Publications in Am. Arch, and Ethn. [Vol. 9 The stress accent of words is often well balanced between several syllables and often marked on one or two. Accent is however less a matter of intensity or loudness of sound, than a rise in pitch and a holding of the accented syllable, which is manifested in lengthening of the vowel or doubling of the pre- ceding or following consonant. Accent is however not deter- mined by organic length of vowels, and often falls on syllables that are intrinsically short. The dwelling of the voice, and its rise of pitch, on the accented syllable, give a peculiar and pleas- ing quality to Yurok speech. Doubling of consonants is frequent, but whether it occurs only in connection with the accent or also organically, is not certain. In place of ww and 11, g'w and II seem to occur. Initial consonant combinations occur, but their scope is limited. The principal observed initially are tsp, tsq, kn, kr, pr, tm, ck, ct, cr, cl, cm, Lm, Lq. These give as the first member of initial combinations q, k, ts, t, p, c, and L, or surds only; and as the second, q, k, t, p, r, 1, n, m, or stops, nasals, and r and 1. Ts and qw are to be regarded as simple sounds. Final combina- tions do not occur except with c as second member. This is found most frequently after k, so that kc may represent what is to the Yurok a simple sound like qw and ts. The imperative suffix -c is however added to stems ending in p, n, r, and other consonants. Endings like ern, erm, erL, ert, consist of the single consonants n, m, L, t, following the single vowel here represented by er. Yurok initial combinations are more numerous than those of Wiyot, but Wiyot possesses more that are final. Vocalic assimilation occurs to some extent: ne-craats, my quiver, wo-croots, his quiver; yots, boat, ne-yots or ne-yets, my boat; erner-heL, at Erner, wo-croots-OL, in his quiver, pa'a-iL, in the water, okapol-iL, in the brush. Other instances will be found among the numerals below. There is a particular ten- dency for er to assimilate other syllables to itself. Most words in which er occurs contain it in from two to four or in all syllables: qerxtsper', neryerwert, cermeryer. When one vowel of a word changes to er, most the others usually become er also : lo'og'e, ler'erg'er, black. 1911] Kroeber: Languages North of San Francisco. 419 STRUCTURE. Reduplication is less marked than duplication : qots-qots, worm; tseix-tsei-uc, mosquito, compare tseix-keni, small; nius-mus, cow; we-tseq '-tseq '-oa, striped pattern. In connected texts neither reduplication nor duplication is frequent. Composition and derivation are abundant. Adjective stems precede noun stems in composition : pelin-tsiek, large-dentalium. In accord with the pronominal structure of the language, com- pounds containing a possessive are common : qoqonewuL-we- tspeg'a, long his ears, mule. "Words are also compounded or derived with nouns as first element : nepui-cnec, otter, nepui, salmon; Lqeliqera, mole, Lqel, earth; merwuci-clei, a species of lizard, which is thought to bite the navel, merwuci. Derived nouns, as in most languages, are formed by suffixes: nep, eat, nepui, food, salmon. Terms of direction are always prefixed, as in Wiyot, Athabascan, Karok, Porno, Wintun, and perhaps other languages of northern California: petskuk-ceg' ep, up-river coyote ; puleku-qwerek, down-stream sharp ; woxpe-kumeu, across- the-ocean widower; pets-ucla, up-river throw. PRONOUN. The Yurok pronominal elements, like the Wiyet, are incor- porative, the independent forms being emphatic or non-syntac- tical. Also as in Wiyot, possessives are prefixed, subjective or objective forms suffixed. The independent pronouns are nek and neka', objective in -k;its, for the first person, qel or qela', objective qelats, for the second. There appears to be no pronoun of the third person. The possessive forms are ne-, no-, my; qe-, qo-, your; we-, wo-, o-, his; m-, someone's, indefinite. The plural seems to be the same. On intransitive verbs the subject is indicated by -k in the first person and -m in the second. In transitive verbs the combined objective and subjective elements determined are -tsek, I you ; -ck, I him ; -tso, I you ; -co, I them ; -xpa, you me ; -m, you him, you them ; -xpen, he me. It Independent Possessive 1. nek ne-, no- 2. qel qe-, qo- 3. we-, wo-, o Indef. me-, m- 420 University of California Publications in Am. Arch, and Ethn. [Vol. 9 would seem from these forms that the objective elements are -p, -ts, -c and the subjective -k, -m, , for the three persons respec- tively. The pronominal elements thus are : Subjective Objective -k -p -m -ts -c The possessives indicate n and q as the essential elements of nek and qel. The Wiyot radicals are evidently the same. Q not occurring in Wiyot, k, in kil and ku-, is the equivalent. Initial n not being permitted in Wiyot, yil and ru probably represent original n. The pronominal forms of Yurok and Wiyot agree in the following points : They are incorporative. Elements added to nouns are pre- fixed, those added to verbs suffixed. The prefix and independent forms are similar to one another, the suffix forms entirely dis- similar, also differing completely among themselves according as they are objective or subjective. The objective suffixes pre- cede the subjective, which are identical whether transitive or intransitive. There is a form, used with body-part terms, denot- ing indefiniteness or absence of possession; it is m- in both languages. The fundamental elements of the possessive and independent forms in both languages seem to be n for the first person and k for the second, the former common, the latter exceptional in American languages and therefore significant. The suffix forms in the two languages however show no similarity. Demonstratives show two stem forms, yo and ki. lyo is this, iyoLko these. Yok is also found. Ku and ki are that, the, denoting reference rather than distance. Ki is also used rela- tively: ki ololekwic,o 'I eqlaxLkome, what men tread-on, the world. With ku and ki compare Wiyot gu-r and gi-c. Kuc and tin are interrogative. Kuc is what, where. Tin-ica is what, what kind; tin-pa, which one. The interrogative par- ticle is hec, postpositive. It is used with verbs; also independ- ently, when it has the meaning : what is it, is it so. 1911] Kroeber: Languages North of San Francisco. 421 NOUN. The Yurok noun like the Wiyot lacks all designation of number or syntactical case. There is a general locative suffix -ir,, -aL, -ei,, -OL, corresponding to Wiyot -akw. In addition a locative -ik has been found in raets-ik, in the fire, Lqel-ik, in the ground, below, hierk-ik, north, Lep-ek, in the house, pets-ik, up stream, pul-ik, puleku-k, down stream. Another locative is -ic: mets-ic, in the fire; \vonoyek-c-ume, sky-in-girl; turip-c-atsin, inhabitant of Turip. Terms of direction themselves are suffixed like locative cases: erner-hiqo, opposite Erner; qenek-pul, down stream from Qenek. Finally there is a suffix or enclitic -meL, by means of, with, on account of, for. This is however used in verb complexes as well as on nouns : tetamoc-ek meL tikwoxpen- ek we-yots, angry-I because broke-I his-boat. VERB. The Yurok and Wiyot verbs are similar in function and structure. Both are distinctly the center of the sentence. In both prefixes predominate except for the expression of pro- nominal relations. Adverbial, modal, temporal, and subordinat- ing ideas are expressed by prefixes. There are also verbal suffixes, but their significance is for the most part not yet clear. Even independent adverbs are to some extent drawn into the verb-complex, which often assumes great length. Such adverbs always precede the verb stem. Many prefixes cannot be dis- tinguished from conjunctive or adverbial particles placed before the verb. They usually precede the verb immediately and are spoken as one word with it. At other times they are separated from it by nouns, adverbs, or numerals. ki-na'eli-hoxkumek, shall two-houses I-build ki-naxkcemi-wanu'layo, shall three-times I- jump-up qdlo- 'oi. ki t s-keno-atsi yuuk. it-seems-a-person has-sat-down-indoors These constructions evidence a compact grouping of the other constituents of the sentence in the verb-complex, but at the same time seem to show that at least some of the modal, tem- poral, or adverbial elements connected with the verb stem are 422 University of California Publications in Am. Arch, and Ethn. [Vol. 9 not so much actual prefixes as proposed particles closely linked with the verb which is the central factor of the sentence. When the subject and object are independent nouns not brought actually into the verb-complex, they regularly follow it. This marks them off from nouns whose general relation to the predicate is adverbial, which normally precede the verb, thus opening the sentence. The general ' ' appositional ' ' nature of the sentence is obvious. The significance of only a part of the verbal prefixes and suffixes that have been determined in Yurok has become clear. Such are : kits-, completed past ki-, future tsa-, imperative nimi-, nimok-, negative kowits-, negative wikiLne-, negative kinek-, when matseki-, kitatse-, if alukumi-, because conini-, because kit-, he who, when, participle wictu-, wiit-, that is who, what, how, relative qolo-, as if, like, appears to tsyu-, all Of suffixes, the following are apparent : -c, imperative -kwilek, verb substantive, similar to Wiyot -er and related endings. Prefixes of undetermined meaning are wil-, menex-, numiL-, tukwile-, kwileki-, niko-, qem-, yokitsnini-, qem-kits-minolini-, ol-, yikun-, pikoxtsi-, mokwile-, okome-, tsigoL-, ca-, ala-, qet-, me-, ha-, moc-, kuni-. Suffixes are -yeg' o or -heg' o, -melek, -exkwun, -pimo, -kwetsok, -wertsek, -uts, -noxpe, -'m. The adjective shows a difference for animate and inanimate, montse, white, animate monterer cokoto, red, animate cerkerter lo'og'e, black, animate ler'erg'er pleli, pelil, large, animate plerer Altogether a more extensive occurrence of subordination is visible in sentence structure than in Wiyot. It may be that this difference is due to the fact that the Yurok texts obtained are a 1911] Kroeber: Languages North of San Francisco. 423 better representation of normal speech and that the sentences in the Wiyot texts are unduly abridged through the informants' inability or unfamiliarity in dictating. NUMERALS. Numeral stems are used with a variety of classifying suffixes. This is a feature not recorded of any other Californian language except Klamath-Modoc, though not uncommon farther north on the Pacific. The observations made on Wiyot reveal at least the presence in that language of similar suffixes, though their extent is undetermined. The following list probably does not exhaust this class of suffixes in Yurok. -iL, in counting -epir, separate dentalium shells" 6 -etani, strings of dentalium shellsii -er, woodpecker scalps"* -erpi, obsidian bladesiia -eriL, white deerskinsn -o, months, dollars -emoiL, nights -en, days, also pernekr hand-measures -emoi, fathoms -ixteli, boats, wagons, conveyances -eli, houses, sweat-houses, nests -emi, times, occasions, years In the addition of these suffixes to the numeral stems there are some apparent irregularities and certain phonetic altera- tions, including several instances of the vocalic harmony, or assimilation from suffix to stem, characteristic of the language. 1 2 3 4 Counting qoore" ni'iL naxkceiL tsoonet, Dentalium-shells qooxtepir nil 'ii [ i i i naxkcepir toonepir Strings of dentalia qootani na'aitani naxkcetani toonetani Woodpecker scalps qererxter ner 'erxker nerxkcer ta 'erner Obsidians ner 'erpi nerxkcerpi White deerskins nerxkceriL Months, dollars qoxto no'o naxkco toono Nights qoxtsemoiL nil 'it Mu ii i . naxkcemoiL tsoonamoiL Days qoore" na'ain naxkcen tsoonen Fathoms qoxtsemoi nit 'amoi naxkcemoi tsoonamoi Boats nil 'aixteli naxkceixteli Houses na'eli naxkceli Times ne'mi naxkcemi tsoonemi Objects of value and mediums of exchange. 424 University of California Publications in Am. Arch, and Ethn. [Vol. 9 As will be seen, the vowel of the stem for two varies between a, i, e, o, and er; in four and one t and ts replace each other; x before t and k is parasitic, and qoore u , the independent word for one, corresponds to a stem qot-. How many is expressed by kuc tsameL; how many months by kuc tamawo; dentalia, kuc tamopir; woodpecker scalps, kuc termerwer; obsidian blades, kue termerpi. TEXTS. WOXPEKUMEU. qe'nek-ic ho" tu-qwe'nomet-ic atsyeg' u'qwin At Qenek he was. Outside the sweathouse 117 he sat. tu-no'L-pu'lukitsro" tep kits-hawe' '-wo 'o'me'pek The sun was low. 118 He had sweated in the sweathouse, wie-tu-awetsyu'qwin qwenome't tu-wic-tu-noL then he sat down outside the sweathouse. There he used to atsyeg' u'qwin wit ! i'ni-oqw ki- we-tsy e' 'g' wolo sit. He kept that his flute lepo'noL mo-no' L-puluki'tsro i'yeger'er 'xcerper ' under the ridge-board. 119 The sun was low. He beat his hair dry. 120 mo-wit-ki'ts-weno 'omo'kciL numi-wo' 'g' ik uki' 'cen Summer had begun, it was the middle of summer. kits-numi-mi'kco'to 'm uki' 'cen kits-wic-o 'locon It was the very middle of summer that he did this. kits-o'-numi-wic-tu-co'to uqeg' e'camewoLek kiconin-ho'l 'em Thus he liked to do. He was very sorry for those who would be ki-6'L tu-wic-tu-e-meL-qe'g' ecamewoL niki'meLtsmeyo'qciL men. Thus he was sorry for them when it began to be evening. tu-no'L-wie-tu-e'-meL-rura'w ' wic-tu-e'-meL-tsig' uwolo Then therefore he began to play 121 the flute on account of that. OLO' 'm we-tsye'g' wolo wi'c-tu-e'-meL-ru'raw ' He took his flute. Therefore he played. 117 qwenomet is the place outside the small exit of the sweathouse. us puluk is down-stream, which at Qenek is nearly west. ' ' The sun was in the west." us leponoL is the curved board, usually the gunwale of a broken boat, that covers the ridge of a sweathouse. 120 her'erxcerp is a stick with which the hair is combed or beaten dry after a bath such as follows sweating. 121 rurawo usually means to sing, but evidently refers to any music. 1911] Kroeber: Languages North of San Francisco. 425 PULEKUQWEREK. pe'kwan qo'li-atsyu kuc-ki-la'ek wa'iqowo"opo-c AtPekwan someone sat. "Where shall I travel!" " Cross the river here. noL-hig' -woon witokwi-nil'L witso-la'ya heL-to-ki-ne'wom Up hill some distance live two. Travel there. Then you will see o'lo'meL kowitso-meL-no'xpew '-m ta"o" ta'wi-ki-laek a house. But do not enter it." o 'lo'meL o'le 'm a house. He said : qo'lin ats-ka'c,a qolo-ni'mok the other is pounding. It seems not are tac-kit-ckui ki-kac,a'-wok heL-weno'o-c ' When it is good that which yon are pounding, give it! " nimo'oxk u heL-weno'o-c " There is nothing ! Give me kits-keno-a'tsiyuuk Lo'ope-c noL-o-ne worn Then he saw atskaho'ritse She is making a basket, we'lin their eyes." ts 'e'xwar "Here!" qolo- 'OL ' Very well, I will travel there." i'-ki-coot "I will go a p-new iinil si-e. qemq o more ! " " It seems a person has sat down indoors." " Catch him wi'ctu '-co'neqw Tims he did. nu'mitsy-u-nne'p he ate everything. meL-ho" pke'tso he started from. ha'k wu c le'yole'k u c he found, gartersnakes, BUZZAED. qetqwo'leg' 6'llekwoL He was still a person. pule'ku Down stream (i.e., north) tu-wictu-me'g' wometso' 'L From there he came. Lmeye'pir rattlesnakes, wictu-hoco' 'n Thus he used to do, werherpqe'rixqe'rni at the head of the river tuo'mmmitsyu Whatever Lqer'wer'c salamanders, Lqwer'ter'q w c kitome'nnik ki-nuwoce'g'onnawoni tu'-na"p frogs, of every sort different kinds he ate. niki'tsyu co'k tu-na'p emcik 'i-wena" ki-uqu'rqL Everything he ate, and also those the foetuses qe"yur tu-qol-o'nneu wo"pu i'nillolik kina'x many that he saw into the river that were thrown. Those ki-wer' 'erg' eriyerwerni we 1 n,tsa' u kc umeyo'maik the young girls, women, pregnant, i' 'mi-ckewo'k-ci' '-meL ki-qo'lic-uqo' 'mtsu' 'mek ku '-umeyo'moik because they did not like anyone to know that they were pregnant, wit wo'xpuk-e olo't'-ni' ku-u'kc kits-qo 'lco" then into the river threw them. The child il.-a