Coast Miwok Oral Tradition

Grammar and Ethnopoetic Organization in a California Context

Oral Tradition, 35/1 (2021):67-86 

Little has been published on the oral traditions of the Coast Miwok that provides any information on the original language and linguistic verbal art of this group.1 The Coast Miwok language was spoken north of San Francisco Bay, largely in an area corresponding to modern Marin County and parts of Sonoma County, California, in two dialects, Bodega and Marin.2 More generally, relatively little has been published on the oral traditions of the San Francisco Bay area, which included speakers of the Miwok and Costanoan / Ohlone branches of the Utian language family, as early missionization led to the loss of the languages and associated oral traditions.3 However, there are existing archival sources for exploring Coast Miwok language and oral traditions. In this article, I assemble and evaluate the available information, and for the first time provide accurate linguistic transcription and annotation of two texts representative of the Coast Miwok tradition of oral narrative. In addition to contributing to a basic knowledge of the oral traditions of the Coast Miwok, the article aims to situate these traditions in the broader context of Central California practices, while also commenting on their general ethnopoetic features.

Not even a single narrative of any sort has ever been published in Coast Miwok.4 There are two small collections of Coast Miwok oral literature recorded in English, one by C. Hart Merriam (1993 [1910]), and the other by Isabel Kelly (1978b). Both include a few scattered Coast Miwok words. Kelly (1978a) also provides some basic information about performance of the narratives, which is similar to that reported for many other areas: narratives were told by both men and women, during winter, in the nighttime. Kelly’s field notes have also been published (Collier and Thalman 1991) and provide some additional information on the tradition, including some linguistic terms (which I provide here retranscribed based on Callaghan’s work). There is a specific word for Coast Miwok mythological-type narratives: ‘akkala (Collier and Thalman

1 For basic information on the Coast Miwok people, see Kelly 1978a. More extensive historical information can be found in Goerke 2007.

2 Information on the language can be found primarily in Callaghan 1970.

3 On the Utian languages generally, see Callaghan 2014.

4 The available texts are a brief prayer (Loeb 1926), a translation of the Lord’s Prayer and another translated Christian prayer (Kroeber 1911; Goerke 2007:208), and a series of Bodega Miwok war songs (Goerke 2007:208-09), which are however so poorly transcribed and loosely translated that they are largely unrecoverable.


1991:421). It is related to ‘akkal, “be old.” In these stories, the primary characters are First Peoples (hukke micca, lit. “before people” (87)). Merriam also reports use of the term ‘ayyaako(“children”) to refer to the First People. Although they are animals, they are different from the “mere” animals of the present, and some have different names in the oral tradition. The everyday word for coyote is ‘oye for example, but in the myths he was called /wuyoki/5 (Bodega dialect) (Collier and Thalman 1991:422-23) or ‘oye ‘oyyiş, “coyote old man” (Marin dialect) (98). These characters of the mythical age were then followed by the şukku ‘inniiko, “new ones” (the present humans) (103). Specific beliefs about storytelling itself included the idea that telling a story during the day would actually shorten the day, and that summer nights were too short to tell the stories (421). More generally, the Coast Miwok traditions as documented by Merriam and Kelly belong to the central California cultural area, where Coyote plays a very prominent role, including that of a Creator in many stories, and the idea of the First Peoples is common.6

The last traditional speaker of the Coast Miwok language was Sarah Ballard, who passed away in 1978. Ballard worked with Catherine Callaghan in the 1960s to produce the 1970 dictionary of the Bodega dialect of Coast Miwok (Callaghan 1970), which also contains grammatical information. Callaghan also recorded ten short Coast Miwok texts and two songs from Ballard, which are deposited in the archives of the Survey of California and Other Indian Languages(SCOIL), but without transcription or translation.7 I have transcribed and translated all of these texts myself, using Callaghan’s dictionary and the grammatical information contained therein, with additional reliance on the closely-related Lake Miwok language in some cases.8 The transcriptions and translations are time-aligned to the original audio recordings, using the ELAN linguistic software (2020), and are deposited in the SCOIL archives along with the original data, as of 2019. All of the texts are thus available for consultation or download by the general public. There are many other texts in the SCOIL archives in other languages (not to mention in other archives around the world) which could be renewed in this way by linguists or others interested in oral traditions and made available to the public, and it is to be hoped that such work will grow in the future. The texts Callaghan recorded from Ballard are:

Title:SCOIL tape #:Total Lines:
1.“The Rattlesnake”6-01425
2.“Barbecuing Meat”6-01612
3.“Thunder and Lightning”6-018144
4.“The Last of the Two Elk”6-0196
5.“The Journey of the Dead”6-02016
6.“The Disobedient Girl,” version 16-02130

5 Where the exact transcription of a word is unclear, I place the term in slashes (/ /) with the best-guess transcription.

6 See Bright 1978b and 1994; Heizer 1978b; and Luthin 2002:513-41 for general discussions of the contents of the traditional verbal narratives of California.

7 The archive can be accessed online at

8 On Lake Miwok, see Freeland 1947 (grammar sketch with texts); Callaghan 1963 (grammar) and 1965 (dictionary).


7.“The Disobedient Girl,” version 26-02119
8.“Wallipo Steals the Fish”6-02222
9.“The Sun Girl”6-02314
10“The Two Lovers”6-02922

Of these texts, numbers 1, 2, and 8 are personal anecdotal accounts (although 8 involves a shamanistic component as well), numbers 6, 7, and 10 are legendary accounts, number 5 is an ethnographic account, and numbers 3, 4, and 9 are mythical narratives. The texts here for the most part have little in common with the material presented in Kelly 1978b or Merriam 1993. An exception is “The Sun Girl.” The same text was collected by Merriam originally in 1910. I give here the text, with linguistic annotation, as spoken by Ballard. I underline sentence-initial lexical elements, whose significance will be explained after the presentation of the text.9


1.         ahh,10 weya ka yutte.
            ahh world PST dark

            The world was dark.

2.         ‘iţi    ka     ‘iş=   ahh hii-n         kooya-n      hanna           ka   cewa.
            then PST this= ahh sun-GEN girl-NOM PST bright

            Then that Sun Girl was the only light.

3.         ke            ‘iţi ‘opu      ka     şu[ţa],  yomik ‘aalla-tto.
            and.then 3S DECL PST located live east-ALLAT

            And this one was lo[cated], lived in the east.

4.          ‘inniiko ka  welak cewa.
            3P PST want shine

            They wanted light.

9 Concerning the orthography, c is pronounced as English ch, ş is a retroflex s, pronounced with the tongue curled back towards the roof of the mouth, ț is a retroflex t, and ‘ is a glottal stop. Coast Miwok has SVO word order, and resembles Latin in having a system of eight noun cases, marked by suffixes. It uses either pronouns or reduced pronominal clitics attached to neighboring words (usually the verb), and demonstratives are likewise either independent words or reduced clitic forms. Verbs can be altered by various suffixes that derive imperative, perfective, inchoative, and other secondary forms.

10 Ahh is simply a hesitation here, so not considered as a line-initial form.


5.         ‘iţi     ka   ‘oye-n,            ‘oye-n             hiyyappa ‘oşşa tayih ka     cuna cuna  hii-n
            then PST coyote-NOM coyote-NOM send two man PST fetch fetch sun-GEN
                        kooya kon= hii-n kooya.
                        girl 3P?= sun-GEN girl

            Then the Coyote made / sent two men go after / get the Sun Girl they, the Sun Girl.11

6.         ‘enak-to      ka     koş=’opyaţi

            far-ALLAT PST 3D=leave.PERF

            They departed for a distant place.

7.         ‘ENAK-TO    ka,
            far-ALLAT PST

            A distant place . . .

8.         ‘oye     ‘enak-to,
            coyote far-ALLAT

            Coyote to a distant place . . .

9.         ka    ‘ikkoş ‘opyaţi. 
             PST 3D leave.PERF

            And they departed.

10.        ‘iţi     ka    ‘iş=hii-n             koola-n     hella   welak na’uuţi-n    ‘ooni   ‘ikkoş şakkaa-ţu
            then PST this=sun-GEN girl-NOM NEG want return-DEP come 3D with-INSTR

            And the Sun Girl did not want to come back with them.

11.        ‘iţi   ka     koş=’aakal ‘oye     coote ‘iş=hii-n          koola-n     hella  welak ‘ooni            
            then PST 2D=tell coyote ? this=sun-GEN girl-NOM NEG want come

            Then they told Coyote the Sun Girl did not want to come

11 Tayih ka is possibly intended as tayihko, “men.”


12.        ‘iţi     ka   ‘iş=oye-n,
            then PST this=coyote-NOM

            Then this Coyote . . .

13.        ‘iţi    ka   ‘iş=’oye-n            hiyyappa  ‘uni      tayyik-ko-n      cuna ‘iţi.
            then PST this=coyote-NOM send many man-PL-NOM fetch 3S

            Then this Coyote, then this Coyote sent several men to go after her.

14.        ‘iţi    ka kon=,     ‘iţi   ka,       ‘iţi   ka   kon=kalen ‘iş=hii-n          kooya.
            then PST 3P= then PST then PST 3P=tie this=sun-GEN girl

            Then they, then, then they tied up this girl.

17.        ke       kon=camma-, ‘iţi-kko ţeecukay,  kon=camma-n   ‘ooni   ‘iţi
            and.then 3P=bring 3S-P ? 3P=bring-DEP come 3S

            Then they brought, they [?], they brought her.

18.        ‘iţi   ka    ‘inniiko-n   welak  ‘u[h]=hii-n      koola-n    cewwaţi[yya] weya
            then PST 3P-NOM want 3S=sun-GEN girl-NOM shine.PERF world

            Then they wanted the Sun Girl to light up the earth.

19.        ‘iţi    ka    hella  podeer;
            then PST NEG able

            [But] she could not do it;

20.        ‘ellee, ‘iţi,
            fish 3S

            fish12 . . . she . . . 

21.        ‘uh=meeh ka     halle          ‘awwuk   pollo
            3S=body PST covered(?) abalone shell

            Her body was covered with(?) abalone shell

12 The narrator apparently doesn’t immediately remember the word for “abalone” and says “fish” instead.


22.        ‘iţi  kaa    ‘awwuk  pollo-n        cewwa.
            then PST abalone shell-NOM shine

            Then the abalone shell gave light.

Callaghan (1970:1) notes that Ballard had not spoken Coast Miwok regularly for several decades at the time they worked together in the 1960s, and had forgotten many words. There is definitely searching or hesitation at some points in the texts. Her basic morphological and syntactic knowledge of the language appears to have remained intact however. Her grammar is regular and consistent throughout the texts, as well as other dictionary example sentences. We should not expect this and the other narratives to be highly elaborated, given the limitations of memory and vocabulary which Ballard experienced, but they are nevertheless useful samples of Coast Miwok language and oral narratives, and are of course the only examples that we have.

The text reveals several interesting features of Coast Miwok, as well as some probable absences of original features. To begin with the latter, the text makes no use of any special narrative marker of reported or myth-time events, unlike most other oral traditions of Native America (K. Kroeber 1997). All other well-documented Miwok languages do have such features. Southern and Central Sierra Miwok make use of special narrative and/or remote past tenses (Broadbent 1964; Freeland 1951). The more closely related Lake Miwok language uses the particle weno (“it is reported”) for this purpose (see Callaghan 1978). Such a particle would presumably have been a high-frequency feature of traditional narratives in Coast Miwok, and part of the speaker’s conscious “discursive awareness” (Kroskrity 2010), and thus something that we would not have expected Callaghan to forget, but it could be that after decades of not speaking regularly she omitted this stylistic feature. Similarly, although there is no direct citation in this text, citations do occur in other of her texts. In those cases, there is no special citational form used. Lake Miwok again has such a form: the particle kaşa, which means “s/he said” (Callaghan 1978).13 Since such a feature is almost universal in oral traditions, it seems most likely that Ballard simply omitted this particle, and thus I am tempted to assume that she omitted some type of attributive particle such as weno as well. Ballard also never uses any particular closing device for her narratives, whereas in Lake Miwok it is common to use the particle ‘aweecu (“that’s all / the limit / the extent of it”) (Callaghan 1978). Note finally that Ballard uses the everyday term ‘oye when talking about Old Man Coyote, rather than the special terms reported above (‘oye ‘oyyiş and so forth) for traditional narratives. Rather than examples of forgetting, however, at least some of these omissions may be indexical of the context of the retellings—as linguistic samples for a non-Miwok academic linguist, rather than attempts to fully perform the stories for an indigenous audience. Similar types of omission have been reported in other such ethnographic performance settings (see Moore 2015 on Chinook, for example).

Based on these apparent absences, we should not take this text as a representative sample of Coast Miwok narrative art as it was practiced at earlier times for an indigenous audience. But it does show several interesting linguistic and artistic features. Most notably, Coast Miwok uses

13 Kroeber (1904) offers a brief text in the Rumsen language of the Costanoan group. Even in this short sample (1904:80-81), there is clear evidence of a citational form, {kaii}.


strict SVO word order. Sentences normally begin with a tense / aspect / modality particle (the past tense particle ka is a pervasive example in the preceding text). There is then an optional marked focus position, prior to the particle. This position can either be empty, be filled by a discourse marker (such as ‘iţi, “then,” in several lines in the text), or on rarer occasions (underlined above) be filled by some other lexical part of speech: a subject noun which “slides” to the left of the particle; a verb accompanied by a pronominal clitic, which both “slide” to the left of the particle; some oblique nominal element (that is, not subject or object); or an adjectival or adverbial element, which can be detached from its normal place in the sentence and then moved to the marked focus position. The only restriction on the marked focus position is that the basic SVO order of the main constituents must be maintained.   

In examining the full range of the narratives, it turns out that the marked focus position is filled by something other than a discourse marker only a small percentage of the time. In particular, this position is used to introduce important new participants or events into the narrative, and it is also used to emphasize certain adverbial and adjectival concepts or oblique participants that would normally be placed after the main SVO constituents of the sentence. Both of these usages can be seen clearly in “The Sun Girl.” In line 1, the world is introduced. In line 4, the people of the world are introduced. In line 6, the great distance to be traveled to get light is emphasized. Then in lines 20 and 21, the body of the Sun Girl, covered by abalone shells which are the source of light for the world, is highlighted. Note however that the Sun Girl herself is not given marked focus treatment when first introduced, nor is Coyote (in this case the mythical creator figure Old Man Coyote).

Except in these special uses of the marked focus position, virtually every sentence starts with ‘iţi ka . . . (“then PST . . .”). The usage of this phrase is very similar to what occurs in Lake Miwok traditional narratives, where miţi ‘ekal (“so then . . .”) is used very commonly to begin lines of narrative—indeed, the two forms appear to be related (Callaghan 1978). In a few cases ke is used. This form is used only when the subject of the sentence is the same as the preceding sentence, and when that subject is expressed as a pronoun or pronominal clitic following ke (as in lines 4 and 17). It is effectively a same-reference marker, and can be translated “and” (though it is used only to link two sentences, not two nouns). It indicates strong continuity of action between two sentences.

It is interesting to contrast the pattern here with what one finds in the sample sentences in the Bodega Miwok dictionary that are not from narratives. Such sentences virtually never begin with ‘iţi, since it is a discourse-level continuation marker. Instead they very commonly begin with a noun in marked focus position—so often in fact that one might assume from the dictionary alone that S + marker + V + O is the unmarked word order in the language. But in fact this is just an artifact of elicitation: when new topics and actions are constantly raised in the context of linguistic field work, these new items often get placed in marked focus position specifically because they are contrastive and new. But in a narrative, the language looks very different—the marked focus position is reserved for key narrative constituents or points of emphasis, and almost always only on their first mention. Thus ‘iţi ka could be considered roughly as a narrative / poetic line marker, while use of content words in the marked focus position serves to create larger narrative chunks.

Below, for the sake of comparison, I offer one more full text—a legendary story:


            THE DISOBEDIENT / BAD GIRL (Second Version)

(Title):  ‘om- ‘omu-n   kooya.
            bad bad-GEN girl

            The bad girl.

1.         ‘inniiko ka    ş[uta],    yomik liwa-n     noo  hinewa-tto.
            3P PST located live water-GEN that side-ALLAT

            There were people living across the water.

2.         ‘iţi    ka   kenne kooya-n                     ‘ame
            then PST one pubescent.girl-NOM menstruate

            Then one pubescent girl . . .

3.         ka ‘u[h]=caa-???   ‘u[h]=caamaţi       ‘uş=huna kocca.
            PST 3S=ha[ve] 3S=have.PERF 3S=own house

            And she had her own [menstrual] hut.

4.         ‘uk-‘oo- ‘uh=’ooma ka   HUJA HII.
            ? ? 3S=fast PST four day

            She was fasting / menstruating for four days.

5.          ‘i- ‘ikko-n ‘atawne,   ka   ‘u[h]=tuppe   ‘alla-tto.
            3P-NOM PST 3S=appear outside-ALLAT

            When they spoke to her, she came outside.14

6.         şuţA-MMI          ‘un=KOCCA-TTO!
            located-IMPER 2S=house-ALLAT

            “Stay at your house!”

14 Temporal subordinate clauses precede main clauses in Coast Miwok, so this line does not involve marked focus position.


7.         ‘eyya=n          wetşiţi         KAA!
            PROHIB=2S open.PERF door

            “Don’t open the door!”

8.         ‘eyya=n         ‘aţawne   manti!
            PROHIB=2S someone

            “Don’t speak to anyone!”

9.         wette-mmi       kaa!
                        close-IMPER  door

            “Close the door!”

10.        ‘u[h]=caamaţi     ka    ‘uş=  ‘uş= ‘uş=huna  traaste.
            3S=have.PERF  PST  3S=  3S    3S=own  utensils

            She had her own dishes / utensils.

11.        ‘iţi     ka    manti-n               hella  pod[eer], manti-n            hella   huke  ‘iţi — 
            then  PST someone-NOM  NEG  able      someone-NOM  NEG   touch 3S 
                        kuleyyi-kko-n       hanna.   
                        woman-P-NOM  only.ones[who]

            Then no one can, no one is to touch them, women alone (menstruating).

12.        kuleyyi-kko  ka    hinak   ‘uh=tuu.
            woman-PL    PST make   3S=food

            The women made her food.

13.        ‘iţi    ka,   ‘iţi    ka      ‘iş-‘ame-n                      waa ‘alla-tto.
            then PST then PST   this=menstruate-NOM  go  outside-ALLAT

            Then, this girl went outside [again].

14.        kee

            And . . .


15.        ‘iţi    ka     ‘uh=waa ‘iş=huyye-tto,
            then PST  3S=go    this=point-ALLAT

            Then she went to that point of land,

16.        kee          ‘iş-şi            ‘omu
            and.then  3S=INSTR bad

            And it was bad [weather].

17.        ‘iţi    ka   ‘iş=’oolok,     ‘omu tayi-n      cama ‘iţi     ‘iS=huyye-tto.
            then PST this=ocean,  bad  man-NOM take 3S   this=point-ALLAT

            Then the badly [crashing] waves carried her to that point of land (lit. “old man ocean”).

18.        ke            ‘uş=hinkaţi               ‘iţi,  ‘oş=şawwatto
            and.then  3S=make/do.PERF  3S   3S=hurt.PERF

            And she did / made it, [but she] was hurt(?) [cf. şawwaţi, “to hurt”].

19.        ‘iţi    ka    ‘el(l)i-ttuma,
            then PST look.around-INCHOAT

            She looked around,

20.        ka     ‘uh=’uţe  ‘uh=şuţa       niş  luppu-tto.
            PST  3S=see   3S=located  this rock-ALLAT

            And [she] saw that she was [alone] on this rock.

21.        ‘e[h]- ke         ‘u[h]=, liwa-n              şuta     nii  hinne-tto       he    noo  hinne-tto,
            ?      and.then  3S=    water-NOM   located this side-ALLAT and  that  side-ALLAT
            And the water was on this side and that side of her,


22.        ka   ‘uh=’oolak ‘ukan.
            PST 3S=cry     enter

            And she started to cry.

23.        ‘iţi      ka    ‘iş=  ‘inni(i)ko na’uuţi-n     ‘ooni.
            Then  PST this= 3P          return-DEP come

            Then they returned home.

24.        ka   kon=hella    ‘uţe   ‘iţi. 
            PST  3P=NEG    see   3S

            And they didn’t see her.

25.        ka      kon-liimATI            ‘iţi.
            PST 3P=search.for.PERF  3S

            And they looked for her.

26.        ‘iţi      ka   kon=’uţe ‘uh=wateh, ‘uh=talah  niş   luppu-tto.
            then PST  3P=see    3S=sit        3S=stand this  rock-ALLAT

            Then they saw [her] sitting, standing by this rock.

27.        ke           ‘uh=’oolak.
            and.then  3S=cry

            And she was crying.

28.        neccuţi,      neccutţi      ‘uume     ‘opu=n       podeer . . . 
            sometimes  sometimes evening  DECL=2S able

            Sometimes, some evenings you can . . .

29.        neccuţi      ‘uume    ‘opu=n         ‘uţe ‘u[h]=talah  ‘iţţi-tto          ‘iş=luppu-tto.
            sometimes evening  DECL=2S  see  3S=stand    3S-ALLAT  this=rock-ALLAT

            Some evenings you can see her standing by that rock.

30.        ke           ‘uh=’oolak.
            and.then  3S=cry

            And she is crying.15

15 This final verb, repeated in lines 27 and 30 (also in line 22), likely has an ironic sense. The word for “ocean” is ‘oolok (line 17), which is obviously very similar to ‘oolak, “cry.”


Note that this text includes a title. Titles lack any tense / aspect / modality marking in Ballard’s stories, unlike other declarative sentences in the language, so they are immediately identifiable, and serve to key the coming narrative—though otherwise Ballard uses no devices such as “long ago . . .” to initiate narratives (Lake Miwok often uses kilackilac, “long ago”). Note also the use of direct citations here that were lacking in the other text, though without any citational framing device, as noted earlier. The citations are however the only lines in the story without any of the line-initial particles listed below. In this text, the use of ‘iţi ka . . . and ke . . . can again be seen, along with a few occasions where (underlined) lexical elements occur in marked focus position. Note on several occasions only ka occurs rather than ‘iţi ka.These all involve continuity of subject or topic (that is, ke could be used in these cases, but the continuity is apparently not seen as being quite strong enough to justify use of ke). Where a new sentence-level subject or topic occurs, ‘iţi ka is always used. Thus Coast Miwok narratives, at least as told by Sarah Ballard, have a series of ranked devices that highlight relative newness as opposed to continuity in the flow of narration, while also marking the initiation of each new main clause:

            Least Continuous          >                      >          Most Continuous

            Lexical element               
            in marked focus:              ‘iţi ka                ka                     ke

Notice again that the Disobedient Girl—like the Sun Girl in the first story—is never in the marked focus position. Instead other elements occur in that position at various points in the story as they become salient. The presence of the central character is so salient that she never needs to make an appearance in this position (especially as she has been introduced by the title). This frees up the marked focus position to serve for introducing new episodes or sequences in the narrative. Thus Ballard’s narratives also show a three-level distinction of discourse-level topics: the most central and continuous characters and topics are mentioned in the title perhaps, but do not occur in marked focus position; second-most-central elements occupy the marked focus position, at least on first appearance; and minor elements never occupy the marked focus position, and also appear only briefly or in one section of the story.

This particular line- and section-marking strategy is likely unique to Coast Miwok in Central California, at least so far as we can determine. Lake Miwok does not have a clearly defined marked focus position in the same way as Coast Miwok, and does not use sentence-initial tense / aspect / modality markers. The Eastern Miwok languages are structurally and morphologically quite different from Western Miwok, and show no features in the documented oral literature closely similar to what we have seen here.16 Unfortunately we have very little documentation of indigenous texts in the Northern Costanoan / Ohlone languages of the San Francisco Bay area.17 The related Costanoan languages are most similar to Coast Miwok in

16 See Broadbent 1964; Freeland and Broadbent 1960; and Berman 1982 for texts. See Cowell 2020 for a poetic analysis of a Central Sierra Miwok text in the original language.

17 There are several missionary translations in a dialect of North Costanoan / Ohlone (Blevins and Golla 2005), but this is obviously not a useful comparison in relation to traditional narratives.


structure among the California languages, with noun case systems and uninflected verbs that combine with nouns or pronouns. There are two texts in Rumsen (South Costanoan) presented in Kroeber 1904:80-81 and 1910:255-58, though the latter one is actually a composite produced by Kroeber himself based on fragments of original Rumsen texts in his notes—and thus clearly not useable as a basis for extended discourse analysis. Two additional texts originally collected by J. P. Harrington are presented in Kaufman 2008, and one of these receives close linguistic analysis in Kaufman 2010. As opposed to Coast Miwok, the texts show much more variation in word order, including several examples of VS and OVS order. Such shifts in word order seem to be a key syntactic and narrative strategy in the Rumsen texts. The texts do show use of a line-initial marker (neej) similar to Coast Miwok, as well as following past tense / irrealis marker (ku), so that many lines begin as neej ku followed by the main clause, closely paralleling the Coast Miwok ‘iţi ka. This marker also varies between neej ku, neej-ink ku, and neej-ink-mur in the texts published by Kaufman (the meanings of -ink and –mur are unknown), suggesting a set of line-initial variations similar to what is seen in Coast Miwok. There is no evidence of lexical elements placed prior to these markers in a focus position in the Costanoan texts, however. Nevertheless further detailed study of other potential parallels in narrative structure and rhetoric between the various Utian languages is certainly warranted. Stylistic variation of the expression “and  then . . .” has been noted for Southern Paiute narratives (Bunte 2002:26) and also Western Mono in California, where the variation is noted as “an authenticating feature of proper performance” and key to textual cohesion (Kroskrity 2015:144). 

The Coast Miwok usage of the focus construction and variations in the line-initial particles has further implications within the broader framework of ethnopoetics, beyond just central California. I have used the term “line marker” to describe the particles in question, with the focus construction denoting larger sections of discourse. But neither of these linguistic devices is limited to narrative—they are both easily found in the dictionary in individual sample sentences not drawn from longer discourses. As such they are among potentially a larger set of linguistic resources available to a narrator, but there is nothing to indicate that they are indexical of traditional narrative per se in the Coast Miwok tradition—unlike such forms as weno or kaşa in Lake Miwok. There has been increasing awareness that grammar generally can serve key narrative or poetic functions at a discursive or structural level, with focusing functions being a key area of interest, even if the forms in question do not rise to the same level of discursive awareness as the traditional markers that often key a performance (see Bunte 2002 on Southern Paiute reduplication and Kroskrity 2010 on Tewa inverse constructions). 

Both Bunte and Kroskrity note that such grammatical features, however, largely escape the conscious awareness of listeners and narrators, in contrast to more highly salient features of traditional narrative such as markers of narrative or mythological past tenses, formulaic openings and closings, and the like. The organizing features used by Ballard appear to be of this type, and to serve highly effectively, even in the absence of any devices that might explicitly key a full traditional performance. If we take seriously Dell Hymes’ concept of indigenous narrative “voice” and his calls to avoid narrative “inequality” (1996), as well as Robert Moore’s call to recognize the ethnographic encounter as its own ethnopoetic genre of narrative and performance (2015), then we must be very careful not to see narratives such as those of Ballard’s as merely incomplete, imperfect, or involving the “omission” of expected elements. It may in fact be the


case that this is what happened—or is some of what happened—in the production of these narratives. But Ballard clearly drew on the grammatical resources of her language to produce brief but highly organized narratives. I think it is likely she was able to do this because of existing narrative traditions in Coast Miwok, which relied on everyday tools for pragmatic focusing in order to produce narrative focus, and on everyday tools for indicating discourse continuity and discontinuity to produce narrative continuity. It would appear, ironically, that the elements of narrative poetics which were more “grammatical” and less available to conscious “discursive awareness” were the features which remained most intact and / or most relied on in the production of these texts, while features such as citational verbs and narrative past markers were the ones dropped—or perhaps were the ones easiest to choose to drop in the context of this particular ethnographic and textual encounter.

Returning to the Coast Miwok texts, one can however see elements of more traditional “performative” rhetorical organization, in particular in groups of four in the second text: the young girl is given four imperative commands in lines 6-9, or her parents go through four actions at the end of the story (lines 23-26). The structure of command, prohibition, prohibition, command in lines 6-9 provides further internal structure. The complementary parallelism of lines 24 and 26—not seeing, then seeing—gives added structure to that overall set of four lines, as does the set of ‘iţi ka, ka, ka, ‘iţi ka as consecutive line markers (which also echo the structure of lines 6-9). Another notable feature is the final word and case marking in line 13, which ironically echo the same word and case in line 6, even as the phrase in line 13 also initiates the second, transgressive half of the story. Likewise the final demonstrative, as well as the noun and its case marking in lines 20 and 26 echo each other, and underline the fundamental predicament with which the story ends. The parallelism of lines 27 and 30 also nicely frames the end of the story, with the use of ke serving here to evoke figuratively the “continuity” or inevitability of the moral outcome of the story. 

Similar features occur in the first story—the first four lines conclude with “dark” in line 1, and “light” in lines 2 and 4, forming a conceptual unit. (Four is reported as the most common sacred number for the Coast Miwok (Collier and Thalman 1991:486-87).) Lines 5 and 13 are closely parallel, with the main difference being that in line 13 the men that Coyote sends are now numerous, producing small grammatical adjustments. Lines 6-9 include three consecutive mentions of a far distance, and conclude with the same final verb in lines 6 and 9 framing this sequence. Note that because the initial position in Coast Miwok sentences is usually either empty or occupied by a fairly abstract discourse marker, and lexical elements occur rarely (and almost never twice) in the marked focus position, lexical parallelism is focused most commonly on the last word or words of a line. Note also that such parallelism, when it involves nouns, also includes parallel case marking on the nouns.

In summary, the two texts here—and the small Coast Miwok corpus generally—show evidence of the same kinds of formal poetic features that have been widely recognized in Native American verbal arts, despite the somewhat challenging nature of the data (Hymes 1981; Sherzer and Woodbury 1987), as well as the same kinds of relations often found between form and meaning (Foley 1991). They also show the unique way in which Sarah Ballard—and likely other Coast Miwok narrators, if we only had the data—made use of the specific morphological and syntactic features of Coast Miwok to develop a line- and section-marking system that was highly


sensitive to both sentence-by-sentence narrative continuity, and also larger-level topical continuity and newness. This system is not found in the other Miwok or Costanoan / Ohlone languages of California for which we have available narrative data, and illustrates the point that each individual language (and of course storyteller) to some extent draws on its own unique grammatical resources for creating narrative patterning (see Bunte 2002), even if general themes and plot lines are shared across several languages in an area, as was certainly the case for areas of California. 

In concluding, it is worth thinking more about the implications of local particularities in the grammar and rhetoric of oral narrative in California. Efforts are increasing to reinvigorate traditional narratives and oral traditions, in both English and the original languages in California generally (O’Neill 2012; Field 2012; Nevins 2017), and in the Coast Miwok (and neighboring Kashaya Pomo) area in particular (Sarris 1993 and 2017). Much more attention needs to be paid to the verbal artistry which can be recovered even from fragmentary traditions. In discussing the traditions of the Yurok, Karok, and Hupa of northwest California, O’Neill (2012) shows that despite quite similar general themes and plots, the three cultural and linguistic groups have kept the stories rigorously separate in key ways, focused on small details of plot or style. This separation has been part of maintaining a strong sense of ethnic identity in a situation where basic lifeways and social structures were quite similar. Related to this were “caricatures . . . about the sounds of the neighboring tongues, much like the ones that English-speaking Americans spread about New Yorkers . . . or about Southerners” (2012:72) and a celebration of “the distinctiveness of their languages” (73). The overall result was a high degree of focus on “the uniqueness of their languages and storytelling traditions, despite quite similar overall cultural patterns and even narrative traditions, broadly speaking (74). O’Neill goes on to show that minute stylistic differences in things such as when and how characters are named, or opening formulas, were seen as important to this uniqueness (78-79), and that such distinctions continue to have high salience for the communities. Margaret Field (2012) reports a similar focus on local dialect (vocabulary) specificity and narrative uniqueness among the Kumiai of Southern California, again with a focus on maintaining salient local identity markers in the context of revitalization, where in fact overall cultural patterns may be quite similar. We should likely expect similar types of localisms for the Bay Area and central California. 

In fact, the Coast Miwok even paid attention to fine-scale details that helped distinguish local identities within the language group. Isabel Kelly notes for example that her two consultants, one a Bodega dialect speaker, the other a Marin dialect speaker, were both very careful to point out that in the former dialect, the word for water was líwa, while in the latter it was kíik (Collier and Thalman 1991:117). The point here is that speakers of one dialect were aware of varying forms in the other dialect, and this metalinguistic awareness was linked to conceptions of ethnic identity. Many similar remarks (related to cultural practices as well as language) recur in Kelly’s notes with regard to both intra- and inter-linguistic boundary maintenance, such as efforts to maintain the secrecy of certain dances from those in neighboring rancherias (Collier and Thalman 1991:324). There is no information to show that unique oral narrative details in particular were a salient marker of ethnic identity among the Coast Miwok. But such unique features certainly existed, and in the context of highly locally-oriented efforts at language and culture revitalization in California, seem worth highlighting. 


I am not aware of the exact status of revitalization efforts among the Coast Miwok at the moment (attempts at contacting the Tribe and selected tribal individuals were unsuccessful during COVID times of 2020-21). At a minimum, this paper and the narratives show that the way Coast Miwok was spoken and “performed” in the broadest sense, on a daily basis, was much more diverse than the documentation found in the 1970 dictionary would suggest. In fact the dictionary, based largely on elicited sentences in the traditional language description model of the 1950s and 1960s (which focused much more on phonology and morphology than syntax and semantics) probably presents a somewhat skewed understanding of the language as a communicative device. This can fortunately be mitigated by use of the narratives, and by an ethnopoetic approach which highlights not just ethnopoetic features narrowly conceived, but the broader communicative nature of the language overall. In this case, ethnopoetic documentation and analysis is not just a secondary addition to basic description of the language (still a common stereotype in linguistics), but is actually fundamental to basic linguistic analysis and understanding in my view. The paper also hopefully offers a model to the community for how future narratives might someday be produced. This is the reason I have made all the narratives available for free download from SCOIL with transcription and translation. They will also be included in a future grammar of the language.

University of Colorado


2 = second person; 3 = third person; ALLAT = allative case; D = dual; DECL = declarative mode; DEP = dependent; GEN = genitive case; IMPER = imperative mode; INCHOAT = inchoative aspect; INSTR = instrumental case; NEG = negative; NOM = nominative case; P = plural; PERF = perfective aspect; PROHIB = prohibitive mode; PST = past tense; S = singular; “=” indicates a clitic, loosely attached to the neighboring word; “-” indicates morpheme boundaries; capitalization in the Miwok text indicates verbal emphasis by the narrator.


Berman 1982
Howard Berman, ed. Freeland’s Central Sierra Miwok Myths. Reports from the Survey of California and Other Indian Languages, 3. Berkeley: Survey of California and Other Indian Languages, University of California.

Blevins and Golla 2005
Juliette Blevins and Victor Golla. “A New Mission Indian Manuscript from the San Francisco Bay Area.” Boletín: The Journal of the California Mission Studies Association, 22.1:33-61.

Bright 1978a
William Bright, ed. Coyote Stories. International Journal of American


Linguistics—Native American Texts Series, 1. Chicago: University of Chicago Press.

Bright 1978b
______. “Introduction.” In Bright 1978a:1-2.

Bright 1994
______. “Oral Literature of California and the Intermountain Region.” In Dictionary of Native American Literature.Ed. by Andrew Wiget. New York: Garland. pp. 47-52.

Broadbent 1964
Sylvia M. Broadbent. The Southern Sierra Miwok Language. University of California Publications in Linguistics, 38. Berkeley: University of California Press.

Bunte 2002
Pamela Bunte. “Verbal Artistry in Southern Paiute Narratives: Reduplication as a Stylistic Process.” Journal of Linguistic Anthropology, 12.1:3-33.

Callaghan 1963
Catherine Callaghan. A Grammar of the Lake Miwok Language. Unpubl. Ph.D. dissertation, University of California, Berkeley.

Callaghan 1965
______. Lake Miwok Dictionary. University of California Publications in Linguistics, 39. Berkeley: University of California Press.

Callaghan 1970
______. Bodega Miwok Dictionary. University of California Publications in Linguistics, 60. Berkeley: University of California Press.

Callaghan 1978
______. “Fire, Flood, and Creation (Lake Miwok).” In Bright 1978a:62-86.

Callaghan 2014
______. Proto Utian Grammar and Dictionary: With Notes on Yokuts. Trends in Linguistics. Documentation, 31. Berlin: De Gruyter Mouton.

Collier and Thalman 1991
Mary E. T. Collier and Sylvia Barker Thalman, eds. Interviews with Tom Smith and Maria Copa: Isabel Kelly’s Ethnographic Notes on the Coast Miwok Indians of Marin and Southern Sonoma Counties, California. San Rafael, CA: Miwok Archaeological Preserve of Marin.

Cowell 2019
Andrew Cowell. “Time-Aligned Annotations of Bodega Miwok Sound Recordings.” Collection Number 2019-18. Survey of California and Other Indian Languages archive, University of California, Berkeley.

Cowell 2020
______. “A Central Sierra Miwok Origins Story: The Theft of the Sun.” English Language Notes, 58.1:132-44.


ELAN 2020
ELAN. Version 6.0. Nijmegen: Max Planck Institute for Psycholinguistics, The Language Archive.

Field 2012
Margaret C. Field. “Kumiai Stories: Bridges between the Oral Tradition and Classroom Practice.” In Kroskrity 2012:115-26.

Foley 1991
John Miles Foley. Immanent Art: From Structure to Meaning in Traditional Oral Epic. Bloomington: Indiana University Press.

Freeland 1947
L. S. Freeland. “Western Miwok Texts with Linguistic Sketch.” International Journal of American Linguistics,13.1:31-46.

Freeland 1951
______. Language of the Sierra Miwok. Indiana University Publications in Anthropology and Linguistics, 6. Baltimore: Waverly Press.

Freeland and Broadbent 1960
L. S. Freeland and Sylvia M. Broadbent. Central Sierra Miwok Dictionary, with Texts. University of California Publications in Linguistics, 23. Berkeley: University of California Press.

Goerke 2007
Betty Goerke. Chief Marin: Leader, Rebel, and Legend: A History of Marin County’s Namesake and His People.Berkeley, CA: Heyday Books.

Heizer 1978a
Robert F. Heizer, ed. Handbook of North American Indians, Volume 8: California. Washington, DC: Smithsonian Institution.

Heizer 1978b
______. “Mythology: Regional Patterns and History of Research.” In Heizer 1978a:654-57.

Hymes 1981
Dell Hymes. “In Vain I Tried to Tell You”: Essays in Native American Ethnopoetics. Studies in Native American Literature, 1. Philadelphia: University of Pennsylvania Press.

Hymes 1996
______. Ethnography, Linguistics, Narrative Inequality: Toward an Under-standing of Voice. London: Taylor & Francis.

Kaufman 2008
David Kaufman. “Rumsen Ohlone Folklore: Two Tales.” Journal of Folklore Research, 45.3:383-91.

Kaufman 2010
______. “Some Observations on Rumsen Ohlone Grammar.” Kansas Working Papers in Linguistics, 31:39-45.

Kelly 1978a
Isabel Kelly. “Coast Miwok.” In Heizer 1978a:414-25.


Kelly 1978b
______. “Some Coast Miwok Tales.” The Journal of California Anthropology, 5.1:21-41.

Kroeber 1904
A. L. Kroeber. “The Languages of the Coast of California South of San Francisco.” University of California Publications in American Archaeology and Ethnology, 2.2:29-80.

Kroeber 1910
______. “The Chumash and Costanoan Languages.” University of California Publications in American Archaeology and Ethnology, 9.2:237-71.

Kroeber 1911
______. “The Languages of the Coast of California North of San Francisco.” University of California Publications in American Archaeology and Ethnology, 9.3:273-435.

K. Kroeber 1997
Karl Kroeber. “Introduction.” In Traditional Literatures of the American Indian: Texts and Interpretations. 2nd ed. Lincoln: University of Nebraska Press. pp. 1-24.

Kroskrity 2010
Paul V. Kroskrity. “The Art of Voice: Understanding the Arizona Tewa Inverse in Its Grammatical, Narrative, and Language-Ideological Contexts.” Anthropological Linguistics, 52.1:49-79.

Kroskrity 2012
______, ed. Telling Stories in the Face of Danger: Language Renewal in Native American Communities. Norman: University of Oklahoma Press.

Kroskrity 2015
______. “Discursive Discriminations in the Representation of Western Mono and Yokuts Stories: Confronting Narrative Inequality and Listening to Indigenous Voices in Central California.” In Kroskrity and Webster 2015:135-63.

Kroskrity and Webster 2015
Paul V. Kroskrity and Anthony K. Webster, eds. The Legacy of Dell Hymes: Ethnopoetics, Narrative Inequality, and Voice. Bloomington: Indiana University Press.

Loeb 1926
Edwin W Loeb. “Pomo Folkways.” University of California Publications in American Archaeology and Ethnology, 19.2:149-405.

Luthin 2002
Herbert W. Luthin, ed. Surviving through the Days: Translations of Native California Stories and Songs: A California Indian Reader. Berkeley: University of California Press.

Merriam 1993 [1910]
C. Hart Merriam, coll. and ed. The Dawn of the World: Myths and Tales of the Miwok Indians of California. Lincoln: University of Nebraska Press.


Moore 2015
Robert Moore. “Reinventing Ethnopoetics.” In Kroskrity and Webster 2015:11-36.

Nevins 2017
M. Eleanor Nevins, ed. World-Making Stories: Maidu Language and Community Renewal on a Shared California Landscape. Lincoln: University of Nebraska Press.

O’Neill 2012
Sean O’Neill. “The Politics of Storytelling in Northwestern California: Ideology, Identity, and Maintaining Narrative Distinction in the Face of Cultural Convergences.” In Kroskrity 2012:60-89.

Sarris 1993
Greg Sarris. Keeping Slug Woman Alive: A Holistic Approach to American Indian Texts. Berkeley: University of California Press.

Sarris 2017
;______. How a Mountain Was Made: Stories. Berkeley, CA: Heyday Books.

Sherzer and Woodbury 1987
Joel Sherzer and Anthony C. Woodbury, eds. Native American Discourse: Poetics and Rhetoric. Cambridge Studies in Oral and Literate Culture, 13. Cambridge: Cambridge University Press.