Tübatulabal (Aztec‑Tanoan, Uto‑Aztecan)

There is a set of verbs in which the regular morphological opposition between telic and atelic aspect (perfective and imperfective) is switched.

1 Background

Every Tübatulabal verb distinguishes two aspect values, telic and atelic, each with a distinct stem. Voegelin (1935: 94) describes the opposition as follows:

The telic is used for an action (e.g., 'to take a bite') or condition (e.g. 'it got green') performed or arrived at in an instant (perfective without tense commitment), and for this reason the action or condition is generally, though not necessarily, felt to be completed at the time of talking. The atelic is sometimes used when an action requires some duration for its performance ('to eat'), but frequently the atelic is quite vague in respect to aspectual meaning.

The difference between telic and atelic has a number of inflectional ramifications:

The atelic stem is morphologically basic; the telic stem is formed by reduplication of the vowel of the initial syllable:

atelic telic
ela- eʔela 'jump'
tɨk- ɨtɨk 'eat'
tana- andana 'get down'
pa:abɨ- a:ba:abɨ 'be tired'
yuʔudz- uyuʔuts 'throw'
(Voegelin 1935: 95, 102)

Voegelin states that this morphological opposition is quite regular. The segmental differences between the two stems are phonologically predictable:

2 Aspectual deponency in Tübatulabal

Some aspectual pairs have the reverse morphological relationship. Voegelin (1935: 95-96) gives a list of 30 such verbs, which he says is nearly complete:

atelic telic
a:dza:ya:w- tsa:ya:u 'yell'
apatsa:h- patsa:h 'shell nuts'
anaŋ- naŋ 'cry'
anab- nap 'throw'
a:na:yuw- na:yuw̥ 'be tired'
aʔay ai 'pick up'
a:ya:n- ya:n 'sing'
acag- ca:k 'roast'
andaŋ- taŋ 'kick'
aha:idž- ha:itc 'chew'
ɨmbɨŋw- pɨŋw̥ 'roll string on thigh'
ɨmɨl:d- mɨl:t- 'scold'
ɨtsɨxk- tsɨxk 'prick'
ɨhɨ:b- hɨ:p 'massage'
ɨhɨ:d- hɨ:t 'pluck feathers'
ɨ:cɨy- cɨ:i 'rock a cradle'
ɪndɪŋwa- tɪŋwa 'summon'
ɪcib- ci:p 'whittle'
i:cilu:b- cilu:p 'split wood'
i:ciug- ciuk 'comb'
ôtôlo:h- tôlo:h- 'groan'
ôcôlo:ŋ- côlo:ŋ- 'snore'
ô:yôm- yô:m- 'copulate'
ʊkʊc- ku:c 'grow'
ʊwuba- wuba 'whip'
ʊyugʊʔ- yugʊʔ 'cut'
ʊndʊmu:ga- tʊmu:ga 'dream'
ʊndʊma:w- tʊma:u 'fail'
ʊtʊc- tu:c 'grind'
ʊnʊŋ- nʊŋ 'pound'

There is no obvious semantic feature that these verbs share that would account for this morphological reversal. There is apparently nothing else unusual about these verbs: the telic and atelic stems behave in the same way as their morphological regular counterparts.

The mismatch evidenced by these forms is apparently recognized by speakers. This is manifested in nominalizations. Only atelic verbs serve as a base for nominalization (via the suffix -i-); thus, in the typical case, nominazations are based on the basic, unreduplicated form, e.g. wa:hay 'to work' ~ wa:hayil '(the) work', wac 'to dig' ~ wacil 'hole' (pp 166-76; -l is the nominative suffix). In the verbs cited above, naturally the reduplicated form should serve as the base for nominalization. Voegelin observes

Whenever this type of reversed formation occurs, it is not entirely certain that the atelic form will be used in nominalization. Informants are rather inconsistent when dealing with these verbs. For the verbal stem cited [anaŋ- 'cry'; MB], the directly nominalized form is theoretically: anaŋi:l the crying. This form is generally given by informants, but occasionally the telic form (which is the basic form in reversed formation verbs) is given instead: naŋi:l the crying. When the informants attention is drawn to the two nominalizations, the latter (based on the telic form) is pronounced incorrect, with the significant information, however, that some people use the incorrect form (naŋi:l ). (p. 167)

Note that there are also some nominal derivational processes which are purely form-based, i.e. they take the unreduplicated form regardless of whether it is atelic or telic (pp. 168-69).

References

Voegelin, Charles F. 1935. Tübatulabal grammar. Berkeley: University of California Press.