DFKI-LT - Hausa Final Vowel Shortening - Phrasal Allomorphy or Inflectional Category?

Berthold Crysmann
Hausa Final Vowel Shortening --- Phrasal Allomorphy or Inflectional Category?
in: Geert Booij, Emiliano Guevara, Angela Ralli, Salvatore Sgroi, Sergio Scalise (eds.):
1 Morphology and Linguistic Typology: On-line Proceedings of the Fourth Mediterranean Morphology Meeting (MMM4) Catania, 21-23 September, Pages 105-125, UniversitÓ degli Studi di Bologna, 2005
In this paper, I will address the phenomenon of final vowel shortening (FVS) in Hausa1. Based on detailed morphological evidence, I shall argue that FVS is but one exponent of a systematic morphosyntactic distinction in the language. Given the systematicity of the distinction together with the diversity of exponence, I shall conclude that a treatment in terms of inflectional morphology is to be preferred over Hayes (1990)’s analysis as Precompiled Phrasal Phonology (PPP). The morphological view will furthermore enable us to connect the Hausa data to a typologically well-established inflectional category, namely marking of the mode of argument realisation, a perspective that will deepen our understanding of Hausa syntax and morphology. The paper is organised as follows: after a brief introduction to the basic pattern and a discussion of Hayes’ account in terms of phrasal allomorphy, I shall present additional data to the extent that FVS cannot be singled out as an isolated allomorphic process. Rather, we shall see that vowel length alternation is subject to close interaction with Hausa stem morphology. Moreover, under a broader empirical perspective, the twofold length distinction will turn out to be only one of many patterns in which an underlyingly tripartite distinction is morphologically neutralised. Next, I shall submit Hayes’s surface-oriented adjacency requirement – a necessary criterion for precompiled phonologies – to some further scrutiny and show that Hausa provides a body of evidence against such a surface-oriented view, supporting instead an analysis in terms of argument structure and lexicalised traceless extraction. In section 4, I shall connect Hausa to strikingly similar phenomena in Chamorro and French, all displaying morphological sensitivity to extraction contexts (Bouma et al., 2001). Furthermore, we shall see that Hausa already provides independent evidence for its membership in the typological class of extraction-marking languages.
Files: BibTeX, 105-125-Crysmann-MMM4.pdf