In this feature, we consider serial verb constructions in which one of the two verbs (almost always the first one) means ‘take’ (or has a closely related meaning such as ‘raise’) and expresses an instrumental role as in (1), or a theme role as in (2).
Following Aikhenvald (2006: 1), we define serial verb constructions as referring to single, monoclausal events which have just one tense, aspect, and polarity value and which do not show any sign of coordination or subordination. Some languages allow for (or require) the repetition of the subject as well as tense and aspect markers on the second verb of the serial verb construction (examples 2, 5, and 8; see also Aikhenvald (2006: 40f.).
‘Take’ serials may have different functions; in our feature, we will focus on instrumental-introducing and theme-introducing ‘take’ serials.
Some languages allow for constructions with an optional marker of coordination, as in Papiamentu kue kuchú kòrta karni (literally ‘take knife cut meat’) ‘cut meat with a knife’ or ‘take a knife and cut meat’ vs. kue kuchú i kòrta karni ‘take a knife and cut meat’ (see Maurer 1988: 256). In such cases, the construction without a marker is not considered a serial verb construction but a variant of the coordinate construction with the coordination marker.
The possibility existing in most APiCS languages of giving ‘take’ serials a biclausal reading (as in the Papiamentu examples) raises the question of whether ‘take’ serials are real serial verb constructions in all the cases described in this chapter, in the sense that they describe a single event and are monoclausal. This is especially problematic if the ‘take’ serial introduces a theme argument of the second verb, allowing for a literal interpretation (value 1), and, generally speaking, if the object of the ‘take’ serial functions, at least semantically, as an instrumental of the second verb (values 3 and 4). It is only in cases where no literal interpretation of the construction is possible (value 2) that we know without any doubt that these are cases of true ‘take’ serials.
We distinguish the following five values; values 1-4 refer to the role of the object of ‘take’:
excl | shrd | all | ||
---|---|---|---|---|
Theme of the second verb, literal interpretation possible | 9 | 22 | 31 | |
Theme of the second verb, literal interpretation impossible | 0 | 12 | 12 | |
Instrument of the second verb, no resumptive pronoun | 3 | 19 | 22 | |
Instrument of the second verb, with resumptive pronoun | 0 | 7 | 7 | |
No ‘take’ serials | 41 | 0 | 41 | |
Representation: | 75 |
Value 1 (the object of ‘take’ corresponds to the theme of the second verb, a literal interpretation is possible) occurs in eight Ibero-Romance-based languages, in fourteen English-based languages, in three French-based languages, as well as in Berbice Dutch, in Sango, in Kikongo-Kituba, in Sri Lankan Malay, in Singapore Bazaar Malay, and in Yimas-Arafundi Pidgin.
Value 2 (the object of ‘take’ corresponds to the theme of the second verb, a literal interpretation is impossible) occurs in five Portuguese-based languages (Angolar, Cape Verdean Creole of Santiago, Fa d’Ambô, Principense, Santome), in six English-based languages (Creolese, Ghanaian Pidgin English, Nigerian Pidgin, Pichi, Sranan, Vincentian Creole), and in Berbice Dutch.
In example (7), a literal interpretation is not possible either; however, the object of ‘take’ does not correspond to the theme complement of the following verb but to its temporal complement:
The following Sranan example illustrates aspect repetition on the second verb in the series:
Value 3 (the object of ‘take’ corresponds to the instrument of the second verb, without resumptive pronoun) occurs in six Portuguese-based languages, in eleven English-based languages, in two French-based languages, as well as in Kikongo-Kituba, Singapore Bazaar Malay, and Chinuk Wawa.
Value 4 (the object of ‘take’ corresponds to the instrument of the second verb, with resumptive pronoun) occurs in five Portuguese-based languages and in two English-based languages.
The possibility of resuming the object of ‘take’ as an argument of the second verb casts some doubt on the status of these constructions as serial verb constructions since the instrumental would be marked twice within the same clause. It is therefore not surprising that the authors of two Cape Verdean varieties and Casamancese Creole translate ‘take’ only literally in their examples (e.g. ‘I took a knife and cut the meat’).
Most APiCS languages with an instrumental serial verb also possess an instrumental adposition. The difference between the two constructions is left unexplained in most cases. An exception to this is Creolese.
In Creolese, the serial verb construction implies that the action is deliberate and planned, which is not the case with the corresponding adpositional construction (Ii waip di teebl wid wan klaat ‘He wiped the table with a cloth’).
This means that more research is needed in order to establish the exact syntactic and semantic status, and possibly the pragmatic status of the different ‘take’ serial verb constructions discussed in this chapter.
Value 5 (no ‘take’ serials) occurs in 54% of the APiCS languages – in creoles, in pidgins, and in bilingual mixed languages in all geographical areas.
Although ‘take’ serials occur mainly in the Atlantic area (23 languages), they do occur in 12 languages outside this area: Chinuk Wawa in North America, Kikongo-Kituba and Sango in continental Africa, Seychelles Creole in the Indian Ocean, Sri Lankan Malay in South India, Papiá Kristang, Batavia Creole, Singapore Bazaar Malay, and Chinese Pidgin English in South East Asia, Gurindji Kriol in Australia, and Yimas-Arafundi Pidgin in the Pacific area.