Tübatulabal // is an extinct Uto-Aztecan language, traditionally spoken in Kern County, California, United States. It is the traditional language of the Tübatulabal, who have now shifted to English. The language originally had three main dialects: Bakalanchi, Pakanapul and Palegawan.
|Region||Kern River, California, United States|
|Ethnicity||900 Tübatulabal (2007)|
|Extinct||2008, with the death of James Andreas|
In English, the name Tübatulabal refers to both the Tübatulabal people and their language. However, in the language itself, the term Tübatulabal refers only to the Tübatulabal people. Its origin is unclear, but it may be related to the noun stem tɨba- "pine nuts". The Tübatulabal term for the Tübatulabal language is pakaːnil.
There are six phonemic vowels in Tübatulabal:
Contrastive short and long versions of each vowel are found in both stressed and unstressed syllables. The vowels have various allophones which occur in different environments, most notably more central lax allophones when the vowels are short and occur in unstressed syllables. i and u can occur as the second member of a diphthong with any other vowel, resulting in ten possible diphthongs (Voegelin reports that ɨu is rare). Phonologically, the members of a diphthong are treated as distinct segments. For example, the common initial reduplication process, which copies the first stem vowel, copies only the first member of a diphthong, e.g.:
- ʔuinul 'the sucker fish'
- ʔuʔuinul 'the many suckers in one place'
Vowel length is contrastive. However, according to (Jensen 1973), in the suffixing morphology length is typically predictable. In most cases, the first suffix is short, the second suffix is long, the third suffix is short, and so on. For example, the verbal stem tɨk- 'to eat' can be expanded to tɨk-ilɔːɡ-ɔ-maːla 'let us go and pretend to eat'. In this word, each suffix alternates in length compared to its neighbors. When arranged differently, the same suffixes will have different lengths. Thus compare maːla 'let us' with the realization of the same morpheme in tɨk-al-aː-mala 'let us go eat'.
Tübatulabal consonants show a basic voicing distinction, with a corresponding alternately voiced phoneme present for almost every obstruent. Tübatulabal voiceless consonants are unaspirated, like those in English after an initial /s/, e.g. as in 'spin', 'stiff', 'skin'.
Non-contrastive allophones of all vowels occur, usually when a vowel follows a nasal consonant, and especially when it also precedes a glottal consonant.
All consonants except the glottal stop can occur as geminates. Gemination is often phonologically predictable. In particular, all consonants except the voiced stops and the glottal stop geminate when following a short vowel. All stops and affricates are geminated in word-final position, regardless of the length of the preceding vowel.
Tübatulabal has predictable word stress, which is tied to morphological constituency and syllable weight. Primary stress falls on the final syllable of the stem. Secondary stress is assigned right to left from the final syllable, falling on every other mora:
- ˌʔɨmbɨŋˌwibaˈʔat "he is wanting to roll string on his thigh" 
- ˌjuːuˌduːˌjuːuˈdat "the fruit is mashing"
Words with the form VːCVCV will be stressed as ˌVːCVˈCV:
- ˌnaːwiˈʃul "the pine-nut pole"
For the purposes of stress assignment, two identical short vowels that are separated only by a glottal stop are treated as a single vowel if and only if they belong to the same morpheme:
- ˌkuʔud͡ʒuˈbil "the little one"
There are three basic word types in Tübatulabal: verbs, nouns, and particles. Verbs may be formed from verbal stems or from noun stems with verbalizing morphology; similarly, nouns can be formed from noun stems or from verbal stems with nominalizing morphology. Particles have their own stems, but they have comparatively little inflection, whereas both verbs and nouns tend to be very morphologically complex.
There are four word-formation processes in Tübatulabal: suffixation, reduplication, conjunction and compounding.
Suffixation is the most common and productive process in agglutinative word-formation. Suffixes form a closed class and occur in a fixed order according to the word type.
There are two kinds of reduplication: full reduplication and partial reduplication. Full reduplication is the less common type and marks the iterative aspect in verbs.
Partial reduplication can occur as initial or final reduplication. Final reduplication is very rare and always expresses the idea of plural allegiance. It is also apparently limited to occurring with noun stems or suffixes that end in wa. Voegelin illustrates with an example:
- tɔhat͡siŋwan 'his hunting partner'
- tɔhat͡siŋwawaːn 'his hunting partner (in the sense that the partner referred to, being very proficient, has many companions in hunting)'
Initial reduplication is far more productive. It is used to express collective plurality in nouns and to express aspect reversal in verbs. Initial reduplication prefixes a copy of the first vowel of the stem (as well as any immediately following nasal), preceded by a fixed ʔ. The underlying stem-initial consonant (if any) may also undergo changes, particularly in voicing and length. Some examples illustrate the reduplication process:
|Base form||Reduplicated form||Base form gloss|
|tana-||ʔandana||to get down|
|paːabɨ-||ʔaːbaːabi||to be tired|
Conjunction involves the combination of a particle with a word of another type. According to Voegelin, the behavior of particles is similar to that of enclitics in other Uto-Aztecan languages but distinct enough from them that it should not be considered to be a kind of cliticization.
Compounding appears to have been a much more productive process at an earlier stage of the language. It now has very limited productivity, and in many cases, it appears to have been completely lexicalized if it occurs.
Each verb stem has an unpredictable inherent aspect value (either telic or atelic; by default, a bare stem is inherently atelic), and an inherent value for transitivity (transitive, intransitive or impersonal). The inherent values can be changed by morphological addition to yield a verb stem with any of the other possible values. Aspect reversal is indicated by initial reduplication. Transitivity change is indicated by the use of one (or more) of a number of derivational suffixes with which verbs are constructed.
The full verb structure can be summarized as (A) + B + (C) + (D), where B is the verb root, and the other positions (all optional) represent classes of morphemes. A indicates initial reduplication, which can occur only once per word. C indicates a class of derivational morphemes, which can be divided into ten ordered positions, each of which allows at most one morpheme per word. D is the final position; there are nine possible morphemes in final position, but only one can occur in any single word.
The C class morphemes are given with examples in the table below. When these morphemes co-occur in a word, they must occur in the order given. Transitivity changing morphemes are marked with *. They have a different effect depending on the inherent transitivity of the verb root, as well as the presence of other transitivity-changing morphology.
|-(i)n||causative *||hɔːhinat||'s/he is coughing (through the agency of a crumb)'|
|-(a)n||benefactive *||weleʔanat kɔːimi||'s/he is crawling to the woman (perhaps in the sense of "he is crawling there for the erotic benefit of the woman")'|
|-(a)la/-(a)ɡiːm/-(a)kin/-(a)min||movement||ʔɨtːɨkːamin||'s/he ate it here and went away'|
|-(i)niːnɨm||distributive||ʔawaʃiniːnɨm||'s/he dug first here, then there'|
|-(i)lɔːk||pretending to||ʔanaŋaːlilɔːɡibaʔat||'s/he wants to go along pretending he is crying'|
|-(i)baʔ||desiderative||ʔamaɡiːibaʔ||'s/he is on the verge of learning about it'|
|-(i)ʃa||future||ʔapaʔaniʃa||'it will get plugged up'|
|-(i)w||passive *||weːhiwat||'s/he is being licked (e.g. kitten by mother cat)'|
|-(i)wɨːt||collective-intensive||ʔapahkaniwɨːdiʃa||'they will speak Tübatulabal'|
|-(a)puw-||similative||wɨʃɨpuwat||'it seems to be ripening'|
The possible verbal final morphemes (class D) are shown below. Unlike the class C morphemes, only one of these final-position morphemes can occur in any single word. Therefore, the ordering of morphemes in this table does not indicate anything about a linear relationship among the morphemes.
|Suffix/suffix type||Gloss||Example word||Gloss|
|Nominalizers||–||kabobaːʔinaːnat͡siŋwajinɨʔɨŋ||'my partner in rattling for it (the dance)'|
|Subordinaters||–||ʔalaːwiʔima tɨkːat||'s/he is eating while talking'|
|Imperatives||–||tɔhaːhai tɔhiːla||'hunt the deer after a while'|
|-(a)t||present tense||ʔɔhtatni||'s/he is asking me'|
|-(a)ma||exhortative||waʃamaːala||'let's dig it'|
|-(a)ha||permissive||wɔːʔiʃɨhatd͡za||'s/he might get jealous'|
|-(i)ukaŋ||past habituative||t͡saːijinaːniukaŋ||'s/he used to make lace'|
|-(aː)haiwɨt||irrealis||muːdakaːhaiwɨt||'s/he should have dodged'|
|-(a)htajat||adversative||pɨːminahtajat||'s/he is making it full (despite the fact that the thing to be filled is very large)'|
All nouns (whether derived from verb stems or noun stems) are obligatorily marked as absolute or relative. Nouns must also be marked with one of the three basic cases: subject, object, or genitive. Relative nouns make a finer distinction between suus and ejus objects and genitives. In addition to this obligatory morphology, nouns may also receive suffixes indicating several secondary cases (inessive, ablative, allative and instrumental) as well as many other derivational suffixes.
Nouns may be divided into three basic classes according to their stem shape and morphological behavior and sometimes according to their semantic contribution as well. The basic test for classification is how the noun occurs when it is absolute. The absolute suffix has a different allomorph when it occurs with a noun from each of these classes. Class A nouns all have vowel-final stems, and add the absolute suffix as -l. Class B noun stems may be vowel-final or consonant-final, but in either case the absolute suffix is -t. Class C is a small class of nouns, many of which are kinship terms or other inalienable nouns. The absolute noun is phonologically null when it occurs with class C nouns.
Each of the classes can be subdivided into two or more classes, depending on phonological differences in the noun stem that lead to divergent behavior in certain case forms. Specifically, class A is divided into A1 nouns (stems end with a long vowel) and A2 nouns (stems end with a short vowel). Class B is divided into five subclasses, depending on whether the stem ends in a short vowel, a long vowel, n, m, or a voiceless consonant. Class C is divided into C1 (nouns which take an overt relative suffix) and C2 (nouns with no overt relative suffix).
|Subject||Object||Genitive||Subject||Suus object||Ejus object||Suus genitive||Ejus genitive|
|B4||pɔm||egg||pɔmt||pɔmda||pɔmdiŋ||pɔmin *||pɔm||pɔmd͡zip||pɔmin *||pɔminin|
|B5||muːʃ||fish spear||muːʃt||muːʃta||muːʃtiŋ||muːʃn *||muːʃ *||muːʃip||muːʃin||muːʃinin|
|C1||tahambiʃ||old man||tahambiʃ||tahambiʃi||tahambiʃiŋ||tahambiʃin *||tahambiʃ||tahambiʃin *||tahambiʃʔin||tahambiʃʔinin|
Morphemes belonging to the particle class are distinguished by the fact that they undergo little or no inflection and suffixation, unlike verbs and nouns. The particle class includes two subclasses of morphemes which behave quite differently: conjunctive particles and independent particles.
Conjunctive particles resemble clitics in that they never appear independently but always lean on another word. However, unlike clitics, conjunctive particles typically bear their own stress, and they do not alter the stress of the word on which they lean. Conjunctive particles include various discourse and modal morphemes as well as the typical pronominal agreement morphemes that occur with verbs.
Independent particles are fully independent words. They include prepositional, modal and exclamatory morphemes, numerals, and one class of pronouns.
The table below shows the pronominal morphemes of Tübatulabal. Like nouns, pronouns distinguish between three cases: subject, object and possessive. (Pronouns do not make a distinction between absolute and relative entities.) Different forms exist for first-, second- and third-person entities. Second- and third-person forms distinguish only singular and plural numbers, but first-person forms distinguish between singular, dual inclusive, dual exclusive, and plural numbers. All pronouns may be expressed bu conjunctive particles. The subject pronouns are unique in that they can also be expressed by an independent particle.
The first-person subject conjunctive forms have special allomorphs when they occur with the exhortative suffix -ma:
The third-person conjunctive form is usually null, but it is expressed by -d͡za after the exhortative or permissive suffixes. (The suffix often undergoes syncope and devoicing, yielding -t͡s.) The second-person conjunctive plural subject form may also syncopate, and medial vowel then shortens as well: -bum. The first-person conjunctive singular subject form may also syncopate, triggering devoicing but no irregular phonology: the suffix then has the form -k.
Subject pronouns typically lean on verbs (if conjunctive) and correspond to grammatical subject: iwikkːɨki "I discarded (it)" (with devoicing); anabaːhaʃta "they can throw it" (with metathesis of the components of the affricate and a change of s > ʃ).
Object pronouns also lean on verbs and indicate any non-possessive oblique function, including transitive objects, ditransitive objects or benefactives, objects of imperative verbs, and subjects of subordinate verbs if not equivalent to the subject of the matrix verb.
Possessive pronouns typically lean on the possessum: haniːnɨʔɨŋ "my house"; ʃɔːɔjin "his wife".
Transcriptions in this article follow the International Phonetic Alphabet (IPA). Much published material concerning Tübatulabal uses the Americanist orthography. In addition, the most important linguistic work on Tübatulabal, the original grammatical description of the language, (Voegelin 1935) harv error: no target: CITEREFVoegelin1935 (help) uses a somewhat different orthography.
Voegelin writes ɨ as ⟨ï⟩ and ɔ as ⟨ô⟩. He also writes ʃ as ⟨c⟩, t͡ʃ as ⟨tc⟩, ʔ as ⟨‘⟩, d͡ʒ as ⟨dž⟩ and j as ⟨y⟩. He also uses a number of special symbols for vocalic allomorphs. ⟨ι⟩ is an allomorph of ⟨i⟩, ⟨μ⟩ is an allomorph of ⟨u⟩, ⟨o⟩ is an allomorph of ⟨ô⟩ (IPA ɔ), and ⟨ŏ⟩ is an allomorph of both ⟨a⟩ and ⟨ô⟩.
The letter ⟨ü⟩ in the name Tübatulabal represents the central unrounded vowel ɨ.
- Walker, Richard Arlin (12 Sep 2018) [25 Aug 2008]. "James ;Jimmy' Andreas Sr. left the gift of his language to his people". Retrieved 29 Jun 2022.
- "Tübatulabal". Ethnologue. Retrieved 2018-05-09.
- See Jensen for discussion of the role of "2.6 Gemination" in Tubatulabal phonology. (1973:61 et seq)
- See Jensen for discussion of the arbitrary behavior of glottal stops in stress assignment. The glottal stop, which is not otherwise counted as a mora, is counted as a mora for the purpose of stress assignment. (Jensen 1973, pp. 76–76).
- The morphological differences between noun classes can probably be traced back to a simple case of allomorphy. At a certain point in the history of an ancestor of Tübatulabal, all class A nouns ended in a vowel, while all class B nouns ended in a consonant. (Class C forms are few in number and have a more complicated origin.) The absolute suffix in this language was *-t, which lenited intervocalically, leading to -l in Tübatulabal class A nouns (compare cognates ʈ͡ʂ in Serrano, r in Tongva.) See (Voegelin 1935) harv error: no target: CITEREFVoegelin1935 (help) and (Manaster Ramer 1992).
- Relative nouns typically require a suffix indicating the possessor or entity to which the suffixed noun is related. Forms in this table are those for a third-person singular possessor -n. The exception is suus forms, which do not allow a possessive suffix (since they are inherently possessed by/related to the grammatical subject of the clause).
- The ejus forms are not given in (Voegelin 1935) harv error: no target: CITEREFVoegelin1935 (help). The forms in this table are hypothesized on the basis of the suffix charts that Voegelin gives. With a few exceptions, all other forms in this table are taken directly from (Voegelin 1935) harv error: no target: CITEREFVoegelin1935 (help). The few forms that are hypothetical are marked with *.
- Cells marked with -- are phonologically null. Cells marked with ? are forms missing without explanation in (Voegelin 1935) harv error: no target: CITEREFVoegelin1935 (help).
- Aion, Nora (2003). Selected Topics in Nootka and Tübatulabal Phonology. PhD dissertation: City University of New York.
- Arvidson, Lucy. Alaawich (Our Language): First Book of Words in the Tübatulabal Language of Southern California
- Crowhurst, Megan (1991). "Demorification in Tübatulabal: Evidence from Initial Reduplication and Stress". NELS. 21: 49–63.
- Gifford, Edward Winslow (1917). Tübatulabal and Kawaiisu kinship terms. Berkeley, CA: University of California Press. Retrieved 2012-08-26.
- Heath, Jeffrey (1981). "Tübatulabal Phonology". Harvard Studies in Phonology. 2.
- Jensen, James R. (1973). Stress and the Verbal Phonology of Tübatulabal (PhD dissertation). Indiana University.
- Manaster Ramer, Alexis (1992). "Proto-Uto-Aztecan Phonology: Evidence From Tübatulabal Noun Morphophonemics". International Journal of American Linguistics. 58 (4): 436–446. doi:10.1086/ijal.58.4.3519778. JSTOR 3519778. S2CID 147964960.
- Voegelin, Charles F. (1935a). "Tübatulabal Grammar". University of California Publications in American Archaeology and Ethnology. 34: 55–190.
- Voegelin, Charles F. (1935b). "Tubatulabal Texts". University of California Publications in American Archaeology and Ethnology. 34: 191–246.
- Voegelin, Charles F. (1958). "Working Dictionary of Tübatulabal". International Journal of American Linguistics. 24 (3): 221–228. doi:10.1086/464459. S2CID 145758965.
- Tubatulabal language overview at the Survey of California and Other Indian Languages
- "Tubatulabal Language" at native-languages.org
- Recordings of Tübatulabal recorded by JP Harrington
- OLAC resources in and about the Tübatulabal language
- Audio recordings of word lists in Tübatulabal from the Northern Uto-Aztecan Collection of Megan Crowhurst at AILLA.
- Tübatulabal, California Language Archives