Work Papers of the Summer Institute of Linguistics, University of North Dakota Session




From the introduction: "In the early stages of analysis, Southeastern Tepehuan (SET) vowel length appeared to be conditioned by accent. Although accent fell in a majority of words on closed syllables, there were so many exceptions that no general statement could be made. Thus accent was relegated to being a feature of the underlying character of roots. Reduplicated forms, however, were a confusing body of unpredictable accent; if accent were phonemic, why did it occur on the root syllable of some forms, and on the reduplicated syllable in others? Some plurals were judged irregular because they seemed to lose or gain whole syllables; vowels appeared or disappeared in various places.

"Native speakers insisted that there was a difference between cos (nest) and coos (he's sleeping), although the linguists tried to explain the two words as being semantically related (one rests in his nest). Historical evidence in this case was the deciding clue to discovery. Related languages and their proto-language reconstructed by Bascom (1965) showed that accent in SET reduplicated stems coincided with length on historical forms, although accent in either the sense of tone or stress did not necessarily occur on the same syllable. What we had considered to be bában (coyotes) was seen to be baabánai in Northern Tepehuan, báabani in Upper Piman, báaban in Lower Piman, and reconstructed as *baabánai in Proto-Tepiman. In this paper it is argued that length rather than accent is underlying, and that accent and vowel alternation in reduplicated forms can be predicted by phonological rules. In section 1 I will describe morpheme structure, in section 2 the rules for reduplication, in section 3 the accent rule, and in section 4 the phonological rules that coordinate with accent to produce phonetic forms."

Included in

Linguistics Commons