Aspect change in verb chains concerns the possibility of marking the second verb in a chain for imperfective aspect (usually with a progressive function) in spite of the fact that the first verb in the chain is marked – or zero-marked, as in many creoles – for perfective aspect. In these cases, the first verb refers to a story-line event, and the second verb refers to a backgrounded event.
Note that in this context, verb chain is understood as being similar to serial verb constructions (absence of overt marker of coordination, subordination, or syntactic dependency of any sort, see Aikhenvald & Dixon 2006: 1); but, in contrast to serial verb constructions, verb chains refer to two (or more) consecutive actions whereby the second verb does not modify the other verb in any way.
It must be stressed that the two events described by the verb chain do not overlap, i.e. the situation referred to by the second verb begins after the situation of the first verb has been completed. However, the situation of the second verb overlaps with the situation(s) referred to by the subsequent verb(s). Consider the following example from Papiamentu:
In the first verb chain, Yan a kohe nan ta bai kas, ta bai kas, literally ‘was going home’, starts after a kohe nan ‘took them’; but Yan, the person referred to by the subject, performs the action of going home (ta bai kas) during the whole paragraph, and so ta bai kas forms the background for all the following situations, described by the verbs bira ‘turn over’, tende ‘hear’, sigui ‘go on’, bolbe tende ‘hear again’, mira ‘see’, bati’e mata ‘hit him dead’, dera ‘bury’, and sali ‘leave’. It is only at the end of the paragraph that Yan, the person referred to by the subject, arrives at home (bai kas ‘went home’).
In other words, the aspect change illustrated in examples (2) and (3) does not take place in a verb chain and is therefore not regarded as an aspect change in a verb chain.
Notice that in European languages like Portuguese or English it is not possible, or at least not common, to have such an aspect change in two coordinated or two paratactic sentences:
In both Portuguese and English, an inceptive aspect verb is used: continuou seu caminho and began walking, or começaram a correr and began running. But the fact that the Portuguese and English translation uses an inceptive aspect verb does not mean that the creole progressive markers have an inceptive aspect function. If they did, they could not form the background for subsequent situations since inceptive verbs refer to the beginning of a situation and not to its duration. In Portuguese, it is the noun phrase seu caminho and the infinitive correr that form the background, and in English the gerunds walking and running.
For this feature, the following three values are distinguished:
|Aspect change in verb chains is possible||17|
|Aspect change in verb chains is not possible||24|
|Verb chaining does not exist||25|
Value 1 (Aspect change in verb chains) is an almost exclusively Atlantic feature occurring in West Africa (8 languages) as well as in the Caribbean (also 8 languages), the exception being Mauritian Creole in the Indian Ocean.
This value is found in eight Ibero-Romance-based languages, in four English-based languages, in four French-based languages, and in Berbice Dutch. It occurs exclusively in creole languages. Examples (5)-(10) illustrate the phenomenon further.
Note that in example (6), the second verb is not only marked for aspect (na), but also for tense (ba).
This almost exclusively Atlantic construction needs more investigation in order to establish its exact syntactic status.