摘要
This paper is a field report on the phonological system of Butanglu, a hitherto unstudied dialect of the Paiwan language in Formosa. It is a special dialect in preseserving, for example, a voiced velar fricative ᴕ which, according to Dahl (1973: 86-96), may be a reflex of PAN*ᴕ, rather than a tongue tip trill r as found in the other dialects of Paiwan. The contrasts between t and č, and between d and ǰ are preserved in Butanglu, yer merged in the other dialects. It is also observed that while the other dialects of Paiwan have only four voiceless stops, either /p, t, k, Ɂ/, or /p, t, k, q/, Butanglu has one extra stop which may be explained as borrowing or the result of secondary development. Thus this dialect has five stops /p, t, k, q, Ɂ/. These characteristics as well as some morphophonemic alternations, such as ə~i and y~Ø, indicate that Butanglu should be an important dialect of the Paiwan language for linguistic investigation.
Butanglu has twenty-eight phonemes /p, b, v, m, t, d, c, s, z, n, ḍ, ḷ, č, ǰ, Ǐ, k, g, ᴕ, ŋ, Ɂ, h, y, w, i, u, ə, a/, and seven morphophonemic alternations: ə~i, ə~Ø, aa~a, uu~u, y~Ø, w~v, and m~n. Although this is a preliminary report, this paper has suggested thirteen rules for the phonological system of this dialect to account for the morphophonemic alternations and phonetic details.
It has been noted that there are very few studies of Paiwan. Naoyoshi Ogawa was the first one who investigated this language from a linguistic point of view. After he had written three papers about the Paiwan language in 1930, Ogawa and his colleague Erin Asai published the monumental work on Formosan languages in 1935. In this book, 165 pages were devoted to Paiwan. His contributions to the study of Paiwan were (1) the classification of Paiwan dialects into two groups: The Northern Paiwan and the Southern Paiwan, (2) the observation of the different reflexes of Proto-Formosan *t₁ and *t₂ between the Northern and Southern groups, and (3) the record of abundant linguistic data. He has been criticized, however, in that these data were not precisely phonemicized and the problems of morphophonemic alternations were not handled.
Raleigh Ferrell was the next scholar to study the language. By using the linguistic data collected from several dialects of Paiwan, Ferrell published a number of works about that language. He proposed several important observations fo the syntactic characteristics of Paiwan language. However, his phonological descriptions in these works had similar defects as those of Ogawa. He did not phonemicize the transcriptions, and did not provide sufficient information about the morphophonemic alternations.
In this paper, the author presents a systematic analysis of the Butanglu dialect in four sections: (1) introduction, (2) phonemes and phonetic description, (3) readjustment system, and (4) morphophonemic rules. Section (1) gives the geographical location of Butanglu and the collecting procedure of linguistic data. Section (2) presents the phonetic realizations of twenty-eight phonemes and several phonetic rules relevant to these phonetic realizations. Section (3) deals with the distribution of phonemes, the syllable structure, the morpheme structure, and the morpheme structure constraints. Section (4) discusses the problem of morphophonemic alternations and suggests a phonological rule for each morphophonemic alternatiion. The rules, both phonetic and phonological, are formalized in terms of distinctive features.