Linguistic Approaches to Bilingualism - Volume 2, Issue 1, 2012

Recent psycholinguistic research demonstrates that using a second language has consequences for the first language (e.g. Dussias, 2003; Van Hell & Dijkstra, 2002) and for domain-general cognitive processes (Bialystok, 2005). This work suggests that the language system is permeable, with cross-language exchange at every level of processing (Malt & Sloman, 2003). Critically, even proficient bilinguals appear unable to switch off the language not in use when they hear, read, or speak one language alone (e.g. Dijkstra, 2005; Kroll, Bobb, & Wodniecka, 2006; Marian & Spivey, 2003), creating cross-language competition. In this article, we describe research that considers how cross-language activation is modulated during spoken production and during the earliest stages of second language learning. We hypothesize that the open nature of the bilingual’s language system may create optimal conditions for new language learning and also for enhanced cognitive control that enables effective selection of the language to be spoken.

This paper proposes a model of morphological variation and development grounded in feature-geometric theory. It tests two hypotheses that follow from this model on a corpus of data from speakers of Spanish as a second language (L2). First, variation is systematic; default, underspecified feature values are adopted when errors occur. This hypothesis is supported for person, number, and finiteness, as 3rd, singular, and nonfinite defaults surface in place of 1st, plural, and finite verbs. Second, developmental trends are observed as nodes are added to the geometry; the unmarked/less specified feature value is successfully produced prior to the marked/more specified one. This hypothesis is partially supported, as accuracy in 3rd person emerges prior to 1st. However, no developmental pattern is found for number. Errors in finiteness are limited to lower-proficiency speakers, whereas intermediate speakers favor 3rd person, finite defaults. Together, these results suggest systematic variation and gradual development in the morphology.

In this paper we investigate whether learners of L2 English show knowledge of the Definiteness Effect (Milsark, 1977), which restricts definite expressions from appearing in the existential there-insertion construction. There are cross-linguistic differences in how restrictions on definiteness play out. In English, definite expressions may not occur in either affirmative or negative existentials (e.g. There is a/*the mouse in my soup; There isn’t a/*the mouse in my soup). In Turkish and Russian, affirmative existentials observe a restriction similar to English, whereas negative existentials do not. We report on a series of experiments conducted with learners of English whose L1s are Turkish and Russian, of intermediate and advanced proficiency. Native speakers also took the test in English, Turkish, and Russian. The task involved acceptability judgments. Subjects were presented with short contexts, each followed by a sentence to be judged as natural/unnatural. Test items included affirmative and negative existentials, as well as items testing apparent exceptions to definiteness restrictions. Results show that both intermediate and advanced L2ers respond like English native speakers, crucially rejecting definites in negative existentials. A comparison with the groups taking the test in Russian and Turkish confirms that judgments in the L2 are quite different from the L1, suggesting that transfer cannot provide the explanation for learner success.

The incorporation of new representations into the mental lexicon has raised numerous questions about the organisational principles that govern the process. A number of studies have argued that similarity between the new L3 items and existing representations in the L1 and L2 is the main incorporating force (Hall & Ecke, 2003; Herwig, 2001). Experimental evidence obtained through a primed picture-naming task with L1 Polish-L2 English learners of L3 Russian supports Hall and Ecke’s Parasitic Model of L3 vocabulary acquisition, displaying a significant main effect for both priming and proficiency. These results complement current models of vocabulary acquisition and lexical access in multilingual speakers.