Contact Linguistics. Chap

Introduction to Contact linguistics. Chapter 9. Creole formation.

“The human mind is the same in every clime; and accordingly we find nearly the same process adopted in the formation of language in every country. The Negroes have been proved to be in no degree inferior to other nations in solidity of judgement, or fertility of imagination.” [Greenfield 1830:51 cited in Holm 1988 page 22]

1. Introduction.

No other contact languages have generated as much debate, disagreement and controversy as those traditionally referred to as “creoles.” They emerged in colonial settings in the New World, the Indian Ocean and West Africa, where European powers subjected African, Asian and other populations to their rule in the course of the 15^th through 19^th centuries. They were created by slaves and other subordinated groups who fashioned materials from the colonial languages and their own mother tongues into new media of communication. Eventually these languages came to function as community vernaculars acquired by children as first languages.

Their birth among the oppressed and enslaved, and their continued association with such disadvantaged groups, explain the low esteem in which most laymen hold them. From the earliest periods of their emergence, they provoked the scorn and derision of European who encountered them on their travels to the colonies. For example, Mrs. Carmichael had this to say about the creole language spoken in St. Vincent in the 19^th century (Carmichael 1833:Vol. 1, 5):

“I could comprehend little or nothing of what they said; for though it was English, it was so uncouth a jargon that to one unaccustomed to hear it, it was almost unintelligent as if they had spoken in any of their native tongues.”

Even today, most laymen continue to condemn these languages as vulgar and uncouth corruptions of standard languages. Their use has been subject to disapproval and even punishment from parents and other authority figures, especially in the schools. Even some scholars of language once held the view that creoles were unworthy of serious scientific study, and until recently, these languages remained “the unwanted stepchildren of linguistic science” (Lounsbury 1968:205).

But linguists have come to view creoles as exciting creations that challenge the conventional wisdom about language change and mixture, and offer fascinating insights into the workings of the human language faculty. Scientific study of these languages dates back to the 19^th century, in particular the work of Van Name (1871), Coelho (1880 – 1886), Schuchardt (1882, 1883, etc.), Hesseling (1897) and others. These early studies explored many of the issues still being hotly debated today, such as the role of substrate influence vs universals in creole formation, the relationship between creoles and first or second language acquisition, and the implications of these languages for theories of language change.
But before we examine these issues in more detail, we must first clarify what we mean by ‘creoles’, and whether there are clear criteria for identifying them as a distinct class of languages.


2. Defining creoles.

The speakers of the languages that linguists call creoles do not all use this term to refer to their form of speech. Many of these languages have names of their own, for instance Patwa (English-lexicon) in Jamaica, Papiamentu (Iberian-lexicon) in the Dutch Antilles, Sranan Tongo “Suriname Tongue” (English-lexicon) in Suriname, Papia Kristang (Portuguese-lexicon) in Malacca and Tayo (French-lexicon) in New Caledonia. Sometimes the popular names match those used by linguists, for instance ‘Kweyòl’ (French) in St. Lucia, ‘Kreyòl’ (French) in Haiti, ‘Creolese’ (English) in Guyana, ‘Krio’ (English) in Sierra Leone, and ‘Creole’ (English) in Belize.

How then did the term ‘creole’ become a cover label for such different outcomes of contact? The story behind this is quite similar to what happened in the case of ‘pidgins.’ The term ‘creole’ was used long before linguists adopted it as a technical label. In fact, it was first applied, not to languages, but to people who were born in the colonies, first Europeans, then persons of mixed descent, and eventually all locally born people. The earliest known attestation of the term is the Spanish word criollo, which appeared in a book published in 1590, and translated into English in 1604 (Coromines 1967:171, cited in Holm 1988:15). The English citation reads: “Some Crollos [sic], for so they call the Spaniards borne at the Indies.” The term was adopted into French as créole, and then into English.
Not surprisingly, the term ‘creole’ and its cognates came to refer to the languages spoken by persons born in the colonies – “the creole(s’) language(s).” Used in this sense, it referred to a fairly diverse set of contact vernaculars. Some of these were quite similar in structure to the dialects introduced by European settlers. Others were more divergent from these source languages, drawing significantly more of their resources from substrate languages. It is partly for this reason that there is no current consensus on how best to define creoles. As DeCamp (1977:3) noted, labels like ‘creole’ and ‘pidgin’ are not comparable to those that classify other groups of languages on the basis of genetic relationship (e.g., ‘Romance’ languages), geographical location (e.g. the Balkan languages), or shared typology (e.g., tone or serializing languages). Of course, all of these classifications are problematic in one way or another. But the problem is especially vexing in the case of creoles.
As with pidgins, the identification of these contact vernaculars is based on a variety of often conflicting criteria, including their putative origins, their communicative functions and their structural characteristics. According to the first two criteria, creoles have traditionally been defined as pidgins that were adopted as native languages (‘nativized’) by newly emergent creole societies. To serve this function, the erstwhile pidgin had to be elaborated in lexicon and grammar. This process of structural expansion, it is argued, involved the ‘restructuring’ or re-constitution of a viable grammar more complex and efficient than that of the pidgin ancestor. The nature of this restructuring will be discussed further below.
There is continuing disagreement concerning the true impetus for this process of elaboration. The ‘nativization’ scenario implies that pidgins are elaborated into creoles by children who adopt the former as a first language. However, there are various situations in West Africa (Nigerian and Ghanaian Pidgin English), Central African (Sango) and Papua New Guinea (Tok Pisin) where erstwhile ‘pidgins’ have been nativized without significant structural change. The ‘nativization’ account also implies that the process of creole formation is rather quick or “abrupt.” Yet recent research on the early stages of creole formation in Suriname and elsewhere suggests that creole grammar emerged over several decades, and that adults as well as children played crucial roles in their development.
These considerations would imply that there is little distinction between creoles and ‘expanded pidgins’ in terms of the nature and speed of their development (See Chapter 8). Both involve similar processes of restructuring over time, and there are no structural criteria that can distinguish one from the other. This is particularly true in cases where the same contact vernacular is used as both a first and a second language by different groups within the same community. (See cases like Tok Pisin and Sango, for example.) Hence some have attempted to distinguish creoles from pidgins on purely functional grounds. For instance, Gilman (1979:274) argues that creoles, unlike pidgins, are languages of reference for an ethnic group. While this may be true, it would once more assign different statuses to varieties that are structurally and developmentally quite similar, for instance Krio and varieties of West African Pidgin English. Similar objections can be made to those definitions that equate “creolization” with “vernacularization” of a pidgin (Mufwene 1997, Pasch 1997).
Much of the confusion over how best to define creoles is due to indeterminacy in the definition of the “pidgins” from which they are claimed to have arisen. Some have tried to clear up this confusion by specifying that creoles are elaborations of “prototypical” pidgins that have become nativized. But such a definition can apply only to a certain subset of creoles – the so-called “radical” creoles like those of Suriname which display extreme divergence from their lexifier languages. And to complicate things further, we have no clear evidence that many creoles (including some “radical” ones) emerged directly from earlier pidgins. The reality is that creoles constitute a motley assortment of contact vernaculars with different histories and lines of development, though of course they still share much in common.
This diversity in the origins and evolution of creoles explains why they have resisted classification on structural or typological grounds as well. In the first place, there are no structural characteristics that all creoles share. And second, there are no structural criteria that can distinguish creoles from other types of language. Attempts to identify a list of such structural diagnostics of creole status have generally failed. This is true even of efforts that focus only on “radical” creoles.
For instance, Bickerton (1981) identified a set of twelve features which he claimed were diagnostic of “prototypical” creoles. These included the use of ‘adjectives’ as verbs; a tripartite opposition in the TMA system between Anterior tense, Nonpunctual aspect and Irrealis mood; a distinct locative copula, and so on. Most of these are discussed in Section 6 below. Bickerton cited Hawaiian Creole English, Sranan Tongo, Jamaican Creole, (rural) Guyanese creole and Haitian Kreyòl as paradigm examples of the creole prototype. But it turns out that none of these creoles has all the definitive characteristics, and all of them exhibit particular structural features not found in the others. Moreover, even a casual overview of other creoles reveals significant differences among them.
A more recent attempt to “vindicate” creoles as a typological class (McWhorter 1998) claimed that there were three structural features that “prototypical” creoles share. But, as Mufwene (2000:67) points out, the criteria in question reduce the set of prototypical creoles to a mere handful. More importantly, each of these “prototypical” creoles lacks one or another of the putative typological features.
This is not to deny that there are also striking similarities across creoles, particularly in the areas mentioned by Bickerton, McWhorter. Some of them can be explained as the result of similar kinds of superstrate and/or substrate influence, or similar processes of simplification and grammaticalization (see section 8 below). Others can be explained as the result of diffusion, as creole speakers migrated from one area to another. The similarities among Western Caribbean creoles like those in Jamaica, Belize and Providence Island are clearly due to such migration. The same is true of the similarities between Bajan, Trinidadian and urban Guyanese, the latter two being largely modeled on the first (Winford 1997a). The diffusion of English-lexicon and French-lexicon creoles throughout the Atlantic area has also been discussed in Huber & Parkvall (1999) and Corne (1999) respectively.
Despite the problems of definition we have pointed to so far, there is still justification for treating creoles as a separate and identifiable class of languages. The grounds for doing so are primarily sociohistorical in nature, as Thomason (1997), Mufwene (2000) and others have argued. From this perspective, creoles are simply contact languages that emerged primarily in plantation settings in various European colonies throughout the world. Such settings shared a number of socio-political and demographic characteristics, including the use of large numbers of slaves who were transplanted from their homelands and placed under the control of a small minority of Europeans. The details of the settlement history and social organization of these plantation societies will be discussed below. Suffice it to say for now that differences in the social settings of each colony led to diversification in the outcomes of the contact between Europeans and the oppressed groups.
In some cases, for instance in most of the Spanish colonies as well as in Brazil and the southern United States, social conditions promoted the acquisition of the language of the dominant group. These colonial dialects of Spanish, Portuguese, English and other languages are not usually referred to as creoles. However, some scholars (e.g. Mufwene 1997) have advocated their inclusion in that category, on the basis that they shared a similar sociohistorical development. In other colonies, the combination of social factors and linguistic inputs favored the emergence of contact vernaculars that were highly divergent from their respective European lexifier languages. The most divergent of these constitute the class of “radical” creoles. Other creoles occupy various mid points in the continuum between colonial dialect and radical creole.
To explain the diversity in the outcomes of creole formation, we must explore the socio-historical background to their emergence, and the social contexts in which they arose.


3. The sociohistorical background to creole formation.


3.1. Portuguese colonization:

European colonial expansion from the 15^th through 19^th centuries brought (speakers of) European languages into contact with a multitude of indigenous languages in the colonized territories. Portugal was the first European power to engage in the quest for a colonial empire, providing in many ways a model to be imitated by other European powers in succeeding centuries.

Portuguese exploration of the North African coast began in the early 1400’s, at first motivated primarily by pursuit of military action against the Muslims they had succeeded in driving from the Iberian Peninsula. In the course of the 15^th century, the Portuguese traveled south along the West African coast, from Morocco (1415) to Cape Verde (1444) and eventually down to Angola and the Cape of Good Hope (1488). They soon realized that there were lucrative opportunities for trade with the various nations they encountered, in commodities like ivory, gold and spices, and later, slaves. They established settlements on islands like Madeira and the Azores in the early 15^th century and later in the Cape Verde islands and the Gulf of Guinea islands, Sao Tomé, Principe and Annobón (1470’s). At first, they engaged in small farming and most of the labor was done by Portuguese peasants. Soon, however, they began importing slaves in ever increasing numbers to cultivate crops and raise livestock in the Cape Verde islands, and to meet the demand for labor on the sugar plantations they established in the Gulf of Guinea islands.
This kind of plantation economy was eventually carried over the Atlantic to Brazil, “discovered” by Cabral in 1500. It became the model for the plantation colonies established by other European powers in the New World and elsewhere.
The trading contacts between the Portuguese and West Africans led to the emergence of a Portuguese-lexicon pidgin that was used extensively along the West African coast from the 15^th through 18^th centuries. This pidgin may have provided a partial model for pidginized forms of versions of other European languages that emerged later on the West African coast. The sociolinguistic situation in the islands where plantations had been established created the right conditions for the earliest known creoles to arise. Such languages – Cape Verdian, Principense, and Annobonese – survive today as the primary vernaculars of these communities. By contrast, the contact situations on the West African mainland did not produce Portuguese-lexicon creoles. The one exception is the creole spoken in Guinea-Bissau, which may have been transported there from the Cape Verde islands (Holm 1989:254).
Portuguese colonial expansion continued beyond Africa to South and Southeast Asia after Vasco da Gama rounded the Cape of Good Hope in 1488, and reached India in 1498. The Portuguese established a vast trading empire from India to Japan, but eventually withdrew from most of Asia. By this time, creole varieties of Portuguese had arisen in several of these Southern Asian settlements. Among those still spoken today are Papia Kristang of Malacca and Daman Creole of Northwest India.
Some scholars have claimed that a Portuguese pidgin was the ancestor of Ternateño, a Spanish-lexicon creole that emerged on the Phillipine island of Ternate in the 16^th century. This creole was later transported to three other locations in the Phillipines: Zamboanga in the southern island of Mindanao, and the towns of Cavite and Ternate on the northern island of Luzon. Scholars remain divided on whether these creoles emerged directly from contact with Spanish, or from an earlier Portuguese pidgin whose lexicon was replaced by Spanish items.
From the 17^th century on, other European powers, Dutch, French and English followed in the wake of the Portuguese, establishing trading contacts with West Africa, particularly in slaves. These supplied labor for the large-scale plantation economies that all these powers (as well as Spain) introduced to their colonies in the Caribbean, North and South America, the Indian Ocean, and parts of the Pacific. Like the Portuguese plantation islands off the West African coast, these New World settings were favorable to the emergence of creoles. To explore the details of these settings, let us turn our attention to the Caribbean colonies.


3.2. The social contexts of creole formation.

The plantation economies established in the Caribbean and elsewhere in the New World and Indian Ocean were based on the forced labor of large numbers of slaves transplanted from Africa and controlled by a small number of Europeans. In general colonization in these areas involved the re-peopling of lands made empty by the extermination of indigenous Caribbean peoples. Hence the newcomers, both European and African, came together in a social, cultural and linguistic vacuum, into which each group introduced some part of its cultural traditions (Mintz 1971:484).

Mintz (op. cit. 481) identifies three general social conditions that influenced the emergence and character of creole languages in the New World:

43. 
    The demographics of each colony, that is, the relative numbers of Europeans, Africans and other groups present in each colony over time.
    
44. 
    The nature of the contact among these groups, and in particular, “the codes of social interaction governing the relative statuses and the relationships of these differing groups in particular societies.”
    
45. 
    The types of community settings within which the groups mixed and interacted.
    

There has been a tendency among some scholars to focus primarily on demographics (population ratios) in attempts to explain creole origins. For instance, Bickerton (1981:4) suggested that creoles typically arose in settings where the dominant group made up no more than 20% of the population. Expanding on this, Baker (1982, 1993) proposed that a major factor was the timing of three crucial developments in the population make-up, which he termed Events 1, 2 and 3. Event 1 was the time when the number of slaves exceeded that of the slave-owners. Event 2 occurred when the number of locally born slaves exceeded that of non-slaves, both locally and foreign born. Event 3 was when the importation of slaves ceased. In this scenario, the period between Events 2 and 3 is most favorable to the emergence of a creole, provided all other circumstances are right (Baker 1993: 138). Bickerton (1984), drawing on Baker’s account, proposed a “pidginization index” which could measure the degree to which a creole diverged from its lexifier language, approximating the grammar dictated by the putative bioprogram. However, Singler (1990a, 1993, 1995) has demonstrated that this approach makes the wrong predictions about creole development.

It is clear that the analysis of population ratios must be complemented by in-depth investigation of the changing contexts and patterns of interaction among the groups concerned. For instance, there were different ratios between Europeans and slaves in different contact settings within the same colony. In the American South, for example, the high ratios of Africans to English settlers on the Carolina coast contrasted with the higher proportions of Europeans elsewhere in the southern colony. The disproportion in the former setting was a key factor in the formation of Gullah, the only English-lexicon creole of N. America. The latter kind of contact led to the emergence of varieties of African American English that were much closer to their English dialectal sources.
Also relevant was the role played by indentured servants brought from Europe to work alongside the slaves. In Barbados, for instance, the high ratios of such workers and settlers from South West England in the first 40 years of settlement led to a contact variety, Bajan, that was closely modeled on those settler dialects. Another crucial demographic factor was the degree of natural increase vs mortality in the plantation colonies. Most of these were characterized by low rates of natural increase and high rates of (especially infant) mortality. This resulted in not only a small number of locally born children but also a continuing importation of new slaves. Suriname represents perhaps the extreme case of this scenario. Contrasting with it were colonies like those in Barbados and the Southern United States, where natural increase was more common. Singler (1992) has argued that the more locally born children there were (especially in the earlier period of settlement), the more the likelihood that the emerging contact language would approximate its lexifier. The contrasting cases of Barbados and Suriname seem to bear this out.
3.3. Community settings and codes of interaction.

The nature of the community settings differed in the same colony over time, as well as from colony to colony. As Singler (1993) has noted, the type of economic activity – crop selection in particular – made for significant differences in patterns of contact. In most cases, the earlier periods of settlement were characterized by small-farm of homestead economies, devoted to cultivation of crops like tobacco. In such settings, there were more equal if not higher proportions of Europeans to Africans, and closer interaction between the two groups. The longer such settings survived, the more likely they were to produce creole varieties closer to the settler dialects. This occurred in colonies like (French) Reunion, and, not surprisingly, Barbados. In both cases, there was a fairly long period of close contact between Europeans and non-Europeans in homestead settings. The second language varieties of the superstrate which emerged in that period prevailed, even when slave importation increased dramatically in later years. By then, they had already become community vernaculars.¹⁷

By contrast, the early introduction of large-scale plantation economies to other colonies made conditions more favorable to the emergence of more radical creoles. The shift to sugar cultivation and increasing slave importation meant less opportunity for the latter to acquire closer approximations to European languages. In addition, the linguistic heterogeneity of the slave groups and their numerical dominance and limited contact with superstrate speakers contributed to the creation of a contact variety that diverged significantly from its lexifier source. This is why Mauritius, colonized by the French in 1721, presents such a sharp contrast to Reunion, settled some 58 years earlier. The MIC that developed in the former colony was and is far more divergent from French, reflecting greater influence from Malagasy and other substrate languages, as well as from internal restructuring.
Various other factors played a role in determining patterns of interaction between Europeans and others. Among these were the codes of social relations regulating such interaction. In general, as Mintz (1971:487) notes, the colonies differed in three respects. First, different European powers exercised different degrees of control over the affairs of their colonies. In general, Spain kept stricter watch than either France or England.
Second, there were differences in the slave codes in each colony. These regulated whether slaves could participate in institutions like the church, the speed with which they attained freedom, and so on. In the Spanish colonies, for example, the rate of manumission was in general rapid and continuous, compared with that in the French and English colonies. The class of freed slaves had closer contact with Europeans, and thus acquired the latters’ language. The codes also determined the sexual and mating practices between the groups. In colonies like Reunion, European male settlers often had non-European spouses. Hence there was a significant number of locally born, free persons of mixed race in the earlier stages of settlement. The emergence of second language varieties of the superstrate in such colonies seemd to be directly related to the linking function of the mixed-race and free groups in such colonies (Hoetink 1967:178). By contrast, in colonies like Mauritius, the strict separation of groups prevented such cohabitation. This helps in part to explain the more divergent linguistic outcomes in these cases (Baker 2000:53, Corne 1982).
Third, there were differences in the ideologies of the dominant classes regarding each colony’s relationship to the pertinent European power. Hence the emergence of a distinct “creole” identity varied from colony to colony. For instance, Mintz (1971: 487) suggests that Spanish colonists identified more fully with their new homes in the Caribbean than either the French or the English. The different political and social climate in the Spanish colonies may explain the paucity of Spanish-lexicon creoles in the New World (but see McWhorter ?? for a different explanation. All of the factors discussed here can help us to understand why different creole cultures emerged in different colonies, with varying degrees of acculturation toward or divergence from European cultural and linguistic practice.
Finally, in each colony, there were differences in privilege and status among different categories of slaves. Domestic slaves had more contact with Europeans and presumably learnt second language varieties of the latters’ languages. Skilled slaves had more freedom of movement than those who labored in the field. In many cases, such differences led to linguistic continua ranging from second language varieties of the superstrate to highly divergent creoles.
Similarly, the creoles that arose in different colonies form a kind of continuum with more “radical” creoles such as those in Suriname furthest removed from their lexifiers, and “intermediate” creoles like Bajan much closer. Most of the English-lexicon creoles of the Caribbean – the so-called “basilectal” varieties of Guyanese Creole, Jamaican Creole etc., fall somewhere in the middle of this continuum.
To sum up, each mix of ecological factors constituted a recipe for different linguistic consequences of contact (Mufwene 2001). As Hymes (1971b:83) put it:

What would be constant would be the fundamental terms, or variables, and variables they would be, such that differences in their values would result in different outcomes.