Date: Sun, 27 Mar 2005 17:38:01 -0700 From: Michael Don Anderson Subject: The Syntax of Ditransitives: Evidence from Clitics
AUTHOR: Anagnostopoulou, Elena TITLE: The Syntax of Ditransitives SUBTITLE: Evidence from Clitics SERIES: Studies in Generative Grammar 54 PUBLISHER: Mouton de Gruyter YEAR: 2003
Michael Don Anderson, Department of Linguistics, University of Arizona
This monograph is intended for linguists with an interest in syntax, in particular ditransitives (Dative and double- object constructions) as they appear in different languages and as supported by clitics and NP movement. The general argument presented is that some well-formedness conditions arise only in certain instances of obligatory cliticization (including double-cliticization). Because these well- formedness conditions vary cross-linguistically, it provides an opportunity to examine the interaction of the morpho-syntactic properties of double- object (DO) constructions and the syntax of clitics -- which is the secondary goal of this publication. Although there is an emphasis on Greek to support the analysis, examples from English, Sesotho, French, Japanese, Spanish, Icelandic, Danish, Norwegian, German and Dutch are also included for typological contrast. An over-view of competing theories with regard to these varying languages and their Dative/DO constructions is provided as well. Anagnostopoulou follows the light applicative (vAPPL) head and main verb decomposition proposed by Marantz (1993), and distinguishes between DO constructions and prepositional Datives in terms of her analysis (implicitly refuting the promotion of indirect-object (IO) to direct object and demotion of direct object to adjunct as proposed by Perlmutter and Rosen 1984; Larson 1988).
Chapter One presents the broad claims of the book as suggested above, while Chapter Two looks at Dative clitics and NP-movement licensing to support those claims. Chapter Three investigates the impact of Case, EPP and Locality on Dative constructions; while Chapter Four points out how clitics obviate some of these Locality effects. Finally, in Chapter Five, she discusses restrictions on Dative constructions given Person-Case relations.
In Chapter Two, the author presents the claim that dative phrases in Greek are affected by the transitivity of the selecting predicate. Dative arguments can be Genitive (DP, PP or clitic/double-clitic) in transitive constructions; while in passives they cannot be Accusative (though she claims that there are no such restrictions on clitic -- including clitic- doubled -- DPs). She shows that this is supported not only by the Greek data, but in English, French, Italian, Sesotho, Chechewa and Dutch as well. In some instances, marginal constructions (Greek passives and unaccusatives; and English passives related to DO constructions) are improved by cliticization (or pronominalization for English). This is in part on what she bases her argument for a contrast between three-way Case/Agreement systems (such as Greek) and two-way Case/Agreement systems (such as English). In the former, unaccusatives and passives function the same with regard to DO constructions; while in the latter, the DO construction is only licensed for passives (unaccusatives rely on PP constructions). This typology is important for her later claims (e.g. that EPP and Case must be checked separately to account for certain quirky constructions).
Chapter Three, which is the real meat of the book, looks first at case theoretic accounts of NP movement in DO (and vAPPL) constructions. She contrasts goal-centered approaches with theme-centered, differentiating between true double-object languages and partial double-object languages. She then discusses extending these approaches to Greek, wherein the goal expresses Genitive or Dative Case and the theme expresses the Accusative. Because these established theories do not in fact account for Greek, she puts forth evidence that suggests that because Greek assigns a three-way Case/Agreement system (the theme gets structural Case while Genitive and Dative goals have inherent Case which is not suppressed in passives and unaccusatives) that this contradicts the assumptions of both theme- centered analyses and goal-centered accounts.
This lengthy chapter is organized as follows: Locality is considered, then Case and the EPP, followed by the implications of c-command, and finally Minimal Domains (under which the Minimal Link Condition (MLC) does not apply within domains but rather only checks for well- formedness across domains). The last section of Chapter Three raises Finite Complements, to account for problems with the Greek verb 'fenete' ('seems').
Locality as defined in (122) below, is an important consideration to much of the analysis presented. The author's claims rely on Locality with regard to relativized features rather than the position of arguments (or their features) (in contrast to Rizzi 1990). Her analysis is independent with regard to movement theories (e.g. Move FF or Agree as the mechanism for feature checking).
(122) If beta c-commands alpha, and t is the target of movement, then beta is closer to t than alpha unless beta is in the same minimal domain as (i) t or (ii) alpha.
She discusses the two main camps of interpretation of Case and EPP features within the Minimalist framework -- those that follow Chomsky (1995), where Case and the EPP are two separately satisfied requirements of a derivation; and those that follow Chomsky (2000a, 2000b, 2001) where the EPP triggers movement and Case is a by-product of Agree. She takes the salient distinction between the two positions to be that in the latter, only active phi-features trigger EPP movement to T, while in the 1995 version, the EPP can be satisfied by categories which lack Case or phi- features (but have some other features which satisfy the requirement). It is this older version of analysis that she adopts.
Anagnostopoulou argues that Icelandic quirkiness is in fact not dependent upon Chomsky's (2000) definition of quirky Case as "theta-related inherent Case with an additional structural Case feature" (p.88) because Japanese Datives would not qualify under this definition (among other considerations). She shows contrast between Greek and Icelandic which argue against, as she puts it, "Case, agreement and EPP/Move being collapsed into a single property 'quirky Case'" (p.88). In fact, she argues that Greek and Icelandic provide evidence that the EPP must be separated from agreement to account for the differences in quirky behavior in certain varying restrictions.
She spends a great deal of time explaining the differences between how Datives and Accusatives surface with regard to inherent and structural Case, especially in passives contrasting with DO constructions. She provides examples of passives from five different languages (English, Japanese, Icelandic, German and Dutch) and places them in a typological grouping wherein she finds that of the ten types of passives from these languages, three require that the EPP and Case be checked separately (by separate arguments). For example, Dative DPs with inherent Case reflect EPP motivated movement to T (Icelandic) along with English IO PPs (specific to inversion constructions). The remaining seven types follow Chomsky's notion that (in quirky constructions) both Case and EPP are satisfied at the same time (i.e., by the same DP). By the end of this section she has presented arguments and examples for the claim that goal DPs and PPs and theme DPs have Case and/or categorical features which can be checked at T.
Under the notion of c-command, she discusses the relationship between A- movement, linear order and hierarchical order, by examining well- formedness restrictions on NP-movement (of goals and themes in passives). She begins her discussion by making four claims: goal passives satisfy Shortest Move by moving the goal (DP1) to T from a position higher than the theme (DP2); theme passives are not allowed in constructions wherein the theme crosses the higher goal DP (violating Shortest Move/Closest Attract); theme passives then are only allowed when they are derived in the same manner as goal passives (i.e., the theme is already in a higher position and moves up from that higher position); and both goals and themes block movement regardless of whether they bear structural or inherent Case. She spends this section of the chapter supporting these claims with examples from English, Japanese and Icelandic or explaining why they don't apply. Furthermore, object shift (OS) is raised as a diagnostic for DO constructions, particularly with regard to Mainland Scandinavian (using examples from Swedish) and Icelandic, which display different properties. This is a useful discussion given her claim that these two languages are assumed to have the same syntax, for which she provides a counter-argument (Swedish allows OS can be a non-local derivation while Icelandic OS cannot). Her discussion (and main concern with c-command) is that she typologically distinguishes between symmetric (i.e. goal or theme may be passivized) and asymmetric passives (wherein the behavior of goals and themes are different) and that this accounts for some of the previously problematic data in various languages.
She makes an important generalization in this section that correlates local versus non-local movement scope with surface order of the objects. She finds that Parallel movement (cf. Richard 1997) corresponds with local derivation (moving the higher DP to a higher position in passives); and non-order preserving OS corresponds with a non-local derivation (raising the lower DP across the higher DP as initially claimed to be illicit).
The section on Minimal Domains primarily focuses on establishing that the MLC does not apply within domains (structuring it within the definition of Locality (122) above) but rather only checks for well-formedness across them. She provides examples of PP constructions that support her notion that movement of two arguments within a domain is "free" and that in those instances, the lower one can move across the higher one without sanction. She furthermore provides independently motivated support for Marantz's (1993) vAPPL construction, based on non-local non-order preserving structures. Her examples of PPs from Greek, French and Italian are accounted for in their distribution by this view of Minimal Domain and Locality. She ties together the claims made about OS and symmetry of passives in the context of vAPPL: passivization and OS are local derivations in asymmetric languages (and multiple movement is order preserving); passivization and OS permit the previously illicit non-local derivation (crossing DPs) in symmetric languages (but multiple movement is non-order preserving here). Crucially she says that IOs in DO constructions are introduced by the vAPPL head merged above the VP containing the theme (following and supporting Marantz 1993).
Chapter Four focuses on how clitics obviate some of the Locality effects noted in Chapter Three, once again relying on the MLC definition (122 above). She looks at clitics in French, Italian and Greek; clitic- doubling in Greek; scrambling in Dutch; A' movement in French, Italian and Greek; clitics and doubling in Spanish; and then how Agree and Move work in relation to pronouns on the Locality restrictions discussed to support her claim for typological distinctions between languages. Here she argues that violations of the MLC are obviated by Dative arguments (when clitics or members of clitic doubling chains) because they fall into the (i) t domain of (122). Her claim is that movement is allowed because the cliticized goal/experiencer is in the same Minimal Domain as t which is the target of movement. Toward the end of this section, she extends arguments in favor of Japanese scrambling (ala Miyagawa 2001) to Dutch scrambling based on cliticization and clitic-doubling. Different issues are raised with clitic-doubling in Spanish, but she argues that the grammaticality of monoclausal double-clitics in this language add further support that clitics obviate MLC violations.
The properties of clitic-doubling follow from the claims she makes about clitics in general, in terms of movement. Basically she states that a chain is formed under movement between the clitic in T and the in situ doubled DP. The higher argument surfaces as a clitic while the lower surfaces as a DP. NP-movement is no longer restricted by the intervening DP which has lost its D-features to cliticization.
Chapter Five focuses on a secondary problem with restrictions on DO constructions -- that of Person constraints. She compares the Person/Case restriction with the Person restriction to establish a unified theory to account for what has been proposed. Features and feature-checking relations are analyzed with regard to pronouns and Datives within the context of these restrictions. In person restrictions, the author argues that verbal phi-features are not eliminated simultaneously -- number is checked separately from person. This occurs whenever the same head is targeted (via Move or Agree) by both a Dative or Genitive IO argument and a lower argument with structural Case. This results in a restriction where the lower argument cannot check person but can only check number. The distinction is that in quirky constructions, the phi- features reside in T while in Person-Case environments, they reside in v-TR. Crucially she states that person checking is only obligatory under Move/Agree, not for in situ 1st or 2nd person pronouns.
This book is a well-written, easy to follow analysis of direct-object constructions across a multitude of languages. It is a useful contribution to the literature because it offers unified arguments which expressly states how it affects competing theories related to the author's claims -- all the while establishing her typological claims. Multiple theories within each of the sub-components of the analysis are considered (such as base-generated versus derivational claims for languages scrambling constructions), and the typology which the author proposes seems well- founded, though there are some few problems which she elects not to discuss (such as proposing that fronting examples in Greek support her claims of the base-order of arguments in Greek, but without any analysis of their construction).
One specific example in which she integrates external theories and provides independent support for one claim over another is found in Japanese scrambling. Scrambling in Japanese has two theoretical camps -- Miyagawa (1997) in favor of base-generated DO constructions and Tada (1989), Saitao (1992), Ura (1996), and Yatsushiro (2001) in favor of derived DO constructions. The author examines both theories within the context of the data and her analysis and offers support in favor of the Miyagawa theory.
As any good analysis should do, Anagnostopoulou anticipates potential objections to some of her claims as she discusses them and then obviates them. For example, one such objection comes from an argument that cliticization along with multiple movement leads to crossing dependencies (Richards 1997). Because this is the type of movement upon which she relies at one point of her analysis of clitics, it is important that she address this potential complaint -- which she does. She proposes that "tucking in" does not occur in her analysis because the nominative raises as an XP while the clitic is a head (or simply a set of features) and that she avoids the problem with crossing dependencies which might otherwise be expected. This type of anticipation and coverage of the syntactic theories relevant to her analysis provide a very effective means for letting the reader consider her claims, compare the data provided with external analysis and come to their own conclusion. One disadvantage to this is that while she covers a broad range of theories, she in some instances offers a more superficial comparison/analysis of these external theories with her own proposals, but it is a minor disadvantage given how well she integrates all the components.
On a final note, it is nice that in this monograph the reader is not left wondering if this analysis only applies to Romance languages or Germanic languages or whether it in fact is more comprehensive; because not only are multiple languages discussed, but multiple language families are included as well.
Chomsky, Noam. 1995. Categories and transformations. In The Minimalist Program, 219-394. Cambridge, Mass.: MIT Press.
Chomsky, Noam. 2000a. Minimalist Inquiries: The framework. In: Roger Martin, David Michaels and Juan Uriagereka (eds.), Step by step, 89-155. Cambridge, Mass.: MIT Press.
Chomsky, Noam. 2000b. Derivation by phase. In: Michael Kenstowicz (ed.), Ken Hale. A Life in Language, 1-51. Cambridge, Mass.: MIT Press.
Chomsky, Noam. 2001. Beyond Explanatory Adequacy. Ms., Department of Linguistics and Philosophy, MIT. [Subsequently published in: Belletti, Adriana (ed.). 2004. Structures and Beyond: The Cartography of Syntactic Structure, volume 3, pp. 104-131. Oxford: Oxford University Press.]
Larson, Richard. 1988. On the Double Object Construction. Linguistic Inquiry 19:335-392.
Marantz, Alec. 1993. Implications of asymmetries in double object constructions. In: Sam Mchombo (ed.), Theoretical aspects of Bantu grammar: 113-150. Stanford, CA: CSLI Publications.
Miyagawa, Shigeru. 1997. Against optional scrambling. Linguistic Inquiry 28: 1-25.
Miyagawa, Shigeru. 2001. The EPP, Scrambling and Wh-in-situ. In: Michael Kenstowicz (ed.) Ken Hale. A Life in Language, 293-338. Cambridge, Mass.: MIT Press.
Perlmutter, David M. and Carol G. Rosen (eds). 1984. Studies in Relational Grammar II. Chicago, Illinois: The University of Chicago Press.
Rizzi, Luigi. 1990. Relativized Minimality. Cambridge, Mass.: MIT Press.
Saito, Mamuro. 1992. Long distance scrambling in Japanese. Journal of East Asian Linguistics 1: 69-118.
Tada, Hiroaki. 1989. Scramblings. Ms., Department of Linguistics and Philosophy, MIT.
Ura, Hiroyuki. 1996. Multiple Feature-Checking: A Theory of Grammatical Function Splitting. PhD. dissertation. Department of Linguistics and Philosophy, MIT.
Yatsushiro, Kazuko. 2001. Case Licensing and VP Structure. Ph.D. dissertation, University of Connecticut.
ABOUT THE REVIEWER:
ABOUT THE REVIEWER
Michael Don Anderson is a Ph.D. student in the Department of Linguistics at the University of Arizona. He obtained his M.A. in General Linguistics at California State University at Fresno. His present interests include diachronic American English syntax (specific to 'have' inversion), idiom compositionality, contemporary productivity of allegedly inert morphemes, and how dialectic variability can be accounted within a Minimalist Framework (rather than strictly accounting for Narrow Syntax).