The study also highlights the constructs of current linguistic theory, arguing for distinctive features and the notion 'onset' and against some of the claims of Optimality Theory and Usage-based accounts.
The importance of Henk Zeevat's new monograph cannot be overstated. [...] I recommend it to anyone who combines interests in language, logic, and computation [...]. David Beaver, University of Texas at Austin
Date: Tue, 28 Dec 2004 14:50:52 -0700 (MST) From: Andrew Carnie <carnie@U.Arizona.EDU> Subject: The Structure of CP and IP: The Cartography of Syntactic Structures, Vol. 2
EDITOR: Rizzi, Luigi TITLE: The Structure of CP and IP SUBTITLE: The Cartography of Syntactic Structures, Vol. 2 SERIES: Oxford Studies in Comparative Syntax PUBLISHER: Oxford University Press YEAR: 2004
Andrew Carnie, University of Arizona
This book, which contains papers presented at the "Workshop on the Cartography of Syntactic Positions and Semantic Types" (Certosa di Pontignano, Siena, November 25- 26, 1999), presents a variety of articles on the architecture of the functional categories in the clause. This book is part of a series of three such volumes that focus on the nature and organization of the functional structure of the clause and other phrases. The other two books: Cinque's (2002) _The structure of DP and IP: The Cartography of Syntactic Structures_, and Belletti's (2004) _Structures and Beyond: The Cartography of Syntactic Structures, Vol. 1_ have been reviewed in LINGUIST issues http://linguistlist.org/issues/14/14-1479.html and http://linguistlist.org/issues/15/15-3497.html respectively. The topic/focus of this volume, not to pun too much, is on topic/focus structure in the CP (complementizer phrase) and IP (inflectional phrase) domains. The papers mostly have grown out of a tradition of work started by Rizzi in his (1997 and subsequent) work on the "fine structure of the left periphery of the clause" in which a variety of functional heads (Topic, Focus, Finite, Force) replace the more traditional CP, as well as out of the older tradition exploring "expanded" INFL which started with Pollock's (1989) paper, and was the height of fashion in early Minimalism.
The book opens with a very interesting survey of the question of clausal cartography by Luigi Rizzi ("On the Cartography of Syntactic Structures"). Rizzi observes that the study of clausal cartography in generative grammar starts with the discussion of affix hopping in _Syntactic Structures_ (Chomsky 1957). X-bar theory brought the question of the nature of functional projections to the forefront resulting in the proposals for CPs, IPs, and DPs, The empirical questions of argument licensing, adjunct licensing, head positions and head movement all provide us with insights into the nature of the functional architecture of the clause. In addition to this historical survey, and a summary of all the papers in the volume, Rizzi provides a very thought-provoking discussion of how minimalist thinking and the reduction of functionality to features has resulted in an apparent "maximization" of the number of functional categories. He observes that this kind of investigation has forced researchers to consider more finely grained semantic analyses including reference to discourse/pragmatic factors such as topicality and focus. It was this that led Rizzi to his influential proposal that the CP structure was more properly divided into Force, Finiteness, Topic and Focus functional projections. This proposal forms the starting point from which many papers in this book develop.
Adriana Belletti's contribution, "Aspects of the Low IP Area", is a beautifully argued article. She claims that parallel to the Focus/Topic structure found in the CP domain there are Focus and Topic projections between the IP and the v/VP. The main empirical domain of inquiry in the paper is the nature of post-verbal subjects. She contrasts Free Inversion (FI) from Stylistic Inversion (SI); the latter type exhibited in French. Using evidence from adverbs, extraction, and negative polarity items, she shows that SI shows the properties we might expect of elements in the CP domain, but subjects in FI appears to be low in the structure. Belletti demonstrates that the subjects in FI constructions behave like foci. She goes on to compare ungrammatical VSO structures to those with a PP complement (VSPP); she gives an explanation of this in terms of case and relativized minimality. She also compares two VOS orders and orders in which the post-verbal subject appears to be topical rather than focal.
As mentioned above, the argumentation in this paper is exquisite and very convincing. Indeed, I intend to assign this paper to my students as an example of how to make arguments about clausal architecture. One point that stuck in my craw, however, was the fairly extensive literature comparing and attempting to explain VSO structures in Spanish and Romanian (and other romance languages), but no reference was made to the quite extensive literature on argument licensing and information structure in VSO languages outside of Romance (not to toot my own horn, but see for example, the contributions in Carnie & Guilfoyle (2000) and Carnie, Harley and Dooley (2005)). This literature contains fairly extensive discussion of argument licensing and information structure in many languages that seem to make use of the "lower" topic/focus domain that Belletti so aptly identifies in Italian.
Paola Beninca' and Cecilia Poletto use evidence from a number of Italian dialects and closely related Romance forms to argue for a modified and more fine-grained CP layer than that proposed by Rizzi. Their paper "Topic, Focus, and V2: Defining the CP sublayers" first argues, on the basis of weak crossover effects and interpretation, that of Rizzi's original [Topic [Focus [Topic [IP]]] structure, the lower of the two topics is more properly analyzed an informational focus position. Using cross- dialectal evidence and verb second (V2) phenomena, they distinguish this position from the higher focus, which they claim is contrastive. They also claim that the Topic projection is also more finely articulated into Hanging Topic, Scene Setting adverbials, (both of which constitute the "Frame"), Left Dislocated Structures and List Interpreted items (which constitute the "theme").
The next paper in the volume, Valentina Bianchi's "Resumptive Relatives and LF chains", is a bit of an odd man out in a volume on functional architecture. Although she makes use of Rizzi's ForceP and proposes a GroundP (which seems to be similar to the "Frame" of Beninca' and Poletto (above)), there is very little about the analysis that bears on the question of functional cartography. Nevertheless the paper is interesting in and of itself. Bianchi surveys different types resumptive strategies in different types of relative clauses, and explains why some types of relative clauses resist resumption and others require it. She draws upon Enc's (1991) notion of referential index, and claims that resumptive pronouns are spell-outs of these indexes on the tail of movement chains.
The paper by Anna Cardinaletti, "Toward a Cartography of Subject Positions", is a useful complement to Belletti's contribution. While Belletti focuses on post-verbal subjects in Italian and related languages, Cardinaletti focuses primarily on preverbal subject positions (although she does have a brief discussion of VP internal and "middle field" subjects as well, including the functional projection where case assignment occurs). Cardinaletti claims that there are at least three distinct functional projections between the CP domain and the surface position of the verb. She distinguishes between a relatively low AgrS, which hosts light pronouns and pro, an intermediate EPP head, which can host locative subjects and expletives, and SubjP, which hosts the "subject of Predication" argument. Like Belletti's paper, the argumentation is extremely elegant, however one wonders the extent to which Cardinaletti and Belletti's accounts are compatible.
Like Bianchi's paper, Carlo Cecchetto's "Remnant Movement in the theory of Phases" only tangentially addresses the question of clausal cartography, by claiming that each of Rizzi's CP articulations can count independently as a "phase-edge" for the CP. The focus of Cecchetto's extremely dense and difficult to follow paper, however, is on remnant movement. He is most interested in distinguishing between cases where remnant movement results in what has traditionally been treated as a relativized minimality effect, to those that are fully grammatical. The standard account is based on the observation that remnant movement over an element extracted from that remnant is ungrammatical when the two movements are of the same kind. Cecchetto marshals a number of arguments against this approach. Unfortunately, I found none of them terribly convincing. Some of his arguments are based on the claim that Japanese scrambling is "semantically vacuous", which, to say the least, is very controversial. His other arguments all rest on the assumption that given the configuration (a) [...[Z ... t-x...]... X ... t-z] is unacceptable because underlyingly the structure is (b) [... ___... [Z ... X...]], where Z is a closer target for the movement to the surface position of X in (a). Note however, there is another interpretation of the ungrammaticality of structures such as (a): note that in (a), X is a closer potential antecedent for the trace than the remnant constituent Z. All of Cecchetto's arguments against a relativized minimality account rest on the fact that Z is the intervening category, not X. Cecchetto provides an account given in terms of phase edges and the Chomsky's phase impenetrability condition (PIC).
"Complementizer Deletion in Italian" by Alessandra Giorgi and Fabio Pianesi examines the close interrelationship of the low CP domain (mood) and the high IP domain (AgrS). They claim that the phenomenon of complementizer deletion (CD) in Italian is neither deletion nor V to C movement. Instead they appeal to a notion of feature "Scattering" (where by features maybe distributed, in order, across a number of different functional categories), and lexicalization. CD deletion occurs when Mood features and Agreement features syncretize on a single head. _Che_ by contrast, appears when Mood and agreement appear in separate heads. The analysis is pleasing and explanatory, but I think it might be refined if it were cast in terms of the theory of Distributed Morphology, and instead of "scattering" the authors could make use of a universal set of functional categories and feature distributions and appeal to head fusion and the principles of vocabulary insertion.
In the first of two papers on clitics, Rita Manzini and Leonardo Savoia ("Clitics: Cooccurrence and Mutual Exclusion Patterns), argue for an approach like Sportiche (1996) that clitics represent functional categories both within DPs and within clauses. They propose a hierarchy of D-op, D, R, Q, P, Loc, and N. P represents discourse reference, limited to objects; Loc is for demonstratives; R is for specifics, D, Q, and N have their usual values (Determiner, Quantifier, and Noun); and D-op is to host partitive "of". These functional categories are merged in their surface position (which may be above C, between C and I and between I and V). Mutually exclusive clitics are not ruled out by a competition-based mechanism (such as Optimality Theory) or by a DM-style Elsewhere principle, instead by simple syntactic parameters that govern the relative order of clitics and the cases where two clitics may not cooccur.
"On the Left Periphery of Some Romance Wh-questions", by Cecilia Poletto and Jean-Yves Pollock, considers the structure of wh-questions and stylistic inversions in French, Italian and Bellunese. They follow Kayne (1998) in assuming that there is no covert movement. Apparent covert relations are actually overt movements with subsequent remnant movement obscuring the initial move. Furthermore, they claim that subject clitic inversion does not involve any head-movement but is also the consequence of remnant movement. They argue for a left periphery consisting of [ Op2 [ Force [ Ground [ Top [ Op1 [IP]]]]]]. The two Op positions attract two different kinds of wh-words. The higher one for wh-phrases that previously would have been analyzed as having moved overtly, and the lower for object wh-in situ. Depending upon the construction, the specifiers of ForceP, GroundP and TopP are the landing sites for various movements and remnants.
Ian Robert's "The C-System in Brythonic Celtic Languages, V2, and the EPP", relates the requirement on Brythonic Celtic (Welsh and Breton) to have preverbal particles to the requirements of V-movement in V2 constructions in the Germanic languages (and the partial V2 nature of Breton). He claims that both phenomena are part of a general requirement that Finite be filled. He also argues that the complementizer particles in Celtic (including Irish) are actually lower (in Finite) than those of English (in force), which explains the apparent variation in adverbial placement in Celtic and Germanic, discussed by McCloskey (1996). This is a nicely argued paper, and one that is causing me to rethink my own analyses of Irish complementizers and copular constructions. I'm a little disappointed however about an important gap in its citations and discussion. The literature on Celtic is not vast and there is a recent paper by Hendrick (2000) that covers much of the same ground (and reaches related, but different conclusions) about the organization of the C system at least; the proposals in that paper should have been addressed here.
The last paper in the volume, by Ur Shlonsky ("Enclisis and Proclisis"), claims that variation among Romance varieties (and extended to various dialects of Berber) in terms of whether clitics are realized as pro- or enclitics, is due to whether the verb has fully checked its features below the clitic head or must do further checking above it. Enclisis is true head-to-head movement. Attaching a verb to a clitic results in the verb (and its features from being inaccessible for further checking.) Proclisis by contrast occurs only when enclisis cannot and results, at least in part, from two separate adjunctions to a single head. Crosslinguistic variation is due to differences in where the cliticization occurs with respect to negation, finite inflection and relevant features in C. This paper is an interesting complement to Manzini and Savoia's paper and I would have liked to have seen (in both papers) a comparison of their approaches.
Overall this book is an important contribution to our understanding of clausal architecture and the number and nature of the CP and IP domains in minimalist syntax. Rizzi's (1997) work has clearly opened up a rich domain of inquiry, as the papers in this volume show.
Before I launch into a couple of conceptual issues that I have with the approach and its virtues and problems, I'd like to point out a couple of minor editorial issues. First, this book is based on a series of papers given at a workshop in 1999. The five-year lag between presentation and publication gives some of the papers a vaguely asynchronic feel. Some of the papers have clearly been updated to reflect more modern thinking; others have not. I'm a great believer that edited volumes, even those that come out of conferences, should make a clear attempt to get the papers to cross-reference each other, and where appropriate discuss each other's analyses. There is almost no cross-referencing between the papers here, even when the papers are clearly closely related in either topic or analysis. Finally, I know from experience that Oxford's proofreaders are excellent, but they seem to have fallen down on the job in a number of the papers in this volume. Particularly frustrating are the diagrams that have been misaligned. For example, the schematic of Benica' and Poletto's analysis (their example 58) is absolutely impossible to understand, as it has been split onto two lines and skewed out of shape. Similarly, throughout there are many cases where bracket labels are not subscripted, which makes them very difficult to distinguish from the heads. In Cecchetto's article the definition of the PIC appears to be part of the main body of the text, which makes it very hard to follow. These may be minor points, but they take away from the important issues and results in the papers.
Turning now to the question of the overall research program that drives this volume: With the exception of Cecchetto, Roberts and Shlonsky (who address Japanese, Celtic and Germanic, and Berber respectively), the bulk of the papers in this volume focus on Romance languages, and in particular on Italian dialects. This should not be entirely surprising given who the authors and editor are, and given that the conference underlying the book was sponsored by the Italian Ministry of Education. However, it does bring two thoughts to my mind. On one hand, it is clear that looking at micro- variation among closely related dialects gives us interesting and comprehensive results. The sophisticated analyses in this volume attest to this. On the other hand, one might wish that a broader crosslinguistic perspective on the issue. This could help confirm the results, and show that the phenomena in question aren't merely Romance centered. I hope the series editors might consider a fourth volume in the series that address variation in more language outside the Indo-European core that is the focus of this book.
This said, I have to laud the editor and contributors on another ground: This book clearly brings into the light a domain of inquiry that has been taboo for too long in mainstream Chomskyan syntax: the role of information structure and discourse/pragmatics on word order and word order alternations. The importance of these factors has too long been ignored and left to functionalists to investigate. This volume is an important step forward in formalizing such notions as topic, focus, theme, rheme, ground, and force into current Chomskyan theorizing.
Carnie, Andrew and Eithne Guilfoyle (2002) (eds.) _The Syntax of Verb Initial Languages_ Oxford: Oxford University Press
Carnie, Andrew, Heidi Harley, and Sheila Dooley, eds. (2005) _Verb First_. Philadelphia: John Benjamins.
Chomsky, Noam (1957) _Syntactic Structures_ The Hague: Mouton.
Enç, Murvet (1991) "The Semantics of Specificity." _Linguistic Inquiry_ 22: 1-25.
Hendrick, Randall (2000) "Celtic Initials" in Carnie and Guilfoyle (eds.) _The Syntax of Verb Initial Languages_ Oxford: Oxford University Press.
Kayne, Richard (1998) "Overt vs. Covert Movement." _Syntax_ 1: 128-191
McCloskey, James (1996) "On the Scope of Verb Movement in Irish" _Natural Language and Linguistic Theory_ 14: 47-104.
Pollock, Jean-Yves (1989) "Verb Movement, Universal Grammar and the Structure of IP" _Linguistic Inquiry_ 20: 365-424.
Rizzi, Luigi (1997) "The Fine Structure of the Left Periphery" in L. Haegeman (ed). _Elements of Grammar_ Dordrecht: Kluwer, 281-387.
Sportiche, Dominique (1996) "Clitic Constructions" in J. Rooryck and L. Zaring (eds.) _Phrase Structure and the Lexicon._ Dordrecht: Kluwer, 213- 276.
ABOUT THE REVIEWER:
ABOUT THE REVIEWER
Andrew Carnie is an Associate Professor of syntactic theory in the department of Linguistics at the University of Arizona, Tucson. His research interests include representations of constituency, case, and VSO languages. He has published 3 edited volumes, and the textbook _Syntax: A Generative Introduction_.