Aims and Objectives

The properties outlined above make this set of languages a particularly well suited area for the understanding of what seems to be a common solution to the problem of quantification but with very interesting low-level parametric differences. Understanding both the common properties and the points of difference will provide invaluable insights into the nature of the syntax-semantics interface.
In the first instance we will address descriptive issues with a view to establish a solid empirical basis out of which relevant generalisations will emerge. More precisely we will aim to establish an inventory of quantificational categories for each of the languages under investigation together with their properties and behaviour in various contexts such as polar and modal contexts, their scopal properties, their status with respect to the notion of distributivity and the distinction between true quantificational force (universal or existential) and arbitrariness (c.f. English free choice any), and also on quantifier floating and its compositional import. Our objective at this stage will be to obtain a coherent characterisation of the notion `quantificational NP' in these languages and which will allow us to specify the parameters along which crosslinguistic variation with respect to quantification can be understood. Having characterised quantificational NPs we will focus on A(dverbial) quantifiers in these languages which will be studied along the same lines as the previously mentioned D(eterminer)-quantifiers, in order to fully specify the typological characteristics of this set of languages including whether their overall behaviour indicates that the Q-word quantification strategy is part of a more general A(dverbial)-quantification pattern as sporadic mentions in the literature seem to suggest.
Once the descriptive stage is completed, our aim will be to provide a theoretical, feature-based, account of the syntactic and semantic properties shared by these otherwise different linguistic systems, a major component of which will be to propose a compositional semantics for the Q-word type of quantification which embraces all the variants realised by the languages in our corpus, in order establish the linguistic parameter(s) responsible for the observed facts. It will also be our aim to relate our work to that already carried out by other researchers on languages of the English/Romance type in order to establish the exact locus of the parametric variation distinguishing the two classes of language.
As the characteristics of the individual languages and their theoretical impact become clear, we will progressively shift our attention to the search for a theoretical account which will allow for the identification of the parametric properties which, on the one hand, distinguish these language from those such as English in which the correspondence between question-words and quantifiers is restricted to certain A-quantifiers (somewhere, everywhere, but not *somewhen, *everywhen), Modern Greek where there is some correspondence between Q-words and quantifiers but where, diachronically, the derivation of quantifiers does not involve conjunctions but a distributive preposition (katà, a locative preposition meaning along, throughout), and, on the other, distinguish the lower-level variations exhibited by the languages in our corpus.