Children's Comprehension of Relative Clauses in an Ergative Language: The Case of Basque

The investigation of the comprehension of L1 relative clauses across different languages has shown that subject relatives (SRs) are acquired earlier and responded to more accurately than object relatives (ORs). Most of this work has been based on SVO nominative-absolutive languages. In this article we present the results obtained in a binary picture-sentence matching task from L1 Basque, a highly inflected, ergative-absolutive, SOV language with prenominal relatives. The results obtained reveal that in Basque, ORs are responded to more accurately than SRs, which indicates that children do not comprehend SRs with more accuracy universally. The results are discussed in the light of different hypotheses put forward in the literature in order to account for the previously observed SR preference. Notes Portions of this article appeared previously as "Comprehension of Relative Clauses in L1 Basque" in Katie Franich, Kate M. Iserman & Lauren L. Keil (eds.), Proceedings of the 34th annual Boston University Conference on Language Development [BUCLD 34], 162–173. Somerville, MA: Cascadilla Press. Copyright is held by the author (©2010 M. Juncal Gutierrez). 1There is an alternative to the analysis outlined above known as the 'raising analysis,' according to which it is the head of the relative clause itself that moves out of its position inside the relative clause (CitationVergnaud 1974; CitationKayne 1994). See Sauerland (2000) for a discussion of both analyses. 2 CitationPolinsky et al. (2011) report no processing preferences in Avar between SRs and ORs. 3RCs with a wh- heading the RC are possible in Basque. However, they are less frequent in natural speech. They are usually learned at school, and they are used in a more formal register mostly by adults. 4In the case of the SR in (6), there is a mismatch between the case marking of the subject gap, which is ergative, and the case marking in the head noun, which is absolutive. This alone could increase the processing difficulty of SRs when compared to ORs. However, as CitationCarreiras et al. (2010) reported, despite having no such case mismatch in their SR and OR test sentences, the results showed that ORs were processed more easily than SRs. 5This hypothesis has been widely discussed in L1 (CitationHakuta 1981; CitationFrazier 1987; CitationKeenan & Hawkins 1987; and CitationFriedmann & Novogrodsky 2004, among others) and L2 (CitationO'Grady et al. 2003; CitationOzeki & Shirai 2007; and CitationYip & Matthews 2007, among others) acquisition studies of relative clauses, as well as in adult processing studies (CitationChen et al. 2010). 6As discussed in J. Hawkins (2007:341). See J. CitationHawkins (2007) for the view that the NPAH does not imply that SRs are acquired first. 7This is especially true for RCs with animate NPs (CitationHsiao & Gibson 2003; CitationKuo & Vasishth 2006). 8The IXA group is a computational linguistics research team from the Department of Computer Languages and Systems at the University of the Basque Country. See CitationCarreiras et al. (2010) for further details. 9We thank a reviewer for pointing this out to us. 10See footnote 5. 11For Babyonyshev & Gibson (1999), only elements that introduce new discourse referents (NPs and main verbs) enter into the calculation of distance between the gap and the filler. 12This task was part of a project in collaboration with Celia Jakubowicz and Laurie Tuller where a wider battery of tests was used to elicit different types of wh-structures. 13We thank Na'ama Friedmann for kindly allowing us to use these materials. 14The materials were adapted to the local dialect of Basque (a variety of Biscayan Basque). Since the differences between Standard Basque and the local dialect do not affect the syntax of relative clauses in any way, the examples are given in the Standard Basque. 15These two subjects behaved above chance level in the SRs. 16This would need to be further investigated. 17Throughout the appendices, the first sentence in italics corresponds to the actual test item administered in the local dialectal form. The version in standard Basque is provided below it.