This paper describes the basic properties of quotation in Russian Sign Language (RSL) using corpus and elicited data. Quotation constructions in RSL consist of source indication (optional), a predicate of speech (optional), and the quote itself. The quote as well as the other constituents of quotation constructions can be marked by non-manual markers (also known as role shift), namely by eye gaze change, head turns, and body turns; however, these markers are optional. Judging by the behavior of indexicals, quotation in naturalistic RSL narratives mostly involves direct speech. Mixed behavior of indexicals is also possible. Interestingly, non-manual marking does not correlate perfectly with the behavior of indexicals. We also find evidence that at least quotes with shifted indexicals are not syntactically embedded. The properties of RSL quotation constructions seem to present problems for some current accounts of role shift in other sign languages.

This chapter presents the Russian Sign Language (RSL) Corpus and demonstrates its capabilities as a research tool by summarizing three corpus-based studies primarily focused on syntactic functions of nonmanual markers. The first study considers question marking in regular wh-questions and in question-answer pairs. It shows that the two constructions have very different nonmanual markers. The second study analyzes marking of topics in RSL, and shows that nonmanual markers of topics are typologically common, but are infrequent in naturalistic corpus data. The third study investigates conditional and concessive constructions in RSL. It demonstrates that these constructions make extensive and frequent use of nonmanual markers, but that no single marker is specialized for the function of expressing conditional or concessive meaning. Instead, complex combinations of multiple markers are employed in these constructions. All three studies also contribute to sign language typology by providing novel descriptions of syntactic and discourse phenomena in RSL.

No abstract

The present research combines three fields of inquiry in sign language linguistics: verbal agreement, person features, and syntactic complexity. These topics have previously been addressed in isolation, but little is known about their interaction. This study attempts to fill this gap by investigating subject agreement in complement clauses in Russian Sign Language. By means of corpus investigation and grammaticality judgments, I found that subject agreement in clausal complements of the control predicates try, love, want, begin, and modal can may be deficient – in particular, it can be reduced to the forms identical to first-person marking even in the case of a third-person subject controller. Deficient subject agreement in complement clauses is thus reminiscent of non-finite verbal forms in spoken languages. I further argue that the choice of first-person forms in deficient agreement reveals a default status of first person in sign languages, which is consistent with proposals regarding the modality-specific properties of first-person reference in these languages.

We describe basic morphosyntactic and semantic properties of question-answer pairs (QAPs) collected from the online corpus of Russian Sign Language (RSL). We identified two classes of QAPs: classical and discourse QAPs, which are different in the semantic relation between the question and answer parts. We discovered that non-manual marking and word order in both types of QAPs are different from other constructions involving wh-signs, namely regular questions and free relative clauses. Guided by the similarity between non-manual marking of QAPs and role shift marking, we hypothesize on a possible grammaticalization process connecting the two constructions.