Island effects in Vietnamese relativization: A formal judgment study

Island effects in Vietnamese relativization: A formal judgment study Nguyen, Thi Quyen Faculty of English Ho Chi Minh City University of Economics and Finance, Vietnam quyennt@uef.edu.vn Abstract This paper examines island effects in Viet- namese relativization using methods in exper- imental syntax developed by Sprouse (2007). Typologically a wh-in-situ language, it is debatable whether Vietnamese employs wh- movement in the formation of relative clauses. If Vietnamese relativization is the result of wh-movement process, relativizing certain el- ements from inside island structures would re- sult in ill-formedness. Using a formal accept- ability judgment task and a factorial design of island effects that incorporates structure (is- land vs. non-island) and length of dependency (matrix vs. embedded) as variables, this pa- per found island effects associated with rel- ativizing elements out of all the island types under investigation. The findings confirm that Vietnamese relative clause formation is constrained by islands and thus solidify the universal status of island phenomenon, i.e. both wh-movement and wh-in-situ languages are sensitive to island violations. It is also suggested that formal acceptability judgment methods provide a useful tool for uncovering subtle linguistic phenomena. 1 Introduction One of the hallmarks of human language is the ex- istence of dependency between two or more ele- ments that can span an unbounded domain, as exem- plified in (1) where the dependency between whoi (filler/antecedent) and ti (trace/gap) spans three clauses. 1. Whoi does John think that Mary suspects Josh killed ti ? While long-distance dependency can be uncon- strained with respect to length or amount of inter- vening materials, there seem to be constraints on the types of structures that can contain the gap. The structures that cannot license a gap are referred to as islands and the low acceptability arising as a re- sult of the gap occurring inside islands is island ef- fect. Empirical investigation into the phenomenon has uncovered a number of island structures, no- tably among which are wh-island, complex NP is- land, subject island, and adjunct island. A number of accounts have been proposed to explain the un- grammaticality of island violations such as Subja- cency Condition (Chomsky, 1973), Barriers theory (Chomsky, 1986) and Conditions on Extraction Do- mains (Huang, 1982; Nunes and Uriagereka, 2000). Nevertheless, as an acceptability-based phe- nomenon, the sources of island phenomena con- tinue to be a topic of debate in linguistic literature, and wh-in-situ (henceforth WIS) languages take on an important role. In WIS languages, not much is revealed through the S-structure of sentences, i.e. whether movement takes place or not, and therefore it is often difficult to empirically determine whether WIS languages are subject to island effects. Some scholars have argued for the presence of island ef- fects (Lee, 1982; Han, 1992; Nishigauchi, 1990), while others have argued against it (Choi, 2006; Hwang, 2007; Ishihara, 2002; Sprouse et al., 2011). The source of such disagreements may lie in the fact that their claims are grounded in informal accept- ability judgments, which are essentially behavioral responses that, without careful control, can …

Content truncated to 500 words