Summary and Keywords
Creole languages have a curious status in linguistics, and at the same time they often have very low prestige in the societies in which they are spoken. These two facts may be related, in part because they circle around notions such as “derived from” or “simplified” instead of “original.” Rather than simply taking the notion of “creole” as a given and trying to account for its properties and origin, this essay tries to explore the ways scholars have dealt with creoles. This involves, in particular, trying to see whether we can define “creoles” as a meaningful class of languages. There is a canonical list of languages that most specialists would not hesitate to call creoles, but the boundaries of the list and the criteria for being listed are vague. It also becomes difficult to distinguish sharply between pidgins and creoles, and likewise the boundaries between some languages claimed to be creoles and their lexifiers are rather vague.
Several possible criteria to distinguish creoles will be discussed. Simply defining them as languages of which we know the point of birth may be a necessary, but not sufficient, criterion. Displacement is also an important criterion, necessary but not sufficient. Mixture is often characteristic of creoles, but not crucial, it is argued. Essential in any case is substantial restructuring of some lexifier language, which may take the form of morphosyntactic simplification, but it is dangerous to assume that simplification always has the same outcome. The combination of these criteria—time of genesis, displacement, mixture, restructuring—contributes to the status of a language as creole, but “creole” is far from a unified notion. There turn out to be several types of creoles, and then a whole bunch of creole-like languages, and they differ in the way these criteria are combined with respect to them.
Thus the proposal is made here to stop looking at creoles as a separate class, but take them as special cases of the general phenomenon that the way languages emerge and are used to a considerable extent determines their properties. This calls for a new, socially informed typology of languages, which will involve all kinds of different types of languages, including pidgins and creoles.
Creole languages have a curious status in linguistics, and at the same time they often have very low prestige in the societies in which they are spoken. These two facts may be related.
To begin with their social status, many creole languages are spoken by descendants of enslaved Africans or of transplanted inhabitants of islands in the Pacific, a provenance that already makes their chances of gaining high prestige slim. Most regions in which they are spoken are not very wealthy, and there often is an upper class in those regions which speaks something other than the creole. In addition, however, creole languages are often perceived as somehow “derived” or “simplified” variants of other languages. This sets creole languages apart (in a negative way) from other languages spoken by non-dominant groups, such as the indigenous languages of the New World. The latter also carry low social prestige, but are at the same time often regarded as “original” and “complex.”
When the World Atlas of Linguistic Structures (WALS) (Haspelmath, Dryer, Gil, & Comrie, 2005) was conceived, creole languages were not included. Linguistic typologists typically shied away from putting creole languages into their samples. Historical linguists do not quite know how to classify them, either; indeed Thomason and Kaufman (1988) defined creoles as languages that did not have direct phylogenetic ancestry, setting them apart from all other languages. Two groups of linguists had less difficulty treating creoles like other languages: formal linguists have generally taken the position that creole languages (but not pidgins) are like other native languages in reflecting features of Universal Grammar. Similarly, sociolinguists embrace the variation encountered in creole speaking communities. In fact, some of the terminology from creole studies, such as the Acrolect/Mesolect/Basilect distinction, made its way into sociolinguistics as well as second language acquisition.
The main question raised here is: how do we define a creole language? Wikipedia gives the following definition online: “A creole language, or simply creole, is a stable natural language that has developed from a pidgin (i.e., a simplified language or simplified mixture of languages used by non-native speakers), becoming nativized by children as their first language, with the accompanying effect of a fully developed vocabulary and system of grammar.” This looks like a fairly clear definition, but it crucially relies on another notion, pidgin, and creole language specialists do not agree on the fact that all creoles emerged out of a pidgin (the so-called pidgin-creole cycle), although some creoles certainly have. This is not to criticize the otherwise interesting and balanced Wikipedia entry, but rather to point to the fact that it is hard to define a creole exactly. Perhaps it is easiest to start with some examples of languages generally acknowledged to be creoles.
The most comprehensive survey of pidgins and creoles, The Atlas of Pidgin and Creole Structures (ApiCS) (Michaelis, Maurer, Haspelmath, & Huber, 2013a) contains information about 76 languages that are included as pidgin and creole languages by the editors. The list is organized in terms of the lexifier language, the language that provides the largest part of the lexicon of a pidgin or creole and that played a role as the socially dominant language in the process of original creation. The Atlas (in the order presented) contains 26 English-lexifier pidgins and creoles,3with Dutch as the main lexifier language, 14 with Portuguese, 6 with Spanish, 9 with French, 3 with Bantu, 2 with Arabic, 3 with Malay, and 10 other. Table 1 provides some examples of these languages:
Table 1. Selected List of Languages Generally Recognized as Creoles
Sranan or Sranantongo
Berbice Dutch Creole
Sri Lanka Portuguese
Principe, West Africa
Curaçao, Aruba, Bonaire, Caribbean
Seychelles, Indian Ocean
Central African Republic
Many of these languages are spoken in former plantation colonies and are associated with the slave trade. There are strong connections between the notion of “creole” and European colonial expansion, although some pidgins and creoles are not directly linked to this expansion.
A much longer list is possible (with other, often smaller and less well documented languages included), but this one will suffice to start approaching the basic question: how does a language enter the club of creoles? What are creole languages? While the status of the above languages as creoles is uncontroversial, it is much harder to define what a creole language is.
The term “creole” was used in a non-scholarly way in the 18th century to refer to specific Caribbean creole languages (of course, it was also a colonial term used to refer to people, generally to people born in the colonies in the New World). It was in the second half of the 19th century that historical linguists started comparing different creoles, and the idea of a class of languages arose. Then, gradually a list of languages with the label “creole” started being drawn up, and criteria were being formulated for being entered on the list, by scholars such as Hugo Schuchardt. Coastal Colombian Spanish, for instance, was not put on the list, after some correspondence with local experts, because it was not different enough from general South American Spanish. By the 1930s, there was enough consensus for John E. Reinecke, based in Honolulu, to start systematically compiling sources on different pidgin and creole languages, an effort that resulted in his pioneering 1938 essay about marginal languages and, ultimately, in a massive 1975 bibliography (Reinecke, DeCamp, Hancock, & Tsuzaki, 1975). Later efforts to list all creoles include Holm (1988b); the long list in Smith (1994) also attempts to enumerate all known mixed languages. The point is that the notion of creole is not a given, but a scholarly construct that developed over the last century, and we should attempt to find out on the basis of which criteria this construct is put together.
A first problem concerns the boundary between pidgin and creole. Theoretically, the line was always assumed to be straightforward: pidgins are languages of intercultural communication without native speakers, and creoles are languages of a specific community with native speakers. However, in actual fact things are not so simple. A pidgin may not have native speakers (having acquired it as their first language) in the strict sense, but may have become an important second language in a multilingual community, learned early on by children in the yard or the street. Furthermore, an erstwhile creole may become repidginized in a process of ethnic redefinition, where speakers shift to a local ethnic language as their first language.
To illustrate this problem, consider a series of varieties, often referred to as Pidgin English, listed in APiCs for West Africa. Of course, situations differ, with Ghana showing the least creole-like use, but the terminology used in the descriptions of these language also indicates that there is a range of nuances between native and non-native.
Table 2. Characterization of West African Pidgins-Creoles in APiCS
Terms used in description of acquisition and use
Used natively/used as a lingua franca
Ghanaian Pidgin English
Functional domain more restricted
The first language/one of the first languages/second language/main language in day to day communication
Cameroon Pidgin English
Range from L1 speakers and fluent L2 speakers to people who Use it only sporadically as a rudimentary and ad hoc means of communication in trade situations
Home or vernacular language/often with a Bubi cultural background, who have adopted Pichi as a primary language
For Portuguese, lexifier Casamancese Creole, also in West Africa, Biagui and Quint (2013) report that the number of second language speakers is much greater than that of mother tongue speakers. “Casamancese Creole is the mother tongue of about 10,000 people. . . . It is also spoken with various degrees of proficiency by several tens of thousands of second language speakers living in the same region.” There are also pidgins that clearly do not serve as any one’s native, home, first, or vernacular language, such as Ndyuka-Trio Pidgin in Surinam (Huttar & Velantie, 1997). Notice that one of the contributors to this pidgin is itself a creole (Ndyuka in Surinam).
A second problem is to define exactly what counts as a separate language. Notice many Creoles have a name or label that is associated with their main lexifier, as in the case of Berbice Dutch Creole in Table 1. How different does the variety have to be from its lexifier to count as separate language, in this case as a creole? Thus the Brazilian Portuguese spoken in some rural communities in Brazil, by people of African descent, has been argued to be a creole language by some, while others have claimed it simply as a variety of Portuguese. The same may be said of Singapore English, Bahamian English, African American Vernacular English. In any case, the band width of variation within languages such as English and Portuguese is vast, and there is also considerable variation within the varieties involved for which a creole status has been claimed. Some scholars have introduced terms like creoloid to deal with these situations, but it is not clear if that really helps. Defining what a separate language is or is not, however, is a problem shared by all researchers proposing a definition of creole.
Furthermore, political circumstances may lead to a variety assuming language status, such as the case of Afrikaans in South Africa, which was declared a separate language in the early 20th century. Thus in Afrikaans, there is a standard variety, clearly different from Dutch, but with some mutual intelligibility and considerable shared lexicon, as well as a range of other varieties with different lexicons and grammatical features, and a pronunciation making mutual comprehension with Dutch totally impossible. In the case of Afrikaans, there are good grammatical reasons as well to consider it a separate language, but for a long time it was simply treated as a variety of Dutch. For a history of the political emancipation of Afrikaans, see Deumert (2004). In other cases, however, local varieties were simply treated as bad forms of some colonial standard and did not make it as separate languages.
4. Time of Genesis
One possible way to define creoles would be as languages of which we can pinpoint the approximate time of birth, in contrast with languages that have very gradually split off from a mother language, and for which we cannot point to a particular time in history for their genesis. We can be fairly certain that Sranantongo came into being in Surinam, as a language used by enslaved Africans on the sugar plantations, somewhere between 1650 and 1680. It would be much harder to state that its lexifier language, English, emerged at date “X,” which would exclude it from the list unless we take the moment of genesis of English to be, for example, the transition between Old English and Middle English.
This definition has one big problem associated with it, which may lead to borderline cases. How do we define what “a particular time” is? In the case of Sranantongo, we have a very specific time frame because the colony was transferred from the English to the Dutch colonizers, and since the dominant lexicon is from English, we know it must have arisen during English colonization. However, in colonies where the same power was present throughout the history of plantation slavery, the potential time window of emergence of the creole spans several centuries. This makes it difficult to distinguish between the case of some Caribbean creoles and, say, the emergence of French, which we could also date to the span of a few centuries in the late Roman period.
The case of French is of interest because it makes us raise the question why we do not call French a creole. French emerged due to the contact of Celtic Latin speakers with Frankish Germanic speakers in the northern part of France. In spite of the many changes, the language did not undergo the type of drastic restructuring one typically associates with a creole, it has been argued. Nonetheless, it is not clear whether the transition from Latin to French was really unbroken. Why is French not more like Spanish and Italian, for instance? Mufwene (2002) claims, without elaborating this argument in structural terms, that the genesis of some Romance languages perhaps should be studied as if they were creoles. Even if we can exclude the Romance languages from the discussion, knowing roughly when they emerged, then, is perhaps a necessary component of a definition of creoles, but not sufficient as a definition. As a reviewer points out, there may be other types of newly created languages for which we can date their genesis fairly precisely, while these are not creoles.
5. Restructuring and Simplicity
Somehow, somewhere, we need to take restructuring into account. Restructuring is a complex notion in itself and is linked to the controversial term of simplification. In all cases where we compare a creole language to its lexifier, we would be very uncomfortable as linguists if the two greatly resembled each other structurally. This can be illustrated with the example of Papiamentu, the creole language of the Caribbean islands of Curaçao, Aruba, and Bonaire (which form part of the Kingdom of the Netherlands), with its lexical roots predominantly in Portuguese and Spanish. Consider a typical Papiamentu sentence such as (adapted from Maurer, 1988, p. 263):
‘She sits in [the] window [and] sees how they are going [away] with her man/husband.’
To take Spanish as the reference point for the sake of the argument, there are a number of ways in which Papiamentu differs from it, while all words in the sentence given (except for nan [3.pl]) are from Portuguese or Spanish. A first set of differences is presented in Table 3.
Table 3. Some Grammatical Differences Between Papiamentu and Spanish Phrased in Terms Of Simplicity
Pronouns and person
Pronouns largely invariant for gender and case, with 3.sg as e, except for possessive su. No person marking on the verb
Variability for gender and case, with forms such as él/le/lo and ella/le/la; also invariant possessive su. Person marking on the verb
No number marking on the verb (cf. bare bai). Number marking on the noun adding the 3.pl pronoun nan
Plural marking on the verb in all persons; plural on the noun with –s
Tense, mood, aspect (tma)
tma marked primarily with the pre-verbal particles ta, a, tabata, and lo
tma marked primarily through verbal inflections
Invariant specific (e) and non-specific (un), as well as null marking (e.g., den bentana ‘in the window’)
Determiners marked for person, number and definiteness: el/la, los/las, un/una, unos/unas; null marking rare
Nominal adjective agreement
Adjectives generally in post-nominal position (with very few exceptions), no agreement in gender and number
Adjectives generally in post-nominal position (with some exceptions), no agreement in gender and number
There are many other differences between the two languages, but the ones in Table 3 are typical for how researchers have distinguished creoles from their lexifiers. (It should also be mentioned that these languages share some features as well, e.g., in the organization of the noun phrase.)
However, a big caveat is needed at this point. Many of the differences between creoles and their lexifiers have been framed in terms of “more or less” distinctions, where typically the creole is assumed to make less distinctions than the lexifier: less gender, less person, less case, less number. Thus, some Afro-Portuguese and Afro-Spanish varieties in different South American countries have been claimed to be creolesmainly because they made fewer distinctions than the lexifier. The more they were like Papiamentu in making few of these distinctions, the more creole-like they were supposed to be. The absence of all these distinctions was assumed to be due to a pidgin phase in their genesis, in which incomplete second language learning led to the acquisition by the speakers of the early creole of invariant verb forms rather than variant ones, etc.
The Eurocentric perspective in the field, gauging creoles in terms of a simplicity yardstick on the basis of the morphosyntactic categories of the European lexifiers, did not take into account which distinctions were made in the creoles not made in the European languages. Once we start to seriously examine the creole languages, the idea of simplicity starts melting away. The list in Table 4 reflects some of the work of Muysken (2000) and others for Papiamentu involving different kinds of differences.
Table 4. Grammatical Differences Where the Papiamentu System Is at Least as Complex as the Spanish System
Frequent chains of serial verbs, as with sinta . . . wak . . . in the example
No serial verbs; rather, gerunds used as co-verbs
Fairly rigid SVO order, with fixed IO DO order
Fairly flexible order with SVO as the basis
Marked focus fronting, involving three constructions, each with their own grammatical conditions
Pragmatically conditioned other orders
A complex set of reflexive patterns, involving animacy considerations, body part—body relations, focus, distance between anaphor, and antecedent, etc.
Very local reflexives formed with the clitic se
A complex set of rules for preposition stranding with or without an agreeing anaphoric element, complex and lexically specified spell out rules, etc.
No preposition stranding
Lexically specified tone with grammatical consequences and tonal sanddhi
In Table 4, a number of features are listed for the Papiamentu system, which is at least as complex as the Spanish system. Some of these complexities may be due to West African structural contributions (see below), others to independent developments or even influence from Dutch.
It is important to note that several of the distinctions listed in Table 4 (e.g., regarding focus) are not morpho-syntactically encoded. Rather, they may result from the combination of separate (partly grammatical) words. In many typological surveys, such as the ones reported on WALS, morphosyntactic encoding is operationalized as morphological encoding, leaving aside juxtaposition on the syntactic level. Likewise, regularities in verbal marking are studied at the level of paradigms, rather than at the level of collocations and syntactic possibilities. This bias has the effect that language with limited or no inflections are, by definition, defined as making fewer distinctions. All or most researchers will agree that certain distinctions that exist in the lexifiers have been lost in the corresponding creole languages; where researchers disagree is whether other distinctions, less familiar and not directly morphologically encoded, exist in the creole languages but not in the lexifiers.
Very few creole languages have been described with the same degree of detail as other languages, and when they are, they often show all kinds of distinctions not made in their lexifiers. However, not all these distinctions are marked morphologically. Indeed, the words in most creoles tend to be short, and there is less affixation than in the European lexifiers. It needs to be seen to which extent this property has deceived people into thinking these languages are simple. Restructuring sometimes takes the form of morpho-syntactic simplification, and sometimes of other change.
This leads to the difficult issue of mixture, and of the contribution of the non-European languages to the creoles. As stated, this contribution was not lexical. While many European-lexifier creoles have words of non-European origin, these tend to form only a small part of the vocabulary, although there are exceptions, such as Berbice Dutch Creole (Smith, Robertson, & Williamson, 1987) and Saramaccan (Smith, 2015a,b). In discussing the contributions of languages from outside of Europe to the European-lexifier creoles, here the focus will be on the Caribbean and the contribution of the languages of West Africa, because of the present author’s own expertise, which is still only limited even for this area. In any case, Table 5 gives an overview of some of the relevant West African features involved. Not always the same West African languages are mentioned as possible contributors, but the Gbe languages figure prominently, as do Akan, Ijo, and Kikongo.
Table 5. Overview of the Main West African Structural Features that may Have Contributed to the Caribbean Creole Languages
Lexical and grammatical tone
Highly probable, but only a few Caribbean creoles have tonal distinctions
Open syllables and paragogic vowels
Particularly present among the English, Dutch, and Ibero-Romance-lexifier creoles, absent with the French creoles. Open syllables are cross-linguistically highly frequent.
Vowel and consonant inventories
Quite possible, but the relevant substrate patterns are also rather unmarked.
Pre-nasal and co-articulated stops
Very likely, but their distribution in the Caribbean is limited
Very likely, but distribution in the Caribbean is limited
Highly probable, but among the creoles, only a few have reduplication to form resultative adjectives. In the West African languages, the reduplication patterns are more complex than in the creoles.
Few if any affixes
Possible. This feature is shared between the Caribbean creoles and many contributing West African languages, but it is not very specific.
Syntax and lexical organization
Very probable. Generally found in the Caribbean creoles, but not always productively used.
Fronting and reduplicating the verb is a feature shared by the Caribbean creoles and related West African languages
Property items as verbs rather than adjectives
Very likely. This features is found productively in a range of Caribbean creoles, though not in all.
Very likely, but again their distribution in the Caribbean is uneven
Very likely, but their distribution in the Caribbean is limited
Body part reflexives
Quite possible, and fairly wide distribution. They are found in some non-standard varieties of the European lexifier languages as well, however.
Double object constructions
Possible. Double object (fixed order indirect object—direct object) constructions are found in English and Dutch, but not French and Spanish. They are very general, however, among the creole languages and often found in West Africa.
Possible. The tma systems of some Caribbean creoles resemble their West African counterparts, but many developments seem due to autonomous grammaticalization patterns.
The amount of West African influence in creoles is, as with many other topics, controversial, but it is fair to say that West African languages have contributed considerably to some creoles (probably not more than a handful), while other Caribbean languages show at least a few structural features from African sources. The amount of West African influence may well be linked to the degree to which European lexifier languages had a significant presence in the colonies.
7. Age and Displacement
Related to the possibility of defining creoles in terms of a definite point of emergence, there is the often-proposed idea that creoles are “young languages,” and that this gives them special features. The reasoning here is that it takes time for languages to acquire all kinds of irregularities and exceptions, and that all non-creole languages carry a lot of debris with them—words with old inflections that have fallen out of use, irregular verbal paradigms, etc. In some sense, creole genesis would lead to the wiping out of all these anomalies, and the creoles would become lean and clean. This idea has some intuitive appeal, but several caveats are in order. We do not know how long it takes for a language to acquire irregularities, and since many Caribbean creoles emerged around 1700 or so, three hundred years is enough time for specific idioms and exceptional meanings to spread. Again, since creoles have limited if any inflectional morphology, this is not where we should be looking, but rather at all kinds of compounds and fixed phrases.
Another criterion that may play a role in distinguishing creoles from other languages is displacement. Creole languages typically emerge far away from the regions where their lexifiers are spoken as native languages, and where the substrates are spoken as native languages: double displacement. Creoles can definitely be viewed as export languages in this respect. If we look at the list of lexifiers mentioned above, they include the major European colonial languages (English, Portuguese, Spanish, Dutch, and French) as well as two language clusters that had been involved in considerable expansion, but only single displacement: Arabic and Malay. In the case of West African pidgins and creoles, there was only single displacement as well; the colonial lexifier languages were brought into contact with much more stable local languages. This resulted in what Bickerton (1981) has called “fort creoles.”
The notion of displacement leads us to wonder about other languages that have undergone considerable expansion in different continents, expansions that have not come under the scrutiny of scholars working in creole studies. In South America, we find that the Arawakan languages have spread enormously, as have the Quechuan languages and some branches of the Tupian languages. Are any of the resulting languages comparable to creoles?
8. The Social Ecology of Language
This question acquires new urgency as there is a new field emerging at the cross-section of sociolinguistics and language typology. Pioneered by Kusters (2003), further developed by McWhorter (2007), and most prominently elaborated by Trudgill (2011), this field explores the relation between the sociolinguistic circumstances in which a language is or was spoken and its typological properties. Kusters looked within four language groups: Arabic, North Germanic, Quechuan, and Swahili. Within each language group he identified one of the most traditional members of the group, a language that had undergone intermediate amounts of restructuring and simplification, and a language that had undergone considerable restructuring. This yielded triples such as Icelandic/Faroese/Bergen Norwegian. He then tried to identify general properties of restructuring, which hold across the different language groups, and embedded analysis of these properties in an Optimality Theoretical framework. The works of Trudgill (2011), McWhorter (2007), and Kusters (2003) are relevant here because they provide another window on creole genesis in two ways.
First of all, some of the most extremely restructured varieties discussed by Kusters have been labelled creoles (Juba Arabic; Manfredi & Petrollino, 2013) or analyzed from the perspective of creole studies (Ecuadorian Quechua; Muysken, 2000). Thus, a model of this type can help us identify creole languages in areas not covered by the traditional canon of creoles. Second, the type of thinking in the work of Kusters (2003), McWhorter (2007), and Trudgill (2011) can help us place creole studies in the wider framework of language contact studies.
This approach can help identify creole languages and also languages that are like creoles in some respects but not in others. A crucial question here concerns the degree to which the structural features of the canonical creoles, the ones listed in Table 1, are actually the accidental result of the meeting of European structural patterns with those of the languages of West Africa, the southeastern parts of Asia, and the Pacific, or indeed a universally conditioned, inherent result of processes of creole genesis. How accidental is the canon of creoles from a typological perspective? This question will require detailed research in all major language families, and will require combined research between language family experts and creole researchers. This is made all the more complicated because the so-called language isolates, languages for which no clear genealogical affiliation can be established, are by necessity excluded from the enquiry, since we cannot prove or disprove their status as creoles. At the same time, perhaps circumstances of their genesis may have contributed to the fact that they are now difficult to classify genealogically.
9. Broader Considerations
Creole studies thus becomes part of a much wider research endeavour, in which properties of different languages are related to their social ecology globally. They are a special, and perhaps very striking, case of the general fact that to some extent at least the properties of languages are linked to their history and the setting in which they are spoken. In such an endeavor, it may be possible as a first step, to create several border areas for languages that are like creoles but differ crucially in some specific respect, as in the set of interlocking circles in Figure 1.
The language types described in Figure 1 are certainly not homogeneous blocks, and the boundaries between these types may be fluid rather than rigid. Table 6 provides further comments on these types, with some potential examples.
Table 6. The combination of the properties of Displacement, mixture, and restructuring in various contact settings
World Englishes and Latin American Spanish, where there are changes but no discernible sub- or adstratal influence
Displacement and mixture
Varieties of such expansion languages, like Indian English and Andean Spanish, where we see traces of local sub- or adstratal languages (Hindi and Quechua, respectively)
Cases of mixed languages such as Michif (Cree/French) and Media Lengua (Quechua/Spanish), where properties of the languages themselves have remained intact
Mixture and restructuring
Cases such as Petjoh (Malay/Dutch), where we see restructuring of both languages as well as mixture
Perhaps the transition from Old to Middle English could be a case where we find extensive restructuring but little clear substratal or adstratal influence.
Displacement and restructuring
Examples of creoles such as, perhaps, Réunion French Creole, where there is little evidence of substratal or adstratal influence
A, B, C
Displacement, mixture, restructuring
Cases such as Saramaccan and Berbice Dutch Creole, where we find all properties interact
These criteria can be also used to help understand the differences between creoles. Given the discussion so far, it is also possible to consider again the different kinds of creoles that were distinguished by Bickerton (1981):
• Maroon creoles, spoken by the descendants of Africans who ran away from captivity on the plantations in different parts of the Caribbean.
• Plantation creoles, spoken by the descendants of the people who remained on the plantations.
• Fort creoles, spoken by the descendants of people who lived near trading posts or forts, often in West Africa.
What distinguishes plantation creoles from maroon creoles can be linked to what may be labeled as “cyclic migration reinforcement” in heritage language studies. An example would be Turkish in northwestern Europe. Since there is continuous cyclic migration between communities in Europe and regions in Turkey, Turkish as a heritage language remains rather stable in Europe. In the same way, the initial displacement of the European colonial languages to the Caribbean was often followed by continuous cyclic migration, leading to the stability of the displaced heritage varieties and in the creoles to which they were linked as lexifiers.
To summarize, the proposal is to stop looking at creoles as a separate class but take them as special cases of the general fact that the way languages emerge and are used helps determine their properties. This calls for a new, socially informed typology of languages, which will involve all kinds of different types of languages, including pidgins and creoles.
Two of the dimensions mentioned in Figure 1 are also mentioned in McWhorter (2007; see also McWhorter, 2012), where he presents a table with nine cells regarding the degree of restructuring or simplification, and the degree of mixture.
10. Critical Analysis of Scholarship
Creole studies have long been characterized by a succession of oppositions between radically different positions, often the result of particular strong claims brought in. Consider first the debates around the assumed similarities among creole languages. One of the reasons creoles have been especially interesting for linguists of different theoretical persuasions is that they appear to have many features in common. Over the years, these similarities have been explained in various ways.
Monogenesis. In the 1960s (Taylor, 1961; Thompson, 1961), the idea was brought forward that creoles had not emerged separately, but were relexified versions of an Ur Portuguese-lexifier Pidgin, probably spoken along the coast of West Africa at the beginning of the slave trade. Relexified in the sense that the Portuguese lexicon had been replaced by English, French, or other lexicons, obscuring their common origin. Linguistic evidence for this monogenetic theory came from Wanderwörter such as pikeniny ‘little, child’ (<Port. pequeninho ‘little (dim.)’) and from structural similarities, such as the system of preverbal tense-mood-aspect particles. There can be little doubt that early Atlantic pidgins were important in feeding into the Caribbean creoles that emerged here and there, but monogenesis in its strong form is not accepted any more.
Bioprogram. One of the reasons was an attractive universalist account for similarities that was proposed by Bickerton (1981). He claimed that innate cognitive abilities of children growing up in pidgin-speaking environments played a crucial role. The children had to create a full creole language out of the rudimentary input they got from their pidgin-speaking parents, and to do this they used resources from their bioprogram. Since the latter is innate, it led to parallel structures in different creoles. Bickerton’s ideas have been criticized on all counts, and his model has few current adherents, but the attempt to look for cognitive explanations for possible similarities has remained appealing.
Shared substrate. Another explanation came around the same time from Alleyne (1980). Although he limited himself to English-lexifier creoles, the implications were broader: shared West African substrate features could be responsible for similarities between creoles in the Atlantic. Linking similarities to substrate was also important in the research program that came out of Lefebvre’s (1998) analysis of Haitian Creole as basically a Gbe language, Fon, relexified with French lexical roots. Whatever the merits of this analysis, it is clear that appealing to West African substrate as a source for similarities is only successful insofar as the substrate features are actually present, and this is only partly the case. However, as such, appealing to shared substrate is not an implausible idea. Needless to say, the story for the Pacific has to be a very different one, but there also may be commonalities between creoles due to shared substrate.
All approaches mentioned have the underlying assumption in common that the creoles share fundamental structural similarities that cannot be explained in terms of the properties of their lexifiers. (It should also be mentioned that a number of their basic typological properties—generally subject-verb-object [SVO] order and prepositions, nominative-accusative alignment—are shared with their lexifiers.) However, it only recently became possible to study this systematically. The major innovation in the discipline will come from work at the Max Planck Institute for Evolutionary Anthropology in Leipzig, where a team headed by Susanne Michaelis and Martin Haspelmath created an Atlas of Pidgin and Creole Structures (APiCS), available as a four volume set and as on online resource (Michaelis, Maurer, Haspelmath, & Huber, 2013a,b).
McWhorter (1998) has proposed to define creoles not in sociohistorical terms, as in the work of Mufwene (2002), but structurally, in terms of a few key typological features. Taking a less radical approach, but going in the same direction, Bakker, Daval-Markussen, Parkvall, & Plag (2011) have argued that creoles are typologically distinct from non-creoles. These studies have the merit of pointing to the fact that in many creoles the number of morphosyntactic distinctions has been reduced with respect to their lexifiers. This is a trend that can be observed not just in the canonical creoles, but also in the expansion varieties discussed by Kusters (2003), McWhorter (2007), and Trudgill (2011).However, their approach suffers from two potential bias problems, already mentioned in the essay. First, the list of pidgins and creoles was established in large part on the basis of the very criteria later used to set them apart as a typological class: structural differences between them and their lexifiers, differences seen in terms of making less morphosyntactic distinctions. Second, the lists of features for which creoles have been coded may also be biased, in stressing the presence or absence of morphosyntactic distinctions. It remains to be seen whether the results of McWhorter (1998, 2011) and Bakker et al. (2011) can be replicated with different feature sets drawn from a data base not specifically geared towards defining creoles. The problem of the language sample bias is unsolvable at present, requiring a much more concerted global search for potential creoles in different language families all over the world, but the feature bias issue can be resolved. Scholars such as Mufwene have criticized what they call the “creole exceptionalism” of McWhorter and Bakker; their criticisms are not based on a clear empirical basis either, but the case for a separate typological profile for creoles needs strengthening. Where agreement can be found across many accounts in the notion that whatever is involved in non-demic expansion will involve structural simplification of some sort with respect to an ancestor language. However, the typological properties of that ancestor language may determine what kind of simplification is found. This may be complete erosion of verb endings, as in most canonical creoles, or rather restructuring of the morphological system towards greater regularity and transparency, without actual loss.
Altogether, the major trend that can be seen in pidgin and creole studies is to bring these languages back into the larger context of contact linguistics and linguistic typology. Rather than studying these languages in isolation, there is an increasing awareness of their place within contact languages in a broader sense.
Another trend in creole studies concerns more philological approaches. New work at the intersection of philology and creole studies is carried out, aimed at creating a solid basis for work on older documents of creoles. How should we read these documents, and to which audience were they addressed? Essentially philological methods are also applied to study creole languages used in the new media, both in migration settings and in their traditional habitat.
There are many interesting books comparing creoles, but Velupillai (2015) definitively replaces older introductions, such as Arends, Muysken, and Smith (1994), which will be used now mainly for references, as is also the case for Holm (1988a,b).
The main journal in the field is Journal of Pidgin and Creole Languages (JCPL), founded in 1986. Recently, the Journal of Language Contact (founded 2008) has also started publishing papers of interest to creolists. For Portuguese and Spanish, there is the Revista de Crioulos de Base Lexical Portuguesa e Espanhola, and in French there is Études Créoles.
Alleyne, M. C. (1980). Comparative Afro-American: An historical-comparative study of English-based Afro-American dialects of the new world. Ann Arbor, MI: Karoma.Find this resource:
Ansaldo, U., Matthews, S., & Lim, L. (2007). Deconstructing Creole: The rationale. In U. Ansaldo, S. Matthews, & L. Lim (Eds.), Deconstructing Creole (pp. 1–20). Amsterdam: Benjamins.Find this resource:
Arends, J., Muysken, P., & Smith, N. (Eds.). (1994). Pidgins and creoles. An introduction (pp. 303–374). Amsterdam: Benjamins.Find this resource:
Bakker, P., Daval-Markussen, A., Parkvall, M., & Plag, I. (2011). Creoles are typologically distinct from non-Creoles. Journal of Pidgin and Creole Languages, 26(1), 5–42.Find this resource:
Biagui, N. B., & Quint, N. (2013). Casamancese Creole structure dataset. In S. M. Michaelis, P. Maurer, M. Haspelmath, & M. Huber (Eds.), Atlas of Pidgin and Creole Language Structures Online. Leipzig: Max Planck Institute for Evolutionary Anthropology.Find this resource:
Bickerton, D. (1981). Roots of language. Ann Arbor, MI: Karoma Press.Find this resource:
DeGraff, M. (2003). Against Creole exceptionalism. Language, 79(2), 391–410.Find this resource:
Deumert, A. (2004). Language standardization and language change: The dynamics of Cape Dutch. Amsterdam: Benjamins.Find this resource:
Dryer, M. S., & Haspelmath, M. (Eds.). (2013). The World Atlas of Language Structures Online. Leipzig: Max Planck Institute for Evolutionary Anthropology.Find this resource:
Faraclas, N. (2013). Nigerian pidgin structure dataset. In S. M. Michaelis, P. Maurer, M. Haspelmath, & M. Huber (Eds.), Atlas of Pidgin and Creole Language Structures Online. Leipzig: Max Planck Institute for Evolutionary Anthropology.Find this resource:
Finney, M. A. (2013). Krio structure dataset. In S. M. Michaelis, P. Maurer, M. Haspelmath, & M. Huber (Eds.), Atlas of Pidgin and Creole Language Structures Online. Leipzig: Max Planck Institute for Evolutionary Anthropology.Find this resource:
Haspelmath, M., Dryer, M. S., Gil, D., & Comrie, B. (Eds.). (2005). The world atlas of language structures. Oxford: Oxford University Press.Find this resource:
Holm, J. (1988a). Pidgins and creoles, Vol. 1: Theory and structure. Cambridge, U.K.: Cambridge University Press.Find this resource:
Holm, J. (1988b). Pidgins and creoles, Vol. 2: Reference survey. Cambridge, U.K.: Cambridge University Press.Find this resource:
Huber, M. (2013). Ghanaian pidgin English structure dataset. In S. M. Michaelis, P. Maurer, M. Haspelmath, & M. Huber (Eds.), Atlas of Pidgin and Creole Language Structures Online. Leipzig: Max Planck Institute for Evolutionary Anthropology.Find this resource:
Huttar, G. L., & Velantie, F. J. (1997). Ndjuka-Tiriyo Pidgin. In S. G. Thomason (Ed.), Contact languages: A wider perspective (pp. 99–124). Creole Language Library 17. Amsterdam/Philadelphia: John Benjamins.Find this resource:
Kusters, W. (2003). Linguistic complexity: The influence of social change on verbal inflection (PhD Diss.). Leiden University. Utrecht: LOT Publications.Find this resource:
Lefebvre, C. (1998). Creole genesis and the acquisition of grammar: The case of Haitian Creole. Cambridge, U.K.: Cambridge University Press.Find this resource:
Manfredi, S., & Petrollino, S. (2013). Juba Arabic structure dataset. In S. M. Michaelis, P. Maurer, M. Haspelmath, & M. Huber (Eds.), Atlas of Pidgin and Creole Language Structures Online. Leipzig: Max Planck Institute for Evolutionary Anthropology.Find this resource:
Maurer, Ph. (1988). Les modifications temporelles et modales du verbe dans le papiamento de Curaçao (Antilles Néerlandaises). Hamburg: Buske.Find this resource:
McWhorter, J. H. (1998). Identifying the creole prototype: Vindicating a typological class. Language, 74(4), 788–818.Find this resource:
McWhorter, J. H. (2007). Language interrupted: Signs of non-native acquisition in standard language grammars. New York: Oxford University Press.Find this resource:
McWhorter, J. H. (2011). Tying up loose ends: The creole prototype after all. Diachronica, 28, 82–117.Find this resource:
McWhorter, J. H. (2012). Case closed? Testing the feature pool hypothesis. Journal of Pidgin and Creole Languages, 27, 171–182.Find this resource:
Michaelis, S. M., Maurer, P., Haspelmath, M., & Huber, M. (Eds.). (2013a). The Atlas and Survey of Pidgin and Creole language structures. Four volumes. Oxford: Oxford University Press.Find this resource:
Michaelis, S. M., Maurer, P., Haspelmath, M., & Huber, M. (Eds.) (2013b). Atlas of Pidgin and Creole Language Structures Online. Leipzig: Max Planck Institute for Evolutionary Anthropology.Find this resource:
Mufwene, S. (2002). The ecology of language evolution. Cambridge, U.K.: Cambridge University Press.Find this resource:
Muysken, P. (2000). Semantic transparency in Lowland Ecuadorian Quechua morphosyntax. Linguistics, 38(5), 973–988.Find this resource:
Reinecke, J. E. (1938). Trade jargons and creole dialects as marginal languages. Social Forces, 17(1), 107–118.Find this resource:
Reinecke, J. E., DeCamp, D., Hancock, I. F., Tsuzaki, S. M., & Wood, R. E. (1975). A bibliography of pidgin and creole languages. Honolulu: University of Hawai’i Press.Find this resource:
Schröder, A. (2013). Cameroon pidgin English structure dataset. In S. M. Michaelis, P. Maurer, M. Haspelmath, & M. Huber (Eds.), Atlas of Pidgin and Creole Language Structures Online. Leipzig: Max Planck Institute for Evolutionary Anthropology.Find this resource:
Smith, N. (1994). An annotated list of Creoles, Pidgins, and mixed languages. In J. Arends, P. Muysken, & N. Smith (Eds.), Pidgins and Creoles. An introduction (pp. 303–374). Amsterdam: Benjamins.Find this resource:
Smith, N. (2015a). A preliminary list of probable Kikongo (Kikoongo) lexical items in the Surinam creoles. In P. Muysken & N. Smith (Eds.), Surviving the middle passage: The West Africa-Surinam Sprachbund. Berlin: Mouton de Gruyter.Find this resource:
Smith, N. (2015b). A preliminary list of probable Gbe lexical items in the Surinam creoles. In P. Muysken & N. Smith (Eds.), Surviving the middle passage: The West Africa-Surinam Sprachbund. Berlin: Mouton de Gruyter.Find this resource:
Smith, N., Robertson, I. E., & Williamson, K. (1987). The Ijo element in Berbice Dutch. Language in Society, 16, 49–90.Find this resource:
Taylor, D. (1961). New languages for old in the West Indies, Comparative Studies in Society and History, 3, 277–288.Find this resource:
Thomason, S. G., & Kaufman, T. (1988). Language contact, creolization, and genetic linguistics. Berkeley: University of California Press.Find this resource:
Thompson, R. W. (1961). A note on some possible affinities between the creole dialects of the old world and those of the new. In R. B. Le Page (Ed.), Creole Language Studies 2. London: Macmillan.Find this resource:
Trudgill, P. (2011). Sociolinguistic typology: Social determinants of linguistic complexity. Oxford: Oxford University Press.Find this resource:
Velupillai, V. (2015). Pidgins, creoles, and mixed languages. An introduction. Amsterdam: Benjamins.Find this resource:
Yakpo, K. (2013). Pichi structure dataset. In S. M. Michaelis, P. Maurer, M. Haspelmath, & M. Huber (Eds.), Atlas of Pidgin and Creole Language Structures Online. Leipzig: Max Planck Institute for Evolutionary Anthropology.Find this resource: