To appear in the Journal of Pidgin and Creole Languages 18.2 (2003)

 

GENETIC LINGUISTICS AND GENETIC CREOLISTICS:

A Response to Sarah G. Thomason’s “Creoles and Genetic Relationships”

 

Salikoko S. Mufwene

University of Chicago

                                                                             

      In this note, I use the term creole as in Mufwene (2001) for some offspring of European languages which developed among primarily non-Europeans in Dutch, English, French, and Portuguese plantation settlement colonies of the 17th-19th centuries and have been disfranchised as vernaculars genetically unrelated to their lexifiers. The term pidgin is likewise used for those that developed in trade colonies of the same European nations, where contacts between Europeans and non-Europeans were originally sporadic. Those inclined to invoke pidgin or creole (PC) prototypes — to which I return below — are likely to identify them within this group, as did McWhorter (1998). Regardless of disagreements we may have about what counts, or does not count, as a PC, we certainly have fewer disagreements about this restricted set, on which I base my discussion below.¹

      Genetic creolistics is used in the title of this note by analogy to genetic linguistics to suggest that the subject matters of both research areas are similar. I would even like to submit that the former should be considered a subfield of the latter, differing from it only in its focus on a specific group of languages. Both research areas are ultimately concerned with mechanisms and ecological conditions of language speciation — although we in creolistics have preferred using the term restructuring, as if the process were peculiar to PCs alone, and perhaps those other varieties imprecisely identified as

 

“mixed” (Mufwene 2001). Some cross-pollination between them should be welcome. Actually, recent  research on the development of creoles militates for reopening the books on some of the ways in which research in genetic linguistics has traditionally been conducted, especially regarding the role that changes in the ecology of a language (both system-internal and external) play in its evolution (Mufwene (2001, Chapter 5).

      Contrary to the myth that creolistics is a young discipline, I’d like to argue that genetic creolistics is just about as old as genetic linguistics is. An important difference lies in the fact that since the beginnings of both subfields of linguistics in the 19^th century, creoles have typically been considered aberrations, irregular developments, therefore not worth the attention and time of serious linguists. Presumably, these varieties would not tell us anything about normal or natural language change. Such assumptions are unfortunately still identifiable in some modified forms in today’s professional linguistics publications. Our enlightened precursor Hugo Schuchardt must be turning with disappointment in his grave on realizing that, like late-19^th-century genetic linguists, some of us today would still ostracize him for wanting to treat PCs as regular, normal, and natural languages, therefore likely to enrich our understanding of Language (Schuchardt 1884, 1924).

      I submit that PCs can tell us a number of unexpected things regarding the architecture, functions, and evolutionary patterns of languages. Actually, due to various issues that have been raised about their developments, we are perhaps now in a privileged position to ask the following question: Does our inability to resolve the issues conclusively stem from the putative abnormality or unnaturalness of the language varieties or from the inadequacy of some working assumptions inherited from the 19^th century? Interestingly, the 19^th century is also where the conceptual foundations of genetic linguistics lie. While we have often been prompted implicitly or explicitly to compare our research methods against this research area (e.g., Chaudenson 2001), what we have learned to date in genetic creolistics can also motivate us to re-examine the practice of genetic linguistics itself, as I do below.

      Reading Sally Thomason’s column (henceforth Thomason 2002) in JPCL 17.101-109, I was struck by how she addresses some of  Mufwene’s (1998, 2001) and DeGraff’s (2001a) data- and theory-based arguments against some of her positions. Without adducing any counter-evidence, she resorts to precisely the same kinds of stipulations that she has been criticized for (especially by myself) — though she denies doing so. The intended meaning of stipulate is of course ‘accept or state a proposition without requiring that it be established by proof’. For example, Thomason invokes “prototypical pidgins and creoles” (2002:105) but she does not give any examples thereof nor tell us how to identify them. I thought I might find this notion explained in Thomason (2001) but no clarification is provided there either, despite what I quote below. Yet, she knows (Thomason 2001:190) that McWhorter (1998) has been criticized not only for submitting such a concept but also regarding the accuracy of the membership of the category (see some commentaries in Linguistic Typology 5, nos. 3&4, 2001; Mufwene 2000, DeGraff 2001b). Presenting the concept as if it were part of the received doctrine in creolistics is more like dodging the issue than addressing it. It is not clear to this particular reader whether Thomason means McWhorter’s best-exemplar prototypes or Mufwene’s (2000, 2001) heuristic prototypes (i.e., “classic creoles” of the Atlantic and Indian Ocean,  which have informed most of our theorizing). Such stipulating makes her position unfalsifiable, even if one were disposed to interpret charitably the following observof hers: 1) “pidgins and creoles that best fit [her] definitions are prototypical” (Thomason 2001:160), and 2) “a prototypical pidgin emerges abruptly, in a new contact situation involving three or more languages, with limited contact and no full bi-/multilingualism among the groups in contact ...” (ibid., 197).²

      Likewise, Thomason invokes “abrupt creoles” (assuming that it is enough to claim that they “developed without going through a stable pidgin stage,” 2002:106) but she does not identify a sample thereof. Thomason (2001) presents no example of this either. This presentation of facts is unclear to those of us who have argued that generally creoles have evolved directly and gradually from their lexifiers, without a transitional pidgin phase. I for one am curious which “abrupt pidgins and creoles” she has in mind. She proceeds to claim: “some mixed languages of both types [i.e., “bilingual mixed languages” and pidgins/creoles] seem to have emerged relatively abruptly, and some mixed languages of both types seem to have developed gradually” (2002: 107), but no examples are given at all to tell us which ones they are. And she punctuates, “Most pidgins still seem likely to have arisen relatively abruptly” (2002:108), but she still adduces no single example to make her position verifiable or falsifiable. If she cannot acknowledge the above statements as stipulations, then I have one more reason for not being convinced by her intimation that “the famous [i.e., ‘time-honored’?] Comparative Method” (CM) is as reliable as she claims.

      Thomason wants us indeed to believe that the CM has been so successful that it “has been altered only in relatively minor ways since” the 1870s and “is envied by many other historical scientists” (2002:102). To my knowledge, there are quite a few historical linguists, apparently “good” (Thomason’s term) and respected ones, who have questioned the “famous” CM. Almost all the contributions to Aikhenwald & Dixon (2001) question its reliability and “the regularity hypothesis of sound change” (Thomason 2002:102). They argue that synchronic structural similarities among languages, on which the CM relies, may be due to common inheritance, mere typological coincidence, or diffusion. Likewise, Campbell (2002:146-147) includes an informative caveat about the CM. Overall, one must independently know the history of population movements and contacts among speakers of the relevant languages in order to sort things out. The position is echoed by Laks (2002). As Thomason (2001) herself acknowledges, contact has played an important part in the histories of all languages. Some of the chapters in Aikhenwald & Dixon (including those by Calvert Watkins, Randy Lapolla, James Matisoff, and Bernd Heine & Tania Kuteva) clearly show that the Stammbaums suggested by the traditional application of the CM do not do justice to the complex ways in which languages are genetically related. (This is precisely the position developed by Laks 2002.)

      There is thus no reason for being complacent about the traditional practice of genetic linguistics. From our struggles in genetic creolistics to understand the covariance of  outcomes of language contact with patterns of social interactions, we can probably learn something on the coevolution of language and society. Genetic linguistics is far from being ahead of genetic creolistics in this respect. I can say this with confidence after discussing the issue with several genetic linguists around the world during the 2001-02 academic year and checking several historical linguistics references before writing this response.

      In the same vein, arguing that language evolution is “chaotic” (in the technical sense of being sensitive to initial and changing ecological conditions), Lightfoot (2002) argues that one cannot rely on the CM alone to establish whether syntactic structures of languages have been inherited from the same ancestor.³ If Thomason, who claims to speak on the behalf of (“good”) historical linguists, does not stipulate her dogma of intercourse-free speciation, then she should have articulated, or summarized, for us her criteria for determining the lower threshold of mixedness that disqualifies a language variety from being genetically related to other languages from which it has inherited, or with which it shares, most of its lexical, and therefore a great deal of the phonological and morphological, stock (ruling out cases of diffusion and borrowing). Hjelmslev’s (1938) so pointedly observed that all languages are mixed in some respect(s) and to some extent. In his footsteps, noting the influence of Frankish in particular, Posner (1996) remarks that if one proceeded literally by the CM, French would not count as a Romance language.

      I have chosen this example because, as Trask (1996:179) observes, the Romance languages are among the best documented cases of genetic classification, because Romanists have had the advantage of having massive documentation about Latin as the proto-language (Classical and Vulgar), which has facilitated their comparisons. (See also Campbell 2002:109.) They are ironically also one of those cases where the socio-economic history is quite well documented and favors invoking language contact as an important ecological factor. It is not by accident that anticipating Bailey & Maroldt (1977) on the development of Middle English, Schuchardt (1983), Meillet (1928), and Valkhoff (1960) argued that the Romance languages developed by “creolization,” though I disagree with their use of the term. As argued in Mufwene (2000, 2001), the concept of ‘creolization’ has no structural grounding (see also Thomason 1997) and it represents no specific restructuring process or combination thereof.⁴

      Just so that some of us will remember that historical/genetic linguists are no more unified than creolists are on a number of issues, let me note that Posner (1985) was among the first heretics to argue that creoles are dialects of their lexifiers. Trask (1996:179) follows suit in stating that there are “two more Romance languages that are spoken only in the Caribbean”: Haitian Creole and Papiamentu. I have no idea whether or not they are “good” historical linguists, but their conclusions are based on similarities in the way Latin speciated into the Romance languages and the way that these diversified into what Posner (1996) identifies as the “neo-Romance” languages of the European colonies.⁵ For the record, they too invoke mutual intelligibility as a criterion, just as do Thomason & Kaufman (1988:168-169).⁶

      An important question here too is whether a language can be related to another in a non-technical sense (see note 3) and yet not be related to it genetically. Can Kituba, Lingala, and English be claimed to be related non-technically to French, simply because they have borrowed a lot of lexical items and phrases from this language? When lay people say that two or more languages are related, do they necessarily mean something different from the technical sense of genetic relatedness as shown by the CM or any other method? Or is it simply the case that they do not have enough information to distinguish common ancestry from “language convergence” (in the sense of Sprachbund)? Assuming that Thomason (2002) is right in suggesting that mixed language varieties are obviously related to their lexifiers in the non-technical sense but not in the technical one, let me note that she has still not addressed the following critical question that I posed above: What is the threshold of mixedness that a language variety cannot cross in order to qualify as genetically related to what for all intents and purposes must be considered its ancestor or a congener? She might also enlighten us more by identifying some of those boundary indeterminacy cases, which any categorization must naturally make allowance for.

      Thomason claims that “When sources of pidgin and creole structural features can be identified with confidence, they usually come from substrate languages and/or from the operation of linguistic universals of various kinds” (2002:105). Coming from a scholar of language contact who has often argued for the role of congruence (under the label of convergence) in the selection of PCs’ structural features, this statement is surprising. In any case, the best demonstrations of the kinds of substrate influence suggested by Thomason would have been provided by relexificationists. If, in the context of French creoles, Chaudenson (1992, 2001) and Corne (1999) have not yet dissuaded her from sweepingly downplaying substantial inheritance from the lexifier, DeGraff (2001a, 2001b) has done an impeccable job of falsifying this stipulation with more elaborate evidence from French and Haitian Creole. Barring faithful inheritance (which I think is disputable even in what Thomason would like us to think of as “non-mixed” languages), one can hardly deny the legacy of the lexifier in the vast majority of cases attributed exclusively to substrate influence, especially in the TMA systems (and negation in Haitian Creole). And, since there was language shift (by Thomason’s own admission) and we can assume that speakers of non-European languages targeted colonial varieties of European languages (however heterogeneous they were), the onus of convincing evidence should weigh more heavily on arguments of substrate influence — which even Chaudenson (1992, 2001) has not ruled out — than on arguments of inheritance from the lexifier. Targeting a particular language entails identifying aspects of its structures that are different from languages that one already speaks and doing one’s best to acquire them, though second language “acquisition” is even more imperfect than first language “acquisition.”⁷ This practical approach is based on the assumption by the learner that the target language is different from what they already speak. (Otherwise there would be no point in learning it, although there is also a concurrent assumption that the target shares some features with what they already speak.) To date, no study of naturalistic second language acquisition has provided evidence of any learner that would only acquire the phonological forms of lexical items in the target language ignoring all the semantic specifications and the combinatoric constraints associated with them, thus relying exclusively on the grammars of languages they already speak. The evidence has lied mostly in the form of interference which suggests partial influence. Consequently, regarding the development of creoles’ structures, the null hypothesis should have been to assume grammatical inheritance from their heterogeneous lexifiers and to explain deviations from them by invoking substrate influence.

      Thomason adds, “Notoriously, few structural features of prototypical pidgins and creoles can be traced directly and unambiguously to the lexifier languages” (105). First of all, this conclusion depends largely on the proportion of PCs’ systems that have been investigated, notwithstanding the fact that the CM, which is presupposed by Thomason’s claim, has hardly been applied, at least not using the right, nonstandard colonial varieties of the relevant European languages. One is also reminded here of the typical dictionaries of PCs, which are really designed as supplements of those of their lexifiers and include very little of the vast lexical materials the PCs and their lexifiers share. Being incomplete documentations of the lexica of the relevant PCs, they give an exaggerated impression of lexical differences between them and their lexifiers. One must, however, bear in mind that similar dictionaries of nonstandard dialects of the same European languages (would) likewise conceal the more numerous lexical materials that they share with their standard varieties. Otherwise, Chaudenson, Corne, and DeGraff lead us precisely to a conclusion quite contrary to Thomason’s. Even Sylvain (1936), who has too often been invoked out of context to support the relexification hypothesis, leads us to their conclusion, viz., that in the particular case of Haitian Creole most of the features she discusses have diverse counterparts in nonstandard French. We should learn to overlook that infamous last sentence of her book — instigated apparently by her major professor — which seems to be the only part of it that most creolists who cite her have read (from quotations by other scholars).

      The above studies are also the closest thing we can invoke about any attempts to apply the CM to PCs, though it is possible that Thomason has applied it (systematically) to some prototypical PC in a study that I do not know. There has just been an embarrassingly dominant tendency to compare structures of PCs to the standard varieties of European languages, precisely those that were not spoken in the settings where these new varieties developed. Likewise, we have also been misled by the significance of heterogeneity of the lexifier, as if the situation were unusual and unique to PCs (and other mixed languages). Like Chomsky (2000) and Mufwene (2001), Lightfoot points out that the situation is in fact also true of cases of language “transmission” not associated with PCs and that “acquisition” with modification is a selective process.

      Neither Baker (1997) nor Thomason (1997) had a compelling reason for claiming that there was no target language in settings in which creoles developed (Mufwene 2000). The alleged “fruitless”-ness of applying the CM to “mixed languages” is nothing but a foregone, a priori conclusion. When a language has inherited about 90% of its vocabulary from its nonstandard lexifier⁸ and shares so many aspects of its semantics and morphosyntax with it, would it not be time to also ask the following questions (again): Are our assumptions about “normal” language “transmission” (under conditions different from those of the development of creoles and pidgins) correct in the first place? Can any individual or population try to speak a particular language non-native to them by selecting only its lexical forms, without paying attention to the morphosyntactic patterns of their usage, and  resort only to their native languages for this aspect of communication? As Chaudenson (2001) asks, is there any documented evidence of this in any other case of naturalistic language appropriation? Let me now add the following question: How can there have been a break in transmission of the lexifier when a PC shares so much of its grammar with its lexifier (typically a colonial koiné), despite modifications that are to be expected from the learning process?⁹ One cannot ignore some basic structural similarities (not due to any universals) which obtain between the relevant European languages and the creoles that developed from them. The following features randomly stated off the top of my head as I write will suffice to illustrate this claim: 1) despite substrate influence, the phonologies of Haitian and Jamaican Creoles largely reflect how their lexifiers were spoken (e.g., the palatalized pronunciation of cat as [kyat] in Jamaican Creole is not particularly substrate, nor that of yo ‘them’ — etymologically related to today’s standard French eux — in Haitian); 2) the positions of adjectives in both creoles and in many others are more similar to what happened in their lexifiers than would have been the case if their grammars were the sole legacy of the substrate languages (this is especially significant because the category ‘Adjective’ is hardly part of the grammars of many substrate languages); 3) English creoles have definite articles where most substrate languages use a distal demonstrative (we don’t know for sure whether the use of the quantifier ‘one’ in the stead of the indefinite article is exclusive substrate influence); 4) the postposed “determiner” la in Haitian Creole does not lack affinity with a similar morpheme which has similar morphosyntax, semantics, and pragmatics in nonstandard (and even) colloquial French varieties, etc.¹⁰ An important question here that has typically been overlooked in the literature on the development of creoles is: Does “substrate influence” mean the same thing as “source of a structural feature?” To me the answer is clearly negative, especially when language “acquisition” is imperfect and linguistic systems are naturally osmotic. As often suggested by Robert Chaudenson, creoles are also precious windows into the structures of colonial varieties of the European languages from which they evolved.

      I could be picky — or pickier?— and dispute many other aspects of Thomason’s column, but I will close only with one last comment. Her statement that “historical linguists are focusing on E-language” (Thomason 2002:106) is confusing. Due to his limited interest in it, Chomsky (1986) actually characterizes this notion ambiguously, viz., 1) as a set of utterances, expressions, linguistic behaviors produced by members of a community, and 2) as what  Paul (1880) calls “Custom Language” (Weinreich et al.’s 1968 translation) — what I identify in Mufwene (2001) as “communal language” (i.e., a population of related idiolects). The former amounts not to some system but to the pool that provides the primary linguistic data that a linguist theorizes from — what Chomsky considers to be an epiphenomenon of (a) language. I am sure that no modern linguist conceives of a language as a set of utterances and that historical linguistics goes beyond using them as data (like other linguists) to infer changes in the system. Yet, although I agree that historical linguistics has focused on communal languages, I cannot resist siding with DeGraff (2001c) and Lightfoot (2002). A communal language is no more real than a species is; it is a construct. No changes occur in it that have not affected at least some of the idiolects from which it is extrapolated (Mufwene 2001). Actually, one can go as far back as Paul (1880) to also note his clarification that the goal of historical linguistics has been to use data in order to highlight changes in the systems that have generated them. The primary reason why I resisted using DeGraff’s “E-Creole” in Mufwene (2001) was to avoid the ambiguity of Chomsky’s characterization of “E-language.”

      In any case, what I see in Thomason’s column is the same thing I have observed several times before in creolistics, viz., an unjustified reluctance among many of us to accept the alternative that PCs are really as natural and normal as other “non-mixed” languages (if there are unmixed ones at all), and they have all evolved by the same kinds of restructuring (qua system reorganization) processes. I was, for instance, shocked to read in Bruyn (1996) and Plag (1997) that since many instances of “grammaticization” in the development of creoles are continuations of processes that started in their lexifiers, these are not really cases of grammaticization as documented in other languages. The following quotation from Bruyn is particularly revealing:

The fact that ben does not function as either a particle or a lexical verb in the early stages of Sranan implies that there has been no language-internal development of a lexical item into a functional one. Rather, there must have been a kind of short-cut from the English participle, which has already some grammatical value, to the function of tense marker in the Creole language. To regard such a development as grammaticalization without further qualification would be unrevealing to the extent that in this case the process must be assumed to have taken place from one language into another, such that patterns of usage in English provided the starting point for a grammatical form in Sranan or other Creoles (30-31).¹¹

      Such a claim seems to be justified primarily by a vested interest in maintaining the discontinuity (or break-in-the-transmission-of-the-lexifier) assumption which is shared by most current hypotheses of the development of creoles (Mufwene, to appear). It amounts to an appalling resistance to contrary conclusions that are clearly supported by the socio-economic history of the territories where creoles have developed. Perhaps it is consistent with Parkvall’s (2001) concern that not assuming the evolutions of creoles to be unique in kind amounts to denying that there is a group of languages called creoles and putatively putting creolistics out of business.

      There is yet every historical, non-linguistic reason for studying the same PCs as an interesting group of languages defined by the ideologies of particular conditions of trade and plantation settlement colonization of the 17th-19th centuries (see Chaudenson’s 1992, 2001 attempt to do so with his three unities of time, space, action). There are also plenty of linguistic reasons to hope that creolistics can make a contribution to general linguistics, in the same way that one can select any arbitrary set of languages and hope to make a contribution to, for instance, linguistic typology and the study of language shift. As a matter of fact, a number of things we have been learning about PCs prompt us loud and clear to re-examine several things we thought we had understood about various aspects of Language, more specifically language evolution in the context of this discussion.

References

Aikhenwald, A. Y. & Dixon, R.M.W. (Eds.). (2001). Areal diffusion and genetic inheritance: Case studies in language change. Oxford: Oxford University Press.

Bailey, C.-J. N. &  Maroldt, K. (1977). The French lineage of English. In J. M. Meisel (Ed.), Pidgins - creoles - languages in contact (pp. 21-53). Tübingen: Narr.

Baker, P. (1995a). Some developmental inferences from the historical studies of pidgins et creoles. In J. Arends (Ed), The early stages of creolization (pp. 1-24). Amsterdam: John Benjamins.

Baker, P. (1995b). Motivation in creole genesis. In P. Baker (Ed.), From contact to creole and beyond (pp. 3-15). London: University of Westminster Press.

Baker, P. (1997). Directionality in pidginization and creolization. In A. K. Spears & D. Winford (Eds.), The structure and status of pidgins and creoles (pp. 91-109). Amsterdam: John Benjamins.

Bolton, K. (2000). Language and hybridization: Pidgin tales from the China coast. Interventions, 5, 35-52.

Bolton, K. (2002). Chinese Englishes: From Canton jargon to global English. World Englishes, 21, 181-199.

Bruyn, A. (1996). On identifying instances of grammaticalization in creole languages. In P. Baker & A. Syea (Eds.), Changing meanings, changing functions: Papers relating to grammaticalization in language contact (pp. 29-46). London: University of Westminster Press.

Campbell, L. (2002). Historical linguistics: An introduction. Cambridge: MIT Press.

Chaudenson, R. (1992). Des îles, des hommes, des langues: essais sur la créolisation linguistique et culturelle. Paris: L'Harmattan.

Chaudenson, R. (2001). Creolization of language and culture. London: Routledge.

Chomsky, N. (1986). Knowledge of language: Its nature, origin, and use. New York: Praeger.

Chomsky, N. (2000). New horizons in the study of language and mind. Cambridge: Cambridge University Press.

Corne, C. (1999). From French to Creole: The development of new vernaculars in the French colonial world. London: University of Westminster Press.

DeGraff, M. (Ed.). (1999). Language creation and language change: Creolization, diachrony, and development. Cambridge, MA: MIT Press.

DeGraff, M. (2001a). On the origin of creoles: A Cartesian critique of neo-Darwinian linguistics. Linguistic Typology, 5, 213-310.

DeGraff, M. (2001b). Morphology in creole genesis: Linguistics and ideology. In M. Kenstowicz (Ed.), Kenneth Hale: A life in language (pp. 53-121). Cambridge: MIT Press.

DeGraff, M. (2001c). Against creole genesis as abnormal transmission. Paper presented at the SPCL section of the Annual Meeting of the Linguistic Society of America, Washington, DC.

Hagge, C. (1993). The language builder: An essay on the human signature in linguistic morphogenesis. Amsterdam: John Benjamins.

Hjelmslev, L. (1938). Etudes sur la notion de parenté linguistique. Revue des Etudes Indo-Européennes, 1, 271-286.

Joseph, B. D. (2001). Is there such a thing as “grammaticalization?” Language Sciences, 23, 163-186.

Laks, B. (2000). Le comparatisme: de la généalogie  la génétique. Langages, 146, 19-45.

Lightfoot, D. W. (2002). Myths and the prehistory of grammar. Journal of Linguistics, 38, 113-136.

McWhorter, J. H. (1998). Identifying the creole prototype: Vindicating a typological class. Language, 74, 788-818.

Meillet, A. (1928). Esquisse d’une histoire de la langue latine. Paris.

Meillet, A. (1929). Le développement des langues. In Continu et discontinu, 119ff. Paris: Bloud & Gay. Reprinted in A. Meillet. 1951. Linguistique historique et linguistique générale. Vol. 2, pp.71-83. Paris: Klincksieck.

Mufwene, S. S. (1989). Some explanations that strike me as incomplete. Column.  Journal of Pidgin and Creole Languages, 4, 117-28.

Mufwene, S. S. (1998). What research on creole genesis can contribute to historical linguistics. In M. Schmid, J. Austin, and D. Stein (Eds.), Historical Linguistics 1997 (pp. 315-338). Amsterdam: John Benjamins.

Mufwene, S. S. (2000). Creolization is a social, not a structural, process. In Ingrid Neumann-Holzschuh and Edgar Schneider (Eds.), Degrees of restructuring in creole languages (pp. 65-84). Amsterdam: John Benjamins.

Mufwene, S. S. (2001). The ecology of language evolution. Cambridge: Cambridge University Press.

Mufwene, S. S. (to appear).  Grammaticization is part of the development of creoles. In Ö. Dahl & V. Velupillai (Eds.), Perspectives on grammaticalization

Owens, J. (1998). Representativeness in the data base: Polemical update for the twenty-first century. Language Sciences, 20, 113-135.

Parkvall, M. (2001). Creolistics and the quest for creoleness: A reply to Claire Lefebvre. Journal of Pidgin and Creole Languages, 16, 147-151.

Paul, H. (1880). Prinzipien der Sprachgeschichte. Halle: Niemeyer.

Plag, I. (1999). Review of Baker & Syea (1996). Journal of Pidgin and Creole Languages, 14, 202-208.

Posner, R. (1985). Creolization as typological change: Some examples from Romance syntax. Diachronica, 2, 167-88.

Posner, R. (1996). The Romance languages. Cambridge: Cambridge University Press.

Schuchardt, H. (1883). Review of Lucien Adam’s Les idiomes négro-aryen et malayo-aryen. Literaturblatt für germanische und romanische Philologie, 4, 236-240.

Schuchardt, H. (1884). Slavo-deutsches und Slavo-italienisches. Graz: Leuschner & Lubensky.

Schuchardt, H. (1924). Hugo Schuchardt — Brevier. Ein Vademekum der allgemeinen Sprachwissenschaft, ed. by Spitzer Léo. Halle: Max Niemeyer.

Sylvain, S. (1936). Le créole haitien: morphologie et syntaxe. Wettern, Belgium: Imprimerie De Meester.

Thomason, S. G. (1997). A typology of contact languages. In A. K. Spears & D. Winford (Eds.), The structure and status of pidgins and creoles (pp. 71-88). Amsterdam: John Benjamins.

Thomason, S. G. (2001). Language contact: An introduction. Washington, D.C.: Georgetown University Press.

Thomason, S. G. (2002). Creoles and genetic relationship. Column. Journal of Pidgin and Creole Languages, 17, 101-109.

Thomason, S. G. & Kaufman, T. (1988). Language contact, creolization, and genetic linguistics. Berkeley: University of California Press.

Trask, R.L. (1996). Historical linguistics. London: Arnold.

Valkhoff, M. (1960). Contributions to the study of creoles. African Studies, 19, 230-244.

Weinreich, U.; Labov, W.; & Herzog, M. I. (1968). Empirical foundations for a theory of language change. In W. P. Lehman & Y. Malkiel (Eds.), Directions for historical linguistics: A symposium (pp. 97-195). Austin: University of Texas Press.