Atlas of Pidgin and Creole Language Structures Online

1. Introduction¹ ⇫ ¶

Lingala (or “Lingála”, if one chooses to mark tone) is the native language of approximately 15 million speakers, while another 10 million use it regularly as a lingua franca. It is spoken in the western and northern Democratic Republic of the Congo (including its capital Kinshasa), in the central and northern Republic of the Congo (including its capital Brazzaville), in northern Angola, and among members of the Congolese and Angolan diaspora in mainly Europe, Canada, and South Africa. In all these locations it continues to acquire new adult speakers as well as native speakers.

2. Sociohistorical background ⇫ ¶

Lingala’s origins are to be found in what is now the Democratic Republic of the Congo. Before the European occupation of this territory in the 1870s and 1880s, Bobangi was an important riverine trade language on the western part of the Congo River. A handful of European officers commanding troops of African workers recruited on the Upper Guinea and Guinea Gulf coasts, the East African coast and Zanzibar, and the Lower-Congo area arrived in the Bobangi region in 1881–1882. The Europeans and their troops acquired an imperfect knowledge of Bobangi, strongly simplifying and restructuring it. Bobangi was the lexifier in this process. The range of other adstrates was diverse: languages spoken by the workers recruited on the West African coasts included English-, French- and Portuguese-based pidgins and creoles, as well as Hausa, Igbo, Yoruba, and other African languages; the Europeans spoke French, English, Dutch, or Portuguese, and pidgin and creole variants of these; East Africans and some Europeans spoke Kiswahili; Kikongo was spoken by the workers recruited in the Lower-Congo area. The lexical input from these adstrates was minor or less than that.

On the banks of the Congo River, Europeans exported the resulting Bobangi pidgin outside the sphere of influence of original Bobangi, obliging newly encountered populations to use it in the restricted forms of mercantile and proto-colonial communication with them. In 1884–1885, it was, as such, brought to Bangala Station, an important state post on the northwestern banks of the Congo River named after local populations erroneously believed to be called “Bangala” (it was renamed Nouvelle-Anvers in 1890, and is now Mankanza). There, it underwent significant expanding influence from local languages, such as Iboko, Mabale, Libinza, Boloki, Losengo, and others. The expanding pidgin came to be so strongly associated with the station that it soon received the glossonym “Bangala”. In the late 1880s and the 1890s, Bangala spread from the station in eastward direction, as far as the present-day city of Kisangani, in northeastward direction, as far as the Sudanese border, and in southward direction, as far as Léopoldville, the later capital of the Belgian Congo (now Kinshasa).

In 1901–1902, Scheutist missionaries at Nouvelle-Anvers recognized Bangala’s usefulness in terms of its geographical spread, but were disillusioned by what they perceived as the pidgin’s “rudimentary”, “broken” structures. Making explicit reference to artificial languages such as Esperanto, they decided actively to enlarge and reform Bangala’s grammar and lexicon.² This language-engineering program was enacted in prescriptive grammar books and dictionaries, textbooks for schools, Bible translations, etc. From the start, the Scheut missionaries also regarded the language name “Bangala” as one of the many “atrocities” to be corrected from above. In some Bantu languages of the region, ba‑ is a plural prefix for ethnonyms, not glossonyms, for which prefixes such as lo‑, li‑, ki‑, or others are used. They therefore altered the name of the language to “Lingala”, a glossonym which appears in no historical source older than 1902. The new language name soon found general acceptance, including in Léopoldville. In the northeastern parts of the Congo, where the Scheutists’ influence was less pronounced and the local languages do not all make use of the same types of prefixes, the old glossonym Bangala survived and, in fact, continues to be used until today.

The engineered variant as a whole was less successfully imposed than the glossonym. It did gain ground in the region around Nouvelle-Anvers, where it soon acquired second-language users as well as native speakers, and from where it spread throughout much of northwestern Congo. The northwest was a region where the Scheutists experienced few difficulties in enforcing everyday language use, among others through the school and mission networks they controlled. There, Lingala is still spoken today in the form the Scheutists designed it at the beginning of the 20th century, or in forms close to it. By contrast, outside of this region, in particular in the capital Léopoldville, the Scheutists’ grip on the language routines of the masses was less tight; as a result, Bangala there underwent only minor effects from the reform project. Developing more independently, it soon acquired its own, local generations of native speakers and continuously received enriching influence from incoming languages. Léopoldville’s population grew rapidly, peaking especially after 1945, with immigrants coming mainly from the Kikongo-speaking Lower-Congo region. The influence Kikongo exerted on (Kinshasa) Lingala in this period was more considerable than the one it had during the pidginization of Bobangi in the 1880s.

Thus, the variant spoken in Kinshasa today is older and closer to the original Bobangi pidgin and its immediate heir Bangala than the one spoken in the Congo’s northwestern regions, which is the result of normative “corrections” and elaborations of Bangala from above. It is the Kinshasa variant which is most influential and continues to spread most significantly (in fact, for the last two decades it has also been encroaching on the terrain of the northwestern variant) and which is covered in the Atlas of Pidgin and Creole Language Structures.

3. Sociolinguistic situation ⇫ ¶

After the Congo’s independence (1960), the Belgian colonial system with French as the official language, backed by four geographically distributed national languages, i.e. Kiswahili, Lingala, Ciluba, and Kikongo, was maintained. Under colonial rule, Lingala had enjoyed a privileged position vis-à-vis the three other national languages, being the language used in the colonial army across the entire territory and the language spoken in the capital Léopoldville. President Mobutu’s reign (1965–1997) reinforced this privileged position. Mobutu strongly centralized the politico-administrative apparatus of then-called Zaire, concentrating all seats of powers in the Lingala-speaking capital Kinshasa. Because of this and because Mobutu was himself from the Lingala-speaking northwestern region, the nation’s single party quite naturally adopted Lingala, next to French, as its political working language. Mobutu also further confirmed Lingala’s role as the recognizable language of the national army. Finally, this epoch witnessed the rise of modern Zairean music, highly popular across the whole of Zaire and, in fact, the entire continent, and sung almost exclusively in (Kinshasa) Lingala. All these factors led (Kinshasa) Lingala to spread to noteworthy degrees. Thus, it also came to be spoken in Zairean cities located in otherwise Kiswahili-, Kikongo-, or Ciluba-speaking regions.

The situation as a whole has remained largely unchanged until the present day, although some new political factors have come into play. Since Mobutu was ousted, in 1997, by rebel groups trained in Uganda, Rwanda, and the eastern Congo, the positions of French and Lingala have been somehow threatened by English and Kiswahili, respectively. Kiswahili has managed to some extent to supplant Lingala as the language of informal communication among the political elite and appears on the bank notes of the new Congolese currency, but it has not led to any form of language shift among the populations of Kinshasa or the other western regions in the Democratic Republic of the Congo. Also, Lingala’s presence in eastern Congolese cities has not diminished, but is, in fact, still growing. In the most recent Congolese constitution, French (not English) is identified as the only official language, and Kikongo, Lingala, Kiswahili, and Ciluba as the national languages, “of which the State ensures promotion without discrimination”. As under Mobutu, they are used in the media, next to French.

In the first half of the 20th century, Lingala spilled over from Léopoldville to neighbouring Brazzaville. From there, as well as from the banks of the river that divides the two countries only politically, it diffused further into the central and northern parts of the Republic of the Congo. The French colonizers introduced French as the official language, while Sango (see Samarin this volume), Lingala, and Kituba (or “Kikongo” and “Munukutuba”, see Mufwene this volume) grew out as regional lingua francas: Sango in the extreme northern corner, Lingala in the centre and north, and Kituba in the more densely populated south. Seven or more times smaller in population than Kinshasa, the city of Brazzaville never ceased to undergo cultural and linguistic influences from its giant neighbour. In this context, Kinshasa Lingala continued to flow into Brazzaville, where it was, incidentally, favourably received by the Marxist political party that ruled between 1979 and 1992, as most of its members came from the Lingala-speaking north. The civil wars that broke out after 1993, with peaks in 1997, opposed two fractions that were, among others, socioculturally indexed by language: Sassou-Nguesso’s militias were mostly Lingala-speakers, the majority of Lissouba’s troops users of Kituba. Since Sassou-Nguesso’s victory over Lissouba late 1997, Lingala has been gaining ground in parts of Brazzaville that were formerly Kituba-speaking. It remains used in certain media across Congo-Brazzaville.

4. Phonology and tonology ⇫ ¶

Kinshasa Lingala uses five vowel phonemes, all of which are oral.

The close-mid back vowel is pronounced slightly higher, approaching the close back vowel.

Contrast between low tones (mostly left unmarked in linguistic texts) and high tones (either marked by an acute accent or also left unmarked) is lexically and grammatically distinctive. Examples of lexically contrastive tone are ngámbo ‘opposite side’ vs. ngambo ‘difficulty’, kokóma ‘to arrive’ vs. kokoma ‘to write’ and sángó ‘priest’ vs. sango ‘news’. Grammatical tone is exemplified in nápésa ‘may I give (subjunctive)’ vs. napésá ‘I have given’, and nabétaka ‘I usually beat’ vs. nabétáká ‘I beat (long ago)’. Low and high tones can combine to rising and falling tones, which are always synchronically or diachronically analyzable and which in actual speech are often realized as simple high tones. For instance, băsí is the result of a contraction of the plural prefix ba‑ with the (irregular) lexical stem ‑ásí ‘women’ (lexical stems as a rule beginning with a consonant).

Semivowels include the voiced labiovelar approximant /w/ and the voiced palatal approximant /j/.

Prenasalization, i.e. the introduction of the consonant by a short nasal sound produced at the same place of articulation, is common: mbóngo [^mbó^ŋgo] ‘money’, kobanda [kobaⁿda] ‘to begin’, and sánzá [sáⁿzá] ‘moon’. In contrast to other variants of Lingala, Kinshasa Lingala has no voiceless prenasalized consonants in word-initial position.

The unvoiced alveolar plosive is palatalized in front of iV: kotia [kot^ja].

The canonical syllable structure is CV (prenasalized consonants having to be interpreted as one C), whence consonant clusters do not occur.

There is no standard or other uniformly accepted spelling for Lingala. Together with the limited teaching in and of Lingala in school education to the advantage of French, this results not only in orthographic variation across writers, but also in high degrees of vacillation in the writing practices of one person. The main linguistic features subject to orthographic vacillation are: the spelling of words of French origin (here, I leave them in the original); the semivowel y (which native speakers write as i, y, or not at all), the semivowel w (written as w, o, u, or not at all), the close-mid back vowel (written as o or u), tone (if indicated, to various degrees of consistency), prenasalized consonants in word-initial position (prenasal written or not), and bound morphemes such as the subject marker and the nominal class prefixes (written separately or in one word with the stem). Outside of these problems, all consonants are written as the IPA characters, except for the palatal approximant, which is written y.

5. Noun phrase ⇫ ¶

The well-known Bantu system of nominal gender classes is reduced to one of inflection classes: nouns are organized in classes, each marked by a distinguishing prefix and forming pairs of singular-plural formation, but class membership does not govern the agreement of modifiers, the subject marker on a verb form, or pronominal pronouns. Examples of the nominal inflection classes are mo‑konzi, ba‑konzi (class pair 1/2) ‘leader’, li‑táma, ma‑táma (class pair 5/6) ‘cheek’, e-lóko, bi-lóko (class pair 7/8) ‘thing’.

A prefix homonymous with the prefix of class 2, ba‑, often occurs in prefix stacking, i.e. it may be preposed to a noun’s singular or plural class prefix (except to the prefix of class 2, *baba-). This procedure not only pluralizes the noun but also marks the (plural) referent as definite or indefinite-specific, while in the case of the ‘regular’, class-based plural the referent remains indefinite-nonspecific. For instance, mo‑báli (class 3) ‘man’, mi‑báli (class 4) ‘men’, ba-mo-báli or ba‑mi‑báli ‘certain men’, ‘the men’; li-loba (class 5) ‘words’, ma-loba (class 6) ‘words’, ba-li-loba or ba-ma-loba ‘specific words’, ‘the words’. For singular noun referents, definiteness or (subtypes of) indefiniteness cannot be signalled.

Word order in the noun phrase, including in possession constructions, is head-initial. The order among the modifiers is possessor - demonstrative - quantifier - qualifier. As mentioned, all modifiers, including the connective, adjectives, numerals, demonstratives, etc., remain invariable with regard to the class of the head noun, and only five adjectives are variable, adjusting to number only: molaí, milaí ‘long’, monéne, minéne ‘big’, mokúsé, mikúsé ‘short’, moké, miké ‘small’, and motáné, mitáné ‘red’.

The connective, used in qualification and possession constructions as well as in some prepositional constructions, is ya (allomorph na, see below).

Na is an example of the many cases of generalization that both content and function words underwent during Bobangi’s pidginization. Na is, as mentioned, an allomorph of the connective ya, but is also the coordinating conjunction ‘and’, and functions as a preposition with a wide range of meanings, including ‘with’, ‘on’, ‘onto’, ‘in’, ‘into’, ‘to’, ‘from’, ‘at’, ‘about’, and ‘concerning’.³

Third-person personal pronouns make an animacy distinction.

If one of the conjuncts in a conjoining construction is a first or second person singular pronoun, an inclusory construction may be used, i.e. the pronoun is pluralized (while its referent remains singular), so as to anticipate the plural reference of the entire phrase.

For pronoun possessors, possession is formed in the same way as for nouns (see (4)), namely by means of the connective followed by the possessor, i.e. a personal pronoun. The connective ya is altered to na in front of a pronoun, a rule which is applicable not only to possessives but to all usage contexts of the connective. Compare ya in (4) to na in mobáli na yé ‘her husband’. There are two exceptions to this rule. First, if the possessor is referred to by means of a pronoun (not a noun), it can be moved in front of the head noun for contrastive stress. In such cases, the connective remains ya: ya yé mobáli ‘her man (not someone else’s)’. Second, in the case of pronominal as opposed to adnominal pronoun possessors, the connective also remains ya.

There are three demonstratives, occurring as adnominals and pronominals: óyo (proximal, ‘this’), wâná (medial and distal, ‘that’), and yangó (anaphoric, ‘aforementioned’). Apart from their use for situational deixis, óyo and wâná can also be used discourse-referentially, while yangó only functions in this capacity. The demonstratives are morphologically invariable, with one exception: in contrastive contexts and only when used pronominally, óyo adjusts to the number of its referent by means of the prefix ba‑.

Cardinal and ordinal numerals differ in the use of the connective for the latter.

6. Verb phrase ⇫ ¶

Lingala’s morphosyntax is to some extent accounted for by synthetic verb forms, composed of various types of bound morphemes. A maximal verb form consists of an initial subject marker, a reflexive object marker, the lexical verb root, root extensions, pre-root or post-root TAM affixes, and a final vowel.

The subject markers make a grammatical person and number distinction. In the third person subject markers, there is an additional animacy distinction, in analogy with the third person pronouns: they are a‑ or ba‑ for all animate and e‑ for all inanimate referents.

Object markers, commonly used in Bantu verb forms to anaphorically or pronominally refer to an object of the verb, are absent, except for the reflexive marker ‑mí‑, which serves for all six grammatical persons.

A process of verbal derivation adds extensions to verb roots in order to enlarge their grammatical or semantic range (see §7). The most frequently used verb extensions are ‑is‑ (causative), ‑el‑ (applicative), ‑an‑ (reciprocative, resultative), and ‑am‑ (passive). Root extensions are tonologically neutral: they receive the tone of the morpheme that follows, also if this is the final vowel.

All TAM markers appear after the root or extended root, except for the future marker ‑ko‑ and the present progressive marker ‑zô‑, which is the contracted variant of a periphrastic form (see below).

The subjunctive is marked by a high toneme on the subject marker, which carries a low toneme in all other moods. The imperative, which can only be used to address one hearer, is recognized by the absence of a subject marker and by a high toneme on the final vowel ‑a. For orders issued to more than one person, as well as for jussive and hortative meanings, the subjunctive is used. Prohibitions are expressed by means of the infinitive followed by the negation particle té.

The tense value of the null form is future.⁴ It contrasts with the “regular” future (‑ko‑) in the mood/modality value: the “regular” future conveys that an event will take place when (explicated or, mostly, implicit) conditions hold, while the null form conveys that all the necessary conditions are already, or will certainly be in place for the event to take place in the future. Compare:

Both ‑í (allomorph ‑é) and ‑á convey a perfect meaning for dynamic verbs and a present meaning for stative verbs. Used on a dynamic verb, ‑í conveys that an action has taken place prior to the moment of speaking and that the endpoint or resulting state of that prior action holds in the present.]´⁵ The time depth between the present and the moment in the past at which the event occurred is neither recent nor remote, but remains unspecified. A dynamic verb form with ‑á also expresses the present endpoint or resulting state of a prior event, but the speaker additionally conveys that the prior event occurred ‘long ago’.

Using ‑í on statives, the speaker communicates that the denoted state holds “generally” in the vast present. Like with dynamic verbs, s/he remains indeterminate as to “since when” it has been holding. With ‑á, s/he conveys that there is a relevant time depth during which the state has been holding, and that this time depth is considerable.

Note that both for dynamics and for statives, the difference between ‑í and ‑á is not one of tense, but of relevance and magnitude of time depth.⁷

There are two past forms, ‑ákí and ‑áká, the former denoting a past for which time depth is undetermined and irrelevant, while the latter an explicitly remote past.

The parallels between the present and the perfect on the one hand and the two pasts on the other may lead one to propose an itemized analysis, splitting the two past suffixes up into the separate morphemes ‑ák‑, ‑í, and ‑á, and positing a fourth ‑Ø‑ morpheme. This would allow one to incorporate the two past forms together with the present and the perfect into a single matrix arrangement, in which ‑ák‑ and ‑Ø‑ are taken to account for aspect and ‑í and ‑á for tense. The advantage of such an approach is cross‑language comparability. Pleading against it is that the distinction between ‑í and ‑á is, as mentioned above, not one of tense (nor one of aspect) but of relevance and size of time depth, and that in terms of aspect, forms with ‑ák‑ can be both imperfective and perfective, the two pasts being used, for instance, for past habits as well as for anteriority.

The habitual ‑aka expresses habits or structural, general-validity statements:

The habitual can also be used with statives. The difference with stative verbs inflected in the present with -í is one of discontinuous (20a) vs. continuous (20b) general validity.

In addition to the forms in Table 5, Lingala also makes frequent use of periphrastic verb forms with specific tense and aspect values. One of them is the progressive, composed of the verb kozala ‘to be’ in auxiliary position followed by an infinitive. In order to position the denoted event in time, the auxiliary is inflected in the -í present, conditional future, or one of the two pasts. In the ‑í present, the auxiliary and the initial morpheme of the infinitive are often contracted to form ‑zô‑.

Other periphrastic forms feature kolinga ‘to want’ (followed by either the infinitive or the subjunctive), kowúta ‘to come from’, or kosíla ‘to end’ (morphologically contracted and followed by a perfect or a past form) in auxiliary position. In all these cases, grammaticalization has bleached the lexical meaning of the auxiliary, giving rise to new tense or aspect functions: prospective for kolinga, after-perfect for kowúta, and already-perfect for kosíla.

7. Simple sentences ⇫ ¶

The word order in the simple sentence is Subject - Verb - Object, except in cases where the reflexive affix -mí- is object. Polar interrogative sentences follow the same word order as declarative ones. In content interrogative sentences, the question word appears in front of the clause if it is its subject, and can appear either in front or at the end of the clause when it is its object.

Subject: Lingala being a pro-drop language, personal pronouns in subject position are only expressed when contrast is needed.

Objects: After ditransitive verbs, the indirect object precedes the direct object. The applicative root extension -el- allows noncore arguments to be integrated into the argument scope of an intransitive or monotransitive verb. The object indicated by the applicative extension comes first.

The causative verbal derivation -is- conveys that the referent of the grammatical subject does not perform the event or undergo the state expressed in the root, but causes it: kokita ‘to come down’ > kokitisa ‘to bring down’, kobonga ‘to be in order’ > kobongisa ‘to make in order’.

Passive voice is expressed by means of the root extension -am-: kokunda ‘to bury’, kokundama ‘to be buried’.

Clause negation is rendered analytically, by means of the particle té, which comes at the end of the clause. Té is also the negative answer ‘No’.

8. Complex sentences ⇫ ¶

Comparative constructions of inequality are rendered at predicate level by means of ‘surpass’ verbs, the parameter of comparison being a nominalized adjective or a noun.

Another possibility is to use the infinitive of a ‘surpass’ verb after the adjective of comparison.

Relative clauses follow the antecedent and are introduced by zero or by the invariable relative marker óyo. In prepositional relative clauses, a resumptive pronoun, positioned at the end of the clause, is obligatory.

Object clauses make use of the complementizers te or que, or zero. The negation particle té comes at the end of each (sub)clause to which it applies.