Date: Tue, 03 May 2005 14:48:30 +0000 From: Ahmad Reza Lotfi Subject: Towards an Elegant Syntax
AUTHOR: Michael Brody TITLE: Towards an Elegant Syntax SERIES: Routledge Leading Linguists PUBLISHER: Routledge (London) YEAR: 2003
Ahmad R. Lotfi, Azad University (Iran)
Towards an Elegant Syntax, henceforth TES, is a collection of 13 articles written by Professor Michael Brody between 1980 and 2001 on issues of interest to the modern syntactician such as elegance, perfection, and minimalism. The book is divided in 4 parts, each of them a reflection of Brody's contributions to some stage in the development of what he terms "elegant syntax". A note on the difference between "elegance" and "perfection", which is published there for the first time, opens the book.
Although Brody followed the minimalist terminology in his earlier publications and used "perfection" to refer to his approach to syntax--it was only towards the end of the 90s that he adopted the term "elegant grammar," now he sees these two as distinct ones: perfection is a notion intended to capture some quality of language itself rather than that of a theory of language. Furthermore, it is an "engineering" term presupposing a task and an evaluation measure. Even the best theory of perfection in language does not need to be elegant itself. Only optimal will do. Theoretical elegance, on the other hand, aims at simplicity in a theory where such concepts as "derivations, economy, merge, move, phrase, projection, [and] c-command ... appear to be either redundant or reducible to much simpler notions (TES, 3)." For Brody, elegance rather than perfection must be the primary objective of linguistic inquiry: perfect syntax (at the best of times) is a re-invention of the "elegant" wheel. That move cannot be an imperfection, and the idea that LF is the basic syntactic level of representation are only two cases that show how "perfect" syntax finally adopts what "elegant" syntax has already discovered.
Part I ("Principles and Parameters") focuses on referential dependency, chains, and empty categories. He argues that contextual definitions for empty categories are consequences of independently motivated principles of grammar. standard GB theory is wrong in attributing the complementary distribution of trace and PRO to the conspiracy of two unrelated modules of grammar. Instead, one can explain both in terms of one single theory, namely Case-checking theory.
Part II ("Beyond Principles and Parameters") sets off with a review of Chomsky's "Knowledge of Language." He criticises his system for the multiple statement of Theta Criterion both at D-structure and at the level (s) where chains are formed. He concludes that chains should not be defined at S-structure but at LF as the basic level of representation. In his Note on the Organization of the Grammar, Brody compares three models of grammar, namely (1) the Ferris Wheel theory with S-structure as the axis around which LF, D-structure, and PF are organised, (2) the Window theory where SS functions as an arbitrary point for syntax-internal mapping between DS and SS, and SS and LF. the mapping between SS and PF, however, is different in its characteristics as syntax is hereby connected to phonology, and (3) Lexico-Logical Form theory whre SS is a mere non- interface level on the LF-PF mapping. In the final article in this section, Brody argues against the theta criterion on chains holding at LF. He attributes the unique theta-position of chains to the Projection Principle instead. PRO, then, is to be taken as an argument.
Part III ("Towards an Elegant Syntax") takes significant steps towards the modification of generative syntax in terms of theoretical elegance. He proposes a minimal theory of phrase structure where such notions as external/internal XPs, empty heads, and abstract lexical elements replace adjunctions and intermediate phrases. Pied-piping of elements in the process of chain formation is now viewed as a consequence of LF requirements (contra Chomsky's account of PF pied-piping). The picture given is more elegant now as no PF requirements force syntactic complications anymore. He also affords a number of simplifications in standard minimalist theories, namely (1) a single syntactic interface level (Lexico-Logical Form, (2) dispensing with dubious distinctions and duplications, (3) eliminating economy conditions, and (4) attributing conditions specific to the Chain/Move relation (such as Uniformity, c- command, Last Resort, MLC, and Procrastinate) to the syntax-external systems with the relation itself as part of syntax proper. Finally, he compares purely derivational, purely representational and mixed theories of syntax. PDTs are just impossible, and mixed theories redundant; hence a PRT.
Articles in Part IV ("Aspects of Mirror Theory") come closer to what Brody considers to be elegant syntax. He elaborates on two core hypotheses of mirror theory, namely Mirror ("X is the complement of Y only if Y-X form a morphological unit--a word" TES, p. 205) and Telescope ("[A] head X in a syntactic tree should be taken to ambiguously represent both th zero-level head(s) and its (their) associated phrasal node(s)" TES, p. 205). This reduces structural relations to specifier ---> head: morphological specifier-head order is then mirrored as the head-complement order in syntax. Excorporation is not permitted, however. As a result, X must roll up into Y. The resulting X-Y unit, by its turn, rolls up into Z, and so on. As morphological words involve domination, problems with c-command relation do not arise. Also the sharing of features between specifier and head makes it possible to dispense with c-command. Antisymmetry is also taken care of as LCA requirements are not violated.
There are priceless lessons for any serious student of syntax to learn from Brody's TES. Although I had already read most of these articles separately, it was only when I read them together and in this order that I noticed how theories are conceived, nurtured, and given a final touch of glory by masterminds. This work is a marvellous course in theory construction: the reader witnesses how Brody's conception of elegant syntax has been shaped and sharpened through the different stages of theoretical/technical development.
However, theoretical elegance will be a mere illusion unless the subject under study is (near-)perfect itself. Physics affords an elegant theory within the limits of the symmetry/perfection with which matter is characterised. In the biological world, such perfection is rather rare unless it is a reflection of the physical (rather than biological) substance of the live matter. Once in the realm of psyche, it becomes even more difficult to come across perfection. For Chomsky, however, the core of language is characterised with a symmetry that makes it an isolated phenomenon in the biological world. It is with regard to this isolated nature of our computational system for human language that his minimalist programme for a theory of syntax makes sense. And what Brody sees as theoretical elegance would vanish in the thin air if it were not for this property of the object of our syntactic theories: perfect syntax. Anyhow, syntax as a natural phenomena (inevitably subject to evolutionary tinkering through ages) cannot be more than near-perfect at its best. What makes elegant/perfect syntax an exciting field of study is our endeavour to close and closer to a perfect model without jeopardising the empirical soundness of our findings. With elegance imposed on a theory while the object of study itself in not that perfect, the scientist may simply lose his sanity.
In his search for elegance, there are also times that Brody sacrifices elegance itself in order to dispense with some awkwardness. When proposing a theory of phrase structure with no intermediate projections and adjunctions (119-126), he resorts to analyses involving new theoretical artefacts such as external/internal XPs, empty heads, and abstract lexical elements. The final product happens to be too ugly to be elegant. Likewise, he dispenses with all syntactic interface levels but one--Lexico- logical Form (LLF)--at the price of including D-structure inside LF, on the one hand, and the enrichment of the lexicon, on the other. This is elegance at the price of elegance!
ABOUT THE REVIEWER:
ABOUT THE REVIEWER
Dr. Ahmad R. Lotfi, Assistant Professor of linguistics at the English Department of Azad University at Khorasgan (Esfahan) where he teaches linguistics to graduate students of TESOL. His research interests include minimalist syntax, second language acquisition studies in generative grammar, and Persian linguistics.