I apologize for the overkill in ascribing the calque trou de nez “nostril” to Parkvall (2000). This confusion stems from my failure to check carefully the original sources cited in Chaudenson (2003, pp. 227–228). The African origin of Haitian Creole (HC) twou nen (misspelled twou-ne) is claimed by Lefebvre (1998, p. 335). Chaudenson is more thorough than I am, given that he cites the precise page numbers for the data he adduces. My comments about Parkvall's failure to consider the potential French origin for the various calques he cites still hold. There appears to be convergence of compounding strategies in the substrate languages and the French target.1
To the unfelicitous and misleading term superstrate, I prefer that of target language. In historical linguistics, a superstrate language is one that is introduced after the establishment of a stabilized form of a target language modified by a substrate language or substrate languages. In the case of French, for instance, Germanic languages constitute superstrates that affected the structure of Vulgar (vernacular) Latin after this language had been restructured by its contact with Celtic varieties spoken by conquered or acculturated Gauls.
My reference to Parkvall's (2000) discussion of presumed substrate-influenced calques was within the context of my support for the important methodological requirement proposed by Chaudenson concerning research on the genesis of French-based creoles—namely, the comparison of Indian Ocean and New World creoles. Because the African influence is generally excluded in the Indian Ocean creoles, any shared features between these and their American congeners must be attributed to their Colonial French target (Chaudenson's and my position) or to the pidginization process. The latter is an alternative to direct substrate influence that Parkvall also considers: “Much of what may look African in Creole semantics may therefore well be an indirect manifestation of former Pidginhood” (p. 113). Concerning this alternative theory, the burden of proof in creole studies lies with those who posit a pidgin origin in the genesis of creoles, inasmuch as the criterion of economy favors substrate or target language influence or a combination of both. So far, no convincing evidence has been advanced by proponents of the theory that assumes the existence of a prior pidgin, either a pidgin transported from Africa or one developed in situ as a means of communication for a highly linguistically diverse servile population. As I stressed in my review of Chaudenson (2003), that author presents convincing demographic evidence for an early stage in the development of French plantation colonies in which numerically inferior or equal incoming slaves had access to the vernacular of the European settlers: Colonial French. Thus, no need existed for a stabilized pidginized variety of that linguistic target. It might well be that with massive importation of slaves within a relatively short period of time at a later stage of colonization, the acculturated or native-born speakers of the modified target—the newly arrived slaves (the bossals)—initially pidginized it. To account for the continuous restructuring that has led to the current forms of French-based creoles, I join Chaudenson in positing the second language acquisition of a target that had become highly variable.
Concerning the calques that form the subject of Parkvall's rejoinder, our positions do not differ markedly. I am prepared to accept, as I did in my review article and elsewhere (Valdman, 2000), the influence of African languages. Indeed, the other calques cited in his book, namely, eye of the foot for ankle (je pye in HC) and the extension of the term for foot (pye) to the lower leg are probably better examples of unequivocal reflexes of African calques. The point that Chaudenson makes, and that I support, is that before ascribing these calques to relexification on the model of substrate languages, one needs to examine the putative target of the creators of French-based creole—namely, Colonial French—as well as other vernacular varieties of the language. Note that in my review I identify another area in which substratal influence might be more probable: prosodics and intonation. In the absence of solid research in this underdeveloped area of the structure of creoles in general and French-based creoles in particular, this claim still remains speculative.
