The biolinguistic enterprise: New perspectives on the evolution and nature of the human language faculty (review)

The essays in this volume, by distinguished authors, cover a wide range of topics. Overall, the quality of work is high. Some chapters lie outside disciplinary boxes and are correspondingly speculative. Others are more traditional linguistics articles with 'bio' introductions. But together, the authors helpfully review huge literatures. The collection reveals connections across projects and shows how research from different fields substantiates the idea that human language is—and is to be studied as—a biological phenomenon.

The introduction sketches some recent history of this idea, including Eric Lenneberg, along with theoretical developments that highlight the question of why the human language faculty (HLF) has the character it does. The editors suggest that Noam Chomsky's minimalist program belongs 'at the center of the revived field of biolinguistics' (5), at least in one sense: given any roughly adequate description of human grammars and their invariant features, we want to know which redescriptions are no less adequate, yet simpler in ways that may reveal the fundamental operations that HLF computes and which aspects of HLF are distinctly human. I emphasize chapters that most closely fit these themes. The editors provide a longer, more balanced overview.

Why are there any languages, and why so many? In 'The biolinguistic program: The current state of its development', ROBERT C. BERWICK and NOAM CHOMSKY address these questions from an 'evo-devo' perspective that stresses commonalities across species and constraints on variation, instead of assuming that organisms can vary in nearly limitless ways subject to winnowing by natural selection. On this view, 'there is but one multi-cellular animal from a sufficiently abstract point of view', with 'superficial variety' resulting partly from varied uses of a 'developmental-genetic toolkit' (24) that has been largely conserved across phyla. Berwick and Chomsky note that work on the role of regulatory circuits ('genetic switches') in maturation was a source of inspiration for the principles and parameters (P&P) model of linguistic variation. And they argue that if the principles reduce to a minimalist core, then the evolution of HLF may likewise reduce to the emergence of lexical items, the combinatorial operation Merge, links to conceptual systems, and externalization—the last being perhaps secondary and late emerging, while the initial core of HLF was useful as an 'internal mental tool' rather than for public communication. In such a case, there might be only one human language, modulo limited variation in externalization.

In another chapter, 'Syntax facit saltum redux: Biolinguistics and the leap to syntax', Berwick asks how distinctive aspect(s) of HLF could have emerged. His discussion of genotype-phenotype bridges and the idea that 'the first adaptive step will likely be the largest in effect' (77), with subsequent tiny steps, is accessible and illuminating. He then reviews some basic linguistic phenomena that suggest an evolutionary leap: recursive generative capacity, constrained displacement (which might impede communication), and restricted grammatical relations (e.g. verb-object but not subject-object units). The proposal is that novel recursion on lexical items interacted with preexisting systems to yield a biologically unique phenotype.

Cedric Boeckx's 'Some reflections on Darwin's problem in the context of Cartesian biolinguistics' focuses on the Mergeability of lexical items. He describes lexicalization as a process in which a concept is endowed with an 'edge feature', said to be a 'catalyst for recursive Merge' (54) and a source of cognitive unification that led to a 'truly general language of thought' (60). Boeckx also views externalization as an 'addendum'. He relates this to poverty-of-stimulus considerations, returning to implications for linguistic variation in another paper, 'Approaching parameters from below', that contrasts two conceptions of parameters. On one view (associated with Mark Baker and Charles Yang), acquisition is a matter of choosing at least one of the fully specified languages that HLF makes available via some parameter hierarchy whose character...