Türkçeyi göçmenlik ortamında yetişkinken öğrenen anadili İtalyanca olanların artgönderim yorumu

Abstract

The research carried out in the recent years
 indicates that the structures at the syntax-discourse interface fall in a
 vulnerable domain for bilinguals (Sorace, 2011 for review). It has been
 proposed that cross-linguistic interference occurs when syntactic features of
 the two languages partially overlap (Müller & Hulk 2001 among others) due
 to the rise of optionality. Subsequent studies (e.g., Sorace & Serratrice,
 2009; Serratrice et al., 2012) found a cross-linguistic interference in
 bilinguals speaking two typologically similar languages. Recently, Sorace
 (2016) has proposed that interference may be due to the cognitive load of processing
 two languages. The present study analyzes the data collected by employing an Acceptability Judgment Test on the
 interpretation of backward anaphora in complex sentences by twelve native
 Italian speakers, who had learned Turkish as adults in immigration setting,
 with twelve matched Italian monolinguals as a control group. It is assumed that
 Italian and Turkish do not differ with respect to the antecedent biases of null
 and overt subject pronouns in the contexts under investigation. The focus of this
 study is on the acceptability of an overt/null subject in intrasentential
 anaphora with three conditions: general sentences, quantifier sentences, and
 subjunctive sentences. Our results show that bilingual speakers reject,
 significantly more, the null subject in an embedded subjunctive sentence as
 referring to the subject in the matrix sentence than the monolinguals. These
 data seem to contradict previous studies (Sorace & Filiaci 2006), wherein
 it was found that monolinguals and bilinguals differ in the interpretation of
 an overt pronoun. A discussion on why a null pronoun is vulnerable in
 Italian-Turkish bilinguals is needed. Though this study reinvigorates the
 hypothesis that the structures at a syntax-discourse interface are vulnerable
 and that bilingual processing cost may contribute to cross-linguistic
 interference.