The prohibitive is the negated counterpart of the affirmative imperative, as in Papiamentu No bai! ‘Don’t go!’. Only the singular prohibitive is considered here, i.e. prohibition addressed to a single addressee. The corresponding WALS chapter is Van der Auwera & al. (2005).
We distinguish the following values:
|Normal imperative construction and normal negator||45||7||52|
|Normal imperative construction and special negator||13||4||17|
|Special imperative construction and normal negator||8||2||10|
|Special imperative construction and special negator||1||1||2|
The prohibitive may have the same form as the affirmative imperative (normal imperative construction) and the same sentential negative construction found in declarative sentences (normal negator) (value 1). This is the most widespread value.
Sometimes the prohibitive uses the normal imperative construction but a special negator (value 2). This occurs in five Ibero-Romance-based languages, in three English-based languages, in three Malay-based languages, in Afrikaans, Chinese Pidgin Russian, Juba Arabic, Michif, Pidgin Hawaiian, and in the bilingual mixed language Gurindji Kriol.
In Singapore Bazaar Malay the special negator jangan is used in prohibitive sentences, as opposed to the negator tak, which is found in declarative sentences; in Afrikaans, the special negator moenie is used instead of the first negator nie, and in Juba Arabic, it is the special negator máta which is used instead of the normal negator ma (but see example 6, Kinubi, which looks similar but was interpreted differently).
The prohibitive may also use the same negator as in declarative sentences, but the prohibitive construction differs from the affirmative imperative (value 3). This value is relatively rare (ten languages). There are three possibilities in which this value may be realized. In the five Upper Guinea Portuguese-based creoles (Guinea-Bissau Kriyol, Casamancese Creole, and the three Cape Verdean varieties) in Cavite Chabacano, in Belizean Creole, and in Kinubi, the negator is used both in declarative and prohibitive sentences, but, in contrast to the affirmative imperative, the subject pronoun is obligatory in the prohibitive.
Another possibility illustrating value 3 is found exclusively in Lingala. In this language, the negator té is used both in declarative and prohibitive sentences; however, in this case, the prohibitive construction differs from the affirmative imperative in that it does not use the imperative suffix -á (with a high tone, as opposed to the final vowel -a, with a low tone).
The last subtype of value 3 occurs only in Pichi. In this language, the prohibitive uses the same negator no as in declarative sentences, but additionally, the subjunctive particle mek may head a prohibitive sentence; in such cases, the subject pronoun is obligatory.
Value 4 (special negator and a special prohibitive construction) occurs only in Haitian Creole and in Mixed Ma’a/Mbugu. In Mixed Ma’a/Mbugu, the prohibitive uses a subject prefix, which is not the case in affirmative imperatives, and the negator used in the prohibitive sentence (si) differs from the one used in the declarative sentence (te).
In Haitian Creole, the other language exhibiting this value, the special prohibitive negator pinga may be used (as opposed to pa, which is used in declarative sentences); in this case, the subject pronoun is obligatory, which is not the case in affirmative imperatives.
Value 1 (normal imperative construction and normal negator, found in 52 languages) is present everywhere, although this value is dominant in the Atlantic (32 languages), especially in the Caribbean. Value 2 (normal imperative and special negator) is predominant in all parts of Asia (13 out of 17 languages) but is almost absent from Africa (exceptions: Afrikaans and Juba Arabic) and the Americas (exceptions: Palenquero and Michif). Value 3 (11 languages) is concentrated on the African side of the Atlantic as well as in Central Africa, and value 4 occurs only in Mixed Ma’a/Mbugu (East Africa) and in Haitian Creole (Caribbean).
Except for value 3, the distribution of the four values in the WALS languages differs considerably from the distribution in the APiCS languages (value 1: 23% vs. 68%; value 2: 37% vs. 22%; value 3: 11% vs. 14%; value 4: 29% vs. 2%).