Abstract

In the context of syntactic research, ‘emergent(ist)’ and ‘generative’ positions are typically interpreted as being diametrically opposed to one another, the former rejecting the notion of Universal Grammar (UG) and the latter rejecting the role of general cognition in shaping language structure. Taking seriously the arguments for a “poor” UG that have, to some extent at least, been part of generative discussion since Chomsky (2005), this paper argues for a novel strongly emergentist approach to syntax that is nevertheless also unambiguously Chomskyan. My objective is to show how the fleshed-out version of Chomsky’s “three factors” model that I am pursuing in the context of the ReCoS project (http://recos-dtal.mml.cam.ac.uk), and which also draws on some key structuralist notions, leads to the expectation of languages (grammars) that necessarily share certain core properties (Chomsky’s “three factors” are UG, the linguistic input, and general cognitive principles not specific to language). These core properties include the fundamental way in which grammars are organized in featural terms, with specific consequences for the types of variation we expect to find crosslinguistically, both synchronically and diachronically. Central to the approach is the way in which acquisition is shaped by the hypothesized conservativity-inducing general cognitive bias to make maximal use of minimal means. I show how its effects are visible beyond syntax, and also demonstrate its implications for matters like the route via which children acquire different components of their grammar, our understanding of the mechanics of acquisitiondriven change, and, most generally, the expected limits of crosslinguistic variation in syntax. What I will argue, then, is that this type of emergentist generative perspective opens up very exciting new possibilities for the investigation of language acquisition, of the mechanisms of language variation through time and space, of how language and more general cognition interact, and also, more generally, for the potential reconciliation of areas of Linguistics that have pursued questions of common interest in undue isolation from one another.

References Chomsky, N. (2005). Three factors in Language Design. Linguistic Inquiry 36: 1 -22.