Morphological re‐analysis in the Bantu nasal class

Summary By way of summary, we have discussed in some detail the issue of the synchronic status of Class 9/10a prefixes in Bantu. We suggested that in some languages these forms may have been restructured so that the historical prefix, or at least part of the prefix, and its effect are incorporated into the noun stem due to the non‐alternating nature of these forms. Various types of evidence which might be cited to defend a synchronic analysis which posits independent prefixes for this class were considered. However, the evidence was shown to be, at best, uncompelling. Further, two proposals concerning the lexical representation of phonological words were considered in the light of the above analysis. Vennemann's claim that the lexicon consists entirely of independent words was shown to be problematic in that it does not allow for the explanation of various analogical developments. We also examined briefly a proposal by Hudson concerning the representation of non‐productive alternations; we suggested that certain patterns of diminutive/ augmentative formation may best be represented under this analysis and regarded as early forms of suppletion. Such representation makes certain claims about the direction which levelling will take; these claims need to be examined in greater detail. Although we have not demonstrated that no synchronic Class 9/10a prefixes are to be found in Bantu, it has been suggested that this may be the case in certain languages. In other languages, which show 9/10a prefixes, we suggested that part of the historical prefixes now figures as part of the noun stem. The implications of this analysis are clear. First, we need to re‐examine the synchronic status of the nasal prefixes of these same noun classes in languages where they are superficially identical. Second, even in cases where the surface facts are ambiguous, the Makonde data, the facts of prefix addition in the diminutive/augmentative, the treatment of loan‐words, etc., suggest that prefix restructuring may be a fairly common phenomenon. The difference between additive and replacive prefixes may represent the difference between derived and suppletive forms synchronically. This raises the general question of whether all of the singular/plural noun class pairs might be suppletive in any language.12 These questions represent interesting directions for future research.