The Republic of Guyana is approximately 215,000 km2 in size. It lies on the north-eastern shoulder of South America and is bordered by the Atlantic Ocean to the north, Brazil to the south and southwest, Suriname to the east, and Venezuela to the west. Standard Guyanese English is the sole official language of the country. Creolese is a widely used vernacular language variety. It is the native language of the majority of the 700,000 inhabitants of the Republic of Guyana and is also spoken by the several hundred thousand Guyanese in the diaspora. Within Guyana, it co-exists with Standard Guyanese English, which is normally learnt through the formal education system.
Toward the end of the eighteenth century, a significant number of planters and slaves of African origin from both the Leeward Islands and Barbados migrated to the then Dutch colonies of Essequibo and Demerara (Harris & de Villiers 1911: 211). Present-day Guyana is made up of what used to be three separate Dutch colonies: Essequibo, Demerara, and Berbice. When the planters and slaves arrived, they brought with them an Anglo-West African language variety. This represented the first introduction of an Anglo-West African variety into an area in which Skepi (or Essequibo Dutch Creole) was the dominant language variety amongst the plantation slave population of African origin (Robertson 1981).
The subsequent linguistic developments within the introduced English-lexicon creole varieties could have been the result of one of two factors. Either the slaves from the Leeward Islands were numerically superior when compared with those who came from Barbados, or they had greater social influence. Whatever the reason, Creolese has ended up with features suggesting a predominant influence from the Leeward Islands. This can be seen from the presence of the first and second person plural pronoun forms awii and a(l)yu, derived etymologically from all we and all you. These parallel cognate forms in the Leeward Islands, and in the Windwards Islands, which were settled from the Leewards. These forms contrast with wi(i) and wuna ~ unu which are characteristic of the English-lexicon creoles of Barbados, Jamaica, Central America, and Suriname.
“Creole continuum” is an expression commonly used by scholars to allude to the language situation of Guyana. This description suggests a linguistic continuum which ranges from the most conservative or basilectal Creolese, on one hand, to the most standard varieties of English on the other, with an infinite number of varieties in between (Bickerton 1975, Rickford 1979, 1987). By contrast, Devonish (1978, 1991, 1998) proposes the existence of a restricted number of intermediate varieties. These are produced by a constrained mixing of features from the acrolect and the basilect. There are, in addition, a small number of features peculiar to the intermediate varieties (the mesolect).
Creolese has a vowel system with twelve phonemic oral vowels and a marginal thirteenth. As shown in the vowel charts in Table 1, there are five simple vowels and seven complex vowels in the language.
(a) simple vowels | ||
i | u | |
e | o | |
a | ||
(b) complex vowels | ||
ii | uu | |
ee | oo | |
aa | ||
ai | ou |
The complex vowels comprise five double vowels, corresponding to the simple vowels, and two diphthongs. The double vowels are simply the result of a lengthened articulation of the equivalent simple vowel.
The CV Phonology approach of Clements & Keyser (1983) distinguishes between complex syllabic nuclei of the structure VC and those of the structure VV. The former would represent long vowels at the phonological level, the latter double vowels. Devonish (1989b: 76-77) identifies the entire range of syllable final consonant clusters possible in Creolese and their distribution with phonetically short and phonetically long vowel nuclei. Based on this data, Devonish (1989b: 77) concludes, “In [Guyanese Creole], unlike in English, the behaviour of syllable final clusters […] is in no way affected by the kind of vowel nucleus which precedes”. This is the basis for the claim that the second element in complex syllabic nuclei in Creolese fills a V-slot. This would mean that such nuclei are phonologically VV, i.e. double vowels, rather than VC, long vowels as in English. Hence, the five simple vowels could be termed the building blocks of the more complex segments.
In Creolese, all the non-low phonetically long vowels have an additional feature. They are more tense, i.e. pronounced much closer to the periphery of the mouth. They are also produced at a point higher in the mouth than their short equivalents.
In the case of diphthongs, the low vowel /a/ combines with the high front vowel /i/ to produce /ai/, and /o/ combines with the high back vowel /u/ to produce /ou/.
Table 2 shows examples of the vowels as used in Creolese.
phonemic | phonetic | example | |
simple vowels | |||
/i/ | [i] | brik | ‘stone’ |
/e/ | [ɛ] | led | ‘lead pencil’ |
/a/ | [a] | bak | ‘back’ |
/o/ | [o] | kop | ‘cup’ |
/u/ | [u] | buk | ‘book’ |
double vowels | |||
/ii/ | [i:] | biif | ‘beef’ |
/ee/ | [e:] | keen | ‘cane’ |
/aa/ | [a:] | baal | ‘ball’ |
/oo/ | [o:] | goot | ‘goat’ |
/uu/ | [u:] | skuul | ‘school’ |
diphthongs | |||
/ai/ | [ai] | klaim | ‘climb’ |
/ou/ | [ou] | kou | ‘cow’ |
In Creolese, there is a marginal phonemic contrast, involving /e/ realized as [ɛ] and /e:/ as [ɛ:]. This contrast, however, is restricted to environments immediately preceding voiced stops. What exists, as a result, is a small number of near minimal pairs such as shown in the following examples:
/led/ [lɛd] ‘lead (pencil)’
/re:d/ [rɛ:d] ‘red’
/zeg/ [zɛg] ‘the name of a dance’
/e:g/ [ɛ:g] ‘egg’
In relation to [ɛ] in the environments preceding voiced stops, we have a new contrast, albeit a very marginal one. This is one in which the longer version, [ɛ:] is neither higher nor more tense than its single equivalent. However, like other complex syllabic nuclei in Creolese, it does not restrict the possibility of two non-liquid consonants occurring in the coda, as in [hɛ:dz] /hɛɛdz/ ‘heads (in the spinning of a coin)’. This demands an analysis similar to that of the other complex syllabic nuclei, i.e. that of VV, a double vowel, rather than VC, a long vowel. In its behaviour, this [ɛɛ] double vowel of restricted distribution in the Creolese vowel system violates the symmetry already described for the rest of the system.
There are thirty consonants in Creolese. The consonants are essentially the same as those recognized for the Jamaican Creole language. For both languages, the same orthographic conventions are used (the Cassidy phonemic representation, cf. Cassidy & Le Page, 1980: xxxix).
IPA | description | Creolese example | ||
p | /p/ | voiceless bilabial stop | put | ‘put’ |
b | /b/ | voiced bilabial stop | biit | ‘beat’ |
t | /t/ | voiceless alveolar stop | tek | ‘take’ |
d | /d/ | voiced alveolar stop | said | ‘side’ |
k | /k/ | voiceless velar stop | kooknot | ‘coconut’ |
g | /g/ | voiced velar stop | goot | ‘goat’ |
ch | /tʃ/ | voiceless alveo-palatal affricate | chail | ‘child’ |
j | /dʒ/ | voiced alveo-palatal affricate | jomp | ‘jump’ |
b | /ɓ/ | voiced bilabial implosive | baala | ‘ball’ |
c | /c/ | voiceless palatal stop | kyaar | ‘car’ |
gy | /ɟ/ | voiced palatal stop | gyaaf | ‘chat’ |
sh | /ʃ/ | voiceless alveopalatal fricative | mashiin | ‘machine’ |
zh | /ʒ/ | voiced alveopalatal fricative | mezho | ‘measure’ |
h | /h/ | voiceless glottal fricative | hool | ‘old’ |
m | /m/ | bilabial nasal | miit | ‘meet’ |
n | /n/ | alveolar nasal | nais | ‘nice’ |
ny | /ɲ/ | palatal nasal | nyuuz | ‘news’ |
ng | /ŋ/ | velar nasal | geng | ‘gang’ |
ɾ | /ɾ/ | alveolar flap (approximant) | prapa | ‘very’ |
r | /r/ | alveolar trill | ron | ‘run’ |
p | /ɸ/ | voiceless bilabial fricative | pan | ‘on’ |
b | /β/ | voiced bilabial fricative | abi | ‘we’ |
f | /f/ | voiceless labio-dental fricative | fiid | ‘feed’ |
v | /v/ | voiced labio-dental fricative | vuot | ‘vote’ |
s | /s/ | voiceless alveolar fricative | sik | ‘sick’ |
z | /z/ | voiced alveolar fricative | zuu | ‘zoo’ |
l | /l/ | alveolar lateral approximant | laaf | ‘laugh’ |
y | /j/ | palatal approximant | yuut | ‘youth’ |
w | /w/ | labio-velar approximant | waam | ‘warm’ |
b | / mb/ | pre-nasalized voiced bilabial plosive | baad | ‘bad’ |
Creolese is in the main an isolating language, with the kind of syntactic information associated with noun phrases (NPs) being signalled by free roots. NPs may contain determiners, possessive and modifying adjectives, and the head noun. There is, in addition, a system of personal and indefinite pronouns which can substitute for NPs.
The syntactic class of noun in Creolese extends across the semantic range beginning with human entities such as maan ‘man’ and uman ‘woman’. The range runs through animals such as daag ‘dog’ and kyat ‘cat’, other living things such as chrii tree’ and graas ‘grass’, to inanimate objects such as boot ‘boat’, and kyaar ‘car’. At the far end of the range are abstractions conceived of as having some physical existence such as wok ‘work’, (s)toorii ‘story, quarrel’, and main ‘mind’.
Bare NPs lacking a determiner may have a generic meaning, designating members of an entire class of entities. An example of a generic NP can be seen in the two occurrences of the phrase wait daag ‘white dogs’ in the proverbial expression:
(1) wait daag no nyam wait daag
white dog neg eat white dog
‘White dogs don’t eat other white dogs.’
In example (1), the NP wait daag does not differentiate between members of the overall class of ‘white dogs’. The reference is to any and all non-specified members of that class. In that sense, we can conceive of generic NPs as consisting of a continuous whole rather than of individual, differentiated members. Creolese does have a determiner slot which can be filled by definite or indefinite articles. These zero-marked NPs signal the generic, contrasting with NPs marked as definite by a determiner.
Creolese has an indefinite article word wan that is distinct from the numeral wáan ‘one’, occupying a position before the head noun and its modifiers. Where speakers wish to refer to an unspecified single member of a class of item, the indefinite form wan is employed. We see the use of wan in (2):
Indefinite NPs are very often employed in the first reference an NP gets in the discourse. Bare noun phrase forms, in addition to the generic interpretation, may also have the same pragmatic reading as noun phrase forms with wan. This is seen in the following example. It consists of a piece of discourse involving a sequence of three sentences at the start of a story. A man, a coconut, a cutlass, a monkey, and a razor are being introduced. Of the five items introduced, three are introduced with wan, and two, including a repetition of the item for ‘coconut’, with the bare NP.
Based on the location of items such as the definite article di (di kooknot ‘the coconut’), we propose that there is a Determiner (Det) slot which occurs at the beginning of the Determiner Phrase (DetP). It is optionally filled. The range of potential fillers includes definite articles, demonstrative adjectives, possessive NPs and possessive pronouns. Evidence that this slot exists even when unfilled comes from observations regarding syntactic reduplication in NPs. This will be discussed in §5.5 under Pluralization.
Demonstrative pronouns largely derive from their demonstrative adjective equivalents. Complex pronominals are restricted to the plural by contrast with the adjectival forms which have a complex form in the singular as well.
Table 4. Demonstratives |
|||||
Dem.pronoun |
dem.adj. |
||||
singular |
da dis |
da N da N dis |
|||
plural |
dem |
dem da |
dem N dem dis N dem ya N |
dem N da ~ dem N de dem N dis dem N ya |
dem da N |
As can be seen from the following examples, Creolese observes a three-way contrast for its demonstratives too:
(4) dis buk ya
‘this book (here)’
(5) da buk de
‘that book (there)’
(6) da chrii yanda/oova so
‘that tree (over there)’
In Creolese, it is possible to have either the possessor preceding the possessum or the possessum followed by the possessor. In the case of animate nouns, the pattern of possession involves the sequence possessor – possessum. This pattern applies whether the possessor is (a) a reduced and cliticized person pronoun (cf. 7a), (b) a full and emphatic form of the personal pronoun (cf. 7b), or (c) a noun phrase headed by a noun (cf. 7c).
(7) a. mi gu, mi put mi paasl
1sg go 1sg put 1sg parcel
‘I went and put down my parcel.’ (Rickford 1987: 230, lines 1125-1126).
b. […] mii tuu gyal dem a daab batm hous
[…] 1sg two girl 3pl asp daub bottom house
‘[…] my two girls were daubing [with mud and cow dung] the yard under the house.’
(Rickford 1987: 229, lines 1114-1115)
c. Jooana muma kaal shi
Joanna mother call 3sg
‘Joanna’s mother called her.’ (Dolphin 1996: 98)
Personal pronouns share an important pragmatic feature. They represent entities known to the interlocutors and are, therefore, all definite. In addition, they distinguish number, there being both singular and plural personal pronouns in the language. The link between definiteness and pluralization is one which is part of the syntax, semantics, and pragmatics of noun phrases as a whole within the language.
Monosyllabic pronouns produce differentiation between independent and dependent personal pronouns. The independent monosyllabic forms have prominent prosody, and usually in addition either a long vowel or an /h/ in the onset to differentiate them from the dependent ones.
subject | object | independent pronouns |
adnominal possessives | |
1sg | mi | mi | mii | mi, mii |
2sg | yu | yu | yuu | yu, yuu |
3sg | ii | ii | hii | hii, ii |
1pl | awii | wi | wi | awii |
2pl | a(l)yu | a(l)yu | a(l)yu | a(l)yu |
3pl | dem | dem | dém | --- |
In strict terms, Creolese does offer as well a compound expression for ‘how’ which is wa fashan. This is, however, very rare in modern speech. More common are the three compound expressions used to express wi said ~ wich paat ‘where’ and wa taim ‘when’.
In Creolese, the pluralizer of the definite NP is dem, which is the 3rd person pronoun. In Creolese, only definite NPs can be pluralized. The definite pluralizer has an alternative form, an dem, which includes the coordinating conjunction an. This suggests behaviour that parallels iterative syntactic reduplication by repeating the definite NP using the pluralizing pronoun. Thus, for a definite plural such as di puliis dem [the police PL] ‘the policemen,’ the interpretation is di puliis (an) di puliis (an) di puliis etc.
The use of the 3PL pronoun as a pluralizer here merely disguises the fact that we are dealing with the iterative syntactic reduplication of definite NPs. We are thus able to further argue that mi broda an mi broda an mi broda represents the underlying form of mi broda (an) dem. The fact that the repetitions are definite, that is they are presumed to be known to the hearer, allows the 3rd person plural pronoun, dem, which is by definition definite, to substitute for all the additional manifestations of mi broda.
There are two types of definite pluralization, the multiplicative and the associative. The latter generally involves the pluralization of NPs with a human referent.
(8) Multiplicative
a. Intensive
di brait schudent doz riid buk buk buk buk til ii taiyad
the bright student does read book book book book until he tired
‘The bright student reads many books until he is tired.’
b. Distributive
di ruum naastii buk buk buk de aal oova
the room nasty book book book is all over
‘The room is nasty. There are books all over (the place).’
(9) Associative
di buk di buk di buk di buk
NP1 = NP2 = NP3 = NP4
di buk an ( )
di buk dem
‘the books’
The linguistic structure is additive, but the semantic interpretation is associative.
(10) Siita dem
Siita them
‘Sita and her friends’
(11) mi faada dem a wok
my father them is work
‘My father and his friends are working.’
NPs can have adjective modifiers. These agree neither in gender nor in number with the noun. The modifying adjective normally precedes the noun, e.g. wan shaap reezaa ‘a sharp razor’. In a construction such as di kozn wo swiit ‘his cousin that was drunk/intoxicated’ (Rickford 1987: 155, line 463), where swiit means ‘intoxicated’, it is only used “in predicate and not in attributive position”, hence ruling out *ii swiit kozn (p. 150).
In describing “verbs” in Creolese, our preference would be for the term “predicator”. This is because the range of items behaving like prototypical verbs includes items that in many other languages would be treated as adjectives. There is, in Creolese, little evidence, when we examine behaviour in the predicate, to support the existence of a separate adjective class. However, in the interest of ensuring comparability across the various descriptions of creole languages, we will, with considerable caution, employ the term verb.
A key feature used in classifying verbs is the way in which their semantic arguments are expressed syntactically. At the syntactic level, this is done by constituents that usually appear in some kind of order and which have meanings such as agent and patient assigned to them. The range of meaning covered by verbs includes processes, motion, action, sensation, emotion, and manipulation (Payne 1997: 57-61). All of these either signal transformations in the state of one of the participants, or states, broadly defined, within which one of the participants exists.
We are able to classify verbs in Creolese into three sub-types, which we will refer to as Verb1, Verb2 and Verb3. We will focus briefly on constituent ordering, before proceeding to analyze the behaviour and functions of each of the sub-types. We will also cover semantic stativity and how it is expressed in the syntactic sub-types. One element of this expression is the role of these sub-types of verb as full lexical modifiers, performing adjectival functions.
Prototypical Verb1 items (i.e. verbs of the Verb1 subtype) are those which, in addition to being transitive, signal a clear change of state of the patient object. The change is viewed as effected by the agent subject. Prototypical members of this class would include items like kuk ‘cook’ and mash ‘mash’. By contrast, there are non-prototypical members of this class, involving patient objects not obviously changed by the agency of the subject. These include si ‘see’, noo ‘know’, gi ‘give’ and sel ‘sell’.
Prototypical Verb2 items signal states which are most easily perceived as temporary, and therefore interpretable as states into which the patient subject has entered. These include colour terms such as re:d ‘red’, wait ‘white’, as well as other temporary states such as (h)at ‘hot’, and kool ‘cold’. By contrast, non-prototypical Verb2 items would include those perceived to be relatively permanent and, therefore, not the result of a process. These include items marking states related to dimensions, such as big ‘big’, smaal ‘small’ and taal ‘tall’, as well as internal states such as baad ‘bad, wicked’ and nais ‘good, attractive’.
In the examples below, we see examples of Verb1 and Verb2 items with patient subjects. Notably, these verbs can all take the emphatic completive aspect marker don. This is used to convey the meaning that the subject has entered the state and exists within it. The implication, therefore, is that the subject is perceived to have, at some point previously, entered the state. This construction is possible even with non-prototypical Verb2 items, where there is a suggestion of the subject being in a permanent state.
(12) Verb1: (a) vs. (b)
a. di klooz dem wash
def clothes pl wash
‘The clothes are washed.’
b. di klooz dem don wash
def clothes pl compl wash
‘The clothes are already washed.’
Verb2 (Prototypical): (c) vs. (d)
c. di floor re:d
def floor red
‘The floor is red.’
d. di floor don re:d
def floor compl red
‘The floor is already red.’
Verb2 (Non-Prototypical – Internal State): (e, g) vs. (f, h)
e. di gyal nais
def girl nice
‘The girl is pretty.’
f. di gyal don nais
def girl compl nice
‘The girl is already pretty.’
g. di maan shaat
def man short
‘The man is short.’
h. di man don shaat
def man compl short
‘The man is already short.’
Unlike Verb1 and Verb2 items, Verb3 items can neither take an object nor a patient subject. Thus, under no circumstances could the subject of (13) be interpretable in any other way than as the agent. The traditional description for this type of verb would be ‘intransitive verb’.
Creolese has a system of thirteen pre-verbal markers relating to tense, aspect, and mood. These are described in the following three subsections (§6.4–6). We begin with aspect markers.
In Creolese, there is a marker whose core function is to signal progressive aspect. This marker, however, has extended roles which signal the inceptive and the habitual as well. There is, in addition, a dedicated habitual aspect marker. These markers all occur in pre-verbal position.
The example below illustrates the use of the progressive marker a.
In the example below, we see a marking the verb sii ‘see’ as habitual.
(15) dem piipl dis nou, jombii a piipl a sii
them people this now, jumbie is people prog see
‘These people now, jumbies are people who can see [through anything].’
(Rickford 1987: 242, line 1259).
The form doz and its variant das function in the role of pre-verbal habitual marker. It is unmarked for tense and, therefore, can be used with reference to either past or present. There is, in addition, however, a specialized past habitual marker yuustu. The following example demonstrates both the use of habitual doz, unmarked for tense, and yuustu, marked for past/anterior tense.
The non-contrastive completive aspect is signalled by a zero pre-verbal marker, as in the following example.
(17) mi Ø put mi paasl wo mii Ø gu bai
1sg Ø put 1sg.poss parcel rel 1sg Ø vol buy
‘I put down the parcel that I’d gone to buy.’ (Rickford 1987: 229-230, lines 1126-1127)
There is, in addition, the contrastive completive aspect, with the overt marker don.
(18) awii manggo don raip laang taim
1pl.poss mango compl ripe long time
‘Our mangoes have ripened some time ago.’
The future markers involve sa, gu, a gu and gain ~ gwain. The modals marking future occur before the verb.
Let us examine the first of these, sa. This is a feature of the most conservative varieties of Creolese, and is demonstrated in the example below.
(19) if Meerii Jeen sa memba dee, den awii sa memba tu
if Mary Jane can member there then we can member too
‘If Mary Jane can become a member there, then we will become members too.’
(Dolphin 1996: 110)
Its use is paralleled by gu~go, the much more common form. As with sa, it occurs before the verb, as below.
The forms involving aspectually marked go, i.e. a go and gain ~ gwain, suggest an inceptive or immediate future as compared with go and sa, at least in some contexts.
(21) Laad, a wen dis kotlis a gu kom out a mi aan?
Lord is when this cutlass is go come out of my hand
‘Lord, when will this cutlass [machete] come out of my hand?’ (Rickford 1987: 163, lines 561-2)
The anterior marker bin is used to mark verbs as having a time reference prior to that of speech time. It precedes the verb.
Rickford (1987) argues that the use of bin in an ‘if’ clause is highly unlikely, contradicting other linguists such as Edwards (1975). The following examples, however, provide evidence to contradict him and support those he sought to contradict.
At the most time-stable end of the verb-noun continuum, involving nouns as predicators, we have the marker a, operating as what would, in traditional linguistic terms, be called the equational copula.
(24) dem piipl dis nou, jombii a piipl a sii
3pl people dem.prox now ghost eq.cop people hab see
‘These people now, jumbies are people who can see [through anything].’
(Rickford 1987: 242, line 1259)
We would argue that the equational copula above is more properly described as a marker of progressive aspect. The subject NP, jombii, is being marked as being in the state associated with the predicator NP, piipl. This use of a contrasts with the spatializing marker a which precedes noun phrases which function as the head of a predicate phrase, by way of being the head of a noun phrase appearing as the predicate of a sentence.
There are two options in relation to locational phrases in the predicate. The first is to require the locational predicator, de, as in the example below.
Locational de extends to adjectival predicates such as gud ‘good’, nais ‘nice’, aarait ‘alright’, baad ‘bad’, beta ‘better’. When this occurs, and focus shifts to the impermanence of the attributes they represent, the interpretation is that of a reference to the state in which the patient or experiencer finds itself. This produces a reading of “condition” or “situation”. Note the contrast in meaning with the example without de.
(26) di blak piipl de far wee beta
the black people there far way better
‘The black people are in a situation which is much better.’ (W.E. in Devonish 1978: 237)
The second option is for the semantically locational phrase to be treated as if it were a permanent attribute. This requires the simple juxtaposition of the subject and the predicate phrase with a zero marker linking them.
The negative no ~ na precedes all other pre-verbal markers, as can be seen in the following examples.
(29) mi no bin don gyaaf wid dem
I neg ant compl talk with them
‘I had not finished talking with them.’
Creolese is largely an isolating language. There is a view that such languages have very strict constituent sequencing rules. In Creolese, however, there is some amount of variability in the sequencing of constituents.
The normal constituent sequence in Creolese is Subject – Verb – Object (SVO), as in (31).
(31) kooknot bring ail
sbj v obj
‘Coconuts produce oil.’
OR: ‘The coconut produces oil.’
There are, however, constructions which seem to deviate from this. They invariably involve verb sequences in which the first verb is a having, holding or carrying verb.
(32) mi gat wan piknii a main
1sg got one child prog mind
‘I am looking after a child.’
This pattern of variability is extended to ditransitive constructions. Below, we see first a sequence of Subject, Verb, Recipient and Theme. This is in keeping with the dominant constituent sequence in Creolese. In the second example, however, we see the verb tek ‘take’ followed by the Theme occurring before the main verb, gi ‘give’, and the Recipient.
(33) mi gi Jaan di buk
1sg give John the book
‘I gave John the book.’ OR: ‘I gave the book to John.’
(34) mi tek di buk gi Jaan
1sg take the book give John
‘I gave the book to John deliberately. (i.e. I didn’t give it to him by accident.)’
One detail is worth noticing in ditransitive ‘give’ constructions involving the use of tek ‘take’. This is not exactly equivalent to that of the double object construction. The former has an implication of deliberateness and planning, which does not exist with the double object construction. This difference is reflected in the translations given.
In Creolese, it is possible to have the interrogative phrase being in either initial or non-initial position in question word constructions. In initial position, the question is open-ended. In sentence final position, there is an implication that the answer is known or is obvious. As a consequence, it often occurs followed by a suggested answer, as in the second of the two examples below.
(35) aan wee alyu yuuzto plee dis?
and where you.PL used.to play this
‘And where did you guys play this?’
(36) yu jraa om wid wo – charkool?
you draw him/it with what charcoal
‘What did you draw it with? Charcoal?’
For yes/no questions, there is no special marking apart from interrogative intonation. This is illustrated in the example below which could, apart from question intonation, be interpreted as a statement.
(37) yu plee pon di greev self?
2sg play on def grave itself
‘You actually played on the grave?’
Relative clauses follow the noun phrases they modify. In examples such as that below, the presence of the relativizer wo ~ wa is optional.
(38) ayu na ga non wo doz oopn mashiin an so?
2pl neg have none rel hab.asp open machine and such
‘Don’t you have any that opens machines and other such things?’
Alternatively, where the relativized noun is the subject of the relative clause, a regular 3rd person pronoun agreeing in number can function as the relativizer.
When the subject is modified by a relative clause, it may be resumed by a personal pronoun:
(40) di man wa de pan di rood ii chupit baad
def man rel loc prep def road 3sg stupid bad
‘The man who is on the road is very stupid.’