Consonant mutation and reduplication in Seereer-Siin

Abstract

In this paper I propose an analysis of Seereer that accounts for the variation found in reduplication by building on the insights of featural affixation theories (Akinlabi 1996, Zoll 1998) and Correspondence Theory (McCarthy & Prince 1995) within the overall constraint-based framework of Optimality Theory (OT). I conclude that unless other conflicting constraints intervene, featural transfer between base and reduplicant systematically takes place. 3 The organisation of the article is as follows: §2 presents the facts of consonant mutation in Seereer and the general approach, §3 consists of an analysis of consonant mutation within a constraint-based framework and §4 provides an account of featural transfer and variation in reduplicative forms. 2 Throughout this article, mutating consonants are indicated by boldface. 3 A similar conclusion is reached by Myers & Carleton (1996) with regard to tonal transfer in Chichewa. Consonant mutation and reduplication in Seereer-Siin 335 Consonant mutation and reduplication in Seereer-Siin 337 In stems that undergo mutation, a given noun class always conditions the same grade, either a, b, or c, regardless of the gradation set or the mutation type (see Appendix 1). Thus, for example, Class 1, which contains human singular nouns, conditions the b-grade. Stems that undergo voicing mutation will appear in voiceless stop-initial forms in that class, while those that undergo continuancy mutation will appear in stop-initial, as opposed to continuant-initial, form. Class 2, which contains human plural nouns, conditions the a-grade, and Class 3b, which contains augmentative singular nouns, conditions the c-grade. By way of example, consider the two nouns in (7). The stem for 'sick person' undergoes voicing mutation, while the stem for 'dead person' undergoes continuancy mutation. In Class 1, the allomorphs occur in their b-grade forms: a voiceless stop, [c], for 'sick person' and a stop, [q], for 'dead person'. In Class 2, the allomorphs occur in their a-grade forms: a voiced stop, [j], for 'sick person', and a continuant, [x], for 'dead person.' In Class 3b they both exhibit prenasalised stops or c-grade forms: [nj] and [NG], respectively. (7) Class 1 b-grade ocir oqon Class 2 a-grade J ir xon Class 3b c-grade ajijir aNGon 'sick person' ' dead person' Voicing mutation Continuancy mutation Consonant mutation and reduplication in Seereer-Siin 363 Sagey, Elizabeth (1986). The representation of features and relations in nonlinear phonology.

Voiceless implosives in Seereer-Siin

“…Crosslinguistically, however, voicing appears to be the only type of phonologically significant phonation type contrast within the class of implosive consonants (Clements & Osu 2002: 300 . Patterns of stem behavior in Seereer-Siin provide evidence that initial implosive consonants in nominal stems are underlyingly voiceless, while in verb stems they are underlyingly voiced (Mc Laughlin 1994. Their phonemic status is further supported by minimal pairs such as those in (2) which illustrate the high functional load carried by the distinctiveness between voiced and voiceless implosives within the context of consonant mutation, namely number in verbs.…”

“…There is a second type of consonant mutation in Seereer-Siin(Fal 1980, Mc Laughlin 1994 which more closely resembles the Fula pattern and in which consonants alternate between a continuant, a stop, and a presnasalized stop. Implosive consonants do not participate in that type of mutation.3 Demolin et al (2002) also mention discussion of the phonetics of these sounds in Waly ColyFaye's 1979 thesis which was, unfortunately, not available for this study.…”

On the typology of metaphony/stress dependent harmony

“…Lexical representations can be impoverished in several ways: they can lack phonological content altogether, as in pro or expletives, or their phonological content can be impoverished: they can lack skeletal (C, V) information, as in the case of featural affixes, or they can lack featural specifications (see Svenonius & Bye 2012 for important general observations). In some cases morphs have a mixed character: they can consist of both segmental and floating material, or segmental and bare skeletal material (Mc Laughlin 2000;Finley 2009 and references therein). The Romance dialect of Calvello (Basilicata, Italy;Gioscio 1985) that I examine in the next section, illustrates both cases.…”

Non-concatenative morphology as epiphenomenon1

“…Nevertheless, the perception persists since a number of influential publications addressing mutation in the context of autosegmental phonology restrict their focus to purely additive mutations (Lieber 1983(Lieber , 1984(Lieber , 1987Akinlabi 1996). 17 Work on mutation carried out within the OT framework has tended to adopt autosegmentalized representations of distinctive features, developing constraints to handle featural autosegments that are 'floating' in the input (Zoll 1998;Myers 1997;McLaughlin 2000;Wolf 2007). Floating features are problematic with respect to getting linearization to come out right in privative feature theories, however, and we therefore follow De Lacy (2006Lacy ( , 2011 in assuming that there are no floating features.…”