Atlas of Pidgin and Creole Language Structures Online

1. Introduction ⇫ ¶

Pichi is an Afro-Caribbean English-lexicon creole spoken on the island of Bioko, Equatorial Guinea. Pichi is an offshoot of Krio, which first arrived in Bioko, the former Fernando Po, with African settlers from Freetown, Sierra Leone in 1827 (Fyfe 1962: 165). It is the second-most widely spoken language of the capital Malabo after Spanish and is used as a primary language in a number of villages and towns along the Coast of Bioko. Pichi is also spoken as a lingua franca throughout Bioko. No official figures exist but by extrapolation from population data, there is reason to assume that Pichi is used by at least a 100,000 people of Equatorial Guinea’s population of 693,000 (UN 2009). This makes Pichi the second most widely spoken non-European language of the country behind Fang (Northwest Bantu). The following overview is based on a corpus of some 50,000 words collected in Bioko between 2003 and 2007 (cf. Yakpo 2009b).

2. Sociohistorical background ⇫ ¶

In 1827, the town of Port Clarence (later renamed to Santa Isabel under Spanish colonial rule and to Malabo after independence in 1968) was founded by the British on the island of Fernando Po, as Bioko was then called. In the first decades after the founding of Port Clarence, settlers from Freetown, Sierra Leone, who spoke the English-lexicon creole language Krio, formed the largest component of the predominantly African population of the town (Martín del Molino 1993: 59). The “Krios”, the Creoles of Freetown, also appear to have provided the cultural and linguistic models for other, numerically smaller groups of Africans in the town (García Cantús 2006: 116–118). In the following decades these groups of diverse origins merged and acquired a distinct cultural and linguistic identity as the “Fernandino” people (Lynn 1984).

Until the end of the 19th century, trade contacts between Port Clarence and Freetown were upheld through regular steamer trips (Martín del Molino 1993: 62ff.). These contacts contributed to reinforcing the link between Krio and its developing daughter language Pichi. Fernandino traders also acted as middlemen in the palm oil trade with the Bube people, the indigenous population of Fernando Po (Lynn 1984: 267), which contributed to the spread of Pichi as a trade lingua franca throughout the island (Martín del Molino 1993: 116). During the 20th century, Bioko witnessed a massive expansion of the cocoa plantation economy. The bulk of manual labour was provided by Southern Nigerian, and to a lesser extent, Liberian, Ghanaian and Cameroonian contract workers, who outnumbered the Bube and Fernandino population of Bioko by the 1950s. The present-day oil boom in Equatorial Guinea has once more made Bioko the destination of migratory movements from the West African sub-region. Although the language has essentially guarded its Krio character, the continuous presence in Bioko of speakers of other varieties of West African Creole and Pidgin Englishes throughout the last hundred and fifty years as well as language shift from Bube (Northwest Bantu) have not gone unnoticed in the phonology, lexicon and grammar of Pichi.

3. Sociolinguistic situation ⇫ ¶

Equatorial Guinea has risen to become sub-Saharan Africa’s third largest oil-exporting nation. The most important factors affecting the use and status of Pichi in this context are: large-scale work-related immigration from West African English-lexicon pidgin/creole-speaking countries of the sub-region and internal migration from mainland Equatorial Guinea to Bioko, the ongoing language shift of Bube speakers to Pichi and Spanish, accelerating urbanization, ethnic hybridization and the ongoing expansion of Spanish-medium education. It remains to be seen how Pichi will adapt to the profound changes that Equatorial Guinea is going through at this moment.

Variation in Pichi may be captured by an – albeit oversimplified – division of speakers into two groups. Group 1 tends to be made up of younger speakers of up to 30 years and encompasses a larger percentage of speakers who acquired Pichi alongside other languages and/or who have been accultured more recently into the Pichi-speaking urban culture of Malabo. Group 2 consists of the Fernandinos, the former commercial and social élite of the island, who use Pichi as their community language and people of diverse cultural backgrounds who grew up in the ambit of Fernandino culture. Group 2 appears to be shrinking at the expense of group 1 (cf. also Morgades Besari 2004). The most conspicuous phonological differences between these two groups follow: For group 1 speakers, there is generally no phonemic contrast between the alveolar fricative /s/ and the postalveolar fricative /ʃ/ (cf. (a) in Table 1) and this systematically applies to all words in positions where group 2 speakers use /ʃ/ (cf. (a)-(b) in Table 1). Additionally, group 1 speakers insert a palatal glide /j/ between /s/ and a following mid vowel where group 2 uses /ʃ/, as shown in (c)–(d):

There is also some variation in the use and acceptance of certain grammatical structures. For example, the speech of group 1 appears to feature fewer types of serial verb constructions. A final area characterized by variation is the depth of Pichi–Spanish language contact. For example, the names of weekdays, numerals and religious terminology are almost exclusively expressed in Spanish by group 1 speakers. In contrast, group 2 speakers appear to have access to both sets of the lexicon. They may employ lunes ‘Monday’ in a code-mixed sentence, but are equally capable of using mɔnde ‘Monday’. Similarly, most group 1 speakers normally employ Spanish numerals above five (e.g. seis ‘six’, siete ‘seven’, ocho ‘eight’) instead of Pichi numerals (e.g. siks ‘six’, sɛven ‘seven’, et ‘eight’). In contrast, group 2 speakers seem to master a wider range of the Pichi numeral system.

4. Phonology ⇫ ¶

Pichi has seven vowel phonemes. Vowel length is not distinctive. There is some lexically determined vowel alternation with alternate forms like kɛr ~ kɛri ~ kari ‘carry, take’, lɛk ~ layk ‘like’; gɛl ~ gal ‘girl’; ùnu ~ ùna ‘you (2pl)’; wɔnt ~ want ‘want’. There is also a tendency towards the reduction of the phonemic contrast between the mid-vowels /e/ and /ɛ/ (e.g. lɛk [lɛ́k] ~ [lék] ‘like’, kɛr [kɛ́r] ~ [kér] ‘carry’) as well as /o/ and /ɔ/ (e.g. nɔto [nɔ́tò] ~ [nótò] ‘neg.foc’, mɔ [mɔ́] ~ [mó] ‘more’).

The consonant inventory counts twenty-three phonemes. The labiovelar plosives /kp/ and /gb/ are marginal and only attested in ideophones, (e.g. gbin ‘sound of a sudden blow’). The voiced labio-dental plosive /v/ is marginally phonemic in words like grèví [grèví] ‘gravy’. In most words, however, /v/ is in free variation with /b/ (e.g. ova [óbà] ~ [óvà] ‘over, be excessive’ and riva [ríbà] ~ [rívà] ‘river’).

Furthermore (cf. §3), for most younger speakers (group 1) the opposition between /s/ and /ʃ/, which is still operational with some older speakers (group 2) has collapsed, with /s/ being pronounced in all positions. Hence, one-time minimal pairs like so [só] ‘sew’ and sho [ʃó] ‘show’ have given way to the homonyms so [so] ‘sew; show’.

Pichi features a rich variety of processes involving nasals/nasalization. For example, the prothesis of /n/ is optional with some words featuring an initial /j/, e.g. yun [jún] ~ [njún] ‘be young’, while other words with a word-final /i/ are optionally subjected to word-final nasalization (realized as [n] or nasalization of the final [i]), e.g. gridi [grídì] ~ [grídìn] ‘be greedy’, hangri [hángrì] ~ [hángrìn] ‘be hungry’. Moreover, a group of high-frequency function words undergoes nasal place assimilation (e.g. the clitic object pronoun =àn [àn] ~ [àm] ‘3sg.obj’. A further process involves the optional insertion of the approximant /j/ between either of the velar consonants /g/ and /k/ and the front vowels /a/ and /ɛ/, e.g. gal [gál] ~ [gjál] ‘girl’, kɛr [kɛ́r] ~ [kjɛ́r] ‘carry’. Another feature of interest is the existence of a voiced uvular fricative /ʁ/ with the variants [r] and [ɣ], e.g. kɛr [kɛ́r] ~ [kɛ́ʁ] ~ [kɛ́ɣ] ‘carry, take’ and rɛs [rɛ́s] ~ [ʁɛ́s] ~ [ɣɛ́s] ‘rice’ (cf. Hancock (1976: 6) on this phoneme in Krio).

The vast majority of words are mono- and bisyllabic. The syllable template is (C)(C)(C)(V)V(C)(C). A vowel constitutes the syllable nucleus – there are no syllabic nasals. Up to two consonants may cluster in codas and three consonants may cluster in onsets. However, most clusters are systematically broken up by deletion and insertion processes, e.g. smɔl [sìmɔ́l] ~ [sùmɔ́l] ‘(be) small’ and hɛlp [hɛ́lɛ̀p] ‘help’.

Pichi is a tone language with two phonemic tones, a high (H) and a low (L) tone, as well as a default low tone over toneless syllables. Hence we find monosyllabic tonal minimal pairs like de [H] ‘locative-existential copula’ vs. dè [L] ‘imperfective marker’. The language also shows the maximal number of possible tonal configurations over bisyllabic words of the same syntactic category, i.e. fiba [HL] ‘fever’, wɔ̀tá [LH] ‘water’, nyɔní [HH] ‘ant’ and bàta [LL] ‘buttocks’. Grammatical tone is also employed for inflection and derivation (cf. §5). Pichi exhibits a mixed prosodic system, a typologically unusual feature that also characterizes other Atlantic creoles (cf. e.g. Berry 1970, Devonish 2002, Good 2004, Rivera Castillo & Faraclas 2005, Rountree 1972). Hence around 95% of the words in the corpus bear a single pitch accent (an H tone). The remaining 5% are tonal words in which all syllables are lexically specified for H or L tones. Tonal words are not normally affected by tonal processes such as tone spreading or deletion, nor do they bear intonational boundary tones. Tones are graphicized in the following way: Monosyllables and penultimate syllables always bear a high tone if they bear no tone mark (e.g. go [gó] ‘go’, waka [wákà] ‘walk’). Conversely, low-toned monosyllables or penultimate syllables always bear a grave accent mark (e.g. ɛ̀f [ɛ̀f] ‘if’, lɛ̀kɛ [lɛ̀kɛ̀] ‘like’). When a high tone occurs elsewhere in the word, it is marked so by an acute accent mark (e.g. màmá [màmá] ‘mother’, nyɔní [nyɔ́ní] ‘ant’. Syllables not covered by these notation rules are always low and remain unmarked (e.g. prɔpàtí [prɔ̀pàtí] ‘property’, banana [bànánà] ‘banana’. Tonal notation applies to the root, hence the components of multimorphemic words separated by a hyphen or an equals sign are marked for tone individually (e.g. us=tin [ús=tín] ‘what’, wàka-stik [wàkà-stík] ‘walking stick’).

5. Morphology ⇫ ¶

Pichi has a largely isolating morphology. However, there is a limited use of inflectional and derivational morphology in which tonal and non-tonal affixes, as well as suppletive forms are made use of. With personal pronouns, the grammatical relations of subject, object and possessive case are marked through constituent order (e.g. for distinguishing subject from possessive case with yù ‘2sg’ and wì ‘1pl’), suppletion (e.g. for distinguishing subject case à ‘1sg.sbj’ from object case mi ‘1sg.emph’) as well as tone to signal inflection (e.g. for distinguishing subject and possessive case dɛ̀n ‘3pl’ from object case dɛn ‘3pl.emph’. In (1), tone alone distinguishes possessive from object case with the 1sg personal pronoun mi (all following Pichi examples from Yakpo 2009b):

Pichi makes use of three affixational processes for derivation. One is a tonal process involving the deletion of the lexical high tone over a word and its replacement by a default low tone in order to form compounds. In (2), the lexical high tone over the first syllable of the modifier noun kɔntri ‘country’ is deleted and replaced by a default low tone (indicated by a grave accent), while the head noun chɔp retains its original high tone:

Reduplication, the second derivational process, also features the tone deletion characteristic of compounding but additionally involves iteration, i.e. the use of a reduplicant. Reduplication is only productive with dynamic verbs and indicates verbal plurality. Compare the reduplicated verb chench ‘change’:

The third process involves suffixation of the formative -wan ‘adv’, which is employed to derive adverbials. The affix -wan is etymologically related to the Pichi cardinal numeral wan ‘one’:

Most of the ideophones on record involve lexicalized iteration, which may be morphological in appearance, hence involve reduplication (e.g. katakatá ‘be hectic’) or appear to be syntactic and involve repetition, as shown with fwífwifwí in (5). Most ideophones are used as manner adverbs:

6. The noun phrase ⇫ ¶

The structure of the noun phrase is represented in the constructed example in Figure 1:

Postnominal modifiers include a pluralizer that is identical with the 3pl dependent subject pronoun. The pluralizer may also express an associative plural in combination with personal names (cf. 6) The 3pl dependent pronoun is also used with impersonal reference in order to background an agent (cf. (1) above):

Determiners like the definite article dì (cf. dì bɔl ‘the ball’ in (12) below), the indefiniteness expressions wan ‘one, a’ (cf. wan blak lapa ‘a black cloth’ in (61) below) and sɔ̀n ‘some, a’ (cf. 7), as well as the proximal and distal demonstratives di(s) ‘this’ (cf. (42) below) and da(n) ‘that’ (cf. 11) precede the nouns they refer to:

The comitative and instrumental preposition wèt ‘with, and’ is the preferred means of coordinating noun phrases:

Five features are important with respect to personal pronouns (see Table 4): person, number, syntactic (in)dependence, case and the pragmatic notion of emphasis.

Except for the suppletive forms mì ‘1sg.poss’ (which substitutes for à ‘1sg.sbj’) and ìn ‘3sg.poss’ (which substitutes for è ‘3sg.sbj’), dependent subject pronouns are also employed in adnominal possessive function. Independent pronouns (glossed as emph) are used in emphatic contexts in both subject and object positions. Hence they may be stressed, focused, topicalized, modified by postposed elements and conjoined. Independent pronouns are also employed as the regular object pronouns of verbs and prepositions. However, in object position the 3rd person dependent clitic object pronoun =àn ‘3sg.obj’ is in complementary distribution with the object pronoun in ‘3sg.emph‘ from the independent/emphatic series. The distribution of the suppletive allomorphs =àn and in is phonologically conditioned by the requirement of a polar pitch configuration over a vowel sequence at the clitic boundary. Hence only verbs or prepositions with a word-final consonant or a word-final vowel with a high tone permit the encliticization of the low-toned vowel-initial object pronoun =àn ‘3sg.obj’ (cf. 9). In turn, verbs and prepositions with a word-final low-toned vowel may only take the high-toned independent object pronoun in ‘3sg.emph’ (cf. 10):

Independent pronouns can be modified by postposed quantifiers (cf. 11), focus and topic markers as well as nouns. The pronominal system may also be extended through compounding. Compound pronouns may feature the numeral tu ‘two’ as the second component and thereby express dual number (cf. 12):

The functional equivalents of indefinite pronouns are common NPs involving generic nouns preceded by the quantifier and indefinite determiner sɔ̀n ‘some, a’ (e.g. sɔ̀n pɔsin/sɔ̀n man ‘somebody’) as well as the quantifiers ɔl ‘all’ (e.g. ɔl tin ‘all things/everything’), ɛni ‘every’ (e.g. ɛni tɛn ‘every time’) and no ‘neg’ (cf. 38-39).

Pichi makes no difference between ‘some’ indefinites used in affirmative and realis modality declaratives, and “free-choice” indefinites (Haspelmath 1997: 48–52) of the ‘any’ type (cf. 13):

(13) Yù fit mek ɛni kayn tin, yù gò si mɔ̀ní.

2sg can make every kind thing 2sg pot see money

You could do any (kind of) thing [in Libreville] (and) you’d earn money.’

7. The verb phrase ⇫ ¶

The core tense-mood-aspect (TMA) system is constituted by particles which express central notions such as imperfective aspect or potential mood. jI In the non-core system, auxiliary verbs express aspectual and modal notions in serial verb constructions. Pichi also makes use of complementizers in order to express modality. The elements of the core TMA system and their position relative to the verb root are provided in Figure 2:

Three lexical aspect classes can be identified on distributional grounds. Stative (stat) verbs (e.g. fit ‘can’, de ‘be.at’, tink ‘think’) do not co-occur with the imperfective marker dè and are always interpreted as stative. Inchoative-stative (ista) verbs (e.g. brok ‘break, be broken’, sìdɔ́n ‘sit down, be seated’, sàbí ‘(get to) know’, blak ‘be(come) black’) may receive both a present state or a past entry-into-state interpretation when they occur as bare verbs in intransitive clauses without further disambiguating information. Dynamic (dyn) verbs (e.g. nak ‘hit’, go ‘go’, kres ‘be crazy’) always receive a past interpretation when they appear unmarked for TMA. The inherent temporal structure of Pichi verbs co-determines the meanings that arise when core aspect marking co-occurs with a verb. These are summarized in Table 5 (an empty space indicates that the corresponding function is not available):

The general imperfective aspect marker dè covers functions generally associated with the imperfective domain such as progressive in (14), habitual in combination with the appropriate adverbial in (15), but also modal functions such as future tense in (16) and hypothetical modality in (17):

The expression of perfective aspect is less uniform. For one, perfective aspect may be expressed by default with dynamic verbs through the assignment of factative TMA (Welmers 1973), i.e. the use of the bare verb (cf. 18)). With non-dynamic verbs, the assignment of factative TMA most commonly yields an imperfective reading, namely present tense (or ongoing state) (19):

Secondly, perfective aspect may be expressed through the use of the narrative perfective marker kan ‘pfv’ with all lexical aspect classes. The marker is, however, specialized for use in the foregrounded sections of narrative discourse as in the following excerpt from a narrative:

Tense is expressed either covertly through the use of aspect marking or overtly by means of the past marker bìn ‘pst’ and the potential mood marker gò. Use of the past marker is not obligatory. It is generally employed in temporally remote, backgrounded, orienting and supportive sections of narratives. Compare the first two occurrences of bìn in (21):

The TMA marker gò is found in statements of intention (cf. 47) and prediction (cf. 13). However, its meaning is usually tinged with modal undertones. In (22), the potential mood expresses an epistemic possibility:

Pichi also makes use of serial verb constructions in order to express aspectual and modal notions. Compare the egressive serial verb construction involving the first verb in series kɔ̀mɔ́t ‘come out’ (cf. 24) and the expression of ability and (root) possibility via the verb fit ‘can’ (cf. 25):

Subjunctive mood is instantiated in the modal complementizer mek ‘sbjv’ and occurs in contexts characterized by the presence of deontic modality, i.e. in directive main clauses such as imperatives and other “mands” (cf. (22) above) as well as in the subordinate clauses of main predicates that induce deontic modality (cf. 26). Subjunctive mood also occurs in purpose clauses (cf. 27):

8. Grammatical relations ⇫ ¶

There are only few strictly intransitive verbs in Pichi (e.g. rɔn ‘run’, day ‘die’, fɛt ‘fight’). A large proportion of verbs is labile and may be used in transitive and intransitive clauses alike. Hence, the undergoer subject of the intransitive clause in (28) becomes an actor subject in the transitive clause in (29). Also note how the unmarked inchoative-stative verb slip ‘lie, sleep’ receives a present state reading in the intransitive clause (28) and how it acquires a past tense reading when used as a dynamic transitive verb in (29):

Double-object constructions featuring transfer and communication verbs like gi ‘give’, das ‘give as a present’ and lan ‘learn, teach’ are the only means of expressing the relation between an agent, a recipient and a theme, as shown in (30). Serial verb constructions of the give-type in order to mark recipient and beneficiary roles are not attested in the language.

There are numerous lexicalized verb–object combinations in Pichi in which syntactic objects occupy non-core semantic roles as diverse as instrument, purpose, source or content (cf. 31). The use of semantically empty deverbal cognate objects for emphasis is also common (cf. 32):

The following two sentences exemplify the use of the form sɛf as a reflexive and reciprocal anaphor in combination with personal pronouns. The form may also function as a focus/emphasis marker with larger constituents and clauses (cf. 41)

The most common type of causative construction involves the use of the causative verb mek ‘make’ and a subordinate subjunctive clause containing the verb of effect. The subjunctive clause is introduced by the modal complementizer and subjunctive marker mek ‘sbjv’:

9. Simple clauses ⇫ ¶

Pichi exhibits a subject–verb word order in intransitive clauses as in (36), and a subject-verb-object order in transitive clauses as in (37). Very often, a coreferential dependent pronoun additionally appears in the predicate which refers to the preceding specific full noun subject. Sentence (36) presents both alternatives:

Pichi negation revolves around the general negator no ‘neg’, which functions as a negative particle in verb negation and as a negative quantifier in NP negation. Sentence negation is characterized by negative concord. Whenever the verb is negated, non-specific NPs in the clause may be preceded by no ‘neg’ in order to add an emphatic sense to the clause as in (38):

Negative concord is obligatory with all negative phrases – syntactic phrases that function as negative indefinites in Pichi and involve the negator no followed by a generic noun, e.g. no say [neg side] ‘nowhere’, no man [neg man] ‘nobody’, no wan de [neg one day] ‘never’. Verb negation also accompanies the use of natin ‘nothing’, the only monomorphemic negative indefinite pronoun (39).

Content questions are formed by way of a mixed question-word system which involves different types of transparent and opaque question elements (cf. Muysken & Smith 1990): (1) simple, monomorphemic elements (e.g. udat ‘who’ and wetin ‘what’); (2) bimorphemic question words composed of either a clitic question particle and a generic noun (e.g. us=pɔsin [which=person] ‘who’, us=say [which=side] ‘where’), or a question particle and a non-generic common noun e.g. us=mòtó ‘which car’; (3) question phrases (e.g. wetin mek [what make] ‘why’).

Question elements are often placed under focus. In (40), a possessor noun is questioned and cleft-focused with the particle nà ‘foc’. In Pichi cleft constructions, the out-of-focus part of the sentence is not usually expressed as a relative clause.

The use of focus structures in the formation of declarative sentences is also commonplace. The reflexive anaphor and emphatic particle sɛf ‘self, emph’ is the most frequently used form in particle focus as in (41). But other elements also play a role, i.e. senwe ‘emph’ (42) or the sentence particle o (cf. (43) below):

Verbs may also be singled out for focus individually in predicate cleft constructions (cf. e.g. Koopman 1984; Muysken 1978):

The most common type of comparative is a “mixed comparative” (Stassen 1985). It involves the use of the comparative particle mɔ together with a serial verb construction in which the verb pas ‘(sur)pass’ functions as the standard marker as in (44):

Equative clauses are characterized by asymmetries and suppletion in the use of personal pronouns, polarity and TMA marking. For one, the two lexically distinct forms nà and nɔto express affirmative (cf. 45) and negative (cf. 46) identity respectively. When an equative clause is overtly marked for TMA the suppletive copula bi ‘be’ is recruited as in (47):

The locative-existential copula de ‘be.at’ may take adverbials and adjectives as complements. This form is used to express relatively transient, less permanent existence in time (cf. 48) and space (cf. 49)

Adjectival complements of de (cf. (50) and (65) below) also render less time-stable properties than their verbal counterparts (cf. 51). However, only the three value property items bad ‘(be) bad’, gud ‘(be) good’ and fayn ‘(be) fine’ may unequivocally be employed as complements of the copula de. Other property concepts in the data are lexicalized as verbs:

Prepositions and locative nouns play a part in expressing spatial relations. Other means include locative verbs (e.g. slip ‘lie’ in (28)-(29)), motion verbs, and motion-direction SVCs (e.g. 62) as well as locative adverbs (e.g. ya ‘here’ in Figure 1). Contrary to locative nouns (52), prepositions may not be employed in the syntactic position of nouns and require explicit mention of the ground, most often a nominal complement (cf. 53):

10. Clause linkage ⇫ ¶

Pichi clause linkage is characterized by a large variety of strategies and forms. The quotative marker se and the morphologically invariant general subordinator we (and its rare variant wɛn) stand out as multifunctional elements with overlapping functions. The subordinator we introduces relative clauses as well as time and manner clauses, and functions as a clause coordinator. In (54), the first two occurrences of we represent uses of we as a clausal coordinator, the third occurrence suggests a temporal reading of we as ‘when’:

The most common means of forming relative clauses also involves the use of subordinator we ‘subord’ and its less common variant wɛn. The use of resumptive pronouns is nearly general in subject relative clauses with specific head nouns as in (55), fairly common in object relative clauses featuring highly transitive verbs like brok in (56) and rare in the relativization of prepositional phrases (cf. 57). Note the possibility of preposition stranding in (57):

The quotative marker se is characterized by a high degree of polyfunctionality (cf. Güldemann 2008). For example, se may be followed by a clause, a phrase or a member of a list as in the following example:

The quotative marker se ‘quot’ may also introduce various types of adverbial clauses. In (59) the quotative marker introduces a manner clause:

A host of other elements are employed to introduce adverbial clauses with more specific meanings, e.g. ɛ̀f(ɛ)/if ‘if’ (cf. (55) above), bìfó ‘before’, lɛ̀k haw ‘as soon as, the way that’ (cf. (14) above), fɔseka ‘due to’, bìkɔs/porque ‘because’ (cf. 60, 22):

Serial verb constructions are less central to event integration than other forms of clause linkage. Amongst other functions, SVCs may be employed to introduce theme and instrument participants as in (61), the ground in a motion event as in (62). SVCs may also provide adverbial modification as in (63). However, most types of SVCs are formed with a restricted number of verbs and may hence be analyzed as types of lexicalized compound verbs:

Other constructions are akin to SVCs but are best analyzed as involving reduced secondary predicates. In a construction like (64), the second predication is always construed as temporally overlapping with the first one. This may lead to differential aspect marking:

11. Contact between Pichi and Spanish ⇫ ¶

Spanish has left a deep imprint on the lexicon and grammar of Pichi (cf. Yakpo 2009a). Code-mixing forms an integral part of the linguistic system of Pichi. In a selected portion of the corpus, a type count revealed that 50% of all nouns, 30% of all verbs and 62% of all numerals used were of Spanish origin. Spanish adjectives (65) and some conjunctions (66) are also regularly found in Pichi sentences:

A considerable number of Spanish words may be considered borrowings. They form an integral part of the Pichi lexicon and are often preferred to their counterparts of Krio origin. Consider the Spanish-origin verbs sube ‘go up’ and baja ‘go down’ in (67). These verbs are more frequent than their Pichi equivalents go ɔp ‘go up’ and go dɔn ‘go down’.

Virtually the entire numeral system, as well as the date nomenclature, have been borrowed into Pichi from Spanish:

A much smaller percentage of words of West African origin from diverse semantic fields was also inherited from Krio, e.g. okobó ‘impotent man’, chàkrá ‘waste; destroy’, wàyó ‘cunning’ (Fyle & Jones 1980). The number of words that originate from Bube, the autochthonous language of Bioko Island and the African language with which Pichi has had the longest period of contact is limited to a handful of items in the corpus (e.g. tòpé ‘palm-wine’).