The given forms of the Serian digit “six” are evidently mere variants of a common original, which seems quite naturally to have been composed of the stem -apka of the numeral “three”, and of both a prefix and a suffix. The prefixes, for there are two, are, to judge from the one in imapkasho, demonstrative in character. It may be compared with im- in imk´, “he”; imke, “that”; imkove, “they”; imki, “that”, in which it appears to be a directive prefix. And the initial n- and sn- may be cognate in origin. But the final -sūk, -’schoch, -shoχ´, -sho, and -shroj, according to the audition or otosis of the collector, must mean “repeated, doubled, again”, etc., or an equivalent. Hence, the Seri number “six” would be literally “three repeated”.

In the Yuman column at least eight different elements are involved in the formation of the digit “six” in the several dialects of the group. The digits “two” and “three” compose the larger portion of the forms, resulting in such outlines as hamhoke, hoomahook, humhoke, humhóque, χemχúk, kumhok. Hamok (10), “three”, is a characteristic form of this digit, and ẖooak (23), habick (4), and kuáka (19), óak (14), uake (2), are characteristic outlines of the digit “two”. Compare these two lists. The final -k; of the numeral “three” is elided in composition, as it is merely a predicative element, as has been indicated in discussing the Yuman digit “three”; hence, ham- or hum-, symbolizing “three”, with the suffixion of such forms as ẖooak, huáka, or uake, “two”, readily becomes humnhoke or hamhoke, literally “two threes”. In such forms as geshbe (2), despé (18), and niugushbai (14) there occurs a common element -shbe, -spé, or -shbai, which evidently signifies “added, over, plus”, just as -eleepai does in m´sig-eleepai (23), “six”, literally “one added, one more than”. The ge- or -g- in (2) is evidently the final g of the Kiliwi form of the numeral one, mesig, m´sig, which may have at one time been the digit “one” in the Tonto (2); so that geshbe or g-eshbe stands for an earlier měsig-eshbe, “six”, literally “one added (to five)”. The term de-spé is evidently a contracted form of siínta-spe, “one added”, as the other similar forms show. Compare ta-sbe-k (10) and siínta, (9) and siyinta (6), in the last two of which the suffix is wanting or at least overlooked by the collector. In ic̲h̲kyum-kabiak (I) the digit kabiak, “three”, occurs, so that ic̲h̲kyum, must mean “repeated, again, iterated”, just as it was shown in the remarks on the digit four. Now, the form maike-sin-kenaich is, perhaps, an ordinal and not a cardinal. The initial maike- signifies “more, over, added, plus”, the final -kenaich is the doubtful part, and the middle portion -sin- is a contracted form of sinta, siínta, “one”, as may be seen in the list of the Yuman forms of the digit “one”. One other form remains to be considered. The Diegueño (14) of Dr Loew has niu-gu-shbai (the syllabication is the writer’s, showing the elements of the combination). An examination of the digits “seven”, “eight”, and “nine” reveals the fact that the initial niu- has the value of “added, over, plus, in addition to”, five. But it has been seen that the ending -shbaí has a like signification. The only reasonable explanation of this anomaly is that like the Tonto (2) g-eshbe, it owes its origin to the term represented by the Kiliwi měsig; and, moreover, it seems to be a dialectic loan-word. If the term geshbe (2) was adopted as meaning six, supplanting, it may be, an earlier form like hamhoke, the force of analogy, to assimilate this to the other forms, namely, of “seven”, “eight”, and “nine”, would affix the regular dialectic prefix niu- (or nio-). These explanations and analyses of the diverse forms of the numeral “six” reveal no relationship between the Serian and the Yuman groups.

SEVEN

It is evident that the forms of the Serian digit “seven” are variants from a common source, and it is equally apparent that the numeral “two” is the basis for the term. The several examples of this numeral are ghá`kum, kahom, kaχ´kum, kookχ´, in which the final -um, or -om appears to be a suffix; in the term for “twenty” Professor MeGee writes ŭntçkō´k; in which the final kō´k is the term denoting “two”, and in which the final -um or -om is wanting, which probably indicates that it is a flexion. Now, it is seen that this numeral “seven” terminates in the syllable -wūū, -ue, and -ui, in direct contrast with, the termination of the digit “two”. The material at hand is too limited to determine whether this final syllable should be -wūū, -ue, -ui, or -kwūū, -kue, -kui. It apparently signifies “added, over, plus”, or some equivalent term. To attain economy of utterance the term denoting “five” was omitted from the original statement, “two added to five”, as the expression of the number seven, and so “two added” became the name of the number “seven”. An initial tom, tum, tŭn, or diŭn occurs in the names for 7, 17, 70, and 700. An evident derivative from the name for “hand”, it denotes “five”. It is a cognate of ŭnt in ksókhŭnt “nine”, literally “four-five”, and also with tanchl in Mr Bartlett’s numbers 12-19; the correct form for “seven”, it would seem, should have been tan`l kaχkue, etc., “five-two-added-on”; its initial t is identical with thet in t-aul (t-anl?), “ten”. The difference in the endings of this prefix—the difference between an m and an n—may easily be explained. In the several vocabularies it is seen that one collector fancied he heard an m sound, while another, equally careful, heard an n sound. The fact appears to be that it is an obscure nasal sound, which may readily be taken either for an m sound or an n sound by the heteroglot. In Bartlett’s list of numerals tan-tasó-que signifies “eleven”, wherein tasó- is the numeral “one”, as given by both M Pinart and Sr Tenochio, tan- the prefix under discussion, and -que the suffix mentioned above, which was regarded as signifying “added, more, plus”.

The first eight terms of the Yuman list are clearly modified forms of a single original combination, which is apparently still retained nearly unchanged in the Yavapai (18) of Corbusier, hěwakě-spé. The signification and function of the final -spé have been discussed in the remarks on the probable derivations and meanings of the Yuman names for “six”. The given conceptual element is evidently the term hěwakě-, “two”. And -spé, as has been ascertained, signifying “added, more, plus”, etc., the expression literally means “two added”, i. e., to five, which is here understood, but unnecessary, since “two added” has acquired the meaning “seven”, originally expressed by the entire proposition. The Kiliwee (23) term ẖooak-eleepai, “seven”, has literally the same meaning as the terms last under discussion. It will be seen that the conceptual element is the term ẖooak;, “two”, which is only another form of hěwakě, treated above. Now, it is mathematically certain that if “two” be an element of the concept “seven”, it must be added to some preceding number that will produce the result sought, and this number is of course five. So it is presumptively certain that the element -eleepai must mean “added, laid onto, superadded, subjoined”. The Hummockhave (8) maik-kewik-enaich is composed of the conceptual element kewik, “two”, the prefix maik- meaning “more, over”, and the suffix -enaich (or -kenaich), which seems to be an ordinal or distributive flexion. So that “two over, added”, is here likewise the expression for the numeral “seven”. The next form, the Diegueño (14) of Dr Loew is another example of the use of the numeral “two” with different flexions, to express the number “seven”. An examination of this Diegueño list of numerals shows that in such a form as nio-khoak, “seven”, the initial nio- is a prefix signifying “added, in addition to”, etc., while the khoak is a form of the numeral “two”. The next ten forms, while apparently derivative from a common source, are difficult of explanation from the material at hand. The same may be said of the last four, three of which are evidently cognate and are very probably shortened forms of the original represented by the first group in the list. Take, for example, a form like (22) hawake-zpé, and drop the final -zpé, as is done in some of the terms in the “eight” list, and also the initial ha-, and the result is a form wake, which in the dialects (6) and (9) would become viiga, víka, which is the form of the digit “two” in these dialects. The form (7) bee-eeka is also merely the digit “two” of this dialect without any index to show that it is not “two” rather than “seven”. The same thing is to be noticed in the Serian lists, in which the form for thirteen is in all respects the same as that for the numeral “eighteen”, both apparently meaning merely “three added”.

EIGHT

The Serian numeral “eight” is expressed by two different terms. The first is based on the numeral three, and the second on the digit four. The former is the remaining factor of an original expression which signified by uttered elements “three added to five (=the full hand)”, but the need for economy of expression led to the suppression of the uttered element denoting “five”, as soon as the shorter “three added” acquired the usual signification of “eight”. The basis of the digit is kō´pka or kapχ´a, “three”, with the suffix -kwūū (-kχue, -que), presumably denoting “added, plus”. This represents the usual method of forming this digit. The second term, kshoχolka, is that which is presumably based on the numeral “four”. This is the form given by M Pinart. But Sr Pimentel, citing Sr Tenochio, writes this osrojoskum, which at first sight appears to be quite different from the other; yet the r of the latter evidently stands for a modified χ and the j for a χ, and making these substitutions the term becomes osχ´oχoskum, which is approximately the form in which Professor McGee and Mr Bartlett wrote this digit in the numeral “eighty”. Now, it is self-evident that if the element “four” constitute a factor in the combination denoting “eight”, it must be added to itself by addition or multiplication, and the result will be the same in either event. The final -olka appears also as -otkŭm, -olchkom, and -oskum, in these Serian vocabularies, either in the numeral “four” or its multiples. The origin and signification of this ending are not clear; but taking into consideration the great variations in the spelling of its recorded forms, especially in so far as the consonant sound preceding the k-sound is concerned, it may not be presumptive to adopt the s-sound (though sχ´ may be more correct) as that which represents approximately at least the true sound, for it varies from l, t, lch, to s. And it has been seen that the final -um is a flexion denotive of serial or consecutive counting and so not a part of the stem. Then it is seen that -s-k- (the last two hyphens representing uncertain vowels) is the termination requiring explanation. Now, it is probable that this termination is identical in meaning and origin with the -sūk, -shoχ, -sho, -schoch, and -shroj (= -shχ´oχ) terminating the forms of the digit “six”. If this identification be correct (and there is no present reason to doubt it), it signifies “repeated, again, duplicated”, as was suspected and stated in the discussion of the forms of the numeral “six”. So granting this derivation to be correct, kshoχolka, then, signifies “four repeated”, which of course denotes “eight”.

In the Yuman list, the first eleven forms are evidently composed of the numeral “three” and a suffix signifying “added, plus, more than”, but the last three of the group want this suffix, a fact due perhaps to the fault of the collector rather than to linguistic development. The terminations -eleepai and -shbe-k and its variants have already been explained when treating of the numeral “seven”. And the twelve forms beginning with chip-hoke (16) are variants from a common original composed of the numerals “two” and “four”. It will be readily seen that chip- in such a form as chip-hoke is a contraction of a form such as tchibabk (14), “four”, chepap (24), “four”, as may be seen in the Yuman list of terms for the digit “four”. Now, the next portion of the term is -hoke, which is but a slightly disguised numeral “two”, as may be seen by reference to the schedules of the numeral “two”. Compare ẖooak (23), huáka (19), uake (2), and hěwáki (18), all signifying “two”. Now, the next term, maike-homok-enaich (8), is a combination of maike, “above, over, more than”, homok, “three”, and the ending -enaich (or -kenaich), which may be either an ordinal or a distributive flexion. The form nio-khamuk (14) is a combination of the prefix nio-, signifying “added, above, or more than”, and the conceptual term khamuk, “three”, the expression signifying “three over, or added to”. The next two examples are evidently irregular, if not spurious. The form pakaikhin-awach is composed of pakai, “seven”, khin-, “one”, and the suffix -awach, “added to”. Now, the last, the Cochimi nyaki-vamivapai, appears to be erroneous. It contains the term nyaki for ginyaki, “hand”, but the remainder of the expression is composed of elements that are not comparable to anything in the meager material at present accessible. The Serian and the Yuman terms herein show no relationship.

NINE

The first three Serian terms for “nine” are evidently forms of a common original, signifying “four added to five”. It is evident that ksō´kh- in (A) ksō´kh-ŭnt is the same element as -ksō´k in ūnçtksō´k, “forty”, and -kschō´k in ŭnz-untçkŭkschō´k, “400”. The element -ŭnt here is a name for “five”. Its literal meaning is “hand”, which may be gathered from the following citations: ŭnol´k=“hand”; mĭ´noŭl´t=“arm”; ŭnulte-mŭ´ka`p=“middle finger”, in which ŭnulte means “finger (or hand)”. These are from the vocabulary of Professor McGee. Then M Pinart records innolχ´, “arm”, intlash “hand”, inol’tis, “finger, index finger”, inol’tip “ring finger”. And Mr Bartlett writes inoyl, “arm”, inossiskersk, “hand”, inosshack, “fingers”. This -ŭnt will be further treated when the numeral “ten” is under discussion.

While it is evident that the first eight forms of the Yuman list are but variants from a common original, it is not, however, so clear what the original signification of the combination was. But as there can not be any question of relationship between these and the Serian terms, this fact will not affect the result of this study. The next terms of the Yuman list are variants of an entirely different combination of elements. The forms (15) humhum-môck and (12) humhamóok may be taken as characteristic of these terms. Now, it is plain that there is here duplication of the stem hum- or ham-, “three”, making the literal sense of the combination to be “three threes”, which of course gave the required meaning. The Cochimi (23) m’sigk-tkmat contains the element m’sig, “one”, and the final tkmat, which appears to mean “lacking, wanting, or less”. And in the Diegueño (14) nitchibab for niotchibab a still different method of expressing “nine” is found. In discussing the numeral “seven” and “eight” the signification of the initial nio- was ascertained to be “added to, over, plus”, and tchibab is of course the numeral “four”. The original expression, then, was “four added to five”, producing the required number, “nine”. The next three forms, though evidently cognate, are, like the first group, not analyzable from the data to be obtained from the meager material at present accessible. The last form is doubtful. These analyses show no relationship between the Serian and the Yuman terms.

TEN

The Serian forms of the numeral “ten” are apparently cognate, being composed, it would seem, of the same elements. Thus they are mere variants of a common original expression, signifying, literally, “two fives”, or what originally was the same thing, “two hands”.

The element khóh- in (A) khólnŭut’ represents ghá‘k:(kha‘k) or kō´k, as it is also written, signifying “two”, and -nŭt’ is the slightly disguised name for “hand” and “finger”, being also transcribed as -nachtl, -nalχ´, -nlχ, and lastly -aul. Compare these carefully with the words denoting “arm, hand, finger”, in this language, and it will be seen that the spelling of khóh- varies in the several vocabularies from khóh-, ho-, χo-, to ka-, respectively. The derivation of the t, or rather tä, in taul of Sr Tenochio, is not evident, but seems to be cognate with the prefix tom-, tum-, tŭn-, or diŭn-, already noticed, making taul thus signify “five added”, i. e., to five, and so producing “ten units”. Such seems to be the evident resolution of the Serian names for the numeral “ten”. But taul may have been miswritten for ta-an`l.

The first four terms of the Yuman list are plainly based on the numeral “five”, expressed by sarap. The form raphawaich (8) is evidently a shortened form of saraphowwaich, literally “two fives”, or, what was the same thing at the beginning, “two hands”. The first term, sarap, signifies “five, finger”, denotively, but its literal or connotive signification is “entire, whole, full, complete, collectively”, a meaning which was suggested in the discussion of the numeral “five”. And howwaich is the form of the digit “two” in this dialect.

The next nine forms are so contracted, irregular, and, perhaps, miswritten that an analysis of them is a matter of doubt and difficulty, but the following ten terms are cognate and signify “two fives (hands)”, or, denotively, “ten”. In the comparative list of names for the “arm, hand, finger”, etc., shah, shawas, shawarra, and eesarlya are a few of the many variants of säl, “arm, hand, finger”, etc. So, in such a form as sahhoke (3) the sah is the name for “hand” and hoke is the numeral “two”, the combination signifying “two fives, hands”, or “ten”. The other nine terms are but variants of the original of this compound. In selgh-iamát (14), selgh for isalgh is the element denoting “hand”, or “five”, while iamat means “added to, upon, over”, there being the subaudition of the element denoting “five”. Hence the original combination meant “five added to five”, or “ten”. This is a strict application of the quinary system.

The Kiliwee term, chepam-mesig (23) signifies literally “one chepam”. If reference be made to the “five” list, it will be seen that there sol-chepam signifies “five”, or, to be exact, is the translation of the term “five”. Now, the element sol- of this compound is a variant of esal, “hand”, while chepam, judging from analogy, must signify “the whole, entire, the complete”, collectively “all”. Moreover, the Kiliwee terms for “fingers (dedos)” and “toes (dedos del pié)”are salchepa and emechepah, respectively, wherein the element chepah is added to esal, “hand”, and to eme, “leg”. Hence it may be inferred that chepam-mesig signifies “one complete count of all the fingers”, and so “ten”. The next is Cochimi, in which naganna means “hand”, and the last term (I) appears to be miswritten. It will be seen from these partial analyses of the names for the digit “ten” that there is no linguistic relationship between the Serian and the Yuman terms.

ELEVEN

The only Seri example of the numeral “eleven” is that which was recorded by Mr Bartlett, who writes it tan-ta-só-que, instead of tan-tasó-que, which exhibits the component elements of this compound. This expression signifies “one added to, or, over, upon”. Its conceptual base is the numeral tasó, “one”. The initial tan- has already been discussed while treating of the numeral “seven”. It was there made a cognate of the initial tom- or tum- of the several examples of that digit, and likewise of tanchl in Mr Bartlett’s numbers 12-19. It would seem that the correct form for “eleven” should be tanchl-tasóque, i. e., “ten-one-added-on”. Where “hand” is the name for “five” and is an element in the name for “ten” there arises confusion, unless there is marked difference between the two expressions.

In the Yuman list the first fourteen examples of the numeral “eleven” have some form of the digit aséentik (sita, siti, sint, shiti), “one”, as the dominant element in the expression, while the elements denoting “added to, more than, plus”, are severally as follows: in the first -nitauk, in four others a variant of -giala, in five others the prefix maga- (umaiga, emmiá, mae); while in some such a flexion is entirely wanting, probably, at least in a majority of the forms, because of misapprehension on the part of the several collectors rather than the abrasion of use. But in mesigk-malha (23) mesigk denotes “one”, and malha “plus, added to”. In the form nie-khin (14), khin signifies “one”, and the prefix nie-, “plus, added”. It will be noticed that the flexion maga (umaiga, mae, emmiá) is a prefix to the element “one”, and so when shahoque, “ten”, is expressed as in (4) it stands between the two notional terms. But in (8) neither “ten” nor an element denotive of addition is expressed.

TWELVE

The only known example of the Seri numeral “twelve” is that which was recorded by Mr Bartlett. He has apparently misapprehended its true pronunciation, for he wrote tanchl-to-que instead of tanchltakahque or tanchltakochque. In his orthography kahom signifies “two”, but the final -om is employed only in serial counting, so that kah- is the stem, which is only a variant of koch in eansl-koch, “twenty”; and tanchl signifies “ten”.

In the first six examples of the Yuman list the element “ten” is not expressed, but only some form of the numeral “two”, with a suffix denoting “added to, over, more than”; in the next three the flexion of addition is prefixed to the element “two”; and in the next two, (19) and (2) respectively, the element “two” is immediately preceded by the very abbreviated and perhaps misapprehended forms of the numeral “ten”; in the next a very questionable form is recorded, for it appears to be an attempt to form a compound signifying “two times six”, but without accomplishing the purpose; yet it may be miswritten for nio-khoak-ěshbe, in which khoak is the element “two”, with a doubled sign of addition, namely, the prefix nio-, already explained, and the suffix -ěshbe, also explained above. In the next two the element denoting “ten” is expressed, with umai-javíc and maga habick as the second part, both meaning “two added”. The last (8) vaike is a highly modified and probably misapprehended form of an earlier havik-ěsbe, “two added”, with a subaudition of the numeral “ten”.

TWENTY

The four examples of the Serian numeral “twenty” are merely combinations of the terms kō´k, koch, kookχ´ and jaukl (for χaukl), all cognate forms, meaning “two”, and the forms ŭntç, eansl, kanlχ´, and taul, all cognate and signifying “ten”.

The Yuman expressions denoting “twenty” are all, with two exceptions, combinations the dialectic elements denotive of “ten” and the forms of the numeral “two”, which have been treated elsewhere in their proper places. The two exceptions are (III) the Cochimi, which signifies “all the fingers and toes”, and (21) the Santa Catalina, which here presents what appears to be a new term for “ten”, for the final word howŭk is the numeral “two”. These analyses do not show relationship between the Serian and the Yuman terms.

THIRTY

FORTY

FIFTY

Comparative Lists of Serian and Yuman Conceptual Terms

Those philologists who have classed the Seri tongue as a dialect of the Yuman stock have laid great stress on the alluring phonetic accordance, supposedly indicative of genetic relationship, between the Laymon (and probably Cochimi) tamá or tammá, “man (homo)”, and the Serian kŭ´tŭmm, ktam or eketam, possibly of the same signification—i. e., “man (homo)”, rather than “man (vir)”; but the accompanying comparative list of vocables purporting to denote “man (homo)” discloses the significant fact that tamá (tammá) belongs only to the Laymon, and (probably) the Cochimi dialects. In Mr Bartlett’s Cochimi record, he wrote delmá, “man, hombre”, and guami (Spanish g), “husband”—that is, “male person”. From certain Laymon texts with interlinear translations in Buschmann’s “Die Spuren der aztekischen Sprache”, etc., the following forms of the vocables in question have been extracted: tammá, “man (homo, Mensch)”; tamma-butel, “this man”; uami-butel, “this man, this male person”; wami-jua, “man (vir, Mann), male person”; wakoe-butel, “this woman”; gui-wuctu-jua, “his woman”; whanu, “small, young, a child”; whanu-wami-jua, “a small, or young, male person”, perhaps “a boy”. Now, wanju or wanyu, “young”, wáhki, “woman” (-aki in wanju-aki, “girl”—i. e., “young woman”); ouami, “(my) husband”, correctly, “(my) male person”; ouĭqua, “(my) wife”, evidently a form of wáhki, “woman”, are all Cochimi vocables. Dr Gabb, in his Cochimi vocabulary, did not record the presumptively correct term denoting “man”; for the word which he has written, wanyuami, and which he has translated “man”, really signifies, “young male person”, rather than “man (homo)”. This is unfortunate, because in Mr Bartlett’s Cochimi, delmá is rendered “man (homo)”, and the Cochimi of Padre Clavigero has tamá, “man”, and the Laymon, tamá, tammá, or tämmá, “man”, and there is seemingly no absolutely satisfactory method of ascertaining whether the l of Mr Bartlett’s delmá, “man”, is genetic or not. But as the Laymon and the Cochimi are apparently cognate dialects, it is probable that the form delmá of Bartlett’s Cochimi and the tamá or tämmá of the Laymon and the Cochimi of Padre Clavigero are cognate vocables. The part of the terms which the two dialects have in common is the final and usually accented -má; in other words, -má is the common conceptual element in the vocables delmá and tamá. This of course rests on the presumption that tamá and delmá are compound terms, having probably genetic relationship. The following facts may aid in discovering the lexica constituting the elements of the two words in question, and these, it is seen, are -má, del-, and ta-. In Dr W. M. Gabb’s record of Cochimi words, collected by him in the vicinity of San Borja and Santa Gertrudis about the “center of the peninsula” of Lower California, the term “Indian” is represented by maha-ti, and “people” by maha. On the same schedule with the Cochimi Dr Gabb recorded a vocabulary of the Kiliwee, dwelling 150 miles “further north” at and near San Quentin. In this dialect, which is Yuman, the word “Indian” is rendered by kimai, and “people” by meẖa-le (preferably meχale335). The apparently genetic accordance between the Kiliwee word for “people” and the Cochimi terms denoting “Indian” and “people” is brought into stronger light by a comparison of the terms for “warrior”; in the Cochimi, mac̲h̲-karai (maχ´-karai), in the Kiliwee, maẖk-pkátai (maχk-pkátai). The unquestioned kinship between these two dialects warrants the inference that these two compound expressions, denotive of the same thing and possessing at least one common element, maχ- or maχ´-, must accord approximately at least, in the signification of their heteromorphic constituents.

In the Kiliwee pah-kute signifies “a chief”, from e-pa, “Indian”, hence “man” (primitively) and kute for (k)e-tai, “large, great”, hence “old”, found in such expressions as sal-kootai, “thumb”, literally “large finger”, and pah-tai, “old”, but literally “old man”. So the name for a chief may be rendered freely “the elder person; the old man (the wise man)”. The Cochimi term mac̲h̲-ka-é, as written by Dr Gabb, denotes “far”, while mac̲h̲-i-kang-i-n̲g̲a means “near”. These vocables may preferably be written thus, maχ´-kaé and maχ´-kañ-iña. The ending -iña is a privative flexion or suffix in Cochimi, forming derivatives with meanings directly adverse to those of the primals; so the literal signification of maχ´-kañ-iña is “not far”, hence “near”; but in maχ´-kaé the final -kaé is the adjective “large, great”, having here an intensive function signifying approximately “more”, while maχ´- is evidently a form of the proximate pronominative found in the terms “thou” and “ye” in this group of languages. In the Laymon kahal ka, “water large (is)”, for a “sea or stream of water”, ka signifies “large, great”; and the Cochimi kättenyi, “few, not much”, is literally kätte- for (k)etai, “large, great, much, many”, and -iñi the privative denoting “not”. And the Laymon metañ, “many, much”, is evidently from m- for ma (a proximate pronominative), eta for the Cochimi etai, “large, great, much, many”, and the final -ñ. Compare Bartlett’s modo, “all, todos”, and modoliñi, “many, much”. Such are some of the forms of the adjective signifying “great, large, much, many”. There is also in the Cochimi an intensive pa, ibal, ibá, which signifies “very”. This explains the presence of the p- sound in the term maχk-pkátai, the Kiliwee for “warrior”.

It has thus been shown that a probable connection exists between the Cochimi terms maha, “people”, and maha-ti, “Indian”, on the one hand, and the maχ-, inferentially signifying “man” in the Cochimi and Kiliwee names for “warrior”, maχ´-karai and maχk-pkátai, and the meχa- in the Kiliwee meχa-le, “people”, on the other. The significance of the initial ta- in tämmá (tamá, tammá, tamal, tammalá) seems to be that of a definitive pronominative; it is found in the Cochimi of Dr Gabb and in the Laymon. Dr Gabb recorded in his vocabulary ta-ip, “good”, but ta-ip-ena, “bad”, the final -ena being the characteristic Cochimi privative suffix; elsewhere written -iñi. So it would seem that the stem is -ip, meaning “good, desirable”. In Kiliwee aχok (Dr Gabb’s aẖok) signifies “flesh, meat”, while aχok-m-gai denotes “deer”, literally “good, desirable meat”, in which m-gai signifies “good, desirable”; it is probably connected with the term ka, “great”, and its variants noted above, and so may also denote “abundance”. Under the word “love” Dr Gabb has m’gai-yip, the free translation of which should read “greatly desirable; abundantly good, well”. Thus -ip, or -yip, signifies “desirable, good, pleasing to the sense”; in Laymon likewise the initial -ta is sometimes wanting, as in wayp-mang, “good (is)”, as distinguished from tahipo-mang, “good (is)”. The final -mang (=mañ) is a term apparently denoting “to exist, to live”, and is possibly cognate with the má (Kiliwee me) in the words discussed above.

This, it would appear, is the origin of the má in tamá, “man”. The individual character of the initial ta is suggested in what has already been said in reference to its absence from such vocables as wayp-mang and m’gai-yip, in which the wayp and the yip are identical with the ip in ta-ip, “good”. This term ta appears as the relative “that” under the form te. It also appears as a prefix in the Cochimi and Laymon numeral “one” and in the adjective te-junoey, “a few”; also in the adjective de-muejueg, “all”; and again in the peculiar numeral “one”, namely du-juenidi.

Such appears to be the analysis of the Cochimi and Laymon tamá, “man”. The form of it recorded by Mr Bartlett, del-má, “man”, compared with his de-ma-nsú, “Indian”, is seemingly a valid confirmation of the foregoing derivation, because this l in de-l-má is probably identical with the final l or lá in tama-l and tamma-lá, “man”, cited above. In the Cochimi for “water”, ca-l, its true character is partly seen; cal oso signifies “river”, but in caa-pa-l (Gabb’s kaχ-pa-ra), “sea”, it becomes a suffix, the element pa signifying “much, great”, and Dr Gabb’s form shows that in the dialect he recorded its form is ra; again in cal ka, “lake”, literally “large water”, it is a suffix. It appears again in Mr Bartlett’s del-mag, “light”, as compared with Dr Gabb’s ma-ahra (=maah-ra), “fire”; it appears evident that the mag of del-mag and the maah of maah-ra are cognate, so that de-l is here found as a prefix, as it is in Mr Bartlett’s de-l-má, “man”. Thus it is that delmá and dema-nsú, “Indian”, of Mr Bartlett and tamá and tammalá of Hervas, Duflot de Mofras, and Miguel del Barco are cognate.

It accordingly appears that the assumed linguistic relationship between the forms discussed above and the Serian kŭ’tŭmm (ktam, tam), “man”, is very improbable, because there are no evidences nor data indicative that the Serian forms have had a common linguistic tradition with the Cochimi and Kiliwee forms discussed above. It seems proper, therefore, to reject such assumed relationship between the Yuman and the Serian vocables in this comparison.

The comparative list of names purporting to signify “woman” in both the Serian and the Yuman tongues reveals not a single phonetic or lexic accordance that may even suggest linguistic kinship between the two groups of vocables.

The comparative list of terms purporting to signify “people” and “Indian” in the Serian and Yuman groups of languages exhibits, in a manner similar to those already examined, the same decisive lack of phonetic accordance between the vocables compared.

This comparison of the Seri and Yuman terms for “head”, to ascertain linguistic relationship, seems barren of any but a negative result. It is true that there is an apparent resemblance between the Seri and the Diegueño terms, and a still more doubtful one between the Seri and the Kutchan. It is significant that the twenty-odd other Yuman dialects employ for “head” an entirely different term. The kinship of the Seri term to either the Kutchan or the Diegueño is therefore nothing more than a possibility, and it seems safe to reject it. The phonetic discordances, and the fact that there has been no evidence adduced to show that the Diegueño term was ever prevalent in the other Yuman dialects, warrant this rejection.

The following analysis may be of service here. A careful comparison of the Diegueño terms for “head”, and “hair” indicates that the form (14) ilta, “head”, is very probably a shortened khalta, “hair”. In the Diegueño, Santa Isabella, and Mesa Grande vocabularies Mr Henshaw recorded several names for “hair” and “head” which may serve to aid in the explanation of the words in the following comparative list. In his Diegueño record lěmĭs and lĭmi, variants evidently of a common original, stand for “hair, feathers, skin, and fish scales”, as in the entries haltau lěmĭs, “rabbit skin”, kasau lěmĭs, “fish scales”, kŭkwaip lěmĭs, “deerskin”, lěmĭs, “feathers” and “hair” of animals; and also yiu-lěmĭs, “eyebrow”, literally, “eye hair”, and ā-lĭmi, “beard”, literally, “mouth hair”, in which yiu for iuu means “eye” and ā for yau, “mouth”. In his Mesa Grande vocabulary, Mr Henshaw recorded h`lta for both “head” and “hair”; in his Hawi Rancheria vocabulary he wrote mă-whl for “head”, and h`lta for “hair”; and lastly, in his Santa Isabella record hŭsta means “hair”, hŭsta-kwarŭr is written for “head” (literally, “hair skin”, meaning “the scalp”); and ŭstú-kŭmō is rendered “skull”. Thus, h`lta, lěmĭs, and hŭsta are terms denoting “hair, fur, skin, feathers, and fish scales”. Yet it is possible that hŭsta is a softened and ill-pronounced cognate of h`lta. In Corbusier’s Yavapai vocabulary “eyebrow” is written yuh-kělěme, and in Dr White’s Tonto word list yŭ-gŭlma, both signifying literally “eye hair”. It is apparently safe, therefore, to regard the element -kělěme or -gŭlma of these two dialects as cognate with the lěmĭs (lĭmi) noticed above. In his Mohave record Mr Corbusier renders his entry himíç, “eyebrow”, literally, “eye hair”; and in the H`taäm or San Tomaseño by Dr Gabb “beard” is written āh-lamĭse, literally, “mouth hair”. “Hair” is written helt’h-yee-mōh, seemingly “head hair”, for “forehead” is rendered by het’l-ŏmȳ, in which helt’h- or het’l seems to be the term denotive of “head”; but in Lieutenant Mowry’s Diegueño this term, which is there written hetltar (for hetltā) signifies “hair”. In Ten Kate’s Maricopa, “beard” is written ya-womis, literally “mouth hair”, -womis being clearly a variant of himiç, which is but a variant of lĭ-mĭth and of -kělěme noticed above. In the Santa Isabella, Mr Henshaw wrote “feathers” lĭ-mĭth.

COMPARATIVE LIST OF DIEGUEÑO AND OTHER YUMAN NAMES FOR “HEAD”, “HAIR”

It seems clear, furthermore, that iltá (14) is merely a curtailed example of khaltá (14), for it is clear that this iltá is a cognate with the h`lta (27), the initial h`-sound of which, Mr Henshaw says, represents a rough guttural utterance (represented herein by the character χ). In (27) of the comparative list h`lta, expresses both “head” and “hair”, thus completing the circuit and making iltá cognate with khaltá, since it is plain that h`alta (χalta) of 24a, hlta of 26, and h`l-ta of 27, the initial sound in each being, as shown above, a rough guttural are related to khaltá. The term hu-ch’lmo (24) is a compound of hu-, “head”, and -ch’lmo, an evident cognate with the element -gŭlma or -kělěme ( =kělěmĭs) noticed above, denoting “hair”; hence, the combination signifies “hair of the head”. In like manner the H`taäm or San Tomaseño form (17) ẖ’lemo may be explained. In this dialect ẖo (=χo) signifies “head”, and an original ẖolemo (=χo-lěmĭs), signifying “hair of the head”, became contracted to the form in question, namely, ẖ’lemo. In the Santa Isabella record of Mr Henshaw hŭsta signifies “hair”, but hŭsta-kwarŭr is given for “head”, while ŭs-tŭk-ŭm-ō is translated “skull”; the last expression should have been written (h)ŭstŭ-kŭmō. Under the caption “robe of rabbit skins”, h`kwĭr is found, but under “skin” in “Parts of the Body” of his schedule, `nyakwăt (26) and n’kwěr (25) are found, both meaning “my skin”; Corbusier’s Mohave record has himát-makwil rendered “skin of man”, but meaning “skin of the body”, himát signifying “body”, and makwil, “skin”. The Mesa Grande term for skin is given as lĭmĭs, a vocable which has already been discussed. So it must be that the foregoing hŭsta-kwarŭr signifies “skin of the hair” or “skin of the head”, if hŭsta is also a synonym for “head”. The final -ŭr in the compound in question is due to the misapprehension of the rolled or trilled r-sound with which the term for skin terminates. The element -kŭmō of the vocable (h)ŭstŭ-kŭmō, rendered “skull”, is also a factor in the Diegueño terms for “head” in numbers (15) and (16) of the comparative list; so that it is highly probable that these terms signify “skull” rather than “head”. And, lastly, it is equally probable that the expression (18) kumpaiya kûwâwâ signifies “hair of the whole head (skull)” rather than “head” only; for the initial kum- is presumptively the cognate of the forms -cumōh and -kŭmō, denoting in the compounds already noted “skull”, while -paiya signifies “all”, and kûwâwâ “hair”. There appears to be a relationship between the terms for “head” and “hair” in (12b) oomwhelthe, “head”, (3) amawhach and mowh´l, “hair”, and (26) mă-whl, “head”. The explanation of the term hu-lchsta, (15), denoting “hair”, is probably to be found in its resolution into hu (χu), “head”, and lchsta for a form of hŭsta, “hair”, discussed above; the term signifies, therefore, “hair of the head”. In like manner huch´lta (24), rendered “head” there, seems rather to mean “hair of the head”, by its reduction to hu, “head”, and ch´lta, for a form of khalta (= χalta), “hair”.

The Serian variants of the term denoting “head”, are respectively (A) a^hleht. (B) ih´lit, and (C) ill´it. These forms certainly have no kinship with the Yuman terms discussed above; they have a totally alien aspect. The Serian terms for “hair” are respectively (A) a^hleht, (B) ina (“feather” rather than “hair”), (C) ill´it kopt´no, and (D) obeke, and while the last has an aspect foreign to the other terms classed as Serian, none of the vocables appear to offer ground upon which to predicate relationship between the Yuman and the Serian. For a further explanation of obeke turn to the discussion of “tooth”.

The comparative list of Serian and Yuman names for the “nose” reveals no evidence of linguistic relationship between the two groups; but an inspection of the Yuman lists for “head”, “hair”, and “nose”, exhibits a close connection between a number of the names for “head”, “nose”, and “beak, bill”.

Eight of the terms for “eye” in the yuman word lists are ído, hidho, or their variants, in five Yuman dialects, Maricopa, Mohave, Hummockhave, Kutchan, and M´mat (virtually in but three, for Hummockhave is but a subdialect of Mohave, and M´mat of Kutchan), and the remaining twenty-one examples are from an entirely different stem or base which is apparently connected with a verb “to see,” one of the forms of which is eyûuk (4), héyuk (7), and iyó-ok (6); the form ído and its several variants is seemingly connected with iúdo (6), “let us see”, apparently an imperative form, in a manner similar to the connection between yú (2), “eye”, and its variants, and the verb form eyûuk just cited.