In American Sign Language (ASL), native signers use eye gaze to mark agreement (Thompson, Emmorey and Kluender, 2006). Such agreement is unique (it is articulated with the eyes) and complex (it occurs with only two out of three verb types, and marks verbal arguments according to a noun phrase accessibility hierarchy). In a language production experiment using head-mounted eye-tracking, we investigated the extent to which eye gaze agreement can be mastered by late second-language (L2) learners. The data showed that proficient late learners (with an average of 18.8 years signing experience) mastered a cross-linguistically prevalent pattern (NP-accessibility) within the eye gaze agreement system but ignored an idiosyncratic feature (marking agreement on only a subset of verbs). Proficient signers produced a grammar for eye gaze agreement that diverged from that of native signers but was nonetheless consistent with language universals. A second experiment examined the eye gaze patterns of novice signers with less than two years of ASL exposure and of English-speaking non-signers. The results provided further evidence that the pattern of acquisition found for proficient L2 learners is directly related to language learning, and does not stem from more general cognitive processes for eye gaze outside the realm of language.

Crosslinguistic transfer in bilingual language acquisition has been widely reported in various linguistic domains (e.g., Döpke, 1998; Nicoladis, 1999; Paradis, 2001). In this study we examined structural overlap (Döpke, 2000; Müller and Hulk, 2001) and dominance (Yip and Matthews, 2000) as explanatory factors for crosslinguistic transfer in Persian–English bilingual children's production of novel compound words. Nineteen Persian monolinguals, sixteen Persian–English bilinguals, and seventeen English monolinguals participated in a novel compound production task. Our results showed crosslinguistic influence of Persian on English and of English on Persian. Bilingual children produced more right-headed compounds in Persian, compared with Persian monolinguals, and in their English task, they produced more left-headed compounds than English monolinguals. Furthermore, Persian-dominant bilinguals tended more towards left-headed compounds in Persian than the English-dominant group. These findings point to both structural overlap and language dominance as factors underlying crosslinguistic transfer.

This paper investigates the development of referring expressions in the narratives of children learning English as a second language (L2). Spoken narratives in English were elicited from sixty Chinese-speaking participants at four ages – five, eight, ten, and young adults – using the wordless picture book Frog, where are you? (Mayer, 1969). Narrative analysis focused on the referring expressions that the L2 speakers used to introduce and maintain reference to story characters in the narratives and on the referential appropriateness of those expressions. We then compared the results of this study with other, similar studies on children learning English as a first language (L1) and found both universal and L2-specific patterns in the development of referring expressions in discourse. On the one hand, regardless of whether English is acquired as an L1 or L2, appropriate use of referring expressions in discourse is developed gradually and is influenced by both discourse function (introduction vs. maintenance) and character type (main vs. secondary). On the other hand, L2 children in our study differed from L1 children in previous studies in the development of referential appropriateness in character introduction, the use of pronominals for referent maintenance, and the timeline of the mastery of appropriate forms for referent introductions versus referent maintenance.

This study shows further evidence for triggered codeswitching. In natural speech from a Dutch–English bilingual, codeswitches occurred more often directly next to a cognate (or “trigger word”) than elsewhere. This evidence from typologically related, cognate languages extends previous evidence for triggering between typologically unrelated languages. With their large proportion of trigger words, the data provide insight into which words can trigger codeswitches; proper nouns, cognate content words with good and moderate form overlap, and cognate function words all induced codeswitching. Further, this study extends the evidence for triggered codeswitching from speech with relatively little codeswitching to speech with a high codeswitching density. In contrast with earlier work, not only words directly following a trigger word but also words directly preceding one were codeswitched more often than other words, suggesting that the scope of triggered codeswitching depends on the frequency of trigger words and of codeswitches in the speech.

This study investigates ultimate attainment at the syntax–discourse interface in adult second-language (L2) acquisition. In total, 91 L1 (first-language) English, L1 Dutch and L1 Russian advanced-to-near-native speakers of German and 63 native controls are tested on an acceptability judgement task and an on-line self-paced reading task. These centre on discourse-related word order optionality in German. Results indicate that convergence at the syntax–discourse interface is in principle possible in adult L2 acquisition, both in off-line knowledge and on-line processing, even for L1 English speakers, whose L1 does not correspond to L2 German in discourse-to-syntax mappings. At the same time, non-convergence of the L1 Dutch groups and differences in the L2 groups' performance between tasks suggest that asymmetries in L1–L2 discourse configurations and computational difficulties in mapping discourse onto syntax constrain L2 performance.

We report the results from two experiments investigating proficient Japanese-speaking learners' processing of reflexive object pronouns in English as a second language (L2). Experiment 1 used a timed grammaticality judgement task to assess learners' sensitivity to binding Principle A under processing pressure, and Experiment 2 investigated the time-course of reflexive anaphor resolution during L2 reading using the eye-movement monitoring technique. Taken together, our results show that despite having demonstrated native-like knowledge of reflexive binding in corresponding untimed tasks, the learners processed English reflexives differently from native speakers in that they took into consideration a matching discourse-prominent but binding-theoretically inappropriate antecedent when first encountering a reflexive. This suggests that unlike what has been reported in corresponding monolingual processing research (Sturt, 2003), initial antecedent search in L2 English is not constrained by binding Principle A.

This study examines word order differentiation in early trilingual development through an analysis of the combinations produced by a Tagalog–Spanish–English trilingual child with an MLU of less than 1.5. Same- and mixed-language combinations were tracked down from diary data and weekly recordings to assess (i) whether word order significantly varied cross-linguistically, and (ii) whether mixed utterances originated from vocabulary gaps rather than from an undifferentiated syntax. The results indicate that (a) argument/predicate sequences were differentially ordered depending on their language and following input-dependent preferences, (b) mixed utterances were generally caused by vocabulary gaps, and (c) they displayed the same order as those single-language combinations produced in the same language context. These findings suggest that evidence for early word order differentiation can be found before the appearance of inflectional morphology and even when three – rather than two – languages are being acquired, indicating that trilingual exposure does not slow down the process of differentiation.