Reviewed by: Creole genesis, attitudes and discourse ed. by John R. Rickford, Suzanne Romaine J. Clancy Clements Creole genesis, attitudes and discourse. Ed. by John R. Rickford and Suzanne Romaine (Creole language library 20.) Amsterdam: John Benjamins, 1999. Pp. viii, 418. $138.00. This collection of papers celebrates the life and work of Charlene ‘Charlie’ J. Sato whose work and interests included, among other things, pidgin and creole languages, language acquisition, and political issues involving minority languages. Focused on the three themes mentioned in its title, the book also includes an introduction (Part A) which highlights the life and work of Sato and contains, as well, two pieces in Hawaiian English. Because of space constraints, I review only some of the contributions. The studies involving creole genesis in Part B cover a wide range of phenomena and languages. Derek Bickerton (‘Pidgins and language mixture’, 31–43) challenges one view regarding the lexical make-up of pidgins, namely, that a pidgin derives its lexicon almost exclusively from one language (cf. Bakker 1995:29). Using textual evidence from pidgin varieties in Hawaii in the late nineteenth and early twentieth centuries, Bickerton argues that Pidgin Hawaiian and Pidgin English were two extremes of the same continuum and that the lexicon of this variety was heterogeneous, though he offers no percentages. In support of his argument, he does cite percentages taken from the pidgins Russenorsk, Chinook Jargon, and the pidgin-derived creoles Saramaccan, Sranan, Guyanais, Berbice Dutch, and Lesser Antillean Creole. Although Bickerton makes a valid point, some 75 additional pidgins still remain to be examined in order to be able to say something definitive about the heterogeneous nature of pidgin lexicons. Sarah Roberts (‘The TMA system of Hawaiian Creole and diffusion’, 45–70) traces the development of the TMA (Tense-Mood-Aspect) markers in Hawaiian Pidgin English (HPE), concentrating on native-born speakers, through six phases from 1778 to the present. She finds that, apart from the past/anterior marker bin, the TMA markers developed in HPE independently of any external influence from other pidgins or creoles. This conclusion goes against Holm’s (1992) Atlantic-to-Pacific diffusion hypothesis, which cannot account for the absence of the Atlantic markers de/da and don in Hawaii, or for the presence of stei ‘progressive’ and pau ‘completive’ which developed locally. However, it is in line with Baker’s (1993) foreigner-talk model in that it accounts for all this and the accommodation of bin in HPE. Three studies examine the nature of the copula in creoles. John Holm et al.’s study (‘Copula patterns in Atlantic and non-Atlantic creoles’, 97–119) looks at copula presence/absence with NP predicates, locative predicates, and adjectival predicates, as well as the function of the copula as a highlighter. The expectation is that these languages would have separate copulas for NPs and locatives, a null copula with adjectives, and a copula as a highlighter if there is substrate influence in these creoles. (Many West African languages display these traits.) Their results reveal that all English-based creoles have different copular elements for NPs and locatives respectively, that they also display a null copula with adjectival predicates, and that four of five of these creoles have a copula as a highlighter. This pattern, however, is not as homogeneous in the Atlantic creoles lexified by French, Dutch, Portuguese, or Spanish: three of eight have a separate copula for NPs and locatives, and two of those cases can be explained by superstrate influence. Moreover, only four of eight have a highlighter. As for the non-Atlantic creoles, they [End Page 210] do not exhibit the expected pattern at all. Thus, the case for substrate influence, while claimed to be present for all the Atlantic creoles considered, is strongest for the English-based creoles. John McWhorter (‘Skeletons in the closet: Anomalies in the behavior of the Saramaccan copula’, 121–42) tests Holm et al.’s substrate hypothesis on Saramaccan (SM), albeit not explicitly so. McWhorter shows that SM uses a separate copula for identificational predicates (e.g. George is my brother ; copula da) on the one hand, and class predicates (George is...
Read full abstract