Left edge topics in Russian and the processing of anaphoric dependencies

Abstract

This paper investigates the cost of processing syntactic vs. extra-syntactic dependencies. The results support the hypothesis that syntactic dependencies require less processing effort than discourse-derived dependencies do (Koornneef 2008, Reuland 2001, 2011. The point is made through the analysis of a novel paradigm in Russian in which a preposed nominal stranding a numeral can show number connectivity (PAUCAL) with a gap following the numeral or can appear in a non-agreeing (PLURAL) form:(1) cathedral-PAUCAL/PLURAL, there were three.PAUCAL__ Numerous syntactic diagnostics confirm that when there is number connectivity, the nominal has been fronted via A'-movement, creating a syntactic A'-chain dependency. In the absence of connectivity, the construction involves a hanging topic related via discourse mechanisms to a base-generated null pronoun. The constructions constitute a minimal pair and Reuland's proposals correctly predict that the A'-movement construction will require less processing effort compared to the hanging topic construction. A self-paced reading study for contrasting pairs as in (1) showed a statistically significant slow down after the gap with the hanging topic as opposed to the moved nominal. We take this to support the claim that a syntactic A'-chain is more easily processed than an anaphoric dependency 3 involving a null pronoun, which must be resolved by discourse-based mechanisms. 4 INTRODUCTIONNatural languages encode anaphoric dependencies in a number of ways. Safir (2004, 2008) introduces the term COCONSTRUAL as a theory-neutral label for any identity relation between two elements, pronounced or not. Coconstruals include antecedent-anaphor relations, filler-gap dependencies, control relations, variable binding, and independent coreference, among others, as illustrated in (1).(1) (a) Mike hurt himself. (c) The construction of syntactic coconstruals requires less effort than the construction of discourse coconstruals.One challenge in testing these claims is to find coconstruals of the different types that nonetheless represent minimal pairs. The goal is to avoid differences in the constructions that might influence the time course of processing, independent of the coconstrual type of interest, so that any processing differences can be that constructions involving movement will be easier to process than similar constructions with no movement. We introduce two syntactic constructions in Russian that will bear on this issue; they appear minimally different on the surface but involve distinct structures.The contrast is illustrated in (9). A nominal is fronted, stranding a modifying numeral. The nominal can appear in a form that agrees in number with the numeral, (9a), or it can appear in a non-agreeing plural form, (9b). We will show that the construction in (9a) involves A'-movement of the fronted element, and thus instantiates a syntactic coconstrual between the nominal and the empty category indicated as a struck-through copy. In contrast, (9b) involves coreference between the f...

The Scrambling Paradox

“…However, this appears to be too strong, as antireconstruction and surface scope relations are found with such sentences (Bailyn 2006). Thus, (63) shows a preference for surface scope (the only option for some speakers), consistent with the analysis of Russian scope in Antonyuk 2015, which is exactly the opposite of what is claimed to be found with Japanese longdistance scrambling (Saito 1992 Polinsky and Potsdam's (2014) account of the topical genitive plural constructions illustrated in footnote 43, in which a high base-generated element is discourse-related to a lower pro element in the structure. Such a theory could explain (63) without sacrificing base-generation: the surface left-dislocated element could serve as the locus of semantic interpretation, allowing for the high scope reading.…”

Concealed pied‐piping in Russian: On left‐branch extraction, parasitic gaps, and the nature of discontinuous nominal phrases

“…PGs in Russian have been reported by at least Franks 1992, Ivlieva 2007, and Polinsky & Potsdam 2014. While in English diagnosing the presence of a PG is relatively straightforward, the fact that argument drop is sometimes available in Russian, as (9) shows, makes it less obvious that a given gap is indeed a PG rather than a dropped NP.…”

“…PGs in Russian have been reported by at leastFranks 1992, Ivlieva 2007, and Polinsky & Potsdam 2014. While in English diagnosing the presence of a PG is relatively straightforward, the fact that argument drop is sometimes available in Russian, as (9) shows, makes it less obvious that a given gap is indeed a PG rather than a dropped NP.Since in this article we will pay special attention to PGs in object positions, we must establish what distinguishes PGs from dropped objects before we can safely use PGs to examine Russian LBE.Ivlieva 2007 notes that Russian PGs are most natural in perfective contexts.…”