Community Research and Development Information Service - CORDIS


RECOS Report Summary

Project ID: 269752
Funded under: FP7-IDEAS-ERC
Country: United Kingdom

Final Report Summary - RECOS (Rethinking Comparative Syntax)

This project aimed to break new ground in syntactic theory by reconceptualising the principles-and-parameters approach to comparative syntax, retaining its strengths and attempting to deal with its perceived weaknesses. The central idea was to treat parameters of syntactic variation as emergent properties, rather than as prespecified aspects of the innate Universal Grammar (UG). The parameters organise themselves into hierarchies, which define the ways in which properties of individually variant categories may act in concert; in the initial formulation of the approach, this was thought to create macroparametric effects from the combined action of many microparameters. The highest position in a hierarchy defines a macroparameter, a major typological property, lower positions define successively more local properties. In the simplest case, parameter-setting in language acquisition starts at the highest position as this is the simplest choice, potentially requiring the postulation of just a single formal feature; acquirers "move down the hierarchy" when confronted with primary linguistic data incompatible with a high setting, with each successive downward movement requiring the postulation of a further formal feature. Hence the hierarchies simultaneously define learning paths and typological properties (although learning paths may also “begin” in other parts of a given hierarchy). To the extent that learning paths define “easier” versus “harder” structures and that language acquisition plays a role in diachronic change, we would also expect the hierarchies to facilitate insight into regular syntactic change.
The main achievements of the project have been: (a) to provide a clear and novel conception of the nature of the cross-linguistic variation which is compatible with contemporary generative syntactic theory; (b) to provide integrated new descriptions of important and well-known cases of cross-linguistic variation in the form of hierarchically organised and implicationally related options, shedding significant new light on the nature of cross-linguistic variation and how this should be accounted for in linguistic theory; (c) a taxonomy of the atoms of variation, related to but in principle distinct from the notion of hierarchy. The mechanism by which language acquirers move “down the hierarchies” was formulated as a minimax optimisation algorithm, initially assuming no formal features/properties in the input, then generalising the initial formal distinction that had been postulated to account for the input, before refining the formal characterisation of the input so that new distinctions attested in the input can be made, and then repeating these operations in the ever more restricted domain defined by the new distinctions. The implications of this idea for language change, and the related question of the relative conservatism of certain morphosyntactic properties in relation to others, were also taken up. As regards the hierarchies themselves, concrete proposals concerning null arguments, word order, agreement systems, aspects of the structure of nominals, verb-movement, types of ergative system, types of negation system, passive, causative and ditransitive constructions have been developed. In general, the project group have, individually and collectively, made ground-breaking new contributions both to the description and explanation of cross-linguistic syntactic variation. The project has taken the field of generative comparative syntax forward in significant new respects.

Reported by

United Kingdom
Follow us on: RSS Facebook Twitter YouTube Managed by the EU Publications Office Top