The Nature of Object Agreement in Hungarian1

Verbal agreement is normally in person, number and gender, but Hungarian verbs agree with their objects in definiteness instead: a Hungarian verb appears in the objective conjugation when it governs a definite object. The sensitivity of the objective conjugation suffixes to the definiteness of the object has been attributed to the supposition that they function as incorporated object pronouns (Szamosi 1974; den Dikken 2006), but we argue instead that they are agreement markers registering the object’s formal, not semantic, definiteness. Evidence comes from anaphoric binding, null anaphora (pro-drop), extraction islands, and the insensitivity of the objective conjugation to any of the factors known to condition the use of affixal and clitic pronominals. We propose that the objective conjugation is triggered by a formal definiteness feature and offer a grammar that determines, for a given complement of a verb, whether it triggers the objective conjugation on the verb. Although the objective conjugation suffixes are not pronominal, they are thought to derive historically from incorporated pronouns (Hajdú 1972), and we suggest that while referentiality and ϕ-features were largely lost, an association with topicality led to a formal condition of object definiteness. The result is an agreement marker that lacks ϕ-features.

Abstract. This article discusses a syntactic puzzle posed by Hale ; namely, the variation in the location of agreement morphology in reduced object-relative clauses. The contribution of this paper is the proposal that phi features are uniformly on the super-ordinate D, and the micro-variation in the location of agreement morphology is an outcome of the difference in the adjunction site of the extracted object. This analysis makes predictions for subject relatives as well as non-reduced relative clauses. The major theoretical implication of this paper is that clause-external phi features are responsible not only for genitive subject-case licensing but also for overt subject-verb agreement in reduced and non-reduced relative clauses.

2007

This work aims at analyzing some empirical data 1 from Italian Sign Language (LIS) related to its CP domain, in the light of Antisimmetry and the so-called Split-CP. It focuses on the possibility that in LIS too, Topicalization, conditional clauses, relative clauses, along with different kind of interrogative and soubordinate clauses are explained LIS-Antisymmetry and split CP

The LACUS forum, 1998

2014

Johnson analyses Gapping as resulting &om Across-the-Board verb movement fiom conjoined VPs: binding domain (contrast wJ(10) for Gapping): 20. Robinl could speak French [CP before shel could Russian] We point this out primarily because Levin gives the above as the principal argument for not showing PG and Gapping as related. We have seen, though, that we can capture the similarities between the two fiom the fact that they both involve ATB Vto-I, while capturing the important daerences with the claim that the ATE3 proceeds fiom dflerent types of conjuncts. 4. Pseudogapping: NOT a Special Case of VPE Lam& (1 995,1999) has offered an interesting analysis of PG as a special case of Verb Phrase Ellipsis (WE). Here, we show several problems for Lasnik's idea that the new analysis does not face. Lasntk shows PG as a two-step process: 1) overt raising of a verbal complement to [Spec, Agr-oP], followed by 2) WE. So a sample derivation becomes: 2 1. Robin could speak French and [TP Kim could [AGR-OP Italian,-311 However, this forces the prediction that any and all languages with PG must have W E as well. This prediction simply does not bear out: 22. German a. Robin konnte Russiche sprechen bevor Kim Franzoesich konnte could Russian speak before French could 'Robin could speak Russian before Kim could French' b. *Robin ksnne Fisch essen, und Kim k(inne auch W' Ebad.) 'Robin can eat fish, and Kim can also' 23. Latvian a Vina var runat angliski, un v i d var italiani (PG good) She can speak English, and he can Italian b. *Vina var runat angliski, un ving var ari (VPE bad) She can speak English, and he can also Our analysis (apparently correctly) connects PG with Gapping, rather than WE, in the implicational universal. Lasnik's analysis faces other empirical problems as well. Lasnik relies on [Spec, Agr-oP] a s a landing site for the overt movement of the surface right remnant. The following examples, though, prove unlikely candidates for such overt raising: 24. a. You behaved shamefully, but I did behaw [ADW bravely] b. This new road will lead to Clovis, and that one will lette [PP to Fresno] c. Robin is likely to win, and Kim is k k l y [IP to lose] d. Pat may believe now that every cloud has a silver lining, but she will tomorrow believe [CP that no good can ever come to people in this evil, evil world] None of the above bracketed elements has Case or Agreement features normally associated with AgrP. If one loosens the concept of the role of Agr-oP @amk appeals to an EPP feature checked there), problems remain. Adverbs do not make good subjects, so the (a) form would not seem to allow raising to [Spec, Agr-oP]. In (d), we see an extraposed clausal complement, which cannot have [Spec, Agr-oP] as its landing site. Furthermore, contrary to the expectations of Las~uk's analysis, PG and W E differ in important empirical ways. For instance, PG shows island effects, whereas W E does not: 25. a. Robin can speak Russian, and I know [a fXend [who can ff3ettk &&kl&m too]] b. *Robin can speak Russian, and I know [a friend [who can ff3ettk Italian]] 26. a. Robin will fascinate the children, and I believe [the claim [that Kim will too]] b. ?*Robin will fascinate the children, and I believe [the claim [that Kim will hsektik the adults]] For us, the illformed PG examples fall out under general constraints on movement. Since W E does not involve movement, no such problem exists. Also, as Levin (1 986:54) notes, WE readily allows for more than one supporting auxiliary, while PG does not. The following contrast: 27. a. Robin has been playing the oboe, and Kim has been phyk&w eeee too b. ?*Robin has been playing the oboe, and Kim has been p h p g the bassoon 28. a. Pat could have been drinking beer, and Kim could have been. . €lmhgk% too b. *Pat could have been drinking beer, and Kim could have been gin Lasnik equates PG with VPE and hence cannot explain the above contrasts. For us, PG involves V-to-1 movement. We take I as including TP and Agr-sP. Note that to amve at the (b) forms above, the ATB Verb movement would have to have as its landing site a projection below IP (perhaps an Asp head position) The degradation follows, then, from a suboptimal landing site. 5. Conclusion Pseudogapping and Gapping are the same, but different. They are the same in that they both involve ATB V-to-I movement; they are different in that PG shows asymmetric ATB movement. Our unification of PG as essentially a marked type of Gapping enables us to make a number of correct predictions, and avoids the set of problems facing's Lasnik's VPE-spirited analysis of PG.

Acta Linguistica Hungarica, 2003

This paper focuses on predicate formation operations which affect the value and determination of lexical properties associated with Hungarian phrasal periphrastic predicates and, hence, on lexeme-formation (Aronoff 1994). Recent work, following the word and paradigm morphological models of Robins (1959), Matthews (1972), among others, has argued that periphrasis or multi-word expression is often best viewed as a type of morphological exponence, i.e., as the product of morphological rather than syntactic operations, contra many current theoretical proposals. In line with this morphological perspective, I argue that, as in inflection, periphrasis is a type of morphological exponence for lexeme-formation. In support of this claim I explore lexeme-formation for several sorts of phrasal predicates in Hungarian (Ackerman 1987; Komlósy 1992; Kiefer-Ladányi 2000, among others), in particular causative formation, causal predicate formation, so-called reiterated activity formation expressed by reduplicated preverbs, and the interaction of these operations with category changing derivation. The general background for the analysis will be the Realization-based Lexicalist Hypothesis (Blevins 2001) and realizational approaches to morphology (Stump 2001) which are compatible with theories subscribing to representational modularity (Jackendoff 1997; 2002). ½º ÁÒØÖÓ Ù Ø ÓÒ Hungarian, like several other Uralic languages (see Kiefer-Honti 2003) contains phrasal predicate constructions in which a syntactically separable preverb (PV) combines with a verbal stem (Vstem). The basic properties of such constructions have been characterized as follows: "In verbal constructions the preverb may keep its original adverbial meaning (e.g., * I thank the participants at the 10th International Conference on Morphology at Szentendre, Hungary as well as two anonymous and very helpful reviewers for comments on an earlier version of this paper.

2014