Componential analysis of Lushai phonology By Alfons Weidert (review)

REVIEWS719 body of data unavailable elsewhere. It also has increased theoretical knowledge of conjunct verbal constructions—but, because of B's obfuscatory prose style, has not done so in a manner which makes it easy for the reader to obtain this knowledge. Bahl has succeeded in one of his major aims, namely convincing his readers of the previously undetected complexity of conjuncts. It is unfortunate that he has not equally well conveyed a full understanding of the nature of the complexity. REFERENCES Burton-Page, B. 1957. Compound and conjunct verbs in Hindi. Bulletin of the School of Oriental and African Studies 19.469-78. Gilchrist, John. 1 796. A grammar of the Hindustani language. Calcutta. [Reprinted by the Scolar Press Ltd., Menston, 1970.] Hook, Peter Edwin. 1974. The compound verb in Hindi. (Michigan series in South and Southeast Asian languages and linguistics, 1). Ann Arbor: University of Michigan, Center for South and Southeast Asian Studies. Kellogg, Rev. S. H. 1875. A grammar of the Hindi language. London. [Reprinted by Routledge & Kegan Paul, London, 1955.] Platts, John T. 1878. A grammar of the Hindustani or Urdu language. London: Oxford University Press. [Reprinted by Munshiram Manoharlal, Delhi, 1967.] [Received 13 July 1977.] Componential analysis of Lushai phonology. (Current issues in linguistic theory, 2.) By Alfons Weidert. Amsterdam: John Benjamins, 1975. Pp. vix, 139. /30.00. Reviewed by F. K. Lehman, University of Illinois, Urbana Lushai is a Tibeto-Burman language of the Kuki-Chin group, spoken for the most part in the mountains on India's border with Burma. There have been, before this, a few partial studies of the phonology, mainly by Bright 1957a,b, Burling 1957, and Henderson 1948; but this is the first comprehensive work on the subject. There exists a good, though non-tone-marked dictionary by Lorrain 1940, but there is no linguistically useful study of the grammar; this reviewer and his students are at present working on one, as the result of a two-year project with a Lushai teacher supported by the Graduate College Research Board of the University of Illinois. We must be grateful to Weidert for this thorough publication, the more so since Lushai is one of the few T-B languages preserving both a fairly full set of syllablefinal consonants (dental, labial, and velar surds; nasals, liquids, and glides) as well as contrastive vocalic length in both open and closed syllables. These facts amount to important comparative evidence for T-B—for which, often on rather incomplete data, these features must be reconstructed. Moreover, the study is especially valuable because of W's exceptional knowledge of Lushai; the internal evidence clearly indicates this, and his being married to a Lushai gives him access to the language at a level few other linguists can attain. The work, however, suffers in two ways. First, it employs a linguistic-phonological framework that appears to be W's private invention—a framework that seems, moreover, largely motivated not so much by a clearly thought-out view of the nature and organization of phonology as by a history of dissatisfaction with generative phonology, similar to that in the work of E. Fudge. (Unhappily, too, 720LANGUAGE, VOLUME 54, NUMBER 3 (1978) W's explanation of his position is indifferently argued in the introduction; it is better stated in his 1977 work on Khamti, a Tai-Shan language.) Second, W's views give rise to a needlessly difficult notation, again his own invention. My objection is not that it is novel or abstract. Rather, I refer to its extensive use of mnemonic devices (largely Greek letters) that stand less for substantive (e.g. phonetically motivated) entities of phonology, real or postulated, than for entities of the analytical framework itself. These devices figure importantly in the system of rules—which, moreover, are pseudo-generative in character; they stand not for processes in a derivation but for mapping relations between levels of analysis (morphophonological, phonemic, and phonetic). In effect, the rules of the analysis do two things: they supply particular specifications, on the basis of mnemonically indicated generalized specifications (indicated either by Greek letters or by blanks), or they catalog morphosyntactic conditions for various phonological alterations, mostly sandhi phenomena, which are themselves...