Step by Step (review)

Abstract

Reviewed by: Step by step ed. by Roger Martin, David Michaels, Juan Uriagereka Kleanthes K. Grohmann Step by step. Ed. by Roger Martin, David Michaels, and Juan Uriagereka. Cambridge, MA: MIT Press, 2000. Pp. xxiv, 357. There is no doubt that Howard Lasnik has been one of the most prominent figures in generative grammar for the past three decades. This collection of essays pays tribute not only to his 25th anniversary at the University of Connecticut in 1997 but also to his far-reaching impact on the field, his positive influence within the community, and his exceptional qualities as a generous, warm-hearted person. The list of his publications and supervised PhD dissertations comprises one part, the preface and acknowledgments in each paper, the other. The following is a brief, if idiosyncratic, summary of what I take to be the main features of each article, sorted by topic. In an introductory chapter to the present volume, Roger martin and Juan Uriagereka speculate on ‘Some possible foundations of the minimalist program’ (1–29). In a nutshell, this is an introduction not to the book itself but to the minimalist program in general. (The editors introduce the content of the collection of papers and the relevance to Howard Lasnik’s work in a separate, five-page preface.) In it, they address conceptual issues of minimalist thinking and present possible consequences of a particular technical implementation of the tools available. The major distinction of ‘types of minimalism’ can be broken down into methodological vs. ontological minimalism. The former has also been called the ‘weak minimalist thesis’ by Noam Chomsky (this volume) and boils down to the well-known Occam’s Razor approach to scientific investigation: ‘the drive for simple and nonredundant theories of the world’ (1). Ontological minimalism is what M & U concentrate on. Chomsky (this volume) views this approach in terms of the ‘strong minimalist thesis’, an approach that centers around the question of how good the faculty of language really is. In an ideal world, the language faculty would be perfect and its design optimal. Against this background, M & U go over a sample of cases that illustrate one answer or the other and then concentrate on a justification of ontological minimalism as the driving force behind the search for an adequate linguistic theory. Noam Chomsky’s ‘Minimalist inquiries: The framework’ (89–155) is the first of two by now widely circulated extensions of the ideas espoused in his book, The minimalist program (1995). Focusing on the aforementioned strong minimalist thesis, he notes the most formidable obstacles for such a view to be the dislocation property of human language and the presence of uninterpretable features on lexical items. These properties of human language seem to be ‘imperfection of language’ or ‘design flaws’. This being said, these two obstacles can be collapsed into one, namely the displacement property P (by virtue of being a necessary device to implement displacement), which in turn would lose its antagonistic flavor if it turns out that ‘P is real, but not an imperfection; it is part of a “best way” (perhaps not unique) to meet design specifications’ (112)—the strong minimalist thesis which, even if wrong, is interesting enough to consider seriously. The technical implementation of P in minimalism differs from the original proposals in some respects. To present a sample of revisions, Chomsky (1) gets rid of the notion of strength of features (while still locating displacement in the morphology), (2) allows feature checking to take place long-distance (via the operation Agree), and (3) explicitly invokes an EPP-feature (whose presence on T, for example, is likely to be universal—the mechanisms to derive the EPP have, however, been revised in the second part; viz. Chomsky 2001). Zeljko Bošković considers (multiple) wh-questions where the wh-expressions are ‘Sometimes in [Spec, CP], sometimes in situ’ (53–87). He focuses on a treatment of French which he refers to as a ‘mixed language’ in that it makes two options available, attested for other languages. One type of multiple wh-question formation is to front one wh-phrase (as in English, for example); another is to leave it...