MICHAEL ACETO
The terms Caribbean Englishes or restructured Englishes are used here as generally synonymous with other terms found in the linguistics literature: English‐derived or ‐based creoles and even dialects of English (Mufwene 2001, 2008). Creolists have never agreed upon a typologically distinct linguistic definition in terms of common structures, features or processes that demarcate so‐called creole languages from other natural human languages (Aceto 1999a; DeGraff 1999; Mufwene 1994, 1996, 2008; McWhorter 1998). Even Bakker, Daval‐Markussen, Parkvall, and Plag (2011), while insisting that purported creoles are typologically distinct (“typologically similar” would probably be uncontroversial; and I will not explore the observation that the conclusions of the article seem unfalsifiable, at least in the sense of Popper 1984), admit that “it is not possible to specify which individual features are responsible for the clusterings” (Bakker et al. 2011: 19) that make the group distinct. This absence of a typological distinction may at first seem troublesome or peculiar, but in fact there is still no principled definition as to what distinguishes a “dialect” from a “language,” either. Yet linguists comfortably and usefully employ both those terms in a general sense, as if they reference specific agreed‐upon linguistic phenomena.
The term “creole” derives more from the sociohistorical circumstances of colonialization surrounding the earliest genesis of these restructured varieties than from any single linguistic feature or cluster of features that might prove to be diagnostic of this group of languages. These are new(er) languages in that they emerged in only the last 400 years or so. Of course, a creole with an English‐derived lexical base is just one possible outcome of linguistic/cultural contact across peoples who originally spoke mutually unintelligible languages. That is, there are creoles with French, Spanish, Portuguese, Dutch, Arabic, and African languages (e.g. Kikongo in Kituba) as the source of most of their basic morphemes. In some areas of the Caribbean, restructured varieties of English coexist within relatively small, well‐defined geographical spaces (e.g. St. Lucia, Carriacou, and Dominica) with chronologically older creoles that do not share the same lexical base (in those cases, French‐derived forms). It is important to note that even today in Latin America the Spanish term criolla means “local” or “native to the area” (as in comida criolla 'local food'). In this way, calling locally born native languages sharing similar features “creoles” makes perfect historical sense.
Most of these new, restructured, or creole varieties in the Atlantic and Pacific regions seem to be the result of disproportionate power relationships in which speakers of one language or, more commonly, a set of languages (e.g. the Kwa and Bantu languages from western and southern Africa in the Atlantic region) are dominated socially, economically, and/or militarily by politically – but, crucially, not numerically – more powerful speakers of another language (in this case, European languages spoken by colonists and settlers involved in the colonization of the Americas). Within the crucible of European colonialization and its lopsided power dynamic emerged these new varieties, with many of the basic lexemes derived from the socially dominant language. The varieties spoken by the colonizers are often called “the superstrate” and the native languages originally spoken by those most exploited by colonialization are often called “the substrate.”
Not all members of the subordinate group spoke the same mutually intelligible substrate languages. One of the unintended consequences of European‐style slavery in the Americas is that it brought dislocated Africans from across 3,000 miles of west African coastline who ordinarily would have never heard each other’s languages into contact with each other in the Americas. This multilingual situation created the need for a lingua franca to bridge communication gaps within African language communities as well as between such communities and speakers of the European languages. For example, within the Anglophone context, children born into these multilingual settings grappled with whatever local varieties of English they heard from colonists and settlers, as well as from slaves who had some degree of familiarity with the language of colonial power. These children restructured the varieties heard in situ further, enforcing structural regularity on the second‐language varieties (in whatever form, pidgin or otherwise) heard spoken by the adults in their communities, drawing on processes made available through the common human language faculty.
Which linguistic features are or are not derived from the superstrates, substrates, or regular linguistic processes is a matter of some differing opinions among researchers. It is uncontroversial that specific words and phonemes heard in Caribbean varieties are derived from substrate languages. For example, dokunu ‘dumpling’ is derived from Twi, a language of Ghana (the Gold Coast during the colonial period), and is heard in several varieties in the Atlantic region; the coarticulated stops /kp/ and /gb/ heard in Saramaccan, a mixed English‐ and Iberian‐derived language of Suriname in northern South America, are also articulated in several West African languages and conspicuously absent from all European languages. However, the source(s) of the creole language’s structure or syntax – substrate, superstrate and/or language universals – is a subject of some healthy debate. The substratist position has dominated creole studies since the 1970s (Holm 1988–1989; Parkvall 2000). The influence of dialects spoken by European colonists on Caribbean varieties has most definitely been a persistent but minority view (Hancock 1994; Mufwene 2001; Niles 1980; Winford 2000). Other researchers insist that features of these new languages are largely influenced by principles of first‐language acquisition/creation (Bickerton 1984; DeGraff 1999). One could argue that all living languages change each time a new grammar is constructed by a child, and therefore, one may view each creation of a grammar as a “restructured” language, to a greater or lesser degree. The complexity of the answer most likely entails that components of all of the above categories (substrates, superstrates, universals in first‐language creation, and even processes in second language acquisition; see Andersen 1983) have played a role in shaping the “look” or sound of specific varieties, as well as languages everywhere. This conclusion raises the question, in the absence of a typologically transparent definition for so‐called “creoles,” of how these languages differ from any other human language.
Restructured Englishes are spoken along the edges of virtually every major ocean or sea in the world because coastal sites were often the points of colonial contact with new territories. Contact between colonizers speaking regional dialects of British English and subsequently colonized peoples occurred largely along coastal shipping and sailing routes during the period of European expansion and colonialization in the seventeenth through the twentieth centuries. Again in this regard, these younger European varieties (whether we call them dialects or creoles) are not different from other human languages: all natural human languages display the effects of cultural/language contact, even if dislocation and slavery are not factors in a given language’s ecology (slavery is one trigger for language contact in that it often brought speakers into contact with each other, but it is not a requirement). What truly makes these languages unique is that they are relatively young varieties (they are about 150–400 years old, while English and Spanish are 1,500 and 2,000 years old, respectively).
Every former British colonial territory in the Caribbean reveals an English variety spoken today, whether one calls it a creole or a dialect. In fact, why some varieties in the Caribbean are designated one or the other is often not based on specific language features but rather on who the speakers are in terms of their ancestry. Many of these languages are distributed across the islands of the Caribbean: Trinidad and Tobago, Grenada, Barbados, St. Vincent, St. Lucia, Dominica, Montserrat, St. Kitts and Nevis, Antigua and Barbuda, Jamaica, and the Cayman Islands. The British Virgin Islands of Tortola, Virgin Gorda, Anegada, and Jost Van Dyke contain largely undocumented varieties. The Dutch Windward Islands of Saba, St. Martin, and St. Eustatius (Aceto 2015) also reveal English dialects. The varieties heard on the US Virgin Islands of St. Croix, St. John, and St. Thomas have received relatively little attention from linguists.
English varieties have been spoken for more than a century by minority populations in several Latin American countries associated with Spanish as a national language: the Dominican Republic; Providencia and San Andres Islands (politically controlled by Colombia); and the Central American nations of Panama, Costa Rica, Nicaragua, Honduras, and Guatemala. Puerto Rico (a territory of the US, as are the US Virgin Islands) has had local varieties of English in contact with Spanish since 1898. Relatively little research has documented these varieties. Belize, an ex‐British colony, is the only nation in Central America in which English is widely spoken and where it is the official language. In North America, creole varieties can be heard in the Gullah‐speaking coastal areas of South Carolina and Georgia, as well as among their descendants in southern Texas and northern Mexico (as well as among Caribbean immigrant communities in, for example, New York City, Miami, and Hartford). Even South America contains English‐derived varieties in the countries of Suriname and Guyana.
Local names for these varieties vary widely. Some may call their language “pidgin,” “creole,” or even “dialect.” Unfortunate qualifiers such as bad, flat, raw, and broken (all reflect negative speaker attitudes often mistakenly and prescriptively encouraged by sites of institutional literacy) often qualify what many speakers also simply call English. Discrete names such as Patois (Jamaica), Guari‐Guari (Panama), and Kokoy (Dominica) are also used.
Some scholars (Bickerton 1984; Thomas & Kaufman 1988) insist that many new Englishes are “creoles” and represent cases of what is called “broken” transmission between the language of power providing the source of basic lexical material and the subsequent emerging English variety. Clear definitions as to what constitutes “broken” are lacking. In Suriname, a regular source of English lexemes was withdrawn as the British traded that colony to the Dutch for New Amsterdam in North America in the late seventeenth century. From then on, Sranan, one English‐derived creole still spoken today, became the source of Anglophone forms for the other emerging varieties Ndyuka and Saramaccan. Perhaps the case of “broken” transmission is accurate and appropriate only in this isolated scenario. English varieties have been spoken consistently in the other former British colonies by colonists, settlers, and slaves (who had varying degrees of access, contact, and familiarity) and their descendants since the colonial period. Metropolitan colonial Englishes have persisted as the ambient language of institutional power (thus, one source of linguistic influence) before and well after the emergence of local vernaculars in all the former British colonies. Standardized varieties associated with literacy continue to exert influence even today.
Clearly the Surinamese varieties are exceptional cases. However, the field of creole studies often considers these languages as the baseline against which other varieties in the Americas are measured. Not surprisingly, other varieties that differ significantly are considered lacking in the same “creole” features, for which, again, no reliable typological inventory of features has been established. The distorted assumption that all Anglophone varieties spoken by the descendants of slaves today once sounded more like one of the Surinamese creoles (or Jamaican, in some research paradigms) is problematic and easily falsified and was often asserted in the past with little corroborating data.
The slave plantation is often considered the prototypical environment for the emergence of creoles. It is important to note that in the Pacific region, creoles and colonial Englishes emerged without slavery playing any role, even if many similar colonial plantation components were in place (often intense contact among speakers of mutually unintelligible languages, dislocated populations moving from one place to another, and labor forces with varying degrees of contact with native speakers of the colonial language of power). In the Atlantic and Pacific regions, a disproportionate demographic ratio between the socially subordinate groups (laborers; African slaves in the Atlantic context) and the numerically smaller superordinate colonial group of power who are the source of most of the lexemes produced an emerging variety that was used, at least initially, as a lingua franca by those adults who spoke non‐European languages, and subsequently as a first language of the children born into this social matrix.
There are several cases of language emergence that resist being classified within the plantation experience, even if they were extensions of colonial practices. Barbuda, an island to the north of Antigua, was populated with slaves for the explicit function of raising food crops and manufacturing goods to supply nearby plantations off the island. The Barbudan slaves often came into contact with no more than a handful of Europeans (Aceto 2002). In west Africa, neither Krio nor Liberian English emerged on plantations. Several Anglophone varieties spoken by Africans and their descendants emerged on small‐scale farms and homesteads who were working alongside perhaps a handful of other Africans and Europeans in the fields. The roles of immigration and movement in search of work, as well as regular small‐boat traffic, following emancipation in the Caribbean are factors that have complicated understanding the earliest emergence of local varieties. This has been largely ignored by most researchers.
Several of the Englishes heard in Central America are the result of intra‐Caribbean migration in search of work since the nineteenth century (Aceto 1995; Holm 1983), for example, Panama, Costa Rica, and Guatemala. Belize and the community of Bluefields in Nicaragua are the result of British colonial competition with the Spanish for territories on the mainland of Central America but are not really part of the plantation experience, either. Trinidadian English is the result of immigration by a variety of ethnic groups who already spoke preexisting Englishes in the nineteenth century. Out of this matrix, a new variety emerged.
At least two of the Surinamese varieties (Ndyuka, Saramaccan) are maroon languages (Smith 1987), which emerged as runaway slaves and rebels formed independent societies in the interior rainforest, away from European institutions. Maroon varieties emerged in remote mountainous areas of Jamaica as well (Bilby 1983). These varieties emerged in even greater isolation from European influences than those on plantations.
Several former French colonial territories ultimately came under British control in the nineteenth century. In St. Lucia, Dominica, Grenada, and Carriacou a French‐derived language had previously emerged. British dominion over these territories for the last two centuries has resulted in new Anglophone varieties significantly influenced by the earlier French varieties, which are declining in terms of number of speakers (Garrett 1999, 2003; Aceto 2010).
European Anglo‐Caribbean varieties (those varieties spoken mostly among the descendants of Irish, Scots, and English colonists and settlers) have been neglected in most research paradigms except for the work of Williams (1985, 1987, 1988, 2003). These Englishes (rarely if ever called “creoles” by researchers) display many similar features with so‐called creole varieties largely spoken by Euro‐Caribbeans on Saba, Bequia, the Cayman Islands, Barbados, and Anguilla (Williams 2003) and may shed light on the European‐derived component heard by Africans or Afro‐Caribbeans who worked alongside many indentured servants, colonists, and settlers.
The concept of decreolization is the result of a highly questionable assumption that uniformity was once the norm among Anglophone colonial varieties, and that any differences heard in varieties today are the result of shifts toward the norms of the institutional standard associated with literacy. Clearly these institutional norms play a role in language contact, variation, and change, especially among speakers of related dialects, but any human language has options for variation and change unrelated to this pressure (Aceto 1999a). Even if the effects of so‐called “decreolization” could be rigorously distinguished from what is regular variation, which all languages exhibit everywhere (whether they are called “creoles” or not), unquestioned acceptance of this concept often obscures the fact that all living languages vary and change, whether in the direction of the lexically related variety associated with institutional literacy or not. Speakers cannot undo the process of creolization (even if that is understood as a linguistic process). All living languages are in the flux of becoming. That the varieties under examination are spoken by communities of speakers of African descent is not sufficient to necessitate a new problematic term like decreolization. Equally vague are terms such as creolization and recreolization. All such proposed constructs raise the question of why one areal group of languages necessitates its own specialized terms for what must be regular human‐language processes.
Creolists in the 1970s and 1980s used to insist that the variation of the purported postcreole continuum was due to “decreolization,” perceived as a unilateral force affecting all Englishes of the Caribbean. Once this strong view of decreolization was questioned and criticized with data which showed that change was not unidirectional, creolists began to largely reject the concept of decreolization, but still held on to the notion of “continuum” (even if decreolization was the explanatory factor), no matter how flawed it was. Nowadays creolists simply assert that a continuum exists uniquely in the Anglophone Caribbean (and why not the Francophone or Hispanophone Caribbean as well?).
Le Page (1998: 91) makes several points on the subject of the continuum: “I came to realize – too late, unfortunately, to stop David DeCamp and others from taking up the ‘continuum’ model – that it was a false representation.” Le Page also writes that DeCamp wrote him that he “regretted the concept had been taken up with such enthusiasm, since he [DeCamp] had never found that it could provide an account of more than 30% of his Jamaican data” (Le Page 1998: 92). Clearly all living languages exhibit variation, but can any variation be arranged along a continuum without forcing the data to fit the continuum model? The insistence by many creolists that something happened in the Anglophone Caribbean that was qualitatively different from the rest of the language‐speaking world (both past and present) is troubling. Le Page is worth quoting again: “I have come to realize more and more the extent to which academics, as they become authorities on subjects, are at the same time creating the universe which they study, the subject matter of their discourse. They evolve their own perceptual framework and then fill it with their percepts” (1998: 48).
This section presents a brief, generalized grammatical and phonological sketch applicable to many restructured varieties in the Caribbean. It must be pointed out that all Anglophone varieties reveal synchronic differences in terms of lexicon, phonology, morphology, and syntax, even if they share general similarities. Features presented here are not exclusively associated with any specific variety. That is, not all the features are derived from Jamaican (often representative of the western varieties) or Guyanese (often representative of the eastern varieties). What follows is an abstraction of possible features based on what I have documented from my research in Panama, Barbuda, St. Eustatius, and Dominica, as well as from archival sources. This treatment does not in any way pretend to be exhaustive. For syntactic features associated with specific varieties, including the Surinamese creoles (which are ignored here), readers should consult the survey presented in Hancock (1987). For a discussion of phones and phonemes in the Atlantic varieties, see Wells (1982). The goal of this section is to highlight the most basic forms heard in a range of Caribbean Anglophone varieties.
The varieties of English that Africans and their descendants in the Americas heard were originally regional, social, and ethnic dialects of British English (including Irish, Scottish, and southwestern English) which were spoken in the seventeenth through the nineteenth centuries. These regional British varieties are the sources of many of the morphemes in the varieties we hear today in the Caribbean. From a diachronic perspective, English‐derived Caribbean varieties are more British oriented not only in their histories but in their phonologies, especially, though in the last century American and Canadian influence can be expected and documented (Van Herk 2003).
Initially these regional British dialect forms were acquired by adults as second‐language varieties of some type for lingua franca functions among those speaking mutually unintelligible African languages and for communication with Europeans. In subsequent generations, these forms would have been embedded in first‐language varieties of English created and expanded upon by children but influenced as well by the second language forms spoken by adults in a given location. These two stages – second‐language forms spoken by adults and subsequent nativization by children in following generations – may also have been influenced by whatever Anglophone forms were heard along the western coast of Africa in slave forts or at subsequent Caribbean trading centers as slaves were distributed to buyers. Both St. Kitts (Baker & Bruyn 1998) and St. Eustatius (Aceto 2015) served this function as sources of slaves sold onward to destinations in the Americas.
There are satisfactory reasons for dividing the Caribbean regions into geographically designated western and eastern varieties on the basis of comparative phonology and syntax (Holm 1988: 445; Wells 1982, 1987; Le Page 1957; Hancock 1987). However, the grounds for this division are largely abstract and impressionistic, since it is my experience, having done fieldwork in both locations, that there are few specific features that one may find exclusively in one region but not in the other. However, several such plausibly relevant candidates are designated below. Creolists are often comfortable with the highly questionable assumption that earlier Anglophone Caribbean varieties were once monolithic, and that contemporary synchronic variation is a more recent (postemancipation) phenomenon. Whether the overlapping patterns between eastern and western varieties represent parallel developments or are due, for example, to intra‐Caribbean migration is an open topic for future research. As has been made clear in dialect studies over the last 50 years, it is not any specific feature that is diagnostic of a variety, but instead the bundle of features that are used in a specific location or community. The Surinamese creoles are ignored here because they are largely unintelligible with the Anglophone varieties spoken outside Suriname. The data are presented in phonetic symbols within a mostly phonemic presentation.
The verbal complex has received significant attention from linguists (Winford 1993), and the forms and distribution of the copula have often been central to this discussion. The three basic lexical functions of the copula (attributive, locative, and nominal) are exhibited in the following forms.
Attributive/Adjectival constructions: /ʃi de gud/, /ʃi aarait /, /ʃi iz gud / ‘she’s good, she’s all right, she’s doing fine’. The verb /de/ or /iz/ is often the copula form, but no overt verb may be realized at all, with only a predicate adjective appearing. The locative forms are often /de/ or /iz/ as in /we im iz/ ‘where is she/he/it?’ Inversion of the copula and the subject is not typically heard for interrogatives in these varieties; rising intonation alone often indicates questions. The equative form of the copula displays the following forms: /a/, /iz/, /bi/ and no overt realization at all: /ʃi a mi sista /, ʃi iz mi sista /, /ʃi bi mi sista /, /ʃi mi sista / ‘She’s my sister’.
The past tense marker, as is the case with most overt grammatical markers in Caribbean varieties is a discrete, free morpheme before the main verb of an utterance. Unmarked nonstative verbs often have a [+past] interpretation, for example, /mi iit aaredi/ ‘I ate already’. Nonetheless, both stative and nonstative verbs may be preverbally marked [+past] with one of range of forms. Depending on the context, an utterance may be interpreted as conveying both simple past and what is sometimes called the past perfect: /mi bin iit/, /mi woz iit/, /mi di(d) iit/, /mi min iit/ ‘I have/had eaten’ or even ‘I ate’. The utterance /mɪ dʌn iit/, with the completive marker /dʌn/, would more closely match ‘I already ate, I have already eaten, I’m done eating’.
The preverbal future tense marker is some reflex of either /go/, /a go/, /goin/ and sometimes /wi/ < will: /dem go dans/, /dem wi dans/, /dem gwain dans/, /dem goin dans/, /dem a go dans/ and even /dem wan dans/. In the last instance, the future marker /wan/ seems to be a grammaticalized form of the verb want. The form gwain is often associated with the western Caribbean.
The preverbal markers /de/, /da/, or /a/, as well as the bound suffix /‐in/ are the features most commonly associated with progressive aspect in Anglophone Caribbean varieties: /di gyal a kaal yu/, /di gyal de kaal yu/, /di gyal kaalin yu/ ‘The girl’s calling you’.
Most of the pronouns noted below may function as subject, object, and possessive pronouns, with the following exceptions: both /ai/ and /a/ are only subject pronouns; /ar/ ‘her’ is an object pronoun with exclusive reference to females; /om/ may refer to ‘him/her/it’ in object position (its distribution is largely limited to the eastern Caribbean). The pronoun /(h)im/ indicates males or females or even nonhuman referents in either subject or object. The plural pronouns /aayu/ and /aawi/ are mostly heard in the eastern Caribbean, while /unu/ is more common in western varieties. An important exception is Barbados, which also reveals a reflex of /unu/.
| Singular | Plural | |
| 1st | /mi/, /a, ai/ (subject only) | /wi, aawi/ |
| 2nd | /yu/ | /unu, aayu, yaal/ |
| 3rd | /(h)i(m)/ ‘he, she, it’ | /de, dem/ |
| /ʃɩ/, /ar/ (object), | ||
| /om, am/ ‘he, she it’ (object) | ||
| /i(t)/ |
The use of a pronoun with distributions as subject, object, and possessive is illustrated by the following: /dem no stie laik dem/ ‘They’re not like them’ and /ʃi doz sii ʃi sista everi en da wiik / ‘She sees her sister every weekend’. The common habitual marker /doz/ is often associated with eastern varieties. Many Caribbean varieties do not overtly mark this distinction; others use /de/ (see progressive aspect above) for it. The pronominal form /unu/ is the only pronoun not derived from a superstrate source.
Possession is marked solely by word order in Caribbean varieties, rather than by a combination of word and bound inflectional morphology: /mi brada uman de de/ ‘My brother’s wife/woman is/was there’.
Anglophone Caribbean varieties often mark infinitives with some reflex of for – /fu/ or /fi/ – as in /unu hafu du it/ ‘You (pl.) have to do it’; /a fiil fi smuok/ ‘I feel like smoking’. The marker to is used as well.
Pluralization is most often marked by a postnominal /dem/ rather than a bound inflectional morpheme (the marker may occur in prenominal position as well, as it does robustly in many dialect varieties), unless a number of more than one has been established previously within the phrase or clause: /di daag dem/ ‘the dogs’, /di daag an dem/ ‘the dogs (and the other dogs)’, /dem daag/ ‘those dogs’; cf. /di tʃri daag/ ‘the three dogs’, in which no redundant plural marker is necessary once a plural number has been established. Nouns like ‘people’ and ‘children’, which are not usually overtly pluralized morphologically in other varieties of English, are often pluralized in Caribbean Englishes: /hau di pipl dem trai fi liv/ ‘How do the people manage to live?’
Negation in the Anglophone Caribbean is designated by a preverbal negator, which is usually some reflex of ‘no’, ‘not’, or ‘never’: /ʃi no siŋ/, /ʃi na(t) siŋ/, /ʃi neva siŋ/ ‘she didn’t sing’. Other negators are derived from ‘don’t’ and ‘ain’t’: /ʃi duon iit/ or /ʃi en iit/ ‘she didn’t eat’.
One of the most heavily researched areas of restructured Englishes of the Caribbean (though, again, this feature is not structurally diagnostic, since it is also found in languages that are not considered creoles) is serial verb constructions (Winford 1993). Verbs may occur serially with no intervening coordinator or infinitival marker: /dem gaan iit/ ‘They went to eat/they went and ate’; /yu waan paas di die wi mi/ ‘Do you want to spend the day with me?’ Other dialects of English reveal similar constructions but usually with imperative forms: ‘Come bring me my food’ and ‘Go get my car’.
An English‐derived or ‐based variety is assigned that designation because the greatest part of its basic, everyday vocabulary is based upon colonial varieties of English heard in the seventeenth through the twentieth centuries. Consequently Caribbean varieties maintain words that may be archaic in other contemporary varieties: /krabit/ ‘mean, disagreeable, rough, cruel’ can be traced to Old English and to usage more recent usage in Scotland; /fieba/ (< favor) ‘to resemble’ (e.g. /ʃi fieba yu/ ‘she resembles you’) is also heard in the American South; /beks/ (< vex) ‘to anger’ (e.g. /wa mek yu beks so/ ‘Why are you so angry?’; /beg/ ‘to ask’ (e.g. /a wan beg yu wan tiŋ/ ‘I want to ask you something’, which is preserved in many other English dialects in the frozen expression I beg your pardon.
Other lexemes reflect historical sailors’ jargon. Since movement by sailing ships was the only mode of transportation to the Western Hemisphere before the twentieth century, the language of sailors influenced passengers (slaves, colonists, settlers, and immigrants), and thus the varieties that emerged at these destinations in the Americas: /haal/ (< haul) ‘to pull’; /gyali/ (< galley) ‘the kitchen (of any household)’.
Of course, many African‐language‐derived words may be found in the lexicon. More than a few words derive from Twi, a language spoken in Ghana and the Lower Guinea Coast (Aceto 1999b): /koŋgosa/ ‘gossip’, /fufu/ ‘a common dish made from of yam and plantains’, /mumu/ ‘dull, dumb, silent’, /potopoto/ ‘mud, muddy’ are common, but there are many others. Other African languages are represented as well: /dʒuk/ ‘to stab, to pierce, to have sex with’ appears to be from Fulani, a language spoken in Nigeria, among other locations on the Guinea Coast.
The influence of African words can be traced beyond straightforward borrowings. Many expressions in the Caribbean which seem to be calques, such as big‐eye ‘greedy’, and day‐clean ‘daybreak’, have a number of correspondences with various languages on the Guinea Coast.
It is often claimed that creoles display few if any bound morphemes. McWhorter (1998: 792) qualifies his claim by stating that creole languages rarely have more than one or two inflectional affixes. However, this feature is not diagnostic of creoles as a grouping, since some noncreole languages reveal few inflectional affixes, for example Mandarin; Modern English has only eight.
In the Anglophone Caribbean, bound morphology is illustrated by the common verbal progressive inflection /‐in/, as in /kaalin/ ‘calling’, and /goin/, which may also function as future marker: /mama kaalin me/ ‘Mama is calling me’ and /mi goin iit/ ‘I’m going to eat’. Comparatives are commonly formed with inflections, as well /fas(t)a/ (< fast + ‐er), and derived forms also make use of bound morphemes /wikidnis/ (< wicked + ‐ness). These are productive affixes, not parts of frozen words or expressions. Many researchers insist that, though these bound morphemes (among others) are robustly heard in the Caribbean, they are not features of ‘deep’ or ‘basilectal’ varieties, since they are parts of the grammars of nearly all varieties of English today.
Many Caribbean Englishes have also created phrasal verbs that do not seem to be heard in contributing dialects of British English or in other contemporary varieties: kiss up ‘to kiss, to make out’, wet up ‘to soak’; cf. show up, cook up, mess up, etc., heard in other dialects.
This section is largely based on Holm (1988–1989), Wells (1982), Aceto and Williams (2003), various specific articles referenced, and the author’s own notes from fieldwork.
The off‐glides [ei] and [ou] of many metropolitan varieties of English are often not heard in the eastern Caribbean, where these sounds most often correspond to [e:] and [o:]. However, Childs, Reaser, and Wolfram (2003) suggest that in some Bahamian communities, the sound [ei] can be heard, perhaps due to proximate influence from the US. In the Leeward Islands, specifically Montserrat (Wells 1982: 587), words that historically had long vowels have been shortened and have no off‐glides, e.g. /de/ ‘day’. In many western Caribbean varieties, these same sounds correspond to those with on‐glides, e.g. /fies/ ‘face’ and /guot/ ‘goat’.
Caribbean varieties of English often display a preference for unreduced vowels, for example, [abɪlɪtɪ] ‘ability’ and [tawɪl] ‘towel’, where other dialects often exhibit schwa. Many varieties in the eastern Caribbean (except Bajan) have no midcentral vowels.
The low front vowel /æ/ found in metropolitan varieties of English in words such map or cat is often realized further back on the tongue as /a/ in the Caribbean: /kyat/. There is often an off‐glide after velar consonants and before this vowel, for example, /gyaadin/ ‘garden’. However, some varieties of English in the Turks and Caicos, as well as Bermuda, exhibit the use of /æ/.
Words commonly found in all dialects of English often reveal different vowels in Caribbean varieties, perhaps as preservations of older regional British pronunciations that have changed since then, e.g. /spail/ ‘spoil’, /bail/ ‘boil’, especially in the western Caribbean.
Except for varieties of English in Barbados, and to some degree in Jamaica and Guyana, postvocalic /r/ is often not heard in the Anglophone Caribbean. Bajan English is recognized by its full rhotic nature at all levels of society. Other dialect areas of the Caribbean are nonrhotic after vowels (e.g. Trinidad and the Bahamas), while others are highly variable (e.g. Guyana). In the nonrhotic dialects, additional phonemes are often created, for example, /nea/ ‘near’, /foa/ ‘four’, but these forms can be heard in range of English dialects not associated with the Caribbean, especially on the eastern seaboard of North America in New England.
Many dialects of Caribbean English (e.g. Bahamian, Bermudan, and Vincentian) may alternate [w], [β] (the voiced bilabial fricative) or [υ] (the voiced labiodental approximant for words which in metropolitan varieties begin with /v/, for example, [wɪlidʒ] ‘village’. This feature could be related to contributing dialect varieties spoken by colonists in the eighteenth century which contain this same alternation (e.g. Cockney), or possibly to early speakers of African languages which lacked the /v/. Some communities may reveal /b/ where other varieties display /v/, for example, /bɛks/ ‘angry’, /rɪba/ ‘river’.
In the Leeward islands (Antigua, Barbuda, St. Kitts, Nevis, Montserrat, and Anguilla), unlike in Jamaican and other western varieties, /h/ is most often not dropped from the beginnings of words. So‐called /h/‐dropping is common in Jamaica and the Bahamas. It also occurs in British Cockney, which is often cited as the source of this feature. In dialects with this feature, new homophones are created, as well as new forms, for example, /iɛr/ ‘hair/air’, /uol/ ‘whole’, /aaf/ ‘half’. On the other hand, dialects with /h/‐dropping often have /h/ as well as /w/ word initially where it is not heard in other dialects of English, for example, /heg/ ‘egg’, /wogli/ ‘ugly’.
The neutralization of /ð/ and /θ/ as /d/ and /t/, respectively (e.g. /tɪŋ/ ‘thing’ and /fada/ ‘father’), is a common feature of many English dialects across the globe, including in the Caribbean. Any dialect with this feature creates new homonyms, for example, /tɪn/ ‘thin‐tin’, /fet/ ‘faith‐fate’, /do/ ‘though‐dough’, /brid/ ‘breathe‐breed’, /tri:/ ‘three‐tree’, /tru/ ‘through‐true’. Words with an /r/ following an alveolar stop are often palatalized, for example, [tʃrii], [tʃru]. In Kokoy, a variety spoken in Dominica, /θ/ often corresponds to /f/, not /t/, for example, /fri/ ‘three’, /fru/ ‘through’ (Aceto 2010). In St. Eustatius, many speakers display both the interdental fricatives as well as their stop correspondences (Aceto 2006). Cutler (2003) and Williams (2003) make similar observations about the English varieties spoken on, respectively, Grand Turk Island and Anguilla.
In many Caribbean varieties, as is common beyond this area as well, word‐final /t/ and /d/ preceded by an obstruent in clusters is often not realized, for example, /lɛf/ ‘left’, /nɛs/ ‘nest’, /ak/ ‘act’, /sɛn/ ‘send’, /bɪl/ ‘build’. However, word‐final clusters of a nasal and a voiceless consonant are heard, for example, /lamp/ ‘lamp’, /tɛnt/ ‘tent/tenth’, /bank/ ‘bank’. In rhotic dialects, clusters in codas are also realized in combination with liquids, for example, /mɪlk/ ‘milk’, /ʃɛlf/ ‘shelf’, /part/ ‘part’, /hard/ ‘hard’. Other consonant cluster combinations occur freely, for example, /aks/ ‘ask’, /baks/ ‘box’, /sɪks/ ‘six’. In some varieties, word‐initial clusters are dispreferred, for example, /taat/ ‘start’, /tan/ ‘stand’, /tap/ ‘stop’.
For linguists working as researchers or in the field gathering data it is understandable or even necessary to highlight the uniqueness of a particular language variety under examination. As one can see from this chapter, the Englishes of the Caribbean share many similarities with forms heard across English‐speaking communities in the Americas in general. In parts of the American South, many long‐term locals, including European Americans, regularly reveal constructions such as ‘she mean’ with no explicit verb, or say < with > with a word‐final [f], [daet] with a word‐initial stop (see above), and verbal forms like ‘my brother’s out a‐walkin’. They also may say ‘carry’ for take, ‘cuss’ for curse, ‘reach’ for arrive, ‘wait on’ for wait (see Deuber 2011: 44 for a list of purported “Caribbeanisms”). Certainly those words are heard in the Caribbean but may be used in other English‐speaking regions as well. Clearly, the history of European contact on the eastern seaboard of what is today the US reveals similar colonial practices, speech communities, as well as population movements to and from their Caribbean counterparts during the British colonial period and thereafter. Once again it is the constellation of forms (words, sounds, structures) that is heard in a well‐defined geographical area that represents the language of that specific speech community, not any one form or even a handful of features that are diagnostic of it.