Morphology and word order in `creolization' and beyond
MICHEL DEGRAFF
degraff@MIT.EDU
Abstract

Linguists have long posited ``abnormal'' (i.e., ``nongenetic'') transmission in Creole genesis, supposedly with ``significant discrepancies'', in opposition to ``normal'' (i.e., ``genetic'') transmission in ``regular'' language change, whereby a language is ``passed down from one speaker generation to the next with changes spread more or less evenly across all parts of the language'' (see, e.g., Thomason and Kaufman 1988). This is what I call ``Creole Exceptionalism''.

In this paper, I select various patterns in morphology and word order in order to question Creole Exceptionalism. Take ``discrepancies'' in one core domain of Creole genesis, namely VP-related morphosyntax. Are such discrepancies at all ``exceptional''?

I start with a sample of VP-related properties in Haitian Creole (HC)--a bona fide Creole on sociohistorical grounds. I compare the morphosyntax of verbs and object pronouns in HC and in some of its major source languages. I also speculate on the development of said morphosyntax from the perspective of second- and first-language acquisition and the role of grammaticalization and reanalysis therein. This, in turn, leads me to examine various theoretical proposals on the morphology-syntax interface vis-à-vis verb and object placement in language change/creation, and to consider germane patterns in Germanic and French diachrony. I also compare the HC patterns with their counterparts from a couple of other Romance-lexifier Creoles, namely Cape Verdean Creole (lexifier: Portuguese) and Palenquero Creole (lexifier: Spanish).

My conclusions are fourfold: (i) Even within a small sample of Romance-lexifier Creoles, there is no structural ``Creole'' uniformity in the VP and its extended projections. (ii) Certain ``discrepancies'' in French and English diachrony seem as ``significant'' as their analogues in Creole diachrony. (iii) This paper's observations argue against the classic (e.g., Bickertonian) Pidgin-to-Creole scenarios whereby pidginization qua structural ``break in transmission'' produces a macaronic and affixless pidgin that subsequently seeds a Creole qua ab ovo creation. (iv) Similarly, there is little evidence from HC to support Lefebvre's relexification hypothesis whereby HC grammar would essentially reflect substratum grammar with the French contribution strictly limited to phonetic strings ``deprived of [syntactic and semantic] features'' and to word-order patterns in lexical projections only (Lefebvre 1998).

The overall conclusion is that Creole languages do not constitute a well-delineated and exceptional class (i.e., there are no special diachronic processes of ``creolization'' and there is no distinct and uniform ``Creole'' typology): ``creolization'' and ``language change'' reflect processes of language development that are uniform across the species.

Keywords: Acquisition; Cape Verdean Creole; Creolization; French; Germanic; Grammaticalization; Haitian Creole; Language change; Morphology-syntax interface; Object placement; Palenquero; Reanalysis; Relexification; Tense-Mood-Aspect; Verb placement; Word order.


degraff@MIT.EDU