Notes on Comparative Penutian

by Howard Berman
Notes on Comparative Penutian
Howard Berman
International Journal of American Linguistics
Start Page: 
End Page: 
Select license: 
Select License


Here, I propose some new Penutian cognate sets and add some cognates to sets proposed by others. The arrangement is in geographical order from north to south.

BE (inanimate). Rude (1997: 137-38) has reconstructed Proto-Sahaptian (PS) *pe 'be situated, be located' (of inanimates) and *we 'be' (general copula). Molala pi'be' (of inanimates) is cognate with PS *pe:'

na.t tam 1u.l~ pi-ssi.

again many egg be-~~s~


'Again there were many eggs there'. (Stevens Savage)

kh-ni-phti-slan nuqun hillim m-pi-si-yi-0.2

f-go-back-FUT Sg2 where house ~ou(s~)-~~-PAsT-BEN-S~~

'You will go back to where you had a house'. (Literally: where a house was for you) (Molala Kate)

In the present tense, pi-is replaced by the suppletive stem wi-, which is cognate with PS *we:

nus luqunc wi-th.

very water be-Sg3

'It's deep water'. (Molala Kate)


In Nez Perce and Sahaptin, the instrumental suffix is -ki. This is probably cognate with the Klamath instrumental suffix -tga. The -t-may be a loca- tive suffix, to which the oblique case suffixes are added (Rude 1997:127, 140). In Molala the instrumental suffix is -ak:

hayks pinhu'.ti nu'.wat k-ld.tin-cyamni ta.p-ak.

then sister-in-law his P1 f-go out-PAST P13 VEN canoe-INST

'Then his sisters-in-law came out in a canoe'. (Stevens Savage)

hayks w&t-ak.

then shoot-PAST P13 arrow-INST

'Then they were shooting with arrows'. (Stevens Savage)

'The Molala examples are taken from the texts of Stevens Savage and Molala Kate, which are kept in the Jacobs Collection at the University of Washington. The following abbreviations are used: BEN benefactive, f feminine, FUT future, INST instrumental, P1 plural, Sg singular, VEN venitive.

The original spelling is piciyi, with the older Americanist use of c for 5. Jacobs probably missed the initial m-because of the preceding final -m of hillim.

[IJAL, vol. 67, no. 3, July 2001, pp. 346-491
Permlsslon to reprint a note printed in this section may be obtained only from the author(s).

Thus, the instrumental in k is shared by all of the Plateau Penutian languages, in contrast to the California Penutian instrumental or comitative in -ni found in Yokuts, Western Miwok, and Maiduan (Berman 1983:402).

CHOP, CUT, BREAK. DeLancey (1989:36) has suggested that the first two mor- phemes in Maidu wo'o-kot-dau 'to chop off the end of a log' and Klamath w-gatt''chop in two, chop down' are cognate. The second morpheme is actually a wide- spread loanword and is probably of imitative origin, based on the sound of some- thing being chopped or brokem3 Note Achumawi -anakat-'to cut (with ax), to chop wood', -alikat-'to cut (with scissors)', Atsugewi ya-7katii-'to cut with knife', Chi- mariko kat-'break, separate', Kiliwa khdt 'break', Serrano kut 'cut', Central Sierra Miwok ko't-wa-'to break a branch', kdp-ug.e-'to break (intr.)', Lake Miwok ko'ta-'to chop with an ax', and Takelma -k!otCk!ad-'break in two'. The presence of this loan- word in Achumawi and Atsugewi is significant because those languages are located between Maidu and Klamath. It is possible that Molala tagktapi-'break (intr.)' (Ber- man 1996:7) contains a metathesized form of this stem. tag-forms agentless verbs, but the leftover -pi-is unexplained.

GIVE. In Berman (1996:18), I suggested that Klamath -oy-, -wy-, -y- 'give a sin- gle object' was cognate with Molala si-m., yi-f., -hay-f. 'give'. Possible cognates to these in Western Miwok are Lake Miwok waya 'give in one installment' (Cal- laghan 1965:151) and Bodega Miwok waye 'give me' (Callaghan 1970:78).

A NEW YoK-UTIAN CORRESPONDENCE. There are a handful of morphemes in Sierra Miwok ending in CCi which correspond to CiC in Yokuts. One example is Central Sierra Miwok -mSi-,Yawelmani -wis 'reciprocal', with the shift of w to m in Miwok when not followed by a voweL4 Note Yokuts -w 'locative', Sierra Miwok -m'indefinite locative' (Berman 19895). Another example is Central Sierra Miwok wiSki7, Proto-Yokuts *7usik' 'heart' (Whistler and Golla 1986:355 and Callaghan 1997:27). For additional examples, see FIRST SINGULAR POSSESSIVE and POSSESSOR below.

FIRST SINGULAR POSSESSIVE. In Yokuts, many kin terms begin with n-followed by a vowel. This is the only prefix in the language. Newman (1944:221) suggested that it may be derived from Penutian **n-'my'. This prefix does not occur when the suffix -ne.f-'decedent' is added: Yawelmani no7om 'mother', 7omne.t 'deceased mother', ne7es 'younger brother', 7esnef 'deceased younger brother', etc. I suggest that, like the n-prefix, Yawelmani -ne.?-originally meant 'my' and is cognate with the Sierra Miwok suffix -n?i-'my'. Yawelmani has the strong grade of the vowel from underlying *-nif-. In both n-and -ne.f-, the earlier possessive meaning has been lost. If this analysis is correct, the structure of the ordinary Yawelmani kin

'The first morpheme may also be borrowed, since it resembles Northern Paiute wi-'with a long object used radially' (DeLancey 1999:72) and Central Yana and Yahi wa-'long object; to handle a long object' (Sapir and Swadesh 1960:163).

In the following list, the sources are as follows: Achumawi-Olmsted (1966); Atsugewi, Chimariko, and Kiliwa-Gursky (1974); Serrano-Miller (1967); Central Sierra Miwok-Free- land and Broadbent (1960); Lake Miwok-Callaghan (1965); and Takelma-Sapir (1909).

Here and in the following paragraphs, Sierra Miwok forms are from Freeland (195 1) and Yokuts forms are from Newman (1944).


terms and the structure of the decedent kin terms are parallel, both consisting of a stem and what was originally a possessive affix.

POSSESSOR. In Sierra Miwok, the suffix -?.ini-is added to nouns and means 'one who has, the possessor of'. It may be cognate with Yokuts -iyin 'intensive pos- sessor'. Since the Yokuts suffix has a y in it, one would expect Sierra Miwok *-?.iyni-.but there is some evidence that inherited *y was lost in Sierra Miwok be- tween i and n. Note Central Sierra Miwok Siyge-'to see', with the mediopassive suffix -ge-, and its derivative Sinti? 'eye'. In Sinti? the y that occurs in Siyye-has been lost between i and n, just as it has been lost in the same environment in -?.ini-. The initial glottal stop in the Sierra Miwok suffix may originally have been an allo- morphic variant, since it also occurs with certain nouns in Wikchamni. Note c'iyliyin 'a bony person' from c'iy 'bone' (Newman 1944:219 and Gamble 1978:95).

HABITUATIVE. The previous set is part of a larger group of cognates which in- cludes the Sierra Miwok habituative. Two of the Yokuts possessor suffixes are -iyin 'intensive possessor' and -min 'entity possessor'. Both have possible cognates in Utian. -iyin may be cognate with Sierra Miwok -?.ini-(see above) and -min matches the Mutsun possessor suffix -min (Berman 1983:406-7 and Callaghan 1997:38). Some Yokuts dialects also have an m-possessor suffix without the final -in: Yawelmani and Chukchansi -m, Ghashowu -?am 'entity possessor'. Yokuts -iyin and -min show the same alternation between i and m found in the Central and Southern Sierra Miwok habituative. In those languages, -i.-is the habituative suffix used with simple verbs and -me.-is the suffix used with complex verbs (Freeland 1951:68). The latter is related to the agentive suffix -me-'one who does by preference, or to excess'. -me-is also used with noun stems, and in at least one example it forms an intensive possessor: Central Sierra Miwok ?dssleme?'a person or a place with many children', with a manipulated form of ?ese'l.i?'child' (Freeland 1951: 152-53). The phonological developments are not clear, but the cognacy of the Yokuts and Sierra Miwok suffixes seems to be guaranteed by the shared i/m alternation and the similar meanings.

ADJUNCTIVE. In Yokuts, the basic form of the consequent adjunctive is -7hiy-, although many dialects lose the or the h. Yawelmani also has a variant -7e.y-, with the strong grade of the vowel. This suffix requires a form of the verb with a short vowel in the initial syllable. The consequent adjunctive refers to any noun other than the agent that is involved in a given activity. Newman's example is Yawelmani xata.7e.y-, from "xata 'to eat'. It has three possible meanings: (1) 'that which is eaten', (2) 'a place or time of eating', and (3) 'an instrumental medium for eating (e.g., a knife, a fork, the mouth, the hands)' (Newman 1944:162). I suggest that this is cognate with the Sierra Miwok instrumental agentive. After a vowel, the suffix is -ya-and the preceding vowel is lengthened. After a consonant, the suffix is -j.a-or -i7ya-, with what Freeland calls an "inorganic i." In both cases, the preceding verb has a short vowel in the initial syllable. Each noun formed with the instrumental agentive suffix has one of the three meanings of the Yokuts suffix: (1) 7iwi.yal 'deer' (7iw.i-'to eat'), (2) Eapiy.a? 'sweat house' (Ea.p-'to sweat'), and (3) st'kilya? 'pencil' (si.k-'to tattoo, to write') (Freeland 1951:153). This proposed cognate set is not a precise phonological match. In particular, the source of the final -a-in Sierra Mi- wok is uncertain. However, the Yokuts and Sierra Miwok suffixes are a good match semantically and they require a similar stem manipulation in the preceding verb, so I think there is a good chance that they will prove to be cognate.
HOWARDBERMAN, Seattle, Washington


BERMAN,HOWARD.1983. Some California Penutian morphological elements. IJAL 49:400-412. . 1989. More California Penutian morphological elements. Southwest Journal of Lin- guistics 9:3-18. . 1996. The position of Molala in Plateau Penutian. IJAL 62:l-30. CALLAGHAN, A. 1965. Lake Miwok Dictionary. UCPL 39. Berkeley and Los An-

CATHERINE geles: University of California Press. . 1970. Bodega Miwok Dictionary. UCPL 60. Berkeley and Los Angeles: University of California Press. . 1997. Evidence for Yok-Utian. IJAL 63: 18-64.

DELANCEY,SCOTT. 1989. Klamath stem structure in genetic and areal perspective. Papers from the 1988 Hokan-Penutian Languages Workshop, ed. Scott DeLancey, pp. 31-39. Eugene: Department of Linguistics, University of Oregon.

. 1999. Lexical prefixes and the bipartite stem construction in Klamath. IJAL 65:56-83. FREELAND,

L. S. 1951. Language of the Sierra Miwok. IJAL Memoir 6. FREELAND,L. S., AND SYLVIA BROADBENT. 1960. Central Sierra Miwok Dictionary with Texts. UCPL 23. Berkeley and Los Angeles: University of California Press. GAMBLE,GEOFFREY.1978. Wikchamni Grammar. UCPL 89. Berkeley and Los Angeles: Uni- versity of California Press. GURSKY,KARL-HEINZ.1974. Der Hoka-Sprachstamm: Eine Bestandsaufnahme des lexikali- schen Beweismaterials. Orbis 23: 170-215. MILLER, WICK R. 1967. Uto-Aztecan Cognate Sets. UCPL 48. Berkeley and Los Angeles:

University of California Press. NEWMAN,STANLEY.1944. Yokuts Language of California. New York: Viking Press. OLMSTED,DAVID L. 1966. Achumawi Dictionary. UCPL 45. Berkeley and Los Angeles: Uni-

versity of California Press. RUDE, NOEL. 1997. On the history of nominal case in Sahaptian. IJAL 63:113-43. SAPIR, EDWARD. 1909. Takelma Texts. University of Pennsylvania Anthropological Publica-

tions, vol. 2, no. 1: 1-263. SAPIR, EDWARD, AND MORRISSWADESH.1960. Yana Dictionary. UCPL 22. Berkeley and Los Angeles: University of California Press. WHISTLER,KENNETH W., AND VICTOR GOLLA. 1986. Proto-Yokuts reconsidered. IJAL 52: 317-58.

  • Recommend Us