Word-based morphology and synthetic compounding By Rudolf P. Botha (review)

480 LANGUAGE, VOLUME 58, NUMBER 2 (1982) the existence of a pitch-accent system that has evolved from tonal origins. H recasts Voegelin's work in a generative, albeit unformalized, framework. The paper takes on more than it can treat properly, but it certainly suggests further lines of research. B. Vine's 'Remarks on African shadow vowels' has such a goal explicitly in mind; and by concentrating on the universal implications of the various manifestations of these unusual voiceless vowels in several related languages, it succeeds in doing much more than surveying the problem. The remaining papers are all worth reading, though space permits me only to name them: D. Birnbaum's 'Rising diphthongs and the Slovak rhythmic law', H. Bochner's 'The / —» o rule in Serbo-Croatian', and M. Laughren's 'An a-s account of tone in Zulu'. Altogether, the volume is quite a bargain. Where else can one find four hundred pages of good phonology for less than ten dollars? [Ellen Kaisse, University of Washington.] Word-based morphology and synthetic compounding. By Rudolf P. Botha. (Stellenbosch papers in linguistics, 5.) Stellenbosch: Dept. of General Linguistics, University of Stellenbosch , 1980. Pp. iii, 167. The aim of Botha's composite work is to present a 'phrase-based' theory of synthetic compounding for Afrikaans. It is an explicit rejection of the lexicalist position: 'Fundamental to every major lexicalist theory is the constraint that W[ord] F[ormation] R[ule]s do not take units larger than words as their bases' (p. 1). Chap. 1 is a short introduction, outlining the basic argumentation and defining some crucial terminology. Chap. 2 is a critique of T. Roeper & M. Siegel's treatment ofsynthetic compounding in English (LI 9.199-260, 1978). In Chap. 3, Botha criticizes Margaret Reece Allen's treatment of synthetic compounding in English (University of Connecticut dissertation, 1978). In Chap. 4, B unveils his own theory. He proposes that synthetic compounds are formed not on the basis of subcategorization frames, as proposed both by Roeper & Siegel and by Allen (which allows them to maintain a word-based theory ofmorphology), but on the basis ofdeepstructure configurations. B motivates this proposal with evidence from Afrikaans, viz. synthetic compounds of the general form O + V + Affix (cf. Eng. gun running, truckdriver). The order of components is the same in Afrikaans. B is able to claim that these are formed from deep-structure configurations, since it is apparently plausible that Afrikaans has underlying SOV word order. This proposal accommodates the traditional treatment of synthetic compounds , and is at least reasonable. Unfortunately, one is led astray by B's criticisms . Nowhere does he suggest that English has underlying SOV order; thus his proposal would not work for the facts of English. Why does he bring them up, since his theory does not cover those data? B does point out a number of what he takes to be 'ill-defined' notions in the other two treatments; however, in the absence of a viable counterproposal, one cannot really evaluate his critiques. [Michael Hammond , UCLA.] Contributions to historical linguistics: Issues and materials. (Cornell linguistic contributions, 3.) Ed. by Frans van Coetsem and Linda R. Waugh. Leiden: Brill, 1980. Pp. ix, 339. /88.00. The seven papers contained in this volume are samples of work by Cornell University faculty members. Included are three papers on Asian languages (Chinese, Tibetan) and four on European languages (French, Romance, Albanian , Germanic). The contributions vary greatly in length, the two extremes being Nicholas C. Bodman, 'Proto-Chinese and Sino-Tibetan: Data towards establishing the nature of the relationship ' (165 pp.), and Robert A. Hall Jr., 'The gradual decline of case in Romance substantives ' (8 pp.) In between we find William H. Baxter III, 'Some proposals on Old Chinese phonology' (32 pp.); John McCoy, 'The reconstruction of upper register nasals and laterals in Proto-Cantonese' (13 pp.); Frederick B. Agard, 'The genealogy ofthe French language: Further toward a theory of chronological language delimitation ' (46 pp.); Gordon M. Messing, 'Politics and national language in Albania' (10 pp.); and Frans van Coetsem, 'Germanic verbal ablaut and its development: A contribution to the theory of internal inflection' (58 pp.). This collection is marked by considerable unevenness . Specialists in...