- Creoles, Their Substrates, and Language Typology ed. by Claire Lefebvre
This is a very focused collective volume. Flanked by the editor's introduction and a concluding commentary by Bernard Comrie on creoles and typology are twenty-five chapters on specific creole-substrate pairings: nine on the Atlantic coast of Africa and the New World, eight on Asia, and eight on Australia and the Pacific Islands. [End Page 305]
Given a choice between superstrate continuity (e.g., French-lexified creoles are dialects of French), substrate continuity (also known as relexification, i.e., creoles superimpose foreign lexicon on substrate grammatical patterns), dynamic blending (a compromise between the previous two), and Bickertonian ex nihilo creation, the volume favors substrate continuity as the primary source of creole morphosyntax. From the introduction: "The data discussed in this book show that creoles massively replicate the typological features of their substrate languages" (p. 29). A closer look shows that the relevant features are usually categorial rather than morphological, especially when substrate languages are morphologically rich. The categories are also subject to simplification, deletion (especially when the superstrate happens to have no semantically related stem), and some blending of substrate and superstrate constructions.
Wherever a single substrate language, or a small set of sister languages with a common typological profile, is identifiable, the dice are loaded in favor of heavy substrate continuity. The Bickertonian model, after all, was restricted to abrupt creoles emerging in a single generation from chaotically multilingual and typologically varied linguistic environments, as (according to Bickerton) in some Caribbean creoles and in Hawaiian creole. But only a handful of the classic creoles were ever serious candidates for this model, and even these have been disputed. The current trend is to extend "creole" generously to stabilized pidgins, and even (as in this volume) loosely to what others call "New Englishes" like Singapore English. This guarantees that substrate effects will be conspicuous.
Many chapters in the volume either point out striking similarities between a creole and a substrate language, or more systematically tabulate and quantify such similarities across a range of categories. Examples of the latter are chapters by Lefebvre, "Substrate Features in the Properties of Verbs in Three Atlantic Creoles" (pp. 127-53), and Jeff Siegel, "Substrate Reinforcement and the Retention of Pan-Pacific Pidgin Features in Modern Contact Varieties" (pp. 531-56). Few chapters propose detailed diachronic trajectories, a defect that can usually be excused by the absence of early pidgin and creole documents and by uncertainties about the formative sociolinguistic contexts. Another problem with some chapters is a failure to distinguish substrate (in the formative period) from adstrate, especially where an already formed creole moved into an area and subsequently underwent normal adstrate influences.
Diachronic analyses of how substrate categories came to be reproduced in a creole are most evident in chapters about English in the Pacific. Most notably, Harold Koch outlines the development of the transitive verb suffix -im (as in kill-im 'kill') in the earliest Australian pidgin and its many offshoots in Australia and Melanesia ("Substrate Influences in New South Wales Pidgin," pp. 489-512). Two suffixes, -im < *(h)im ~ *(th)em and -it < *it, initially competed, with -im winning out as the pidgin spread inland to Aboriginal languages with restricted phonotactic systems that disallowed final stops. The more interesting (and debatable) contention is that a transitive suffix in the pidgin was motivated by a sharp binary transitivity opposition in Aboriginal languages, expressed by ergative-absolutive case-marking. I find this unconvincingly teleological, given that Aboriginal languages lack transitive markers in verbs (pp. 502-3), that the pidgin follows subject-verb-object constituent order with no structural case-marking even of pronouns, and that the oldest recorded forms of the pidgin allowed bare transitive verbs with neither a transitive suffix nor an overt object (pp. 498-501). This suggests that the generalization of -im may have had more to do with phonotactics (in effect, euphony); as Koch points out earlier, adding -im (or -it) made English verbs with final clusters...