Prof. John Hawkins talar under rubriken "Language Universals and the Performance-Grammar Correspondence Hypothesis"

Tid och plats: Måndagen 27 oktober 2008, kl. 14-16, B 3




Language Universals and the Performance-Grammar Correspondence Hypothesis
John A. Hawkins,
UC Davis & Cambridge University

A widely held view in linguistic theory is that performance factors, including ease of processing, have had little or no impact on the formulation of grammatical rules. Most generative linguists follow Chomsky (1965) in assuming that the competence grammar is an integral part of a performance model, which must be regularly consulted in production and comprehension, whereas grammar itself has not been shaped by performance. In this talk I present evidence for an alternative view, drawing on typological patterns across grammars and comparing them with performance variation data in selected languages. The patterns of preference in performance within languages possessing several structures of a given type (e.g. alternative words orders or relative clause types) appear to be the same patterns found in the fixed conventions of grammars, in languages with fewer structures of the same type (e.g. with more fixed word orders). These preferences, and the quantitative distribution in the selection of less preferred structures, show striking correspondences. This leads to the "Performance-Grammar Correspondence Hypothesis" (PGCH), according to which grammars have been profoundly shaped by processing (Hawkins 1994, 2004). I give a brief overview of some of these correspondences, followed by a more detailed illustration of this research method at work (involving adjacency effects in performance and grammars). I argue that this kind of empirically-based performance account provides us with an independently motivated explanation for grammatical universals, as well as a better description of the cross-linguistic facts, since it accounts for general patterns, for exceptions to these, and for universal patterns that are not predicted by grammar-only principles. The PGCH may turn out to be incorrect in some cases. But there is a lot of support for it, and my reason for proposing it is to encourage others to gather relevant data and take seriously the possibility that grammars have been profoundly shaped by performance.

Chomsky, N. (1965) Aspects of the Theory of Syntax, MIT Press, Cambridge, Mass.
Hawkins, J.A. (1994) A Performance Theory of Order and Constituency, CUP, Cambridge.
Hawkins, J.A. (2004) Efficiency and Complexity in Grammars, OUP, Oxford.