Contact Linguistics. Chap

Exercise: The following is an extract from modern Sranan Tongo. Try to identify those aspects of phonology morphology and syntax that distinguish it from its lexifier, English. For instance, what kinds of phonological changes have occurred in English words? How are time and aspectual reference conveyed? etc.


<sub>A. Ma fa a pikin fu Oom N ø du dede dan?
But how the child of uncle N do die then
'But how did Uncle N’s child actually die, then?'

B. Ayi, a wan moi vraag, noh. Wel, mi ben de a Novar.
Yes, COP one nice question, TAG. well, I PAST be LOC Novar
Mi ø de a Novar, mi ø de nanga wan man, ben (e) taki tori nanga wan man.
I be LOC Nover, I be with one man, PAST IMP talk story with one man

('Yes, it’s a good question, no? Well, I was at Novar. I was at Novar, I was with a guy, I was talking to a guy.')

Dan wan man ø ry kon nanga wan bromfiets, ma a man di ø kon teki mi,
Then one man ride come with one bike, but the man REL come take me
en nanga den suma dati no ø de bun. Dan a man ø tan a strati,
he and the-pl people DEM NEG COP good. Then the man stay LOC street
dan a e bari kari mi ‘D, D, Oom N, Oom N, Oom N'.
then he IMP shout call me 'D, D uncle N. uncle N uncle N

('Then a guy came on a moped, but the guy who came to pick me up, he and those people (I was talking to) were not on speaking terms. So the guy stayed in the street’ then he called me 'D, D, Uncle N, Uncle N, Uncle N')


Dan mi e taki taki a man ø abi mi fanowdu, san a man no e kon.
Then I IMP talk COMP the man have me need why the man NEG IMP come
Mi no e go. Okay, dan a man, a man di mi nanga en e taki,
I NEG IMP go. okay, then the man, the man REL I and he IMP talk,
a man ø taki, yu kan go want a man no e kon. A no o kon dya.
the man say, you can go because the man NEG IMP come. He NEG FUT come here


('So I was saying that the guy needed me, why wasn’t the guy coming to me? I’m not going. Okay, then the guy, the guy I was talking to, the guy said, you can go for he’s not going to come. He won’t come here.')


Di mi ø go a man ø taigi mi taki taki ‘Oom N, sidon a baka a fiets.
When I go the man tell me say say 'Uncle N, sit LOC back the bike
Oto ø naki a boi fu yu.’ Mi ø taki ‘san! oto ø naki en.’
Car hit the boy of you. I say 'what, car hit him'
Mi ø taki efu oto naki en a o dede. Merkwaardig....
I say if car hit him he FUT die. Oddly


('When I went, the guy told me 'Uncle N, sit on the back of the bike. A car has knocked down your son.’ I said ‘what! a car knocked him down?’ I said if a car has knocked him down he’ll die. Oddly enough....</sub>


6. Some aspects of creole grammar.

Given the wide differences across creoles, it is difficult to generalize about their structural characteristics. This is true even if we limit our attention to the more radical creoles. However, certain general features can be identified.

6.1. Creole phonology.

In general, creole phonology appears to be the result of varying degrees of reinterpretation of superstrate phonology in terms of substrate phonetic categories and phonological rules. Relatively little research has been done so far on this aspect of creole structure, far less its sources. Hence we have only a vague picture of the substrate contribution.

It would appear that creoles generally preserve sounds that are similar in the input languages. Among the consonants, these include a variety of stops such as /p, t, k, b, d, g/, and nasals like /m, n/. Where there is a mismatch between superstrate and substrate sounds, the former are generally substituted by their closest equivalents in the latter. This kind of L1 influence is of course characteristic of second language acquisition in general. It helps explain, for instance, the substitution of /b/ for /v/, /t/ for /Q/, d for /D/, /l/ for /r/ and so on. Some radical creoles such as the Surinamese creoles and Krio (Sierra Leone) preserve West African words containing coarticulated and prenasalized stops (e.g. /gb, kp, mb/ etc.) which are fairly common in their substrates(Holm 1988:127-130).
However, it is also clear that creoles owe certain aspects of their phonology to their superstrate sources too. For instance, several distinctive features of Caribbean English-lexicon creoles can be traced back to regional English dialects. These include loss of initial /h/ in word like ‘ospital,, the palatization of /k.g/ in words like /kyari/ “carry”, /gyal/ girl etc., insertion of the glide /w/ in words like bwai “boy”, gwine “going”, etc. Even features such as the substitution of /b/ for /v/ in words like bery “very”, /t/ for /Q/ and /d/ for /D/, were also typical of some regional dialects of English, and of nautical English too (Lalla & D’Costa 1990: 51- 60, 100,). Using contemporary Standard English phonology as a point of comparison with English-lexicon creole phonology would therefore be quite misleading with regard to the superstrate sources of the latter.
Creole vowels similarly tend to be those common to their source languages, though some vowel systems are more complex than others are. Practically all creoles have at least the five vowels /i, e, a, o, u/, though some have /E/ and /O/ in addition. Holm (1988:114) suggests that “the seven vowel system found throughout much of West Africa is also the basic system of the Atlantic creoles.” Again, however, some aspects of creole vowel systems appear to derive from superstrate sources. With reference to English-lexicon creoles, Lalla & D’Costa (1990:60-64 107) mention such features as the use of /a/ for /O/ in words like drop, /ie/ for /e/ in words like game, /ai/ for /Oi/ in words like boil, etc.
There is evidence of substrate influence on creole phonotactics too. A general feature is the simplification of both initial and final consonant clusters. In some English-lexicon creoles, initial /s/ is lost in clusters like /sp/, /st/ and /sk/, yielding, or instance, puun ‘spoon’, tan ‘stand’ and kin ‘skin’. Interestingly, French-lexicon creoles generally preserve such initial clusters (Parkvall 1999:32). Similarly, /t, d, k/ are lost in final clusters like /st, sk/ and /ld/, yielding laas ‘last’, ool ‘old’ des ‘desk’ etc. This phenomenon is also common in French creoles. The simplification of final consonant clusters also has models in English regional dialects (Lalla & D’Costa 1990: 100, 107).
Also typical of many Atlantic creoles is the use of paragoge (the addition of a vowel, especially to European stems that end in a consonant). This yields words in Sranan like luku ‘look’, kiri ‘kill’, etc, containing a CVCV syllable pattern typical of many West African languages. Lalla & D’Costa (1990:65) suggest that “syllabic structure in early JC was West African” citing now obsolete words like wharra “why”, killee, preachy, etc. found in earlier JC texts. Here again, they suggest that this pattern might have been reinforced by a similar syllabic structure found in southwest dialectal English words such as hearee “to hear”, caree “to care” etc.
Finally, certain suprasegmental features found in some Atlantic creoles seem to have their source in substrate influence. It has been claimed that some Caribbean English-lexicon creoles preserve vestiges of West African tone systems, though they are not tone languages in the strict sense (Carter 1979:3, 1987; Devonish 1989). Also of interest is the apparent retention in some Atlantic creoles of tone or pitch based distinctions between “intensive” and “distributive” reduplications, Thus in Jamaican we find smaal smaal (even pitch) “very small” vs smaal smaal (falling pitch) “smallish” (Carter 1987, Gooden 2001).
In general, much more research needs to be done before we have an adequate understanding of creole phonology and its origins.


6.2. Creole lexicon.

There are three major sources of creole lexicon: input from the superstrate, which supplies the bulk of the vocabulary, retentions from substrate languages, and internal innovations.

Many superstrate items have their phonological shapes and meanings fairly well preserved in creoles. Most of these come from the regional dialects introduced by settlers and servants in the early period of colonization. But a significant number of items also came from so-called “nautical” language, as used by sailors and other seafarers. In Jamaican creole, for instance, we find such survivals as jerk “ barbecued” (originally “to salt and smoke dry meat”), cow “any horned cattle”, kotlash “machete” and tsokop “chock full” (Lalla & D’Costa 1990:101). Other items from English dialects include buss “kiss”, maaga “thin” (eerie “smart, excellent” the possible source of the popular Rastafarian term irie [airi] “excellent, very fine” (Lalla & D’Costa p. 226). Many JC compounds also derive from English dialectal sources, including morning-time, self-same “identical” was-mout “liquor” and others. Several such items are also found in Suriname, suggesting a similar nautical or dialectal English source. They include no so “otherwise”, too much “very” before time “long ago”, etc.
Superstrate-derived lexical items often undergo semantic and even categorial change, sometimes under substrate influence. For instance, in JC, wind (vb) has come to mean ‘twist and turn [the hips] provocatively”, while sweet is a verb meaning “please greatly.” In the Surinamese creoles, English prepositions like round and up are reanalysed as the verbs lontu “go around” and opo “lift” respectively. The Atlantic creole cognates of foot and hand refer to the entire limb, including ‘arm’ and “leg” respectively, not just to the extremity. These kinds of semantic reanalysis are also found in extended pidgins. Thus Tok Pisin as (arse) means not just “buttocks,” but also “cause, foundation”. Similarly, bel means not just ‘belly,’ but also “seat of the emotions.” These parallel the range of meaning expressed by equivalent words in the substrate.
Creoles also derive a significant number of words from their substrates, and to a lesser extent from other languages, usually via pidgin varieties of the latter. These tend to be lost over time, so earlier creole texts provide a clearer picture of the contribution from these sources. Lalla & D’Costa (1990:77) identify 250 words in earlier (17^th to 19^th century) JC texts, which required etymological commentary. Of these, 180 or three fourths were from English dialects, 36 were West African, 14 “Hispanic” and 25 of unknown or multiple origin. Modern JC preserves a fair number of the West African items, including backra “white person, boss” (< Ibo, Efik mbakara “white man”) kongkongsa “deceit, deceitful” (< Twi Nkongkongsá “falsehood, deceit”); nyam “eat, food” (multiple W. African sources); tata “father, elder” (multiple W. African sources) and so on.
Another interesting aspect of creole lexicon is the formation of compounds modeled on both superstrate and substrate patterns. In JC, compounds derived from Early Modern English dialects provided not just a source, but also a model for lexical expansion. Thus we find innovations like man-crab “male crab”; good-belly “good-natured”, hard-ears “stubborn”, pick-mouth “troublesome” etc. Some compounds are clearly calques on W. African equivalents, and are widely distributed across Atlantic English lexicon creoles. They include CEC forms like bad-mouth “to speak ill of” (cf. Mandingo da-jugu, Hausa mugum-baki “bad-mouth”) eye-water “tears”, suck-teeth “a disapproving sound” (multiple W. African sources) and cut-eye “a scornful look” (Rickford & Rickford 2000:95). A final word formation strategy that has its sources partly in the substrates and partly in internal motivation is reduplication, discussed in the previous section. The richness of creole derivational patterns belies the myth that creoles lack bound morphology, an idea that no doubt arose because creoles generally lack inflectional morphology, (See DeGraff 2001 for a discussion of the richness of Haitian Creole morphology).


6.3. Creole morphology and morphosyntax.

The general lack of inflectional morphology in creoles is the result of processes of simplification that emergent creoles share with pidgins, or derive from pidgin-like ancestors. Hence categories such as case, agreement, etc., are lacking in these languages. Simplification also has consequences for various aspects of creole morphosyntax. For instance, case and gender distinctions are reduced or eliminated in the pronominal system, though number distinctions are preserved. Table 2 illustrates the pronominal system of Jamaican Creole, which is generally representative of Caribbean English Creoles (CEC’s) outside of Suriname.

Table 2: The pronominal system of Jamaican Creole.

Singular Subject Object Possessive

1^st person mi mi mi/fi mi
2^nd person yu yu yu/fi yu
3^rd person im im im/fi im

Plural
1^st person wi wi wi/fi wi

2^nd person unu unu unu/fi unu
3^rd person dem dem dem/fi dem

Some creoles have more complex systems than this. For instance, in their singular pronoun paradigm Principense and other Gulf of Guinea creoles make distinctions between emphatic and non-emphatic forms, as well as between stressed and unstressed forms (Holm 1988:202). This appears to be due to substrate influence, as we saw in the case of Melanesian Pidgin in the previous chapter.

Creole systems of copular predication are also organized quite differently from those of their lexifiers. Most creoles employ overt (and distinct) copulas only in predicate nominal and locative constructions. These are invariant forms, with no distinctions of number and agreement. The following examples from Jamaican Creole illustrate:

(1) Jan a di liida.

John COP the leader. “John is the leader.”

(2) Mieri de a skuul

Mary COP LOC school “Mary’s at school.”

Moreover, creoles generally lack copulas in ascriptive-type predication, that is, those corresponding to predicate adjective constructions in their lexifiers. Example (3) illustrates:

(3) Di mango raip

“The mango is ripe.”

The property items (corresponding to adjectives) in these constructions tend to behave more like verbs. They can be preceded by TMA markers, undergo predicate cleft, occur in comparative constructions with serial paas or moro ‘exceed” and display other verbal properties.

The following sentences from Sranan illustrate:

(4) A pikin e bigi.

The child IMP big
“The child is getting big.”
FOC happy the man happy.
“The man is HAPPY.”
(6) Kofi langa moro yu.
Kofi tall surpass you.
“Kofi is taller than you.”

The degree of verbal behavior varies according to the semantic class of the property item, and from creole to creole (Winford 1997b:256).

6.4. Creole TMA systems.

Perhaps the best known and most debated area of creole morphosyntax is their Tense, Mood and Aspect (TMA) systems. With very few exceptions, creoles express temporal, aspectual and modal meanings via preverbal markers. There are striking similarities in the sources, functions and distribution of these functional heads across creoles, regardless of what their lexifiers were.

Practically all of the Atlantic English-lexicon creoles, for instance, employ a Past tense derived from been, a Future derived from go and a Terminative Perfect expressed by done (or, in the Surinamese creoles, kaba < Portuguese acabar ‘finish’). In addition, the unmarked verb has very similar interpretations and functions (dictated by the context) in all these creoles. Table 3 gives a brief overview of some basic temporal and aspectual categories in three of these creoles.


Table 3. Some Tense/Aspect categories in three Atlantic English-lexicon creoles.


Tense categories Jamaican Guyanese Sranan

Relative Past (b)en bin ben

Future wi go (g)o
Prospective a go a go e go