Asymmetries in relative clause comprehension in three European sign languages

Abstract

Relativization is a robust subordinating type across languages, displaying important typological variability concerning the position of the nominal head that the relative clause modifies, and sign languages are no exception. It has been widely assumed since Keenan & Comrie (1977) that the subject position is more accessible to relativization than object and oblique positions. The main aim of this paper is to investigate the extension of this famous generalization both across modalities (sign as opposed to spoken languages) and across relativization typologies (internally as opposed to externally headed relatives), and to verify how it interacts with age of first language exposure. We here report the results of a sentence-to-picture matching task assessing the comprehension of subject and object relative clauses (RCs) in three sign languages: French Sign Language (LSF), Catalan Sign Language (LSC), and Italian Sign Language (LIS). The results are that object RCs are never easier to comprehend than subject RCs. Remarkably, this is independent from the type of relative clause (internally or externally headed). As for the impact of age of exposure, we found that native signers outperform non-native signers and that a delay in language exposure emphasizes the subject/object asymmetry. Our results introduce a new potential diagnostic for LF movement: the existence of a Subject Advantage in comprehension can be used as a reliable and measurable cue for the existence of long-distance dependencies, including covert ones.

Nominal Word Order Typology in Signed Languages

“…While the LSM data comes from my own fieldwork, supplemented by Cruz Aldrete's (2008) descriptive grammar, the majority of the data used in this typological study comes from published manuscripts, including grammars of particular signed languages, small scale typological studies of other phenomena in signed languages (e.g. Branchini, 2014;Wilbur, 2017;Hauser et al, 2021), and larger scale typological studies and handbooks of signed languages (e.g. Zeshan, 2006;Pfau et al, 2012;Zeshan and Palfreyman, 2017).…”

Challenges in the research of relative clause constructions in signed languages to date

“…" (p. 62) Showing two similar animate agents in stimuli may also lead to extensive use of subject relativization which could be another drawback. In order to minimize this effect, some researchers added third animate agents to observe the characteristics of object relativization (e.g., Hauser et al 2021). By showing pictures including three referents, signers are given a choice between character perspective and observer perspective.…”

“…By showing pictures including three referents, signers are given a choice between character perspective and observer perspective. However, for the comprehension tasks, it is very useful to check the typological differences around subject preference, same as Hauser et al (2021) did by showing the participants signed video input and testing comprehension.…”

“…To understand the subject advantage of LSF RCs, she conducts visual world paradigm using eye tracking with LSF videos. In addition, Hauser et al (2021) also use similar picture elicitation to compare RCs in three different sign languages, LIS, LSC and LSF, and use a comprehension task to see what sort of utterances were understood.…”

“…. Some comprehension tasks (e.g.,Hauser et al 2021) regarding RCs use similar processes and therefore these tasks are discussed in the category named elicitation methods.2. The authors use the abbreviation LSB, however, the current use of language for Brazilian Sign Language is Libras Nunes and Quadros (2008).…”