Local and Non-Local Binding of Reflexives and Pronominals in Russian Object Control Infinitives

This paper proposes a new analysis of the structure of the Spanish control verb construction in which the matrix verb selects the infinitival clause. Starting from the fact that there is a difference in syntactic phenomena, that is, clitic climbing, between control verb constructions, it turns out to be related to the type of matrix verb. It is argued that both infinitival clause selected by the volitional verb and infinitival clause selected by the non-volitional verb project CP category. I suggest that syntactic difference captured in these two constructions is due to Merge applied in the derivation process. Following Epstein et al.(2016), I assume that internal Merge and external Merge are freely applied. Under the concept of phase cancellation proposed by Chomsky(2015) and Epstein et al.(2016), I argue that C becomes visible in the derivation for infinitival clause selected by non-volitional verb and C becomes invisible in the derivation for infinitival clause selected by volitional verb.

The aim of the current is to investigate long-distance agreement (LDA) in Urdu infinitive clauses specifically emphasizing on DPs case licensing, agreement, and movement, employing Bošković’s model (2003, 2006, 2007). Previous studies articulated that long-distance agreement (LDA) is mediated by case-driven in long-distance movement, short distance movement in diverse range of languages including Hindi Urdu, Kashmiri, and Tsez to the clause edge, thereby enabling agreement across domains. Such outcomes established a close link between case and agreement, with movement providing the structural configuration necessary for agreement. Employing qualitative research design, Urdu demonstrates a different pattern in infinitival clauses as in Urdu, they are structurally reduced (vP/VP rather than CP/TP), which blocks accessibility for agreement under the Phase Impenetrability Condition (PIC). Consequently, matrix verbs, seskha and passand and auxiliaries, hai, license ergative or accusative case but fail to establish φ feature agreement with embedded infinitival or the object DPs. On the other hand, move could theoretically circumvent PIC utilizing successive cyclic movement, the Urdu instances show that such movement is not activated, blocking agreement. These findings deviate from previous Hindi Urdu accounts that reveal the LDA triggered by embedded objects and instead support Bošković’s (2007) distinction: Agree is locality bound and blocked in reduced clauses, while Move offers a potential escape route that remains unused in Urdu nonfinite structures. This study contributes to the broader theoretical debate by representing that Urdu gives strong empirical evidence for phase theory while showing heterogeneity in agreement.

We report results from two eye-movement experiments that examined how differences in working memory (WM) capacity affect readers' application of structural constraints on reflexive anaphor resolution during sentence comprehension. We examined whether binding Principle A, a syntactic constraint on the interpretation of reflexives, is reducible to a memory friendly “recency” strategy, and whether WM capacity influences the degree to which readers create anaphoric dependencies ruled out by binding theory. Our results indicate that low and high WM span readers applied Principle A early during processing. However, contrary to previous findings, low span readers also showed immediate intrusion effects of a linearly closer but structurally inaccessible competitor antecedent. We interpret these findings as indicating that although the relative prominence of potential antecedents in WM can affect online anaphor resolution, Principle A is not reducible to a processing or linear distance based “least effort” constraint.

Scandinavian has a reflexive pronoun and a reflexive possessive for the 3rd person, and a reciprocal pronoun for all persons. Regular binding domains are finite and non-finite clauses, small clauses, and noun phrases with a verbal content and a genitive ‘agent’. There are also less expected binding relations within NPs, possibly involving an invisible binder. Within VP an indirect object may bind a direct object. Even non-c-commanding binders within VP do exist. Non-local binding into small clauses and infinitival clauses is frequent. Some varieties, especially Norwegian, also allow long distance binding, i.e. binding into finite subordinate clauses. At this point, there is a great deal of variation in acceptability, and definite rules are hard to identify.

External input is necessary to acquire language. Consequently, the comprehension of various constituents of language, such as lexical items or syntactic and semantic structures should emerge at the same time as or even precede their production. However, in the case of pronouns this general assumption does not seem to hold. On the contrary, while children at the age of four use pronouns and reflexives appropriately during production (de Villiers, et al. 2006), a number of comprehension studies across different languages found chance performance in pronoun trials up to the age of seven, which co-occurs with a high level of accuracy in reflexive trials (for an overview see e.g. Conroy, et al. 2009; Elbourne 2005). From a theoretical perspective, only syntactic rules are assumed to affect the interpretation of reflexives, which receive their interpretation from a previously mentioned antecedent within the same sentence. Pronouns, on the other hand, require semantic and pragmatic considerations to identify the correct antecedent and their interpretation is affected by phenomena such as focus and prominence (e.g. Chomsky 1981; Hendriks and Spenader 2006; Reinhart 1983). This implies more complex resolution processes to find the correct interpretation of pronouns, which in turn poses a higher demand during language comprehension than reflexives. This has previously been pointed out by for example Bloom and colleagues (1994) and Grodzinsky and Reinhart (1993). Can this additional demand during pronoun comprehension alone explain the year-long asymmetry in production and comprehension and the strong advantage of reflexives? As noted by Matthews and colleagues (2009), no linguistic account so far can satisfyingly capture children's behaviour in studies regarding pronoun acquisition. An alternative or additional explanation of previous results would be an impact of the experimental setting itself on children's performances. The methods usually employed to investigate child language comprehension, such as act-out-tasks, generally require the generation of an action in response to a previously comprehended sentence (for an overview of tasks used and the variety of outcomes see Conroy, et al. 2009). This naturally poses additional demands on

Anaphora interpretations in Spanish utterances and the neo-Gricean pragmatic theory

The difference between the strong and weak constraints four-dimensional variational (4DVAR) analyses is examined using the representer method formulation, which expresses the analysis as the sum of a first guess and a finite linear combination of representer functions. The latter are computed analytically for a single observation under both strong and weak constraints assumptions. Even though the strong constraints representer coefficients are different from their weak constraints counterparts, that difference is unable to help the strong constraints compensate for the loss of information that the weak constraints includes. Numerical experiments carried out in the Agulhas retroflection for single and multiobservation assimilations clearly show that the weak constraint 4DVAR produces analyses that fit the observations with significantly higher accuracy than the strong constraints.

This article addresses the question of L1 transfer in L2 acquisition of reflexive binding. It incorporates recent research on Binding Theory which focuses on the relationship between morphological complexity of anaphors and the occurrence of long-distance binding of reflexives (cf. Yang, 1983; Pica, 1987; Hellan, 1988; Battistella, 1989; Huang and Tang, 1989; Cole et al., 1990; Progovac, 1992). Reflexives typically fall into two categories: simple (X0) reflexives that may take long-distance antecedents and complex (XP) refle xives that may not. Acquisition of the English binding pattern by native speakers of Serbo-Croatian requires recognition of the morphological com plexity of English reflexives. Prior to reanalysis, learners are predicted to transfer the L1 X0 anaphor type and incorrectly assign long-distance antece dents to English XP reflexives. The interpretation of English reflexives by native speakers of Serbo- Croatian was investigated using two types of written sentence comprehension tasks. A picture identification task and a multiple-choice questionnaire were administered to intermediate ( n = 20) and advanced (n = 20) L2 learners and a group of English native speaker controls (n = 20). Results consistent across task type support the transfer hypothesis and suggest learners have access to Universal Grammar in second language acquisition.

Language behavior by mildly handicapped and nonretarded children on complement clauses

Findings on perception of missing subjects in complement clauses are reported in two experiments with school-aged children and adults. The complement clauses were embedded into four matrix verbs in Greek, such as Ask (ask information), Promise, Tell (order/command), and Tell (give information) equivalent to English in syntactic and semantic constraints. The data from Experiment 1 show the following: (1) Perception of missing subjects in Ask, Promise, and Tell complement clauses is significantly higher than in Tell, but Guttman coefficient of scalability (.58) was slightly lower than the required one (over .60) to document a developmental sequence between the four constructions. (2) In general, these results, as other findings, contradict with Minimal Distance Principle (MDP) advocated by Chomsky (1969, 1972). (3) Differences in perceiving the source (i.e., NP) in Ask and Promise and the goal/recipient (i.e., NP) as subject in the complement clauses on Tell and Tell cannot consistently be explained by the Semantic Role Principle (SRP) postulated by Maratsos (Lederberg & Maratsos, 1981; Maratsos, 1974) either. (4) Results from Experiment 2 with three age groups, despite minor differences, confirm the results in Experiment 1, suggesting that comprehension of complement clauses with Ask, Promise, and Tell is a prerequisite to comprehension of complement clauses with Tell, according to Guttman coefficients of scalability (over .60 for all groups). (5) The findings are discussed within the framework of the SRP, but the emphasis is placed on the interaction of semantic and pragmatic presuppositions over processing the four construction types.

All languages have expressions, typically pronominals and anaphors, that may or must depend for their interpretation on another expression, their antecedent. When such a dependency is subject to structural conditions, it reflects binding. Although there is considerable variation in binding patterns cross-linguistically, in fact, variation is along a limited set of parameters. The Germanic languages exemplify some of the main factors involved. In Germanic, third-person pronominals generally do not allow binding by a co-argument. However, in Frisian and Afrikaans, they do, being embedded in a richer structure than meets the eye. In Continental West Germanic and Scandinavian, anaphors come in two types: simplex anaphors (SE-anaphors)—deficient for number and gender—and complex anaphors (SELF-anaphors). These typically consist of a pronominal or SE-anaphor combined with an element like Dutch zelf ‘self’ or one of its cognates. In all the Germanic languages SELF-anaphors are bound in their local domain—approximately the domain of their nearest subject—except in a few identifiable positions, where they are interpreted logophorically. That is, they accept a non-local antecedent, provided this element holds the perspective of the sentence. The distribution of SE-anaphors involves three different conditions. First, they can be bound by a co-argument only if the verb belongs to a restricted class, which allows syntactic detransitivization. Second, in general, SE-anaphors allow non-local binding. But the conditions differ among subgroups. In Dutch and German, they can only be bound non-locally when contained in a causative or perception verb complement or a small clause. In Mainland Scandinavian, non-local binding is, in principle, available to all infinitival clauses (subject to some dialectal variation). For instance, in some varieties of Norwegian, referentiality of intervening subjects restricts binding; in other varieties, the restricting factor is not “finiteness” but “being specified for tense.” Third, in Icelandic long-distance antecedents beyond the infinitival domain are licensed by a subjunctive, together with the requirement that the antecedent holds the perspective. Faroese largely patterns like Icelandic, although lacking a subjunctive. However, the class of verbs that allow this pattern coincides with the class of verbs in Icelandic that have a subjunctive complement. Non-local binding of SE-anaphors is sensitive to the requirement that the antecedent be animate, but the languages show differences in the details. Unlike the West Germanic languages, the Scandinavian languages all have a possessive reflexive in third person. In general, their distribution appears to be quite close to that of SE-anaphors, but this is subject to dialectal variation, with various differences in the details.

This article provides experimental evidence for the existence of long passive in Dutch and proposes a syntactic analysis of the findings. Long passive, a control-like complementation configuration with matrix passive and promotion of the embedded object to matrix subject, has long been considered ungrammatical in Dutch, but naturally occurring examples call for reconsideration. We report on a judgment experiment, which yielded two main findings: (a) long passive in Dutch is subject to considerable speaker variation and fully acceptable to a number of participants, and (b) its acceptability depends on the class of the matrix verb. We propose that long passive (qua implicit control and long object promotion) is in itself unproblematic and that the observed interspeaker variation is due to the incompatibility of passive participles with the so-called infinitivus pro participio effect (the occurrence of an infinitive-like form in place of a participle in the perfect) and the (un)availability of a strategy to obviate this clash. This provides a new perspective on the syntax underlying the infinitivus pro participio effect. We link the differences between matrix verb classes to structural properties of the infinitival clauses that the verbs belonging to these classes embed, offering novel insights into the syntax of the infinitival dependents of Dutch aspectual verbs, which, we argue, include a (covert) PP layer, similarly to obligatory control adjuncts.

Agreement and movement in Icelandic raising constructions

This paper investigates the phenomenon of the replacement of Accusative case marking on the direct object of a transitive infinitive (or, rarely, participle) by the Genitive when the non-finite clause is embedded under a negated matrix verb. Basing myself on data collected from native speakers, corpora and the Internet, I show that the phenomenon of long-distance Genitive of Negation in Lithuanian is acceptable (and often obligatory) with various kinds of matrix verbs: subject control verbs, object control verbs with Dative, Genitive and Accusative objects, and some complex noun + verb predicates. In some of these instances, Genitive of Negation can affect more than one direct object. Besides that, the case-marking rule is virtually unbounded in its application, being able to target deeply embedded direct objects, provided that there is a chain of infinitival clauses. The application of this rule shows considerable variation, which depends on the type of the matrix verb, on the degree of syntactic embedding, on word order and also to a large extent on individual preferences of speakers. From an areal perspective Lithuanian is shown to pattern with the more conservative Slavic languages (Polish and Slovene), Latgalian and the Baltic Finnic languages Estonian and Finnish, rather than with the closely related Latvian, which, like Czech, has abolished Genitive of Negation almost completely.

Postal (1974) (with roots much earlier) analyzed ECM subjects as raising into higher clause. Chomsky (1973, 1981) instead argued that the boundary of an infinitival clause is so weak that it is as if its subject is in the higher clause for Case and other purposes. Davis (1984) proposed that it is infinitival Infl rather than the higher verb that licenses accusative Case on its specifier, similar to a classic analysis of the Accusative-Infinitive construction in Latin, except that in English, the accusative assigning feature must be inherited from the matrix verb. I discuss a variety of ECM-type phenomena in English, concluding that for some speakers, ECM subjects can, but need not, raise into the higher clause. Then I show how an extension of Davis’s account covers the range of phenomena considered.