Toward a Unified Computational Model of Quantificational Scope Readings

Toward a Unified Computational Model of Quantificational Scope Readings* Prakash Mondal Department of Humanities and Social Sciences, Indian Institute of Technology Delhi, Room No. A-33, Nilgiri House, New Delhi 110016, Hauz Khaus prakash.mondal@research.iiit.ac.in Abstract. This paper is aimed at uncovering a unifying computational grounding beneath a diverse range of cases of quantificational scope effects both within and across languages. Since quantificational scope readings are quite variable and interspersed with issues of modularity and interfaces of grammar, an underlying and universal generalization is certainly hard to come by. Research on quantification is not new at all; studies and research done on quantification have not yet been able to arrive at a useful but universally valid and satisfactorily unified account of how quantificational readings are derived at all, let alone computationally. Here in this paper, a preliminary sketch of a unified three-tier computational model will be drawn up to show how quantificational scope readings across languages can be computed and derived. For this purpose, principles drawn from recent derivational accounts of quantificational scope will be aligned properly to eliminate their incompatibilities with each other. Keywords: a unifying computational grounding, quantificational scope, computational model. 1 Introduction Quantification is one of the hottest topics in linguistic research where the issue of how syntax maps into semantics is invariably associated with the question of how constrained the syntax-semantics interface is. If quantification is a linguistic phenomenon which is operative at the syntax-semantics interface, the question that naturally arises is: how much contribution does syntax as opposed to semantics or vice versa make toward a heterogeneous set of scopal effects that we find in quantificational readings within and across languages? When do syntax and semantics interact for variable scope effects in quantificational readings? Issues of such kind are certainly convoluted given that quantificational interpretation is not a unified phenomenon (Ruys and Winter, 2010; Szabolcsi, 2010). Given the backdrop above, we find that quantification has been handled from a number of perspectives- set-theoretic, representational, derivational etc. (see for details, Ruys and Winter, 2010). Here in this paper, present derivational accounts of quantificational scope will first be considered. The principles proposed in those accounts have computational significance in the operational design of the language faculty. But they will be shown to be incompatible and inconsistent with each other. A unifying account that eliminates their inconsistencies but aligns them beautifully in a computational model so that the correct generalizations about quantificational scope readings across languages can be captured is therefore needed. 2 An Outline of the Background Landscape An initial but brief survey of some of the current derivational approaches towards quantificational scope readings will first be presented. And then some problems associated with them will also be drawn up in later sections just to show how it calls for a unifying account of computational operations underlying quantificational scope. * Copyright 2010 by Prakash Mondal PACLIC 24 Proceedings 489 Fox (2003) has argued that QNPs (Quantificational Noun Phrases) move mainly because they need their argument requirement to be satisfied. …

Content truncated to 500 words