ABSTRACTS The present grant provides funds to allow the PI (Prof. Larry Hyman) and Dr. Francis Katamba to complete their joint research on the phonological and grammatical bases of Luganda prosody. Supported by National Science Foundation researchers have carefully documented and jointly published on a number of issues concerning the interfacing of phonology and grammar in Luganda. The project has centered around three areas: 1)the analysis of prosodic features (quantity, tone, accent, and intonation, which impact, respectively, on theories of syllabicity, tonal representation, metrical representation and phonetic implementation); 2)the analysis of prosodic domains required by these features (the phonological word, clitic group, tone group, tone phrase, and intonational phrase); and 3)the nature of the syntax-phonology interface (i.e. determining whether-or under what conditions-the phonology may directly access the syntax, rather than indirectly accessing the syntax by mediation of the prosodic domains). As is documented in the proposal, the current project has resulted in breakthroughs in each of these areas. First, by closely scrutinizing the complex data of Luganda, the researchers have discovered that the "facts" (which have been cited by a number of theoretical phonologists) are not always exactly as the standard references indicate. Second, the discovery of these facts has led the researchers to theoretical innovations which they have explored and propose to investigate further in this final phase of the project. These innovations include a refinement of moraic theory (particularly as it relates to complensatory lengthening and the question of what is a tone-bearing unit); a new contour representation of underlying tone in Luganda (and perhaps a subclass of so-called pitch-accent systems); a cyclic approach to morphology and phonology that allows affixes to be "infixed" rather than added to the periphery of their base; extension of overlapping prosodic domains to the lexicon; and testing of these results against the lexical prosodic domains required by the segmental phonology (e.g. for vowel height harmony, consonant mutation, consonant nasalization (the "Ganda (for ?Meinhof's! Law"), etc. Renewed funding from the N.S.F. will thus allow the PI and Dr. Katamba to complete their productive collaboration and attend to the discoveries that have arisen thus far in the project.
