More Evidence for Yok-Utian: A Reanalysis of the Dixon and Kroeber Sets

by Catherine A. Callaghan
More Evidence for Yok-Utian: A Reanalysis of the Dixon and Kroeber Sets
Catherine A. Callaghan
International Journal of American Linguistics
Start Page: 
End Page: 
Select license: 
Select License


1. Introduction. In the early part of the twentieth century, Roland B. Dixon and Alfred L. Kroeber undertook a mass comparison of California Indian languages, in hopes of uncovering evidence for deeper genetic rela- tionship among some of the families of languages. Their research resulted in a new classification of California Indian languages (Dixon and Kroeber 1913), for which they later published their evidence (1919).

Penutian was one of the proposed new superfamilies, whose name was coined from pen (similar to words for 'two' in Yokuts, Wintun, and Maidun) and uti (similar to words for 'two' in Miwok and Costanoan). Dixon and Kroeber believed that the Pen languages (Penian) and the Uti languages (Utian) comprised subgroups within Penutian. The Utian subgroup has since been substantiated (Callaghan 1988a; 1992), and there is increasing evidence supporting Yokuts and Utian (Yok-Utian) as a deeper genetic unit (Callaghan 1997).2

This article is a revision of a paper I presented at the SSILA meeting at the Linguistic Society of America on January 11, 1998, in New York City. I wish to thank William Shipley for sharing his Proto-Maidun corpus with me, Alice Shepherd for allowing me to cite her Proto-Wintun material, Kenneth Whistler for letting me copy his Patwin notes, and Kenneth Whistler and Victor Golla for allowing me to copy their Proto-Yokuts reconstructions.

Utian consists of two genetically related families, Miwok and Costanoan. On the basis of lexical items, structural similarity, and sound correspondences, the Miwok languages may be grouped as follows:

I. Eastern Miwok (Mie)

A. Sierra Miwok (Mis)

    Northern Sierra Miwok (Mins)

    Central Sierra Miwok (Mics)

    Southern Sierra Miwok (Miss)

B. Plains Miwok (Mip)

C. Bay Miwok (Miba)

11. Western Miwok (Miw)

A. Coast Miwok (Mic). Coast Miwok was probably a single language with variant dialects.

    Bodega Miwok (Mib)

    Marin Miwok (Mim)

B. Lake Miwok (Mil)

[IJAL, vol. 67, no. 3, July 2001, pp. 313-451 02001 by The University of Ch~cago.All rights reserved 0020-7071/2001/6703-0003$02.00

We cannot overestimate the difficulty Dixon and Kroeber faced in gather- ing their data, often from obscure sources. They should be praised for keep- ing the meanings of words largely constant, a virtue frequently neglected by those undertaking deep comparison today. Unfortunately, Dixon and Kroe- ber's phonological criteria were much looser. They often segmented off the first CV-of a word if the remaining portion showed greater similarity to other words in their proposed set, despite the fact that the Penutian lan- guages have very few prefixes. They allowed unrestricted metathesis, and some words bore little in common besides the position of a single conso- nant. Since the proposed Penutian superfamily quickly became a spring- board for more extended genetic classifications, it is in order to update and re-evaluate the original sets.

The Costanoan languages can be grouped as follows, although the relationship among the Northern Costanoan languages is more like a "Stammbusch" than a Stammbaum.

I. Southern Costanoan (Cos)

A. Mutsun (Csjb)

B. Rumsen (Cru)

    Northern Costanoan (Con) A. Soledad (Csol) (Chalon)? Soledad may be a transitional language between Northern and Southern Costanoan (Okrand 1989). B. Santa Cruz (Cscr) (Awaswas) C. Santa Clara (Cscl) (Tamyen) D. East Bay (Ceb) (Chocheiio). This was apparently a single language with several dialects from Mission San Jose to Richmond.

    San Francisco (Csf ) (Ramaytush)
    Ill. Karkin (Ckar)

I have normalized modern transcriptions. j is [y], y is [i], and raised dot (.) represents length. Voiceless unaspirated stops and affricates are written /p, t, t, k, 1, c, El, and aspirated stops are followed by an h. Aspiration is not phonemic word-finally in Proto-Yokuts or in the first member of a Yokuts consonant cluster, where it is predictable. c is [ts -s] in Lake Miwok and [E -ts -s] in Plains Miwok. In other Miwok languages, c is [El.

Additional Utian abbreviations are PMis Proto-Sierra-Miwok, PMie Proto-Eastern-Miwok, PMiw Proto-Western-Miwok, PMi Proto-Miwok, PCos Proto-southern-Costanoan, PCon Proto-Northern-Costanoan, PCo Proto-Costanoan, and PU Proto-Utian. Also Mm Maidu, MK Konkaw, Mn Nisenan, PM Proto-Maidun, Ww Wintu, Wn Nomlaki, Wph Hill Patwin, Wpcc Cache Creek Patwin, and PW Proto-Wintun.

Yokuts abbreviations are Ypal Palewyami, Yhom Hometwali, Ytu Tulamni, Yyd Yawdan- chi, Ywik Wikchamni, Ychoy Choynimni, Ygash Gashowu, Ychuk Chukchansi, Ydum Dumna, Ychaw Chawchilla, Ytach Tachi, Ywol Wo'lasi, Ychy Choynok, Yyaw Yawelmani, Yfnv Far Northern Valley Yokuts, PY Proto-Yokuts, PYbv Proto-Buena-Vista-Yokuts,PYgen Proto- General-Yokuts, PYnim Proto-Nim-Yokuts, PYn Proto-Northern-Yokuts, PYv Proto-Valley- Yokuts, PYsv Proto-Southern-Valley-Yokuts,PYnv Proto-Northern-Valley-Yokuts,and PYU Proto-Yok-Utian.

2. Methodology. This paper is such an update on the material in the Dixon and Kroeber article (1919) supporting their original Penutian classification. They compared the Miwok, Costanoan, Yokuts, Wintun, and Maidun languages of Central California without first reconstructing forms at the family level, but in compiling their sets, they did cite obvious cog- nates from different member languages of a family. In this update, I have substituted proto-forms whenever possible, i.e., Proto-Miwok (PMi) for words in contemporary Miwok languages, Proto-Costanoan (PCo), Proto- Utian (PU), Proto-Wintun (PW), and Proto-Maidun (PM) when appropriate. In other cases, I have replaced Dixon and Kroeber's forms with modern transcriptions in contemporary languages. Where such transcriptions were unavailable to me, I retained the forms as they appeared in the original article, followed by (D&K), except that I usually did not cite every member of a set of obvious cognates. Occasionally, I substituted the transcriptions of the original sources when they had been altered by Dixon and Kroeber.

My numbers follow Dixon and Kroeber's numeration for the original 171 sets. Higher numbers refer to the additional sets at the end of their section on Penutian. In some cases, one of their sets consists of two or more un- related subsets, which I have indicated accordingly. I follow with an opin- ion about which part of each set or subset (if any) is valid; e.g., D (discard), U (supports Utian), YU (supports Yok-Utian), YU-W-M (supports Yok-Utian-Wintun-Maidun, Dixon and Kroeber's "Penutian"), Y-WM (supports Dixon and Kroeber's "Pen"), etc. For example, I judge that the first set (Person)consists of two unrelated subsets. In the first subset, I sub- stituted Proto-Miwok, Proto-Costanoan, and Proto-Utian forms for those in contemporary Utian languages. In the second subset, I substituted Proto- Yokuts and Proto-Maidun forms for contemporary forms, and I judge that they match; that is, they should be set aside for further consideration. I

Unmarked Yokuts reconstructions are from the Whistler-Golla slip file. Items followed by GG were reconstructed by Geoffrey Gamble, with the level of the reconstruction by the Whistler-Golla system. CC indicates a Yokuts reconstruction by Catherine A. Callaghan. Updated Yokuts items are from Shipley (n.d.).

CS indicates a Proto-Costanoan form reconstructed by Catherine Schambach (1977). San Francisco Costanoan is from Schoolcraft (1860). Additional abbreviations are A Fray Felipe Arroyo de la Cuesta (cited in Mason 1916), CHM C. Hart Merriam, GM Gregory Mengarini, HWH H. W. Henshaw (Heizer 1955), JC Juan Comelias (in Taylor 1860), KW Kenneth Whistler, P Alphonse Pinart, RS Richard Smith, SB S. A. Barrett, VG Victor Golla, and WS William Shipley (1997).

A query indicates that a reconstruction is questionable or inclusion of an item is dubious on semantic grounds. Parentheses frame a related item which may not bear directly on the comparison. Items between square brackets were normalized from old transcriptions.

updated the transcriptions of Dixon and Kroeber's Wintu and Patwin forms, and I judged that they do not match either subset.

In evaluating these sets, I have tried to err in the direction of inclusion, but I have excluded items that share only one phoneme or, worse, one point of articulation. I have disallowed initial expansions, arbitrary prefixes, and apocope of the first syllable unless there is support in the form of internal evidence within the families, especially since there are very few prefixes in these languages, although stem compounding does occur. I have limited metathesis to the interchange of the second vowel and the third consonant, or to plausible consonantal interchanges parallel to those synchronically observable. I have rejected most instances of assimilation, but I have al- lowed what I consider to be plausible sound changes. Of course, I have been much more selective in the case of Utian, or even Yok-Utian, where sound correspondences can be ~tated.~

For example, in set 83, Dixon and Kroeber (1919:70) claim that Miwok ki.k 'water' is cognate with Yokuts ?ilik''water', reasoning that Yokuts 2- is an expansion and that Miwok ki.k has undergone assimilation, presumably from some proto-form of the shape *li(.)k. Of course, I cannot disprove this scenario, but there is no independent reason for assuming it.

On the other hand, when comparing Yokuts nu-?ath 'older sister' (1919:68) with words in Wintun, Miwok, and Costanoan, they properly seg- mented nu-, because Yokuts nu- -no- occurs in other kinship terms; indeed, it is one of the few prefixes in Yokuts. If Yokuts and Utian are ultimately related, there is evidence that this prefix originally meant 'my' (Callaghan 1997:41).

In set 143, Dixon and Kroeber (1919:60) were correct in the segmenta- tion of Csjb nu-p.i 'that one' (note PCo *nu(.) 'that', PCos *ne-p.e 'this one', and Csjb 'that one'), enabling valid comparison with PW *pi *'third- person demonstrative sg. subj.'. They were also justified in the segmentation of Mm he-kot.0 'to break, crack of itself' (1919:69), which is a compound stem, but they erred in set 134 (1919:60) when they segmented 70- in Mib ?omd..tak 'big' in order to compare it with such forms as Csjb matini, mati1.i 'large, great' A. The proper segmentation is ?orno.-tak, where -tak is an intensifier. Mib ?6mo- is probably related to Mib 76mu 'bad'.

3. Dixon and Kroeber's resemblant sets. The updated resemblant sets now follow:

1 discuss individual sound developments within Utian and Yokuts only when they are pertinent to Yok-Utian comparisons. Those interested in these developments should consult Callaghan (1972; 1988~) and Whistler and Golla (1986).

1. Person. (1) PMi *miw.y-, PCo *mukw.e(h), PU *mikw.y -*mukw.y 'person, Indian'. The Utian items are cognate, as Dixon and Kroeber claim.

(2) PYnim *ma? 'person, human' matches PM *maj 'person' almost exactly. Ww wi.-, wi- 'person, man' and Wpcc win 'people, person; someone (in- def.)' do not match either (1) or (2). (1) U, (2) Y-M

    Man (1). PMiw *tdji~, PCo *ta.riS -*, PU *ta(.)li~ ? 'man (male)'.4 PY *tha?aih ?? 'person'. 'The Utian forms are probably cognate, and the Proto-Yokuts form may be cognate. Ybv toxy 'man' (D&K) is problem- atical, and Mip saweh 'man (male)' is definitely not cognate. YU

    Man (2). PMie *nan.a--*'man (male), husband', PYv *no.nlo-i**no.nla- 'man (male)'. These may well be cognate unless deep diffusion is at work (note Mono [Shoshonean] nana 'man' [D&K]). YU

    Boy, youth (1). PMiw *he.nas 'boy, youth' is not cognate with PCo *~in.i'child, offspring', and neither one is cognate with PMis *sali- 'boy, young man' or Mip sali-nai (D&K). Cscr ala-cu (D&K) is not cognate with any of the above. Wn k'aina (D&K); Wph, WPcc se'rita 'young man', Wph Curbj 'boy' are not cognate with each other, but PMis "sali- may match Wp se'rita. U-W ?

    Boy, youth (2). PCo *kot(.)o -*koc'(.)o 'boy, youth' is probably not cognate with PYnim * i**no.fa (dim. *no.$o- i**no.$a) 'youth, young man, boy' or Yp k'uhuhu.teml 'boy'. Note Miss koco- 'son, nephew', PU *kot(.)o -*koc'(.)o 'boy, youth', and apparent diminutive symbolism within Utian. U

    Woman, girl (1). Discussed later.

    Woman, girl (2). PMis *70$.a- 'woman, female, wife' is not related to Mip ?gsa.?ah- -?ysy.?yh 'woman, wife', which is derived from a stem meaning 'bad' (note PMis *7y;.y- 'bad'). PCo *?ac'ja- 'girl' is a different and probably unrelated set. D

    Girl. (1) PMiw * 'girl, young lady' may be cognate with PYgen *kuju.lum ? 'girl' or PYv *ka.?i-tha 'girl' but does not match Mn konoj 'girls'. (2) Csf Ca tra 'girl or maid' may be cognate with Ypal ketephal' 'woman'. (1) YU, (2) YU

    Baby. We have similar and related sets within Miwok; PMis *?esel.y- 'child', PMis "7ec'elel.y- 'baby' (diminutive symbolism), PMi "7es.e -*?e.s 'child' (vocative?). PMiw *?e'laj 'child' is most probably a loan from Wpcc ?ilaj 'baby, kid, child' (note PW *?elaj -*?ilaj KW 'baby, child, small'). This suspicion is heightened by the fact that the Lake Miwok vocative of 7e'laj 'child' is suppletive; ?e'.~ 'Child! Daughter-in-law!' (addressing adult children), and this suppletive form is cognate with the Eastern Miwok

PU *i > PCo *i -*e in the second syllable if the vowel in the first syllable was /a/.This rule ceased to operate in Proto-Costanoan times. See also sets 138 We and 85 Earth.

forms. Note also Mib 7a'j 'child'. None of these is related to Csf 0 cloosh cush 'infant or child', PM *kono 'girl, infant' or Mk tsilak'a 'baby' (D&K). D

    Old man. Ceb hunfac' 'old man' is not cognate with Miss hum.ele- t.i- 'little old man', PY *moko . . . 'old' CC, or PYnim *moxol < **moxo.l- 'to be old (man)'. But note PCo *mak.u(h) -*mak.o(h) 'husband', which may be cognate with the Proto-Yokuts form. PW *k'ij-'man to be old' KW > 'old man' does not match Mk kano 'old man' (D&K) or Mn 7oskon 'old man'. YU

    Head (1). Mip tolo- 'head, hair' may be cognate with PYgen *to?ol < **to?al 'head'. Some support comes from Ckar tacalaths A Itakla-f/ ? 'head'. Wpcc dut 'head' and Mn Col 'head' also match. YU-W-M

lla. Head (2). (1) PMiw *7ufu 'whiskers', Cru 7ut 'hair, head', PU *7ufu 'hair, head' may be cognate with PYgen *7ojow < **7ofaw 'hair, head'. In this regard, note that the deep underlying Yokuts vowel in the second sylla- ble could have been either ***o or ***a. (2) PMis *han.a- 'head' is a totally unrelated set, which may match Wpcc ?aria 'head (of person)' and PM *?on0 'hair on head', but note PM *70- 'with head' (instrumental prefix).

(1) YU, (2) U-W-M

    Head (3). Mib m61u 'head, hair' is probably not cognate with PCo *mo.fel -*mo.hel 'head'. None of these matches PY *philiw ?, PYnim *thiliw ?, or PYv *phijiw, all meaning 'forehead'? D

    Hair, forehead. Ceb tim.ah 'forehead' may match Ww tamoj 'head hair, scalp', although the meaning is divergent, but it does not match PM *syn 'forehead, face'. U-W

    Ear (1). Discussed later.

    Ear (2). PW *mat 'ear' does not match PM *bono 'ear'. D

    Eye (1). PMi *~ynfy -*;yt.y (-*~yt?) 'eye', PCo *hinte -*hi(.)n 'eye', PU *Synfy 'eye' match PM *hi(.)n 'eye, (small) seed' very well, but not PW *tum -*tus 'eye, face' or Wn Lui 'eye' (D&K). U-M

    Eye (2). PY *sasa- 'eye' goes well with Wpcc sa. 'eye', and Wpr sas 'eye' (D&K). Y-W

    Nose (1). PY *fhytjyk' < **thytjik' 'nose' (Ychuk sinik' 'nose') might go with PW *sono 'nose' and Wpcc linik 'nose, bill of bird' (note PW *liniq 'snot'), but only the first consonant of Chukchansi matches Mk sumo 'nose' (D&K). Y-W

    Nose (2). PMi *hu.k -*huk.u, PCo *(h)u.s, *(h)us-e-, PU *hu.k *huk.y 'nose' match Mm hiky 'nose: smeller' (note PM *hi. 'to smell some- thing'). U-M

    Mouth. (1) PY *sama- 'mouth' goes with PM *sim 'mouth'. (2) PW *go1 'mouth' may match Mk kombo (D&K) 'mouth'. (1) Y-M, (2) W-M

    Teeth. PMi *ky(.)t -*kyt.y, PCo *si(.)t -*sit.e, PU *ky(.)t -*kyt.y 'tooth, teeth' match PW *si 'teeth' and Mm Cik"tooth', but not PM *Caws. 'tooth, jaw', PY *thi.lij ? 'teeth, tooth', or Ytk faiii /taqi/ 'teeth'. U-W-M

    Tongue (1). PYnim *thalxath 'tongue' matches PW *tahal 'tongue' and Mn 7al 'tongue', but not PM *?en 'tongue'. Y-W-M

    Tongue (2). PMie * 'tongue' and PMiw *lent@ -*lemtip ? 'tongue' may be interrelated through metathesis. PCo *las.i(h) -*las.e(h) 'tongue' is probably also involved at some level. Note PMie *lakty- *lakjy- 'to lick'. U

    Beard. PY *ta.rn1u? > PYn *ta.m'uth 'beard' does not match PW *;eke; 'hair7, Wph sdpsep 'beard whiskers', Mm sim-pani 'beard, mustache' (< sim 'mouth'), or Mn masaw 'beard, mustache'. D

    Chin. PY *?aw'a? > PYnim *?aw'a.?i- 'chin' may be cognate with Cru auix P, d-wis HWH /?aw.ix -?aw.isl 'chin'. Note also PMi *hew.e 'chin'. These forms may match Mm Cawa 'chin, jaw' and PW *q'aba 'chin', but not PYv *?uk'~.~ul

? 'chin', or Mk yappa 'chin' (D&K). YU-W-M

    Neck. Discussed later.

    Shoulder. PY *k'aphsal -*k'aphsaJ (CC), PYnim *faphal -*fathal 'shoulder' do not match PW *sala 'arm, shoulder'. The last Yokuts form superficially matches PM *da-dak' 'shoulder', probably derived from PM *dak 'flat surface'. An apparent match disappears through etymology. D

    Hand (1). Three sets are involved: (1) PMi *?, PCo *?i?.u(s), PU *? 'hand'. (2) PMi *.ti.s -*.tis.y 'fingernail, toenail, claw', is proba- bly cognate with PCo *tu(.)r -*tu.ri? 'fingernail, toenail, claw'. (3) PW *sem 'hand, finger' may go with PM *jim 'arm (elbow to shoulder)', but not PM *ma 'arm (fingertip to elbow: hand)' (Maidun forms not cited by D&K). (1) U, (2) U, (3) W-M ?

    Hand (2). PYgen *phuthuy ? 'hand, arm' may be cognate with PMis *pa?an 'arm, lower arm', but not Mics wo.qot-u- 'arm'. Cru put? means 'thumb, big toe' and is probably not cognate. YU

    Arm. PMis *taw.a? 'shoulder blade, shoulder (cut of meat), includ- ing part of the ribs and upper forelimb' is cognate with PMiw *ta'wlik 'arm', but Mins 'arm' and Mip 'arm' are unrelated. I suspect Cscl talme-s 'arm' (D&K) is also unrelated. D

    Nails. Discussed later.

    Breasts. PMi *mu.s, PCo *mu(.)?, PU *mu.s 'breast, milk' bear no similarity to PY *mi.nif 'breast, teat, milk' apart from the first consonant, although there is an i:u alternation within Utian (see Belly, Fingernail under Hand) as well as within Yok-Utian. PW *?i(.)m-it 'breast, milk' does not match either, but PM *min 'breast, suckle, milk' matches the Proto- Yokuts form.5 Y-M

Dixon and Kroeber apparently segmented the first CV-in the Wintun forms as some type of "extension," although their only real evidence was the fact that Wintun . . .mit more closely resembled the Yokuts words for 'breast' than imit (D&K). Etymological analysis shows that

    Belly. This is a complex set. PMiw *puluk 'belly, intestines' may be cognate with PYv *palik' 'belly', but not Ykr olok', lak'i-n 'belly' (D&K). Note also Mip put.ul 'guts, intestines', PCo *putlu--*push-'pertaining to the belly', PU *puf.ul- *puflu- -*pu~lu-'belly, intestines' as well as PCos *pi.ti-(n)-*pit.i-(n))'abdomen, belly, tripe'. Wpcc bus 'belly' and PW *pot 'gut' are both matches. This is an especially interesting set, since the Proto- Wintun form and the Cache Creek Patwin word match different alternates of the Proto-Utian form. YU-W

    Knee (1). PY *k'uju.sul 'knee, elbow', PYsv *k'uju/ 'knee' is a differ- ent set from Yn opuk 'knee' (D&K), YKr upuc 'knee' (D&K), and PYn *. . .pug. . . ? < **. . .pus . . ., PYv *puSon 'knee' < **pusu.n. The latter set might go with Ww phujeq 'knee' and Mm pokb's 'knee' (but not with Wpcc phdmak 'knee'). Y-W-M

    Knee (2). PMie *hon.oj -*ho.noj 'knee' may match the Wph 7a'nak 'knee, elbow'. U-W ?

    Knee (3). Mil thok'dl.u 'knee' is probably a loanword, but I have not been able to trace its source. It does not resemble any other Miwok word for 'knee' and is probably not cognate with Cru to.10:-'knee'. D

    Foot, leg (1). Discussed later.

    Foot (2). PMi *kolo, PCo *kolo CS, PU *kolo 'foot' may be cognate with PY *kalas.a-'lower leg' CC (not cited by D&K). These forms match Wn kole 'foot' (D&K). YU-W

    Heart (1). PY *hoghog 'heart' matches PM *hon 'heart, to breathe'. Y -M

    Heart (2). Ceb mi.njih 'heart' bears no resemblance to PW * 'heart' except in one place of articulation. D

    Bone. PY *Cij 'bone' does not sufficiently resemble PCo *fad.i 'bone' for inclusion, and it does not resemble PMie *kyc'.yc' 'bone' or Mib rnz2.i 'bone' at all. Loss of an initial syllable is very rare in Yok-Utian (but see Knee [I], Five [I]), and the only "augments" that occur are occasional reduplicative prefixes in Eastern Miwok, most of which are Simplex Stems in compounds. Final syllables are sometimes lost (see Eye [I]). D

    Blood (1). Discussed later.

    Blood (2). PW *sa.q 'blood' and Ww 'blood' do not match PM *sede -*sedej 'blood'. Ww 'blood' is based on a stem tedmeaning 'to be red' < PW *tede 'red'. D

    Liver. Discussed later.

    Fat. PY *hi.xa-'fat' resembles PW *hu.m -"hu.1 'fat' and Mm hjt 'fat, grease' only in the first phoneme. D

    Skin. PYnim *$uluj 'hide, skin' differs from Yv daplan 'skin' (D&K), Ykr fapia, fofop 'skin' (D&K), Yp palai 'skin' (D&K), and Ybv gapac 'skin' (D&K), which fall into four sets. Yp palai 'skin' (D&K) may match PM *po 'skin'. Y-M ?

    Faeces (1). PMie *kyna-'faeces' resembles PW *$eni 'faeces' if one allows palatalization before /el. U-W

    Faeces (2). PYnim *pitik' 'faeces' matches PM *pi/ 'shit' very well. Y -M

    Urine. Csjb c'ala 'urine, to urinate' (not cited by D&K) may be cog- nate with PYnim *$ulonl'urine' GG. PYnim *$ujo. 'to urinate' GG matches PW *&.mu-'urine, to urinate' and PM *du 'urine'. YU-W-M

    Sinew. PY *phikhil'sinew' is a match for PM *pak 'sinew'. Y-M

    Horn. PMi *kil.i -*, PCo *Ei.ri, PU *kil.i -* -*c' 'horn, antlers' may be cognate with PYn *khoje? < **khoji.? 'horn' (not cited by D&K). These forms match PW *k'ili 'horn', but PY *?y~yl''horn' does not match. YU-W

    Tail (1). Mil kd.k, Cru ko.k, PU *ko.k 'tail' (not cited in D&K) may be cognate with PY *k'u/ > PYsv *k'u/uS'tail'. These forms may go with PW *sutu 'tail'. YU-W

    Tail (2). Ybv tapak (D&K) does not match PM *buk 'tail', since I do not allow initial expansions or loss of the first syllable without supporting evidence. D

    Navel. PW *naq 'navel' matches Mk naka (D&K) 'navel'. W-M

    Chief. Mip ceka-'chief' matches PW *se.qtu ? 'chief', but not Mn hu.k' 'chief '. U-W

    Shaman, ceremonial initiate. PMis *?al.ini-'shaman' is probably not cognate with PY *?ayfhiw? 'shaman', which may be cognate with PMis *jaq.e-'to sleep, lie down' (not cited here by D&K). (See 166 Sleep [2].) Mip ?>'doctor' (from Smsi-'to doctor') is unrelated. Mil jomta 'doctor' is probably a loanword from Patwin. PW *jom 'doctor, poison' matches PM *jom 'spiritual power', but such words are culturally loaded and subject to borrowing. W-M ?

    House (1). PCo *rukw.a 'house' CS does not match PW *qewel 'house' or PM *hy 'conical bark house'. D

    House (2). PMi *koc'(.)a'house' is totally unrelated to PMis *?u.c'u'house, space', which is a lengthened grade of PMis *?uc'.u-'to stay, be sit- uated'. Neither one is cognate with PY *fhi-(< **fhi.-)'house'. One cannot arbitrarily chop off the first syllable of Utian words to make them match. D

    Bow. Discussed later.

    Arrow (1). PW *no-'arrow' matches PM *nok' 'arrow'. Note Ww n0.t 'arrow, arrowhead', Wpcc nok'o 'arrow'. W-M

    Arrow (2). Mics tywki-, PMiw *tuw(.)u< **,tyw(.)y'to shoot' may be cognate with PY *fuju < **fuji'to shoot' and PYnim *fujof< **fujo.s 'arrow'. Csjb tujos 'arrow' may be cognate or a deep loanword. Cscl tawi-sem 'arrow' (D&K) may be cognate, but not Cru tep.s 'arrow' or Csjb temo 'arrow'. YU

    Knife. Miss hope-'knife' (probably Tope-in some dialects) is not cognate with Mil cica 'quartz, flint, obsidian', and neither one is cognate with PCo *tipa-'knife'. None of these match Mm Cam.i 'knife' or Mk tsumi 'knife' (D&K). D

        Boat (1). Miss wo.te-'boat' is a loanword from Spanish bote 'boat'. PY *?ow'in ? 'boat' is not cognate with Csf Wah lee 'boat' or Cscl walin 'boat' (D&K). D

        63a. Boat (2). Mil nu.h 'boat' is most probably a Patwin loanword, as Dixon and Kroeber admit. PW *nuh 'boat' matches Northern Maidun no 'boat' (D&K). W-M
    Boat (3). Mib ~a'ka'boat' matches PM *jak' 'bridge, boat', but not Mn wakaj 'boat' (< Spanish barca 'boat'?). U-M ?

    Pipe (1). Ceb 'pipe' is a match for Wpcc top0 'pipe' and Mip topo-kela (D&K), although the first part of the Plains Miwok form looks like a loan from Patwin. U-W

    Pipe (2). PY *pam'o.--*pa?om-'to smoke (tobacco, etc.) may be cognate with PMie *pa?my- (*pa?ym- Stem 2) 'to smoke (tobacco or pipe)'. An Eastern Miwok instrumental agentive form, *pa?ym-.a-'pipe', could be derived from Stem 2, although this does not seem to be the present-day word for 'pipe'. Wph bdmit 'Indian smoking pipe' and PM *pan 'tobacco' may match. YU-W-M

    Pestle (1). No Yok-Utian. Neither Ww sataq 'pestle, pounding rock', Wn cotok 'pestle' (D&K), Wp taki 'pestle' (D&K), nor Wpcc fusa 'pestle' matches Mk suni 'pestle' (D&K). D

    Pestle (2). PYnim *phaluj-*phalwoj? 'pestle' may be cognate with Mic pa, paya 'pestle' (D&K), but not Mip hopa-'pestle'. Mm bayi 'pestle' (D&K) and Mn baj 'pestle' also match. YU-M ?

    Sun (1). PMi *hi.?, PCo *(h)i?-men,PU *hi.? 'sun' may be cognate with PY *'today'. These forms may match Ww sas 'sun, moon, month, clock', but not Wp sun 'sun' (D&K) (note PW *sani 'daytime'). YU-W ?

    Sun (2). PY *?ophi?? 'moon' > PYn 'luminary'. Cf. PYnim *70pholo'sun, day' ?, PYv oph'luminary'. None of these matches PM *pot'luminary, sun, moon'. D

    Moon. PMi *kome-'moon' is cognate with PCo *kolma -*kolme 'moon'. U

    Star. Discussed later.

    Night (1); 74. Night (2). Under Night (I), Dixon and Kroeber list only Miwok forms, probably a typographical error, so I am considering Night (1) and Night (2) together. (1) PMi *kawyl 'night' does not go with Ww ke.n-wani -ken-wani 'night time, evening', which means something like 'moving down'. Note PW *ken 'down'. (2) PY *thoj'on < **thoj'an *thoj'no-3 < **thoj'na-7 'night' does not even look like Wn leni 'night' (D&K) or Wpcc sin01 'night'. (1) D, (2) D

their segmentation (i-mit) was wrong. Proper segmentation (PW *?i.m-it) removes the appar- ent similarity.

    Cloud. Mib pilaw 'cloud' is probably not cognate with PY *k'ylyj ? 'cloud', and certainly not with Ybv piya-p'iya (D&K) 'cloud', and Mim ya- kal 'cloud' (D&K, segmentation arbitrary) does not go with any of them. PY *k'ylyj ? 'cloud' does not match Ww 4a.h 'cloud', but it may go with Wpcc k'ir 'cloud'. Y-W

    Lightning. Mins wi1.e~-y- 'lightning', Cscr wilep 'lightning' (D&K), PU *wil.ep 'lightning' may be cognate with PYnim *walam -*walma'lightning'. Ww walu.qi 'lightning' and PM *wis-p'il 'lightning' also match.


    Rain. Discussed later.

    Snow. PMie * 'snow' is not cognate with Mil tbna 'snow', and these forms show no resemblance to either Ceb wakan 'snow' or PW *jo.l 'snow'. PM *ko. 'snow' resembles the Miwok form only in the first conso- nant. D

    Fire. (1) PMi *wyke 'fire' may be cognate with PY *?oSith 'fire'. Note also PMis *wyk-ty- 'to catch fire, burn'; PMi * 'to burn', Csjb 'to catch fire, light a fire', PU * 'to burn'. (2) PCo *~ot.o(w) 'fire' is a different set. PM *sa 'fire' shares only one resemblant phoneme.

(1) YU, (2) D

    Smoke. Discussed later.

    Ash. Mil 'white ashes from a forest fire' (not in D&K) matches PW *phut 'ashes, dust' and Mm didusi 'ashes'. PY *haphaS 'ash, dirt' ? does not match. U-W-M

    Coal. PYgen *sal'u ?? 'coal, charcoal' might match PW *q'al 'charcoal', but not Wpcc wali 'hot coals'. Y-W

    Water (1). PMi *ki.k -*kik.y, PCo *si.(j), *sij.e- < **si.s -**si~.e?, PU *ki.k -*kik.y 'water'. This set is not cognate with PY *?ilik'. U

    Water (2). PW *mem 'water' superficially matches PM *mo-m 'wa- ter'. However, the latter form is bimorphemic, and PM *mo- means 'action with mouth' (an instrumental prefix; note PM *mo(.), *mo.ho 'to drink'). Cognate relationship between the Wintun and Maidun forms is unlikely. D

    Earth. Discussed later.

    Mountain (1). PY *tulaw 'up, high' ? > *tulul 'mountain, foothill'. The Yokuts form matches PW *tot -*c'o.l ? 'mountain'. Y-W

    Mountain (2). PMis *leme- 'mountain' may be cognate with PY *lomith 'mountain', but these forms do not match PM *jaman 'mountain'. YU

    Stone. Neither PY *xalul' ? 'rock' nor PYv *.file1 < **sili.l 'rock' matches PW *sos 'rock', Ww son 'rock(s), stone(s)', Wp todui, tului 'stone' (D&K), or Wpcc k'oddj 'rock, money'. D

    Salt. PW *wet 'salt' does not match PM *ba 'salt'. D

    Wood. PMis *~y~.y-

'wood' is probably not cognate with PY *hyt'ys 'wood', but it might match Ww ;u.s 'wood, stock, club, log'. Wn tcok"wood' (D&K) and Wpcc t'ok 'wood, a stick, firewood' appear to be an unrelated set (note PW *Eo.q 'tree'). PM *&a 'tree, wood, stick' does not seem to match. U-W

    Grass, brush. PYnim *ja.wil 'grass, brush' may be cognate with Csf Yah wah 'grass' (miscited by D&K as yama). Cscr wai (D&K) is not cognate. YU

    Sugar pine. PMi *san(.)ak 'mountain pine' (not cited by D&K) may be cognate with PYnim *;ogo.xi.f < **&oga.xis?? 'sugar pine', or the word may be a deep loanword from PMis *sag.ak 'sugar pine' into Yokuts. The Yok-Utian forms may match PW *sumu 'sugar pine' and PM *sumu 'sugar pine', although the near identity of the Wintun and Maidun forms suggests borrowing. YU-W-M ?

    Yellow pine. Ychuk ?' 'yellow pine' and Ytk idiiit (D&K) 'yel- low pine' matches PM *Pini 'yellow pine', but not Ww k'u1em.i 'sugar pine, yellow pine, Ponderosa pine'. Y-M

    Digger pine. PYnim *thog < **tho.g -**thu.q 'digger pine tree' and PYnim *tho.gaS 'digger pine nut' match PM *to.n 'digger pine', but not Ww c'ati 'ripe digger pine cones or nuts', Wn tc'oko 'digger pine' (D&K), Wpr c'usak 'digger pine', but it might match Wpcc t'uwa 'pine tree'. Wpr c'usak 'digger pine' includes -sak, presumably from Miwok (Whistler 1977). Note PMis *sak.y- 'digger pine'. Y-W-M

    Buckeye. PYnim *topig 'buckeye' does not match Wn pasa 'buck- eye' (D&K) or Mn polo 'buckeye, buckeye ball'. D

    Manzanita (1). PY *?apthuw 'manzanita' may match Mm 7kkpy 'large manzanita', but not Ww 'manzanita wood' or phaj 'yellow manzanita berries'. Y-M

    Manzanita (2). Wp Peje, Peja 'manzanita' are loanwords from Miwok (Whistler 1977). Note PMi *?eje -*?ej.e 'manzanita (berries)'. D

    Tule (1). Mil kd.1 'round tule' may be cognate with PYv *k'oji~ ? 'tule sp.'. Note Ypal kololi-s 'tule' (D&K). Wp guye (D&K) might match, but Wp sa-kulul 'tule' (D&K) probably does not match. YU-W ?

    Tule (2). There are two sets here. (1) Mim Lo-kos 'big round tule' CHM may match Wph taka 'big round tules'. (2) Mim Tappa 'tule' SB may go with PW *iop -*top 'bulrush' (Scirpus robustus) KW. (1) U-W,

(2) u-W

    lble (3). PYv *pumuk' 'round tule' does not seem to go with anything. Mil patpat 'cattail, flat tule' is probably a loanword from Wph pa.tpat 'cattails' (note PW *patpat 'tule'). Note also Ykr patak 'tule' (D&K) and Yv poton 'tule' (D&K), which match the Proto-Wintun form. Wp p'oca'k 'tule' (D&K) and Mk pok-poko 'tule' (D&K) might also go. Y-W-M

    Dog. PMis *c'uku-'dog', Mip cua-'dog' < Spanish chucho 'dog', Cscl tcutcu 'dog' (D&K), also from Spanish; PY *Cy.sys 'dog' ? borrowed?, PW *suku 'dog', PM *sy 'dog, pet' (Ccr hi-tcas 'dog' [D&K] does not be- long). Words for 'dog' were so heavily borrowed throughout California (Kroeber 1941) that this set should be discarded. D

    Coyote. Three sets are involved. (1) PMis *?a?e.-li-'coyote' is probably a loanword from PYn *ho.Su-< **'coyote' or some similar word. The Yokuts form matches Mn hu.saj 'coyote' (Shipley 1997:4) very closely. I suspect borrowing is involved. (2) PMi *?ole 'coyote' matches PM *?ole. 'coyote' perfectly, suggesting deep borrowing, but we must keep it because of extension and time depth. (3) Ybv hulapinsa-s (D&K) matches PW *hul ? 'wolf' and Mm helijene 'wolf'. Note also PMi *hu.n -hun.u 'wolf', not cited by D&K. 'Coyote' is a culturally loaded term, subject to borrowing and taboo replacement. (1) D, (2) U-M ?, (3) YU-W-M

    Fox. Mim Ah-wah-ke 'gray fox' CHM may be cognate with PYnim *?aw'Cha-'fox', which matches PW *ha.w 'fox' and PM *haw 'gray fox'.


    Bear. Discussed later.

    Grizzly bear. Ww wimaj, wimah 'grizzly bear' does not match PM * 'grizzly bear'. D

    Puma. Discussed later.

    Deer. PW *no.p 'deer' does not match Mn dyde 'deer' RS. D

    Elk. PYnim *Sohkoj < **sohkaj 'elk' GG may go with Wph lokdja 'tule elk'. Y-W

    Jackrabbit. PMie *pepla.-li-'jackrabbit' is not cognate with Mim aule 'jackrabbit' SB. Neither one is cognate with PYnim *topol 'jackrabbit'. Ww pat-keles 'jackrabbit: outside-long one' does not match PM *pal 'cottontail' despite some superficial similarity, because of the etymol- ogy of the Wintu form. Likewise, Mn bo(.)je'rabbit' RS does not match. D

    Bird. More than one set is involved. (1) Mins mice.-mati-'bird', Csjb mi.ti-s 'fledgling', PCo *mit.e-n -*mif.i-n 'to grow', PU *mi(.)fe-'bird, grow, fledgling' may be cognate with PYnim *mithe.< **mifhi'bird, grow' GG. (2) Ychuk &xrna'-? 'bird' VG, may match PW *&(?it) 'bird' KW (not cited by D&K). PMis *c'ic'ka-'bird' and PCo *c'i.rit-min 'western meadow- lark' are probably unrelated, also the other Yokuts forms cited by D&K.

(1) YU, (2) Y-W

111. Buzzard. PMie c'y.?y-'buzzard' does not go with any of the other forms. PMis *'buzzard' could be cognate with PYn *hoth 'buzzard', but not PYv *k'o?f-'buzzard'. Mil hu.~'buzzard' is probably a loanword from Patwin. Note PW *hu.s 'buzzard, turkey vulture'. The possible Yok- Utian cognates match PW hu.s 'buzzard' and PM *hy.s -*hu.s 'buzzard'.


    Crane (1). Mil wa'.k 'crane', Mins Wa'h-ah-su CHM [wa?.as-y-] ? 'great blue heron' and PYnim *wa.xif 'crane' are all onomatopoeic and subject to borrowing. Mn k'ywe 'fish crane' does not go, nor does Ww qa.t-, which may mean 'duck' rather than 'crane'. D

    Crane (2). PCos *ture; 'small brown crane' is not onomatopoeic and it does match Ww torehas 'sandhill crane'. U-W

    lhrtle. (1) PMie *?aw(.)an.a-'turtle', PCo *?awniE-min 'turtle' does not sufficiently resemble Yn on 'turtle' (D&K) or PYnim *pzwa.xith ? 'turtle', but the Utian forms do resemble Ychuk ?ow' 'turtle' (not cited by D&K). (2) There is a closer resemblance between Yn on 'turtle' (D&K) and PW *?an-'turtle'. (1) YU, (2) Y-W

    Rattlesnake (1). (1) PY *fhy?yl 'rattlesnake' matches PW *$iwil 'rattlesnake'. (2) Mil hol6maj 'rattlesnake' matches Mk cola 'rattlesnake' (D&K) and Mn sola 'rattlesnake' RS. (1) Y-W, (2) U-M

    Rattlesnake (2). Miss law.a.ti-'rattlesnake' may match Ww da.q 'Western rattlesnake'. U-W ?

    Frog. PMi *wafa.k 'frog', PCos *wakrac'-min, Ychuk wa!akpz?aj' 'frog', PW "wataq 'frog'. These words are onomatopoeic and susceptible to borrowing. D

    Fish (1). (1) PMis "1api.s-aj 'fish' may be cognate with PY *lophifh 'fish'. (2) Mib 3el.e.-we 'fish' may be cognate with Cscr helai 'fish' (D&K).

(3) PMi *py.le 'roe' matches PW *pu 'roe'. Mip pu 'fish' and Mil pci 'roe' are probably recent loanwords from Patwin. (1) YU, (2) U ?, (3) U-W

    Fish (2). Cru kin.irs' 'large headed fish' P does not match PW *c'i.r 'fish', Ww diket 'fish' (D&K), or Wn tcit 'fish'. D

    Salmon. Discussed later.

    Trout. (1) PY *tallim'trout' may be cognate with PMie "falpa-? 'trout' and possibly Mim tBlo 'trout' SB, but not Mib lewim 'trout' SB. (2) Mil hzi.1 'trout' matches Ww sula -sulat 'trout', but not Wp ciya-tir, mol 'trout' (D&K), or Mm palik 'shiner, a kind of small fish', or Mn pula 'fish'.

(1) YU, (2) u-W

    Fly. PCo *mu.muri -*mum.uri 'fly' is not cognate with PY *pan'iskaj ?? 'fly' or PYn *mo.nlaj ? 'fly'. None of these matches Ww Zilit, xilit 'fly' or Ww lulit 'fly' (D&K), Wp homota 'fly' (D&K), or Mk amelulu 'fly' (D&K). But Mip'fly' (not in D&K) does match Wp homota 'fly' (D&K), although these words are onomatopoeic. U-W ?

    Spider (1). Csjb ti.ras'-min means 'one having a big rump'. It does not go with PW *kerek 'spider'. D

    Spider (2). PYnim *ma.Za-?? 'spider' > PYn *me.Za- 'spider' matches Mm maka'? 'spider'. Y-M

    Louse. Words for head louse and body louse are usually different.

(1) Neither PMi *ket.y -*ke.t 'head louse' nor PCo *kah.aj -*kaj 'head louse' is cognate with PY * 'head louse', which matches PM *di 'head louse'. (2) PM *pedes 'body louse' matches PW *pher- 'louse' and PYnim *phathath? 'body louse', but not Ww donos 'head louse'. (1) Y-M, (2) Y-W-M

    Flea. PMi *ky(.)kys 'flea' is not cognate with PCo *po.r 'flea', and neither one is cognate with PY *pa.k'il 'flea'. PMi *ky(.)kys 'flea' matches PW *k'ok'o 'flea' and possibly PM *i'yk'ys 'flea', if we assume assimilation in the Miwok and Wintun forms. Note Wp tunkuc 'flea' (D&K). U-W-M

    Grasshopper. (1) Miss ?ag.ut 'grasshopper' may match PM *?en(e.) 'grasshopper', but it is not cognate with PYnim *?a.n- 'grasshopper' > PYn */a.najwu/ ? 'grasshopper' CC. (2) The Yokuts forms do not match Ww nep(ut) 'grasshopper', but they may match Wpcc taram 'grasshopper (any kind), small grasshopper'. (1) U-M, (2) Y-W

    Worm. Dixon and Kroeber cite reflexes of PYnim *k'a.thuw ? 'worm, bug' and PYn *k'e.w 'worm'. (1) PYnim *khthuw ? 'worm' matches PW *k'es 'worm'. (2) PYn *k'e.w 'worm' matches PM *k'aj(e) 'worm'. Mn pep 'worm (D&K) does not match any of them. (1) Y-W, (2) Y-M

    Name. PMis *?oj.a- 'to name' (not cited by D&K) may be cognate with PYnim *hojo. < **hoja 'to name'. PM *ja 'name' does not match. Note PY *wa.sith ? 'a name', PYnim *hojo.woS 'a name', Mip ?ow.a- 'to name', ?owa.-se- 'a name', and Mib ?&we < *?dwe(?) 'a name'. YU

    White. Discussed later.

    Black (1). PMie *kulu- 'black' may be cognate with Ceb 5irke.-wis -Silke(.)-wis 'black' and Csf Shol co te 'black'. It also matches PW *Eul and *sil, both of which mean 'black', as well as PM *sil 'black' and Mn kylym 'black, dark (weather, etc.)'. Wn ku?a 'black' (D&K) probably does not match. U-W-M

    Black (2). PMi *mulu- 'black', PCo *muru-.t 'night' where -t is a noun formative suffix, PU *mulu 'black' are probably not cognate with PYn *motok' < **motak' -*mothk'i- 'black, to blacken'. The Utian forms match Wph multhu, multhi 'dark' (color) and Wn molok 'black' (D&K). U-W

    Red. Discussed later.

    Large. Dixon and Kroeber presumably segment 70-off Mib ?om6..tak 'big' to make it fit the other forms. The correct segmentation is ?omd.-tak, where -tak is an intensifier. 76~~0-

is probably related to Mib %mu 'bad'. JPH could not re-elicit Csjb matini, mati1.i 'large, great' A. PYgen *me/ ? (< **mi./-) > PYv *ma/. . . ? 'large, powerful, good' does not match Mn muk'uje 'big', Ww qomos 'big thing(s), old people', Ww bohema 'to be big', or Wn komaso 'large' (D&K). D

    Small. Miss cin.imi-'small', Mics fun.i-ci-'little', and Mip 7itit.i'little, small' are unrelated to each other. PYn *k'ot'i-?? means 'large, big', not 'small'. Csjb ku.ti-s 'a little bit' may match PW *k'u-'small' and Mm k'iit' 'creature', a generic term for small animals, birds, etc., but it does not match Mk nukti 'small' (D&K). Note Mics kuci-'good'. As Dixon and Kroeber state, Mil k'u5.i 'little" is probably a loanword from Hill Patwin (note Wpcc k'u5ij 'small'). U-W-M

    I. Sets involving (1) *nu-, (2) *ni-,and (3) *kan are involved in a complex and interrelated way in both Yokuts and Utian (Callaghan 1997:41). *ni-matches PW *ni-'I' and PM *ni '1'. Wpcc 2u 'I' does not match. (1) YU, (2) YU-W-M, (3) YU

    Thou. PMi *mi(.) 'thou' is probably related to PCo *me(.)'thou', and both may be related to PY *ma-?'thou' GG. Support comes from PY *man' -*min 'your sg.' CC, *manan 'your pl.' GG, and PU *mi.ny -*me.ny 'thy'. The Proto-Miwok form matches PW *mi 'you sg. subj.' and PM *mi'you sg.', as well as much of California. YU-W-M

    We. PMi *mas.i -*ma.? 'we7, PCo *mak.i -*mak.e 'we incl. appo- sitional' ?, PCo *maksi -*mak~e'we excl. appositional' ?. PU *maksi 'we' may be cognate with PYnim *mak' 'we du. incl.' CC and PYnim *ma.k-in 'our du. incl.' < ** ?? CC. YU

    This (1). PMi *?i. -*7i 'that; he, she, it', Ceb 3i 'his, her, its', PU *?i -*?i 'that, 3 sg.' may be cognate with PY *xi 'this near' GG and PYnv *hi?-*he?< **hi. 'this near' GG. PW *?e 'third-person proximal sg. subj.' and Mn hede 'this' also match. YU-W-M

    This (2). PMi *ne -*ne 'this', PCo *ne -*ne 'this, here', PU *ne. -*ne 'this, here' may match Mn 7uni'this, this one' despite the shape of the first syllable, which I would normally not allow one to pare off. Histori- cally, demonstrative pronouns are frequently compounded in Costanoan, and I am trying to be generous. U-M ?

    That (1). Dixon and Kroeber list reflexes of several Yokuts stems; PY *ki -*ka -*ko 'this far' GG and Ytu xuntu GG 'this invisible'. The last form might match Mn hod0 'yonder'. Y-M ?

    That (2). PMi *no -*no 'that, there', PCo *nu. -*nu 'that, there', PU *no(.) -*nu(.) 'that, there' probably do not match PM *?dn 'that yonder'. U

    That (3). PCos *nup.i 'that (one)' can be analyzed as a compound of PCo *nu(.)'that' plus *-p.i. Note also PCo *ne(.)CS 'this' and PCo *ne-p.e 'this one', Ceb pi.n 'that (one)', and PCo *pim 'that (one)'. PCos *-p.i matches PW *pi 'third-person demonstrative sg. subj.', but not PM *my(.)'that7.U-W

    Who. (I) PMie *manu-'who, someone', PMi *ma'n-ti'who, some- one', and Ceb ma.-to 'who?' may ultimately derive from PU *ma 'is it? and PU *ma.n -*man (?) 'that is', which matches Mk mene 'who' (D&K).

(2) PCos *(h)am-pi'who?' is a different stem, and it might match PM *homo'who, which (of a choice), where?'. None of these stems matches PY *wath'who?' or Wn popi 'who' (D&K)or Wp popa 'who' (D&K). (1) U-M,


    What. PMi *hin-ti 'what, something', Csjb hin-ti-s ? 'what?', PU *hin-ti 'what? may be cognate with PY *had-'what?'. Note PYn *had-ta 'what?' CC. PW *he-'demonstrative-interrogative' and PM *hesi 'what; something, any' CC might also match. YU-W-M

    Up (1). I no longer think PMi *li.le 'up' is cognate with PCo *ri(.)ni(h)'up; and neither one matches PW *?ole.l 'up, above'. D

    Up (2). PCos *tapre 'above' is probably not cognate with PYgen *thiphin'above, up; sky'. PW *panti 'top' does not match, but PM *hipin 'above' matches the Yokuts form. Y-M

    Down. Mip wani-t 'down, below' is probably not cognate with Cru winmu-j or 'below', and neither is cognate with PY *xoniw 'downstream, west'. None of these matches PW *ken -*kenti 'down' or Mm kanai-na (D&K). The last two forms resemble each other. W-M

    Tomorrow (1). Yhom him !a 'tomorrow', Ywik hijamhus' huyamxuf -hijamxas' 'tomorrow', Ychuk hi.khaw 'tomorrow', Yyaw hi757 'tomorrow' may match Ww hima 'the next day, tomorrow' and Mk himaduk 'tomorrow' (D&K).Y -W-M

    Tomorrow (2). Wp bina 'tomorrow' (D&K) may go with Mm be'nek 'tomorrow', but not with Yv bayi-u 'tomorrow' (D&K).W-M

    One. PMi *ken.e 'one' may match PW *k'ete.t 'one7, but it is not cognate with PY *jit < **ji.t 'one' and does not match PM *wyk' 'one' or Mm sy'tti 'one'. U-W

    Two (1). PY *pogij 'two' matches PW *pan-palet/l 'two' KW and PM * 'two'. Y-W-M

    Two (2). PMi *?oti'two', PCo *?o~i-s'two', PU *?o'ofi'two' may be cog- nate with PY *hoth-in PY *hoth-pogoj-*hath-pa& your' (< **h~t~-~or~ij 'two twos' ? CC). YU

    Three. PCo *kap(.)e-s-*kapha -*kapha-n 'three' is not cognate with PY * in 'three', this being the only instance of U k : Y ; in initial position in Dixon and Kroeber's corpus. Neither one matches PW *panot 'three', but the Yokuts form matches PM *sa.-py 'three'. Y-M

    Five (1). PCon *mi~.ur< **(h)im-is.u-s? 'five: one hand' ? is not cognate with PY *jit-sygyl'five' ('one hand'?). These forms do not match Ww $an-sem 'five, left hand' or Mk ma-tsani 'five7(D&K).D

    Five (2). PMis *ma~.o-k.a-'five' may match PM *ma.-wyk' 'five: one hand' if I'm generous. Mip kas.0-k.0-'five' probably represents remote assimilation. U-M ?

    Six. PCo *sak.en 'six' is not cognate with PMis *tem.o-k.a-'six', or Mip teme.-pu-'six' < teme-'big', or PY *Coliphij'six'. None of these forms matches Mm sdj-Cok'o 'six: 'both threes'. D

    Eat (1). Cru xamp 'to eat' P and kat 'to eat' (D&K) are close enough to PYv *xatha'to eat' CC that we should not discard this set. YU ?

    Eat (2). PMis "7yw.y-'to eat' is not cognate with PCo "7am.a 'to eat' CS, nor do these forms match PW *ba 'to eat' or PM *pe -*pa 'to eat', but the Wintun and Maidun forms match each other. W-M

    Drink. PMi *?u~.u'to drink', PCo *?ukw.e'to drink' CS may be cognate with PY *?ukun < **?ukin 'to drink'. None of these matches Wp elu 'to drink' (D&K) or PW *tethi-*tethu ? 'to drink'. YU

    Run. Discussed later.

    Dance (1). PMis *kala.-g-y-'to dance, kick backwards' and Mins kalty 'to dance, tramp' are not cognate with either PCos *c'it.e 'to dance' CS or Cscl to-ken 'to dance' (D&K), but they may be cognate with PY "ka1as.a-'lower leg' CC; see Foot (2). PYnim *kham'-'to dance' may be cognate with Mib ka'.wul 'to dance', and it matches PM *k'a(.)min'to dance', but it does not match PW *?on0 'to dance' or Wpcc thon 'to dance'. YU-M

    Dance (2). PY *wothij 'to dance' matches Mm wetLm 'to dance'. Y-M

    Sing. PMis *myl.i-'to sing' may match Wpcc muhu 'to sing'. u-W ?

    Sleep (1). PYnim *k'aniw ?? 'to lie' (vocalism unclear) may match PW *khina'to sleep' and Wpcc k'ana. 'to sleep', be asleep, go to sleep'. Y-W

    Sleep (2). (1) PMi *?e.c'y 'to sleep', PCo *?e.te 'to sleep' CS, PU *?e.fy 'to sleep' are unrelated to the Yokuts forms. (2) PMis *jag.e-*jag.a-'to sleep, lie down' may be cognate with PYgen *?igith? 'to sleep', PYv *?anath-w-'to dream', and PY *?agthiw'shaman'. (1) U, (2)YU

    See. (1) PMis *syj-ge-'to see' may be cognate with PYsv *$i 'to see'. Note PU Syn-ty 'eye' and PY *sasa-'eye'. Note PM *hi(.)n 'eye, small seed'. (2) Wpcc $o.w 'to glance at, look at, see' matches PM *be 'to see'.

(1)YU, (2)W-M

    Kill. PCo *nim.i 'to fight, kill, lie down' does not match PW *xb.m-u 'to kill'. D

    Give. PMiw *waja -*waje 'to give' may be cognate with PY *wa?in 'to give'. These forms match PM *mej 'to give'. YU-M

    Laugh. PMis *hyja.k-y-'to laugh' and Mib ha'waj 'to laugh' may be cognate with PY *hafa 'to laugh'. YU

    Cry, mourn. PYn *wa.xil 'to cry' may match PW *wac'-u *wac'-i 'to cry' and PM *wak 'to cry out'. These forms are onomatopoeic.

Y-W-M ?
Tabular Sets

Dixon and Kroeber (1919:55) lay out the following sets in column form, ". . . with a view to bringing out parallelisms that otherwise might fail to impress without detailed analysis and discussion." Unfortunately, they arrange the forms in such a way that any two of them can be compared if they share a resemblant syllable. For example, Yokuts muk'es 'woman' is laid out above Mib kule'jih 'woman', so that Yokuts -te-is compared with Miwok ku-, which is improper methodology without independent reasons for such alignment. Prefixation is almost nonexistent in Utian and very rare in Yokuts. Loss of an initial syllable rarely occurs (but see set 155 Five [I]), and initial "phonetic increments" almost never occur in these languages either. Stem compounding may occur, and final syllables may be lost in some Eastern Miwok derivational processes.

Remote consonantal metathesis and assimilation do occur, as Dixon and Kroeber claim. Examples are PCo *c'i.rit-min 'western meadowlark' > Cru ti.ric'-im 'western meadowlark', and Mip kas.0-k.0-'five', probably < PMie *ma~.o-k.a-'five'. Regular vowel loss takes place in Yokuts under certain circumstances, but not in the first syllable. Miwok IM : Costanoan Is/ is a valid correspondence in the environment of high vowels. Dixon and Kroeber noted this correspondence, although they did not determine the conditioning factors. Miwok 1-1-1 : Costanoan I-r-I is another valid corre- spondence. Otherwise, Dixon and Kroeber have little supporting evidence for the types of changes they posit.

6. Woman, girl (1). At least three different sets are involved. (1) PMiw *pdc'.is'woman, old woman' might match Ww p'o.qta 'woman, female' and Wpcc pdkita 'woman', although I have suspected that the Proto-Western- Miwok form is a deep loanword from some Patwin dialect that underwent palatalization. (2) Csjb mukur-ma < *muku-s-ma 'woman, old woman' may be cognate with PYgen *muk'es < **muk'i.s 'woman'. The relationship of PYv **< ** -**'woman' to this set is uncertain. (3) Mib kul(.)e'jih 'woman' is probably related to PMiw *klil-e~'wife', which matches PM *kyle 'woman, wife'. (1)U-W ?, (2)YU,


14. Ear. PMis *folko~-*,to.kos 'eat', PCo *tuk~u-$? 'ear', PU *fo(l)ku~ -*fok~u-? 'ear' may be cognate with PY *thuk''eat'. PMi *?alok'ear', Mip'ear', and Csjb ?o.Eo 'ear' are probably unrelated. Note also PYbv *~ukolGG 'ear'. YU

20. Neck. PMiw *hele'.ki 'neck' is not cognate with Ceb ran.aj 'neck, nape', Csjb ka,t.ak 'nape', and probably not cognate with PCo *horkos 'throat'. PMis *my?-9-'to swallow' may be borrowed from PYnim 'to swallow'. Note PYnim *rnyk'yS'throat, neck'. Mins, Mip'nape' is not cognate with these forms, but it may match PW *toq 'neck'. PM *kuj 'neck, throat, to swallow' does not match any of the above forms. U-W

31. Nails. Mib pic.i 'nail, claws' is probably a loan from Mn biCi 'nail, claw, hoof', as Dixon and Kroeber suggest. Note PM *biCi. 'nail, claw, hoof ', Mn biCi. -Cibi 'nail, claw, hoof', an instance of remote consonantal metathesis. (2) As noted in 28 Hand (I), PMi *,ti.s -*,tis.y 'claw, nail, hoof' is probably cognate with PCo *tu.r -*tu.ris 'claws, nails', despite the fact that they do not share a segmental phoneme, which underscores the fact that shallow comparison should be performed as carefully as possible. Ybv gotcoyic 'nails' (D&K) and PW *kahaj 'nail(s), claw(s)' do not match.

(1) D, (2) D

32. Foot, leg (1). PMis *hate-'foot, tracks' may be cognate with Cscl hattam 'foot' GM and Cscr hatash 'feet' JC (note Cscr hd-tac 'sole of foot' HWH). Both may be cognate with Yhom hatasij 'heel' JPH (not in D&K). Note PMis *ha.~y-'to step, step on, stamp' and PYv *ha.thim ? 'to dance'. PYgen *wutog 'foot' is unrelated, as is PMis *tug(.)u-'thigh'. YU ??

42. Blood (1). PCo *patja-n 'blood' may be cognate with PYv *phajax 'blood'. PY *'blood' is unrelated, as is PMi *kiEaw 'blood'. YU

44. Liver. PMi *kyl.a -* 'liver', PCo *sira 'liver', PU *kyl.a *ky(.)la 'liver' are not cognate with PY *tiph < ** 'liver' (Ykr dalapic 'liver' [D&K] does not match). The Utian forms do match PM *kyl.a 'liver' and PW *kila 'liver' (note Wpcc Ehkla 'liver'). This is one of the best of Dixon and Kroeber's Penutian sets. Note that palatalization has occurred in Cache Creek Patwin. If the explanation is borrowing, it must be very deep. U-W-M

50. Bow. The bow and arrow were relatively late in California. Several sets are involved here, and none can be reconstructed to the deepest level within families, with the exception of Maidun and Wintun. (1) PMiw * 'bow' is unrelated to Mip tanuk.a--tonuka-'bow, bow and arrow', which is probably a loanword from Costanoan. Note Cscl tanukam 'bow' GM and Csf Pa nu ka 'bow', also Mip kaca--kyca-'bow'. PW *nun 'bow' does not match, but Ww kol, kulu.1 'bow, gun' matches PMiw * 'bow'. (2) Dixon and Kroeber give Ceb conok 'bow', which resembles Mics'bow'. I have no explanation other than chance or borrowing. (3) PYnim *ialiph 'bow (sinew-backed)' may be borrowed around. It is not cognate with Ypal kelgal 'bow7 or PM *pandak, *panda$ 'bow', which may in turn be a ken- ning term (note Mm pa'n 'to make string or rope by rolling material on the thigh' and ddk"to stick or glue something'). (1) U-W, (2) D, (3) D

72. Star. (1) Mics &le-'Morning Star', Mil td7le 'star', PMi *ty?le 'star7 ? match Wp tilamen (D&K), but not Ww luju.q 'star' or 8atu.s 'star'.

(2) PY *Caja.tas -*Cojo.tis -*Cajtas 'star' CC matches Ww 8atu.s 'star'. Miss ca1a.t~-'star' is probably a deep loan from Yokuts. (3) Ww luju.k 'star7 matches Mm lylj 'star'. (1) U-W, (2) Y-W, (3) W-M

77. Rain. PMis *nyk.a-'rain, to rain' is unrelated to Miss 7u.muc-a'year', Mip'rain', and PMiw "7u.p~-*?up.a 'rain, to rain'. PCo *7amne 'to rain' CS, PCos * 'rain' are unrelated, as are Ww luhes 'rain' and Wph 'raining'. D

80. Smoke. PMi "ka.1, *ka.lis, PCo *ka(.)r,PCos *ka.ris -*ka.res, PU *ka.l -*ka.lis 'smoke from a fire' are unrelated to PMis *'smoke

from a fire', PY *mu.iak' 'smoke'; PM *suk' 'smoke (of fire)'; PW *nu.q 'smoke' and PW *mo.q ? 'smoke'. U

85. Earth. More than one set is involved. (1) PMi "wa1.i -* 'out- side, area, world', Ceb warep 'earth', PU *wal.i(p) -*wa(.)li(p) 'earth, world, outside'. The Proto-Utian form may be cognate with PYn *wa.lJa 'sky' (note PMis *wali.s-y- 'to dawn'), and the Yok-Utian set matches Wph wilak 'plain, valley, ground, world'. (2) PCo *kari 'outside' CS matches PM *k'aw 'earth, ground, dirt'. (3) PMis *py.laj 'field, valley, a flat place', PCo *pire 'world, land, earth', PU * 'ground, field, valley' may be cog- nate with PYgen *pa?an' ? 'world'. PY *haphas 'ash, dirt' ? is unrelated, as is Csjb mun 'loose earth, dirt, dust'. (1) YU-W, (2) U-M, (3) YU

104. Bear. Again, several sets are involved. (1) PCo *?ores < **?ole? 'bear' is unrelated to PYn *to.nohoph ? 'bear'. It may be cognate with Yn, Yv ului 'bear (D&K) or Yn uyum 'bear' (D&K). (2) Ypal molaj' 'bear' may match PM * 'brown bear'. Ypal tens'i? 'grizzly bear' is unrelated. PMiw *kule 'bear, grizzly bear' is probably based on a stem meaning 'black' (note PMie *kulu- 'black' and PMie *?ysy.-mati- -*?ysy-m.afi'bear, grizzly bear: bad creature', probably another euphemism resulting from an avoidance taboo). Wp tcuku, sikai 'bear' (D&K), PW *$il 'bear', and Wpcc sildj 'bear (generic), black bear' are unrelated to the preceding sets. (1) YU ??, (2) Y-M ??

106. Puma. Mil Pupu'k~e, ?zip.ukse 'mountain lion' is unrelated to Cru xe.ks, PY *wyhy.syth, PM *pekun, PW *pate, all meaning 'mountain lion (puma)'. Wp wok'en (D&K) 'mountain lion' matches the Proto-Yokuts form.


120. Salmon. PCo *(h)u.raka 'salmon' is unrelated to Ceb Ee.ri 'type of black-skinned fish', which is probably a loanword from Spanish el chere or cheri. Ccl tcipal 'salmon' (D&K), PW *nur 'salmon', and Ww Eejta 'fall salmon' are also unrelated, but the Proto-Costanoan form does match Wph hur 'salmon (both varieties)'. U-W

130. White. (1) Cru paxla-st 'white' might be cognate with Miss pasa- s.i- 'white' (not in D&K). Note PMi *pup- -*poto- 'white'. (2) Csjb tcalka -pal(.)ka 'white' A may be cognate with PYnim "$0101 (-*Eolol ?) 'white7, and this set matches Wp tcaloki 'white' (D&K), but not Ww xaja 'to be white (as snow)', Wp sum 'white' (D&K), or Wn Luyuke 'white' (D&K). PMis *kele-1.i- 'white' derives from PMie * 'snow' and is unrelated, as are Mm kdkkok 'white', Mk edali 'white' (D&K), and Mn kow 'silver (white)'. (3) Yhom /away 'white' matches Wpcc iama. 'white'. (1) U ?,

(2) YU-w, (3) Y-W

133. Red. PMie *wete- 'red' is not cognate with PMi *julu- 'red, light colored', Cru juri-t- 'redness', PU *july- 'red', which is related to the set PMis *jul.a- 'fire, a blaze, light', PMi *jul- 'to blaze, burn', Csjb jul-ke 'to blaze, burn', and PU *jul- 'to blaze, burn'. None of these sets is cognate with PYn *hapilk'aj ? 'red' or PYnim *paZhikin ? 'red', Yn tcoyikin 'red' (D&K), Yv goliklai 'red' (D&K), Ybv xaw'al 'red', Ypal holotsgai 'red' (D&K), or with PW *tede 'red', Wpcc tu'luk?a 'red', Wp warak (D&K), or with PM *la$ 'red'. U

161. Run. PYgen *hiwith 'to walk' (not in D&K) is probably a deep loan from PMis *hywa.t-y-'to run'. I had considered them possible cognates in my Yok-Utian article (Callaghan 1997), but a second look at Western Miwok cognates-Mib 'to run', Mil hicaw 'to run', PMiw *hic'.aw 'to run7-leads me to reconstruct PMi * 'to run' and to postulate me- tathesis in Proto-Sierra-Miwok. Such reasoning makes borrowing of the metathesized form into Yokuts mote' probable than cognate relationship. PYnim *lawith'to run' is not part of this set. Ws witili 'to run' (D&K) < PW *witil 'to be fast, do fast' does not match the preceding forms. (Note that Mib -ti in hi~wa-ti'to run, run away' is a perfective suffix, not part of the stem, removing partial similarity.) Mm we16 'to run' may match the Wintun form, but not Mk yewo 'to run' (D&K). W-M ?
Yokuts-Maidun Direction Terms

    North. PY *xosim 'north' does not match PM *tos 'north'. PYv *thoxilmeans 'west', not 'north'. D

    South. PY *xomoth 'south' matches Mn komow 'south'. But note Mm komd 'north' and PM *komow 'direction', based on words meaning 'north', 'south', and 'east'. Y-M

    East, upstream. PY *nu.thu-'east' matches Mn notow 'east'. Y-M

    West, downstream. Ywik ta.thu 'west' < PYgen *lathiw ? 'west' does not match Mn taj -ta(.)waj -ta.w.aj 'west'. Note Mm trij 'west'. D

Yokuts-Costanoan Resemblant Forms

    Not find. Ytk wejili 'to not find' (D&K) is not cognate with Csjb waian 'to miss, err' < 'miss the mark' ? A because of semantic divergence. D

    Cheat. Yyaw < *ju.mi 'to cuckold someone' may be cognate with Csjb jum.e 'to lie, cheat, deceive, mislead'. YU

    Swallow. PYnim *magy < **my+k'i (**my.k'u ?) 'to swallow' may be cognate with Csjb mujku 'to swallow without chewing (soft substances)'. Note PMis *my?ty-'to swallow', whose stem (my?-)might be a loan from Yokuts. Note also Ceb muj.e 'to eat pinole'. YU

    Hear. PYnim * 'to hear' is not cognate with Csjb namti 'to hear, understand'. D

    Hang. PYnim *lalik 'to hang' is not cognate with Csjb 'located on top' because of semantic divergence. D

    Fear. PYnim *p'a.xa 'to fear' may be cognate with Csjb pojko-n 'to get scared, be frightened'. YU

    Drown. Ywik r!oguniu 'to drown' is not cognate with Csjb 'to drown' A. D

    Cold, to be. Ytk dotc 'to be cold' (D&K) is probably not cognate with Csjb tursi 'to be cold (weather)'. I cannot be sure until I find a better transcription of the Yokuts. D

    Throw at. Ytk to 'to throw at' (D&K) is probably not cognate with Csjb fu 'to strike in the eye' A, whose basic meaning apparently pertains to the eye. Note Csjb tuknu 'to signal with the eye(s)'. D

    Embrace. PYnim *kom-? 'to embrace' is probably not cognate with Csjb kapla 'to embrace'. D

    Urinate. PYnim *&jo < **Cuju 'to urinate' GG may be cognate with Csjb 2ala 'urine, to urinate'. YU

Penutian Kinship Terms

    Older sister. PMis *te.te-'older sister' is not phonologically close enough to PCo * 'sister, older sister' or to PYgen *nu-?ath'older sis- ter' CC to be considered a probable cognate. Wph ?lichu 'older sister' and Wp otu-n 'older sister' (D&K) do not match the Utian forms, but they may match the Yokuts form. Y-W

    Younger sibling. PMis *;ale-'younger brother' is not cognate with PCo *taw~i--*taw~e-younger brother', but it may be cognate with Csjb tare 'younger brother or sister' A if we posit 1;-I through diminutive sound symbolism in Sierra Miwok. PMis *?ate-'younger sibling' is not related to these forms, and Wph Ea.n 'younger sibling' does not match any of them. U

    Grandfather. PMie *'grandfather', PCo * 'grandfather', PU * 'grandfather' may be cognate with PYgen *no-phoph 'father'. Note Ygash 'father, voc.' (Newman 1944:222). PW *?'grandfather' also matches. YU-W

    Grandmother. PMis *?am(.)a-'grandmother' matches PW *?ama'grandmother'. I could not find Csjb ama (father's parents) in Mason (1916) or in the analyzed portion of Harrington's Mutsun notes. U-W

    Mother. PCo *?an.a-n 'mother' does not match Wpcc na.khe'. 'mother, voc.' or Wpcc ne.n 'mother'. Mics 7an.i~-y-means 'maternal aunt younger than mother (and perhaps other female relations)', not specifically 'stepmother', and it is probably not cognate with the Costanoan form. D

    Mother's brother. PMi *ka.ka 'mother's brother' may be cognate with PYnim *?a.ka~'mother's brother'. YU

    Child's parent-in-law. PMis *maksi-'co-parents-in-law, co-mothers- in-law' is probably not cognate with PCo *mak.u(h) -*mak.o(h)'husband' because of a substantial difference in meaning. Ytk makci 'child's parent-in- law' (D&K) is probably a loanword from Sierra Miwok. D

    Father. PMiw "7ap.i -*?api 'father' is probably cognate with PMis *?ypy-'father', suggesting PMi *?ap(.)y 'father', which is in turn cog- nate with PCo "7ap.a 'father' CS, yielding PU "7ap.y 'father'. These forms may be cognate with Ytk opo-yo 'father, voc.' (D&K). YU

    Son. PMis *?agsi-'son' is probably not cognate with PCo *? 'son'. But the Costanoan form may be cognate with PYnim *?e.nas' < **?i.nas 'grandfather, man's grandchild'. YU

    Mother's mother. PYsv *khami$ 'grandmother' is probably not cognate with Mics kuma?sa-'aunt, mother's brother's wife' because of semantic divergence. D

Yokuts-Maidun Resemblant Forms

    Tear, to. PYnim *khy-jy'to tear' may match Mm he-kot-o 'to break, crack of itself' ?, although there is considerable difference in meaning. Mm he-is an instrumental prefix meaning 'to move down, off, or apart'. Y-M ?

    Smell, to (tV). PYgen **hag- (<**hyg-) -*s/~i.g. . . CC 'to smell' may be cognate with PU *huk.y ? 'to smell something'. This possible Yok-Utian set matches PM *hi 'to smell something'. YU-M

    Fly, to. PYnim *hajin ? 'to fly' may match Mm khj 'to fly'. Y-M

    Cheat, to. Ywik motot 'to cheat' is probably not cognate with Mins m0c.a-'to hide something, save' because of disparity in meaning. Neither one matches Mm pedd 'to steal'. D

    Cry, to. Ywik ?ahan 'to cry out' matches Mm ?a. . . 'to say'. Y-M

    Burn, to. PYnim *hlu 'to bum' may match Mm ty'ti 'to bum something'. Y -M

    Bend, to. PYnim *nukhum' ? 'to bend' does not match Mm not 'to bend' (D&K). D

Miwok-Yokuts Resemblant Forms

    Kick, to. PMie *tekmy-'to kick with the toe' does not match Ywik taj 'to kick'. D

    Hang, to. PMis *jy?ki-'to hang something up' does not match Ywik tot 'to hang' or PYnim *lalik 'to hang'. D

    Hear, to. PMis *tolej-y-'to hear, understand' does not match Ywik tag 'to hear' < PYnim * 'to hear'. D

    Hit with the hand, to. PMis *kut.e--*kut.e-'to push, shove, push down' does not match PYnim */&lo'to hit, to throw'. D

    Throw, to. PMis *?o.pu-'to throw and hit, hit with the fist' does not match Ywik #hm' 'thro~'.~


    Pay, to. PMie *tiw.a-'to buy' may be cognate with Ywik fawtaf 'paying'. YU ?

    Shout, to. PMis *kaw-ye-'to shout' does not match Ytk tcaw, tcaii 'to shout' (D&K). D

    Pull, to. Mins kup.a--kyp.a-'to pull, drag, lead, pull down' does not match Ytk xapit 'to pull out' (D&K). D

Gamble (1978) uses # to indicate a base
Miwok-Maidun Resemblant Forms

    Whistle, to. PMis *kuja.k-y-'to whistle with the mouth' does not match PM *hu.i'whistle' or Mm hlik 'whistle'. D

    Press, to. Mins tupi 'to press' (D&K) may match Mm k'i-tap 'to squeeze'. Note Mm k'i 'to pinch'. U-M ?

    Break, to. Mins tas-wa-'to break' matches Mm tus 'to break' (D&K). U-M

    See, to. Mins hete-7! 'Look!' does not match PM *Ce 'to see'. D

    Die, to. PMis *jyn.a--*jin.a-'to kill' does not match PM * 'to die' or Mm he'no 'to die in bunches'. D

4. Discussion. Dixon and Kroeber's 171 sets actually number 173, since 1 la and 63a are also included. There are several additional sets, most of them involving only two families, which I have added and numbered consecutively, beginning with 172 and yielding a total of 218 sets.

I have discarded 66 of these sets which I believe show insufficient evidence of resemblant forms linking at least two of the language families. In this regard, I have been more stringent with proposed Utian and Yok- Utian sets. For example, I reject PY *'three' and PCo *kap(.)e-s*kapha -*kapha-n 'three' as possible cognates, since there is no other ex- ample of a Y $ : U k correspondence in initial position in my corpus, but I would have admitted it between two other families.

I cannot overemphasize the necessity of performing etymologies within families to weed out spurious resemblances. An example is PCo *~ak.en 'six', which superficially resembles Mm saj-dok'o 'six', but which is actually a stem compound meaning 'both threes', where Mm -dok'o means 'both'.

Of the remaining sets, some consist of two or three subsets, yielding 182 resemblant sets. Examining table 1, we note the low number (13) of sets that link all four families Dixon and Kroeber sought to relate, counting Miwok and Costanoan as a single family (Utian). By contrast, 38 sets in- volve only Yokuts and Utian, more than any number connecting two of the other families exclusively, and 64 sets include Yokuts and Utian resemblant forms, again a greater number than for other two-way comparison^.^ More importantly, the sound correspondences Dixon and Kroeber discovered

'The number of resemblant forms between Wintun and other languages is high. Whistler (1977) presents evidence for the slow encroachment of the culturally dominant Patwin into

Yok-Utian     38     Utian-Maidun     14
Yok-Utian-Wintun     9     Utian-Wintun-Maidun     7
Yok-Utian-Maidun     4     Yokuts-Wintun     16
Yok-Utian-Wintun-Maidun     13     Yokuts-Maidun     23
Utian     14     Yokuts-Wintun-Maidun     8
Utian-Wintun     24     Wintun-Maidun     12
Total Yok-Utian     64     Total Yokuts-Wintun     46
Total Utian-Wintun     53     Total Yokuts-Maidun     48
Total Utian-Maidun     38     Total Wintun-Maidun     40

among their sets appear to be sporadic, even using their full corpus, except in the case of Miwok and Costanoan.

4.1. More evidence for Yok-Utian. Initial comparison of Proto-Yokuts and Proto-Utian yielded several regular and recurrent sound correspon- dences (Callaghan 1997), and a reanalysis of the Dixon Kroeber corpus has greatly augmented the number of resemblant forms and added two new correspondences. This material plus additional Yok-Utian sets now follows. For the sake of clarity, I have included additional sets that illustrate the new correspondences.
English Yokuts Utian

breakfast PYnim *wa.?al 'breakfast' Mib w&let 'to have breakfast' (Mins wa.1~-'breakfast' loan from Yokuts?)

girl PYv *ka.?i-tha'girl' PMiw *ko'.la 'girl, young lady' PYgen *kuju.lum ? 'girl'

give, to PY *walin 'to give' PMiw *wcija -*wcije 'to give' > PYnim *wa& 'give' (< **wala -**wale ? (PC0 *(h)ara'to give')

head PYgen *to701? 'head' Mip tolo-'head, hair'

lake PYgen *pa7as 'lake' PMis *pol.aj 'pond, pool, lake' Mil pdl-pol 'pond, lake'

man (male) PY *fha7afh?? 'person' PMiw *tciji: 'man male'
PCo *fa,ri~'man male'
PU *fa(.)lij'man male'

Miwok territory, and Callaghan (1964; 19876) lists Patwin loanwords into Lake Miwok. Borrowing could easily account for many resemblant forms between Wintun and other lan- guage families.

shine, to world

break, to


fish man (male)



blow with the mouth


steal. to


dance, step, to ground squirrel PYsv *ta.7ac 'to shine (sun)' ?

PYgen *da?an' ? 'world'

PYnim *fha,thaj'to break'

PYn *holh(< **hu.yh?) 'buzzard'

PY *lophirh'fish'

PY *fha?a'afh?? 'person'

PYnim *da,thuw? 'worm'

Yokuts and Utian f, p,f -Yokuts

PY *ta.m'u; > PYn *ta,mufh 'beard'

PY *duju.sul 'knee, elbow', PYsv *dujui'knee'

PYnim -*7o.s-? 'to steal'

PYv *ha.thim ? 'to dance, sing'

PY *sixthil -*?ixi.thal 'ground squirrel' > PYn *?ithxil(through metathesis)

Mil i'u'lip 'to shine, glisten'

Mins py-laj-y-'field' PCo *pire 'world, land, earth', Csjb pira 'bury' PU *py(.)la(j)'world', land, earth'

PMis *fa!-'to break'
(Mil th6d.a 'break off ')

PMis *'buzzard'

PMis *lapi.s-aj 'fish'

PMiw *triji: 'man male'

PCo *fa.ris 'man male'

PU *!a(,)li?'man male'

PMie **ke.~y.
s, s


Mil 7dra <to cheat7
Csjb ha.,ta 'steal'

Miss tolo.ko-,t
Mics tolo.ko-s-u-

PMis *ha-N-'to step, stamp'

PMiw *?(re 'ground squirrel'

Csjb sirikna 'young ground squirrel'

Although three series of stops and affricates can be reconstructed, there is evidence that as- piration and glottalization are secondary (Callaghan 1997:25).

woman, girl

bear cheat, to dance, to

eat, to ?

father fear, to fox grass, brush

heel mountain paying pestle see, to son turtle swallow, to

PY *hath--*both-in *hath-pagij

*hoth-pogoj'four' (PY *payij-*pogoj-*poqij'two') PYU *-f-> PY *-t-, *-th-1 -i ??? PY *hoth-< PYU *hoyi 'two' ???

Ypal ketephal' 'woman' < PY *kathipha19??

PU *?o~i-*Pot-'two'
Mil 7o!-d!.a 'four: two twos'

Csf Ca tra 'girl or maid'
Additional Resemblant Sets


Yn, Yv ului (D&K) Yn uyum (D&K) 'bear' Yyaw < **ju.mi 'to cuckold' PYnirn *kham'-'to dance' PYv *xatha'to eat' CC

Ytk opo-yo 'father, voc.' (D&K)
PYnirn *pa.xa 'to fear'

PYnirn *?aw'Pa-'fox'

PYnim *ja.wil 'grass, brush'
Yhom hatasij 'heel'
PY *lomith'mountain'
Ywik thawtaS'paying'
PYnirn *phaluj? -*phalwoj?


PYsv *?il'i 'to see'
(PY *sasa-'eye')
PYnim *?e.naS < **%.nus

'grandfather', man's grandchild' Ychuk low' 'turtle'

PYnim *ma.$y < **my& (**my,!& ?)

PCo *lore? < **?ole?'bear'

Csjb jum.e 'to lie, cheat, deceive'

Mib kd.wul 'to dance' Cru xamp 'to eat' P Cru kat 'to eat' (D&K)

PU *?ap.y 'father' Csjb pojko-n 'get scared, frightened' Mim Ah-wdh-ke 'gray fox'

CHM Csf Yah wah 'grass' PMis *hate 'foot' PMis *leme-'mountain' PMie *tiw.a-'to buy' Mic pa, paya 'pestle' (D&K)

(< *pala ?)

PMis *?yj-ye-'to see'
(PU *syn-ty'eye')
PCo *7in,i?'son' < **? ?

(PMis *7aysi-'son')

PMie *7aw(,)ana-'turtle'
PCo *7awni?-min ? 'turtle'
Csjb mujku 'to swallow

without chewing' Ceb muje 'to eat pinole'

Palewyami /el may derive from *i, *y, *u, or *a (Whistler and Golla 1986:337).
this     PY *xi(.) 'this near' GG     PMi *?i. -*li 'that; he, she, it' Ceb 7i 'his, her, its' PU *7i(.) 'that, 3 sg.'
tule (round)     PYv *$ojis ? 'tule sp.' Ypal kololi-s 'tule' (D&K)
    Mil kd.1 'round tule'
white     PYnim *to101 (-*c'olol ?) < **Cola1 (-**cola1 ?) 'white'     Csjb tcalka pal(.)ka
'white' A

5. Conclusions. A reanalysis of the Dixon and Kroeber material un- covered 27 new resemblant sets between Yokuts and Utian, but it offered only sporadic support for links with Wintun and Maidun. The biggest sur- prise was the recognition of two new sound correspondences, which encour- aged me to review my earlier material. PY *-7-: PU *-1-is the only recurrent correspondence so far involving medial glottal stop, and it occurs after Yokuts la, 01. It is too early to determine if this correspondence is in complementary distribution with another correspondence. The second sound correspondence-PY *t, th, *i : PU *$, *s-is also restricted to noninitial position, and it is supported by corresponding alternations within Yokuts and Utian.

The additional resemblant sets include a possible Costanoan cognate for PYnim *ma.k'y < **my.tY 'to swallow', Csjb mujku 'to swallow without chewing (soft substances)', which is more viable than PMis *my?-?y-'to swallow', whose stem (my?-)may be a loan from Yokuts. The vowels of Csjb mujku fit PY *y : PU *u, which occurs in the environment of labials.

A second look at the set for Run leads me to think PYgen *hiwith'to walk' is probably a deep loan from PMis *hywa.t-y-'to run' instead of a possible cognate. PMiw *hi&aw'to run' suggests PMi * 'to run' with metathesis in Sierra Miwok. Continued research of this kind yields insight into the complex problem of distinguishing loanwords from potential cognates.

Although reanalysis of the Dixon-Kroeber corpus has produced addi- tional evidence for kinship between Yokuts and Utian, genetic relationship has not yet been substantiated. Yokuts and Utian have been spoken side by side for centuries, if not millennia, and the trade routes were east-west. It is conceivable that the resemblances could have resulted from long-term diffusion. What has been vindicated is the fact that my methodology of stage-by-stage comparison from shallow to deep can yield detailed results, even for deep levels.

Although I cannot yet claim genetic relationship, I will indulge myself in a little fantasy. I have found four possible examples of Yokuts -t-, -th- : Utian -!-,enabling me to propose a sound law: PYU *-i-> PY *t, *-th-/-i. If this fantasy is valid, I can reconstruct PYU *ho!i 'two', which means, I suppose, that we should refer to the Yok-Utian family as "Hotian."

5.1. Relevance for historical Amerindian studies. Dixon and Kroeber (1919:69) were aware that the sound correspondences within their Penutian resemblant sets were sporadic rather than systematic, and they attributed this fact to ignorance of the scope of ". . . expansion, contractions, assimi- lation, metathesis, analogizing, and reduplication."

Subsequent studies have revealed that the particular expansions and segmentations which Dixon and Kroeber proposed were rarely justified, and that the sound correspondences, especially the nonidentical correspon- dences, among the remaining sets were sporadic except in the case of Utian and Yok-Utian. The typological similarities that Dixon and Kroeber also noted, yuch as case systems and absence of prefixes or stem-initial con- sonant clusters, are intriguing but they do not constitute proof of genetic relationship.

None of these observations proves that the languages in question are un- related, since there is no known way for determining that natural spoken languages are unrelated. All we can claim is that the evidence in Dixon and Kroeber (1919) is insufficient for asserting genetic relationship, even when updated.

Frachtenberg (1918:176-77) understood the deficiencies of their pro- posal early on: ". . . Dixon's and Kroeber's presentation of Penutian and Hokan are exceedingly inadequate, both methodologically and in regards to subject matter."1° Unfortunately, he followed this statement by assert- ing, "The absence of conclusive evidence concerning Penutian and Hoken is the more unfortunate, as there exist strong reasons to believe that the Takelman, Kusan, Siuslawan, Yakonan, Kalapuyan, and (perhaps) Chi- nookan languages spoken in Oregon may be proven to be Penutian sister- tongues." Frachtenberg thus set the precedent for assuming the reality of "Penutian" without proof and incorporating, at least tentatively, even more languages into this poorly defined family, which was continued in Sapir's (1929) master classification of Amerindian languages north of Mexico into six superfamilies, one of which was "Penutian."

In the nineteenth century, various languages and families of languages were proposed as being Indo-European, meaning that they could be sys- tematically related to a core of languages: Latin, Greek, Sanskrit, and Germanic. At the same time, a close-knit set of correspondences was be- ing elaborated among the "core" languages. Pitkin and Shipley (1958) and later Shipley (1965) made heroic efforts to discover such correspondences among the languages classified as California Penutian. Shipley (1983-84)

lo Frachtenberg (1918:176-77) had examined Dixon and Kroeber's corpus of comparative vocabularies during a visit to San Francisco.

expressed his dissatisfaction with the results and declared that the term "Penutian" had outlived its usefulness.

The notion that California Penutian constitutes a genetic unit has already been discarded by most specialists in the field, even those who believe that the families are ultimately related at a deeper level. This article further demonstrates that there is no set of systematic correspondences linking the so-called California Penutian languages. The result is that scholars who argue for a wider affiliation of languages under the umbrella of "Penutian" are left without a core set of languages linked by sound correspondences, and consequently without a clear answer to the question of what it means to claim that a certain language or language family is "Penutian."

Fortunately, Goddard (1996:l-16) has opted for a conservative classi- fication for the new Handbook of North American Indians, which undoes the Sapir classification and lists Utian, Yokuts, Maidun, and Wintun in- dependently. Hopefully, this new classification will help dispel any mis- conception by nonspecialists that genetic relationship among California Penutian languages has been substantiated.


BROADBENT, M. 1964. The Southern Sierra Miwok Language. UCPL 38. Berkeley and

SYLVIA Los Angeles: University of California Press. CALLAGHAN, A. 1964. Phonemic borrowing in Late Miwok. Studies in California


Linguistics, UCPL 34, ed. William Bright, pp. 46-53. Berkeley and Los Angeles: Univer-

sity of California Press.

.1965. Lake Miwok Dictionary. UCPL 39. Berkeley and Los Angeles: University of

California Press.

. 1970. Bodega Miwok Dictionary. UCPL 60. Berkeley and Los Angeles: University

of California Press.

. 1971. Saclan: a re-examination. Anthropological Linguistics 13:448-56.

. 1972. Proto Miwok phonology. General Linguistics 12: 1-3 1.

. 1984. Plains Miwok Dictionary. UCPL 105. Berkeley and Los Angeles: University of

California Press.

. 1986. Proto Utian independent pronouns. Southern Illinois University Occasional

Papers on Linguistics 13: 12-24.

. 1987~. Northern Sierra Miwok Dictionary. UCPL 110. Berkeley and Los Angeles:

University of California Press.

. 19876. Lake Miwok naturalization of borrowed phonemes. Working Papers in Lin-

guistics 35, ed. Brian D. Joseph and Arnold M. Zwicky, pp. 84-93. Columbus: Department

of Linguistics, Ohio State University.

. 1988~. Proto Utian stems. In Honor of Mary Haas: From the Haas Festival Con-

ference on Native American Linguistics, ed. William Shipley, pp. 53-75. Berlin: Mouton

de Gruyter.

. 1988b. Karkin revisited. IJAL 54:436-52.

. 1988c. The state of Miwok /I$. Ms.

. 1990. Proto-Costanoan numerals. IJAL 56:121-33.

. 1991. Utian and the Swadesh list. Southern Illinois University Occasional Papers on

Linguistics 16:218-37. . 1992. The riddle of Rumsen. IJAL 58:36-48. . 1993. Proto Utian interrogative pronouns. American Indian Linguistics and Eth-

nography in Honor of Laurence C. Thompson, ed. Anthony Mattina and Timothy Montler, pp. 329-41. University of Montana Occasional Papers in Linguistics, no. 10. Missoula: University of Montana.

. 1994. Proto-Miwok numerals. IJAL 60:161-76. . 1997. Evidence for Yok-Utian. IJAL 63:18-64. CALLAGHAN, A,, AND ZINNY S. BOND. n.d. Marin Miwok dictionary. Ms.

CATHERINE COLLARD,THOMASLEE. 1968. Yokuts grammar: Chukchansi. Ph.D. dissertation, University of California, Berkeley. DIXON, ROLAND B., AND A. L. KROEBER. 1913. New linguistic families in California. Ameri- can Anthropologist, n.s., 15:647-55. . 1919. Linguistic families of California. University of California Publications in American Archaeology and Ethnology 16:47-118. FRACHTENBERG,

LEO J. 1918. Comparative studies in Takelman, Kalapuyan and Chinookan lexicography: a preliminary paper. IJAL 1:175-82. FREELAND,L. S., AND SYLVIAM. BROADBENT. 1960. Central Sierra Miwok Dictionary with Texts. UCPL 23. Berkeley and Los Angeles: University of California Press. GAMBLE,GEOFFREY.1976. Aspirated stops in Proto-Yokuts. Ms. . 1978. Wikchamni Grammar. UCPL 89. Berkeley and Los Angeles: University of

California Press. . 1988. Reconstructed Yokuts pronouns. Diachronica 5:59-71. . 1991~.Palewyami: a Yokuts key. A Festschrift for William F. Shipley, ed. Sandra

Chung and Jorge Hankamer, pp. 61-81. Santa Cruz, Calif.: Syntax Research Center. . 19916. Yok-Utian: old California Penutian? Paper presented at the Hokan-Penutian Conference, Santa Cruz, California.

. 1993. Yokuts imperative and demonstrative pronouns. American Indian Linguistics and Ethnography in Honor of Laurence C. Thompson, ed. Anthony Mattina and Timothy Montler, pp. 385-96. University of Montana Occasional Papers in Linguistics, no. 10. Mis- soula: University of Montana.

GODDARD,IVES. 1996. Introduction. Handbook of North American Indians, vol. 17, pp. 1-16. Washington, D.C.: Smithsonian Institution.

GOLLA, VICTOR K. 1964. Comparative Yokuts phonology. Studies in California Linguistics, UCPL 34, ed. William Bright, pp. 54-56. Berkeley and Los Angeles: University of Califor- nia Press.

. 1984. Far Northern Valley Yokuts numerals. Paper presented at the Hokan-Penutian Conference, Berkeley, California. HARRINGTON,

JOHN P. 1985. John k? Harrington Papers 2: Northern and Central California. Washington, D.C.: National Anthropological Archives, Smithsonian Institution.

HEIZER,R. F. 1952. California Indian Linguistic Records: The Mission Indian Vocabularies of Alphonse Pinart. Anthropological Records, vol. 15, no. 1. Berkeley and Los Angeles: Uni- versity of California Press.

. 1955. California Indian Linguistic Records: The Mission Indian Vocabularies of

H. W. Henshaw. Anthropological Records, vol. 15, no. 2. Berkeley and Los Angeles: Uni- versity of California Press.

KROEBER,A. L. 1910. The Chumash and Costanoan languages. University of California Pub- lications in American Archaeology and Ethnography 9:237-71.

. 1941. Culture Element Distributions: XV-Salt, Dogs, Tobacco. Anthropological Records, vol. 6, no. 1. Berkeley and Los Angeles: University of California Press. . 1959. Northern Yokuts. Anthropological Linguistics 1, no. 8:l-19.

MASON, J. ALDEN. 1916. The Mutsun dialect of Costanoan based on the vocabulary of de la Cuesta. University of California Publications in American Archaeology and Ethnology 11:399-472.

MENGARINI, 1877. Santa Clara vocabulary. Contributions to North American Eth-

nology, vol. 3, pp. 538-49. Washington, D.C.: Government Printing Office.

MERRIAM,C. HART. 1979. Indian Names for Plants and Animals among Californian and Other Western North American Tribes. Assembled and annotated by Robert F. Heizer. Socorro, N.M.: Ballena Press.

NEWMAN, STANLEY. 1944. Yokuts Language of California. Viking Fund Publications in Anthropology, no. 2. New York: Viking Fund. OKRAND,MARC. 1977. Mutsun grammar. Ph.D. dissertation, University of California, Berkeley. . 1989. More on Karkin and Costanoan. IJAL 55:254-58. PITKIN, HARVEY, AND WILLIAMSHIPLEY.1958. A comparative survey of California Penutian. IJAL 24:174-88. SAPIR, EDWARD. 1929. Central and North American languages. Encyclopaedia Britannica, 14th ed., vol. 5, pp. 138-41. SCHAMBACH, 1977. A preliminary reconstruction of Proto-Costanoan. Ms.


ALICE. 1981. Wintu Dictionary. Report no. 2, Survey of California and Other Indian Languages. Berkeley: Department of Linguistics, University of California. . 1997. Tentative Proto-Wintun forms (draft). Ms. SCHOOLCRAFT, R. 1860. Archives of Aboriginal Knowledge. Vol. 2. Philadelphia: J. B.

HENRY Lippincott & Co. SHIPLEY,WILLIAME 1963. Maidu Texts and Dictionary. Nisenan Texts and Dictionary. UCPL

35. Berkeley and Los Angeles: University of California Press. . 1983-84. Penutian among the ruins: a personal assessment. Journal of the Steward

Anthropological Society 15:59-63.
. 1997. Proto Maidun lexicon (draft). Ms.
. n.d. Yokuts vocabularies. Ms.

SMITH, RICHARD. n.d. Nisenan-English and English-Nisenan dictionary. Ms. TAYLOR, ALEXANDER

S. 1860. California Notes: Indianology of California. The California

Farmer, September 4. ULDALL, HANS J~RGEN, AND WILLIAMSHIPLEY.1966. Nisenan Texts and Dictionary. UCPL

46. Berkeley and Los Angeles: University of California Press.

WHISTLER, KENNETH W. 1977. Wintun prehistory: an interpretation based on linguistic re- construction of plant and animal nomenclature. Proceedings of the Third Meeting of the Berkeley Linguistics Society, pp. 157-74. Berkeley, Calif.: Berkeley Linguistics Society.

. 1979. Patwin numerals: A reconstruction and its cultural implications. Paper read at University of California, Santa Barbara. . 1980. Proto-Wintun kin classification: a case study in reconstruction of a complex semantic system. Ph.D. dissertation, University of California, Berkeley. . n.d. Patwin notes. WHISTLER, KENNETH W., AND VICTOR GOLLA. 1986. Proto-Yokuts reconsidered. IJAL 52:317-58. . n.d. Proto Yokuts reconstructions.

  • Recommend Us