LINGUIST List 24.2646
Mon Jul 01 2013
Review: Syntax; Dutch; English: Aelbrecht (2010)
Editor for this issue: Rajiv Rao
Meredith Johnson <majohnson25
The Syntactic Licensing of Ellipsis
E-mail this message to a friend
Discuss this message
Book announced at http://linguistlist.org/issues/21/21-643.html
AUTHOR: Lobke Aelbrecht
TITLE: The Syntactic Licensing of Ellipsis
SERIES TITLE: Linguistik Aktuell/Linguistics Today 149
PUBLISHER: John Benjamins
REVIEWER: Meredith Johnson, University of Wisconsin Madison
This book proposes a novel theory of ellipsis licensing, based primarily on data from Dutch modal complement ellipsis (MCE). In MCE constructions, the infinitival complement of a modal is elided, stranding the modal and its subject. The author makes two major claims: first, that ellipsis is licensed through an Agree relationship between an [E]-feature and the ellipsis licensing head; and second, that ellipsis occurs immediately after the licensing head is merged, rendering the ellipsis site inaccessible for further syntactic operations.
In Chapter 1, the author provides an overview of the various analyses of ellipsis phenomena. In the first part of the chapter, Aelbrecht distinguishes between structural and non-structural approaches to ellipsis. Next, the author introduces the two restrictions that any theory of ellipsis must account for: recoverability and licensing. In this work, Aelbrecht adopts the Phonetic Form (PF) deletion approach to ellipsis and addresses the issue of ellipsis licensing.
In Chapter 2, Aelbrecht describes MCE, a previously unstudied form of ellipsis. In Section 2.1, the author starts with an overview of the Dutch modal system. Aelbrecht shows that Dutch distinguishes epistemic, deontic and dynamic modal verbs. She argues that, syntactically, deontic and epistemic modals are raising verbs, while dynamic modals are control verbs. Furthermore, she presents evidence that modals head a ModP that takes a Tense Phrase (TP) infinitival complement.
Section 2.2 provides a detailed description of MCE. MCE is licensed by only root (i.e. deontic and dynamic) modals; ellipsis with epistemic modals results in ungrammaticality. Aelbrecht argues that MCE targets the complement of TP. She shows that MCE deletes verbs, objects, VP-level adjuncts, negation and auxiliaries, while subjects and temporal adjuncts survive the process. Interestingly, MCE displays an extraction asymmetry; subjects can move out of the ellipsis site, while objects cannot. Subject extraction is found with deontic modals; because they are raising verbs, their subjects are base-generated in the specifier of the embedded vP. Since the subjects of deontic modals survive ellipsis, they must be able to move out of the ellipsis site. In contrast, object extraction is blocked; both wh-movement and scrambling are banned with MCE.
Chapter 3 provides justification for the author’s two main claims about the licensing of ellipsis. Aelbrecht’s first claim is that ellipsis is licensed by an Agree between an [E]-feature and the ellipsis licensing head. Once an Agree relation is established, the complement of the head bearing the [E]-feature is elided. This analysis is essentially a refinement of Merchant’s (2001) approach to ellipsis, in which the presence of an [E]-feature on a head triggers ellipsis of its complement. In order to justify this analysis, the author provides evidence from a variety of ellipsis phenomena where the ellipsis licensor does not have to be adjacent to the ellipsis site.
The second claim is that ellipsis occurs as soon as the licensing head is merged. The effect is twofold: the ellipsis site is frozen for further syntactic operations; and lexical insertion at PF is blocked. This approach differs from previous analyses, such as Merchant (2001), who argues that ellipsis occurs post-syntactically. Aelbrecht shows that her “derivational ellipsis” analysis accounts for extraction asymmetries between subjects and objects. She shows that if the [E]-feature is located on T and the Mod head licenses ellipsis, then subjects have a position to which they can move between the ellipsis site and licensor; namely, Spec, TP. In contrast, objects do not have a similar escape hatch. Wh objects move to Spec, CP, which is above the licensing head; therefore, they cannot move out of the ellipsis site before the [E]-feature is checked and no further syntactic operations can apply. Similarly, scrambled objects move to a position higher than the modal (as evidenced by the surface word order), and thus also will not be able to move out of the ellipsis site.
In Chapter 4, Aelbrecht shows that data from four other elliptical phenomena is consistent with her analysis of ellipsis licensing. Section 4.1 discusses sluicing. The author follows van Craenenbroeck (2004), who argues that sluicing structures involve a double CP structure. The head of the higher CP (identified as ForceP) licenses ellipsis of the embedded TP; because of this, the lower CP (identified as FocP) intervenes between the licensing head and the ellipsis site, providing an escape hatch for wh-extraction.
Section 4.2 deals with verb phrase ellipsis. Aelbrecht follows Sag (1976), Zagona (1988) and Lobeck (1995) (among many others) and assumes that T is the licensing head. Furthermore, she assumes that vP is the constituent targeted by verb phrase ellipsis (Johnson 2001, Merchant 2008). Her theory predicts that extraction out of the ellipsis site should be possible, as the projection VoiceP sits between the licensing head and ellipsis site. This prediction is borne out; wh-extraction and adjunct extraction are both possible with verb phrase ellipsis.
Pseudogapping is the subject of Section 4.3. Pseudogapping is identical to verb phrase ellipsis, except that the object moves out of the verb phrase before ellipsis takes place. Aelbrecht follows Gengel (2007b), and assumes that the displaced object undergoes movement to a focus position between TP and vP. Since the landing site of focus movement is between the licenser and ellipsis site, extraction is possible.
The last section discusses the British English ‘do’ construction. Unlike verb phrase ellipsis, the ‘do’ construction does not allow wh-extraction or pseudogapping; however, subject raising is still possible. Aelbrecht argues that the ‘do’ construction is licensed by ‘do’ itself, which occupies v, and that it elides a VP constituent. This explains the limited extraction possibilities; the licenser triggers ellipsis of its complement, so there is no landing site for displaced constituents. Subjects of unaccusative verbs are able to move out of the ellipsis site because they are attracted by the licenser of ellipsis, v. When v is merged, it can attract the subject to its specifier before triggering ellipsis of the VP.
Chapter 5 concludes by reviewing the author’s two main claims about the licensing of ellipsis. The author outlines two goals for further research: seeing if other ellipsis phenomena can be accounted for under this analysis; and finding an answer for the longstanding question of why only certain syntactic heads are capable of licensing ellipsis.
A strong merit of this book is that Chapter 2 provides a very thorough description of a previously unstudied form of ellipsis. For that reason alone, this book is worth reading. On the theoretical side, Aelbrecht describes very clearly in Chapter 3 why the extraction asymmetry in MCE poses a problem for previous analyses of ellipsis. Furthermore, the new analysis proposed by the author neatly accounts for the data.
I was not left convinced that the analysis presented in the book can be extended to account for all ellipsis phenomena. Of course, ellipsis phenomena are incredibly diverse, so developing an analysis of all instances of ellipsis was outside the scope of this book. Rather than viewing the following as veritable “shortcomings” of this work, I would instead frame them as issues to consider in future research.
While this new analysis of ellipsis certainly seems to cover the data from MCE, I was left wondering if the new data warranted such a departure from other theories of ellipsis. The book would have benefited from a stronger rebuttal against approaches that equate phase heads with ellipsis licensors, such as Gengel (2007b). In Chapter 3, the author argues that the extraction data from MCE pose a problem for phase-based approaches. Aelbrecht notes that an object that undergoes wh-movement or scrambling in non-elliptical constructions would have to move out of the vP before the C head is merged. Under a phasal approach to ellipsis, the fact that objects cannot survive ellipsis cannot readily be explained. There are other logically possible, although potentially undesirable, ways to account for this asymmetry in a phase-based approach to ellipsis. For example, one could argue that ellipsis bleeds A’-movement of objects, or that the domains that constitute phases are subject to cross-linguistic variation (as argued for by Bošković 2005 and Müller 2007, among others ).
It is also unclear how this new approach could account for the licensing of antecedent-contained deletion (ACD). Aelbrecht argues explicitly that both covert and overt movements are blocked from the ellipsis site after the licensing head is merged; however, most accounts of ACD resolution rely on either Quantifier Raising (QR) (e.g. Fiengo and May 1994) or covert-rightward movement (Fox 2002) to obviate the infinite-regress problem. It’s possible that a cover-rightward movement analysis would work if that movement occurred before the ellipsis licensing head is merged, but this would need to be made explicit in future work.
The most obvious target audience for this book is syntacticians interested in ellipsis phenomena. However, this work addresses many other issues, making it a useful read for those interested in phases, Agree operations, scrambling and other forms of A’-movement, and the syntax of modal constructions.
Baltin, Mark. 2007. Deletion versus pro-forms: A false dichotomy? Ms, New York University.
Bošković, Željko. 2005. On the locality of left branch extraction and the structure of NP. Studia Linguistica 59(1): 1-45.
van Craenenbroeck, Jeroen. 2004. Ellipsis in Dutch Dialects. Utrecht: LOT Dissertation Series.
Fiengo, Robert and Robert May. 1994. Indices and Identity. Cambridge: MIT Press.
Fox, Danny. 2002. Antecedent-Contained Deletion and the Copy Theory of Movement. Linguistic Inquiry 33(1): 63-96.
Gengel, Kirsten. 2007a. Phases and ellipsis. In Proceedings of the 37th meeting of the North East Linguistic Society, Emily Efner & Martin Walkow (eds.). Amherst MA: GLSA.
Gengel, Kirsten. 2007b. Focus and Ellipsis: A Generative Analysis of Pseudogapping and other Elliptical Structures. PhD dissertation, University of Stuttgart.
Johnson, Kyle. 2001. What VP-ellipsis can do, and what it can’t, but not why. In The Handbook of Contemporary Syntactic Theory, Mark Baltin & Chris Collins (eds), 439-479. Oxford: Blackwell.
Lobeck, Anne. 1995. Ellipsis. Functional Heads, Licensing and Identification. Oxford: OUP.
Merchant, Jason. 2001. The Syntax of Silence. Sluicing, Islands and the Theory of Ellipsis. Oxford: OUP.
Merchant, Jason. 2008. An asymmetry in voice mismatches in VP-ellipsis and pseudogapping. Linguistic Inquiry 39(1): 169-179.
Müller, Gereon. 2007. On Deriving CED Effects from the PIC. Linguistic Inquiry 41(1): 35-82.
Sag, Ivan. 1976. Deletion and Logical Form. PhD dissertation, MIT.
Zagona, Karen. 1988. Proper government of antecedentless VPs in English and Spanish. Natural Language and Linguistic Theory 6: 95-128.
ABOUT THE REVIEWER
Meredith Johnson is a PhD student at the University of Wisconsin-Madison. Her main research interests are in syntactic theory; specifically, ellipsis phenomena, restructuring constructions, and the syntax of Algonquian languages.
Page Updated: 01-Jul-2013