Volume 14, Issue 1

Gurindji Kriol is a mixed language spoken in northern Australia. It is derived from Gurindji, a Pama-Nyungan language, and Kriol, an English-lexifier creole language. Despite these clear sources, Gurindji Kriol contains grammatical systems which are not found in Gurindji or Kriol, for example asymmetrical serial verb constructions. The origin of these constructions is unclear given that Kriol only contains a very limited set of serial verb constructions and they are not found in Gurindji. The development of asymmetrical serial verb constructions is examined and it is suggested that they are a product of the more restricted Kriol serial verb construction developing and expanding under the influence of the Gurindji complex verb. The formation of this construction was a part of the more general genesis of the mixed language which was derived from code-switching.

Diverse evidence from Bardi and other Australian languages is used here to argue for a multifunctional view of complex predication, one which includes event classification but is not limited to it. These data bear directly on the question of whether complex predicate formation involves the transfer of arguments from the light verb to the preverb, or whether complex predicates are better analysed as a unification of preverb and light verb. The multifunctionality of the construction, as well as the syntax of Bardi and other languages, provides arguments against the argument transfer analysis.

This article presents a crosslinguistic study of the association of event-oriented modal categories, particularly the potential and the necessitive, with non-canonical case marking of the subject and voice-related constructions. It is estimated that in at least a quarter of the languages of the world such an association exists. Voice-related constructions are much more strongly associated with the potential than with the necessitive, and the source constructions are overwhelmingly passive and spontaneous voice. Accordingly, a grammaticalization path from passive and spontaneous to potential can be posited. Furthermore, it was found that the expression of modality with non-canonical case marking patterns often originates in, or is associated with, possessive constructions, usually resulting in necessitive constructions. Possible motivations and contexts for the development of the constructions under investigation are suggested.

An important executive function in “language design” is marking the boundaries of embedded clauses, and of clauses (and clause sequences) in the scopal domain of operators such as conditional ‘if’ and negation. Crosslinguistic comparison reveals a range of devices that have this effect, facilitating parsing by listeners. These include (i) prosody, (ii) constituent ordering, (iii) coordination, (iv) boundary-markingmorphemes, and (v) continuousmorphological indexation. The last two are emphasized in this article, with data from Australian and West African languages. Such boundary-marking functions deserve greater recognition in grammatical typology, especially since acoustic analogues have been well-studied by phoneticians.