The phrase ,,Ich mach dich Messer“ (lit.: I make you knife) is a ritualised threat used in a variety of contemporary German that is popularly known as ,Kanak Sprak' (kanak language), but is also referred to as ,Kiez-Sprache' (,hood language') by its speakers. Kiez-Sprache is a multi-ethnic variety that combines features of a youth language with those of a contact language, and has counterparts in other European countries (e.g. Rinkeby-Svenska in Sweden, straattaal in the Netherlands, or københavnsk multietnolekt in Denmark). So far, the interest in Kiez-Sprache has been mainly from sociological and sociolinguistic perspectives, and less so from grammatical ones.

After a brief discussion of Kiez-Sprache in general, I investigate the status of a phrase like ,,Ich mach dich Messer“ within the grammatical system of Kiez-Sprache as well as from the point of view of Standard German. Drawing on a corpus of spontaneous speech samples, I show that this phrase does not stand alone, but rather exemplifies a productive type of construction in Kiez-Sprache that is characterised by bare nouns and semantically bleached verbs. I argue that this construction reflects a linguistic division of labour between syntax and semantics that is supported by a pattern that Standard German provides for light verb constructions. Given these relationships, a phrase like ,,Ich mach dich Messer“ should not be regarded as a random grammatical simplification, and in particular not as an isolated case of determiner drop, but rather as a systematic phenomenon that indicates the grammatical productivity of this new variety.

The first part of the present paper deals with the syntax and semantics of the representation of Tense, Mode and Aspe ct (TMA) in Haitian Creole. While the first section offers an overview of the relevant data as presented in different descriptive works on Haitian Creole, the second and third sections develop a semantic and syntactic description of the Haitian Tempus particle following the account of Giorgi & Pianesi (1997). The second part of the paper discusses the origin of the Haitian TMA-System. I first present different accounts who argue that the substratum languages have had a major influence on the origin of creole languages. Then, after summarizing Bickerton's argumentation against the substratum hypothesis, I show that current results of research in second language acquisition refute the substratum theory in a categorical fashion. Following this line of reasoning, I argue that creolization must be treated as an instance of first language acquisition and I discuss the Lexical Learning Hypothesis (LLH) of Bickerton (1986). Finally it will be shown that the syntactic theory of Giorgi & Pianesi (1997) implies proposals concerning first language acquisition that can replace some problematic concepts of the LLH while permitting a straightforward account for the origin of the Haitian TMA-System.

The paper examines an old debate concerning the correct analysis of non-constituent conjuncts in sentences like John gave [the books to Mary] and [the records to Sue]. Five possible analyses of such coordinate structures are briefly reviewed: the large conjunct, the right-branching, the gapping, the constituent conjuncts, and the non-constituent conjuncts approaches. It is argued that the latter, i.e. in terms of non-constituent conjuncts, is the best to account for the differences between conjunct-internal and conjunct-external sharing. Subject-verb agreement data demonstrate that conjunct-external shared material obeys the limitation of strict phonological identity across the conjuncts, whereas conjunct-internal shared material behaves leniently in this regard. The conclusion reached is significant. Coordination is no longer in any way a valid diagnostic for constituent structure.