Copyright © 1999 Massachusetts Institute of Technology
Creoles constitute a unique language group. Other groups consist either of languages derived from a hypothesized protolanguage (as Welsh, English, Greek, and Sanskrit derive from Proto Indo-European) or a single historical ancestor (as Portuguese, Spanish, and Italian derive from Latin). Creoles, however, have no clear affiliation. Although the term creole has been used to characterize any language with an appearance of language mixture, languages generally accepted as creoles have all arisen in recent centuries from contacts between speakers of unrelated languages, usually as indirect results of European colonialism. They are found throughout the tropics, especially where large numbers of workers have been imported as slaves (or, occasionally, indentured laborers) to work on European-owned plantations. Because more work has been done on plantation creoles than on other types (those that developed in racially mixed communities on African and Asian coasts, or those derived initially from maritime contacts in the Pacific), most of what follows will apply primarily to plantation creoles.
Until recently, few scholars (e.g., Schuchardt 1882-91; Hesseling 1933) took these languages seriously; most treated them as deformed versions of European languages. Even to scholars who accepted them as true languages, their origins were controversial: some saw them as radical developments of French, English, and other languages, others as non-European languages thinly disguised by European vocabularies, still others as descendants of an Ur-creole perhaps springing from Afro-Portuguese contacts in the fifteenth and sixteenth centuries, perhaps even dating back to the medieval Lingua Franca. Other issues, such as whether creoles were necessarily preceded by, and derived from, some radically impoverished quasi-language (jargon or early-stage pidgin), or whether children or adults played the major role in their creation, were debated with equal heat but equally little agreement.
More than a century ago, Coelho (1880-86) pointed out that creoles showed structural similarities much greater than would be predicted given their wide distribution, their varied histories, and the large number of languages spoken in the contact situation where they arose. However, creoles would have held little interest for cognitive science had it not been proposed that, due to their mode of origin, they reflected universals of language more directly than "normal" languages.
This hypothesis of a species-specific biological program for language, providing default settings where acquisitional input is greatly reduced and/or deformed (often referred to as "bioprogram theory"), assumes a version of GENERATIVE GRAMMAR proposed by Borer (1983), and now widely accepted by generativists, in which parametric variation in SYNTAX arises solely from variability in grammatical MORPHOLOGY. In the (severely depleted) morphology of creoles, many (often most) grammatical morphemes are distinct from those of any language spoken in the contact situation. This indicates that grammatical morphemes, lost in the pidginization process, are replaced by the creation of new morphemes, often from semantically "bleached" referential items (thus the verb for "go" marks irrealis mode, locative verbs mark imperfective aspect, etc.). Some grammatical functions required immediate recreation of new morphemes, whereas morphemes for other functions were recreated centuries later, if at all (Bickerton 1988). The implication of these facts for the study of universal grammar remain, surprisingly, unexplored.
Bioprogram theory has obvious implications for other fields of inquiry. In LANGUAGE ACQUISITION, it suggests that creoles should be acquired more rapidly and with fewer errors than non-creoles, a prediction that has only recently begun to be tested (Adone 1994). In innateness studies (see INNATENESS OF LANGUAGE), it suggests in addition to syntactic principles a default SEMANTICS yielding highly specific analyses of TENSE AND ASPECT, modality, negation, articles, and purposive constructions, among others (an element missing from generative accounts of innateness of language). In the EVOLUTION OF LANGUAGE it suggests that syntactically structured language could have emerged abruptly from a structureless protolanguage (Bickerton 1990a).
Although bioprogram theory has had "an explosive impact . . . upon all aspects of the field" (McWhorter 1996), other approaches to creole genesis continue to flourish. Perhaps the most currently popular alternative is substratism, which claims that languages spoken by non-European ancestors of creole speakers served as sources for characteristically creole structures such as serial verb constructions (e.g., equivalents of "I carry X come give Y" for "I bring X to Y") and focusing of verbs by fronting and copying (e.g., equivalents of "is break he break the glass" for "he broke the glass" as opposed to merely cracking it). Alleyne (1980), Boretsky (1983), and Holm (1988-89), among others, exemplify this approach. However, substratists attribute these and other creole features to the Kwa language group in West Africa, heavily represented in some creole-speaking areas (Haiti, Surinam), but more lightly, if at all, in others (the Gulf of Guinea, Mauritius) and not at all in still others (Hawaii, the Seychelles).
Substratism's mirror-image, superstratism -- the belief that creoles derive their syntax from colloquial versions of European language -- is nowadays largely confined to French creolists (e.g., Chaudenson 1992). More widespread is a "componential" approach (Hancock 1986, Mufwene 1986) claiming that different mixtures of substratal, superstratal, and universal features contributed to different creoles, the mixture being determined in each case by social, historical, and demographic factors. However, because no one has yet proposed a formula for determining the relative contributions of the various components, this approach is at present virtually unfalsifiable, and it constitutes a research program rather than a theory.
Controversies have thus far centered around creole origins. But only recently, after decades of speculation and conjecture, have serious attempts been made to gather historical data (e.g., Arends 1995; Baker 1996). Not all this work is of equal value; claims of a scarcity of children in early colonies are refuted by contemporary statistics (Postma 1990; Bickerton 1990b). Documentation for the earliest stages of creole languages remains sparse, except for Hawaii, where rich data for all phases of the pidgin-creole cycle has been unearthed by Roberts (1995, 1998). These data confirm, at least for Hawaii, the main claims of bioprogram theory: that the creole was created (a) from a primitive, structureless pidgin (b) in a single generation (c) by children rather than adults. Hopefully, ongoing historical research on other creoles will determine the extent to which these conform to Hawaii's pattern.
Adone, D. (1994). The Acquisition of Mauritian Creole. Amsterdam: John Benjamins.
Alleyne, M. (1980). Comparative Afro-American. Ann Arbor, MI: Karoma.
Arends, J., Ed. (1995). The Early Stages of Creolization. Amsterdam: John Benjamins.
Baker, P., Ed. (1996). From Contact to Creole and Beyond. London: University of Westminster Press.
Bickerton, D. (1981). Roots of Language. Ann Arbor, MI: Karoma.
Bickerton, D. (1984). The language bioprogram hypothesis. Be-havioral and Brain Sciences 7.
Bickerton, D. (1988). Creole languages and the bioprogram. In F. J. Newmeyer, Ed., Linguistics: The Cambridge Survey, vol. 2. Cambridge: Cambridge University Press, pp. 267-284.
Bickerton, D. (1990a). Language and Species. Chicago: University of Chicago Press.
Bickerton, D. (1990b). Haitian demographics and creole genesis. Canadian Journal of Linguistics 35:217-219.
Borer, H. (1983). Parametric Syntax. Dordrecht: Foris.
Boretsky, N. (1983). Kreolsprachen, Substrate und Sprachwandel. Weisbaden: Otto Harrassowitz.
Chaudenson, R. (1992). Des Iles, des Hommes, des Langues. Paris: L'Harmattan.
Coelho, F. A. (1880-86). Os dialectos romanicos o neo-latinos na Africa, Asia e America. Boletin da Sociedade de Geografia de Lisboa 2: 129 - 196; 3: 451 - 478; 6: 705 - 755.
Hancock, I. (1986). The domestic hypothesis, diffusion and componentiality: an account of Atlantic Anglophone Creole origins. In P. Muysken and N. Smith, Eds., Substrata Versus Universals in Creole Genesis. Amsterdam: John Benjamins.
Hesseling, D. C. (1933). Hoe onstond de eigenaardige vorm van het Kreols? Neophilologus 18:209-215.
Holm, J. (1988-89). Pidgins and Creoles. 2 vols. Cambridge: Cambridge University Press.
McWhorter, J. (1996). Review of Pidgins and Creoles, An Introduction, edited by J. Arends, P. Muysken, and N. Smith. Journal of Pidgin and Creole Languages 11:145-151.
Mufwene, S. (1986). The universalist and substrate hypotheses complement one another. In P. Muysken and N. Smith, Eds., Substrata Versus Universals in Creole Genesis. Amsterdam: John Benjamins.
Postma, J. M. (1990). The Dutch in the Atlantic Slave Trade, 1600-1815. Cambridge: Cambridge University Press.
Roberts, S. J. (1995). Pidgin Hawaiian: a sociohistorical study. Journal of Pidgin and Creole Languages 10:1-56.
Roberts, S. J. (1998). The role of diffusion in creole genesis. To appear in Language.
Schuchardt, H. (1882-91). Kreolische Studien. Vienna: G. Gerold's Sohn/K. Tempsky.
Copyright © 1999 Massachusetts Institute of Technology