| Popis: |
One of the basic properties of natural language that linguists must explain—one of two facts Chomsky et al. (2019) call “non-negotiable” parts of any linguistic theory—is displacement, namely the fact that words can be interpreted as if they occupied a position other than where they are pronounced, as in (1). (1) What did you read ______? What is especially puzzling about displacement is that there are many cases where it seems like it should be possible, but it is not. For example, (2) is possible, but not (3). (2) What did you read a book about ______? (3) * What did you read Gretchen’s book about ______? (cf. I read Gretchen’s book about Internet language.) Constructions that prohibit displacement are called syntactic islands, and they have been the topic of intense research since the term was coined by Ross (1967), yet many puzzles remain (see Boeckx 2012; Citko 2016; Szabolcsi & Lohndal 2017 for an overview). One puzzle has to do with cross-linguistic variation: some islands appear to be universal, but others vary by language. For instance, English 'whether' is an island to displacement (4), but the Spanish equivalent 'si' is not (5). (4) * Who don’t you know whether ______read a book? (5) ¿ Quién no sabes si ______ leyó un libro? who no you.know whether read a book ‘Who don’t you know whether read a book?’ In the last ten years, the experimental investigation of syntactic islands has undergone a boom, with the establishment of a widely accepted experimental protocol developed by Sprouse and colleagues (Sprouse et al. 2016; Sprouse, Wagers & Phillips 2012) for isolating the effects of islands on sentence judgments. Yet despite the boom, islands in Spanish remain understudied (Pañeda et al. 2020). We aim to fill this gap by using a formal acceptability judgment experiment to test five types of syntactic islands in Spanish. We employ Sprouse et al.’s factorial paradigm to test island constructions that are claimed to be universal (Complex NP islands) and those that are claimed to differ for Spanish and English (whether-islands, wh-islands, subject islands). Our main research question is therefore: Do native speakers of Mexican Spanish display syntactic island effects in their judgments of Complex NP islands, whether-islands, wh-islands, and subject islands? Within this broad question, we set out to address some specific claims in the literature about the factors at play in how these island effects are realized. For instance, for wh-islands, it has been claimed that restrictions on displacement vary according to the wh-word and the extraction site (Torrego 1984). For subject islands, it has been claimed that the position of the subject (pre-verbal versus post-verbal) matters (Gallego & Uriagereka 2007). Finally, it has been claimed that processing is important for all island types, such that it is valuable to compare island effects to individual working memory capacity (Pham et al. 2020). We can therefore identify several more specific research questions, presented in (6). (6) Research questions 1. Do Mexican Spanish speakers display syntactic island effects in their judgments of Complex NP Islands (when extracting subjects or objects)? 2. Do Mexican Spanish speakers display syntactic island effects in their judgments of wh-islands when extracting subjects or objects over a wh-argument? 3. Do Mexican Spanish speakers display syntactic island effects in their judgments of wh-islands when extracting subjects or objects over a wh-adjunct? 4. Do Mexican Spanish speakers display syntactic island effects in their judgments of whether-islands (when extracting subjects or objects)? 5. Do Mexican Spanish speakers display syntactic island effects in their judgments of subject islands when the subject is in pre-verbal position? 6. Do Mexican Spanish speakers display syntactic island effects in their judgments of subject islands when the subject is in post-verbal position? 7. Do island effects on judgments vary by individual working memory capacity as measured by a backward digit span task? 8. Do island effects on judgments vary by individual vocabulary size as measured by a lexical decision task? |
