Modeling the competence grammar of heritage speakers who exhibit low proficiency in their L1 represents a significant challenge for generative and experimental approaches to bilingual linguistic research. In this paper we revisit the core tenets of the incomplete acquisition hypothesis as developed in recent scholarship (in particular byMontrul (2002et seq.) andPolinsky (1997,2006)). Although we adopt many of these fundamental aspects of this research program, in this article we develop an alternative model that provides a more accurate depiction of the process that leads to what these scholars describe as the (later) effects of incomplete acquisition, thus improving the predictive power of this research program.

In this paper, I present an exploratory study on cross-linguistic interference among Quechua-Spanish bilingual children living in a language contact situation. The study focuses on convergence in the tense, aspectual and evidentiality systems of the two languages. While in Quechua past tense features are strongly linked to evidentiality in the matrix of features associated with the functional category Tense, in Spanish, past tense features are linked to aspectual features. The study presents evidence that supports the Functional Convergence Hypothesis according to which syntactic convergence among bilingual speakers is favored when the matrix of features associated with a functional category is partially divergent, as is the case for Tense in Spanish and Quechua. The Spanish results indicate that among bilinguals past tense is associated with evidentiality features and contrast sharply with the results of the monolingual comparison group. Bilingual Quechua results exhibit an incipient emergence of discourse-oriented background and foreground distinctions, similar to those found in Spanish in association with aspectual morphology.

This study examines the acquisition of the featural constraints on clitic and null distribution in Spanish among simultaneous and sequential Chinese-Spanish bilinguals from Peru. A truth value judgment task targeted the referential meaning of null objects in a negation context. Objects were elicited via two clitic elicitation tasks that targeted anaphoric contexts and left-dislocated topics. An acceptability task tested sensitivity to left-dislocated object drop. Although simultaneous bilinguals were mostly undistinguishable from monolinguals, the late learners differed from both of these groups across tasks. Age of arrival led to different outcomes, with late learners showing more deficits than the child learners. Late learners avoided using clitics and relied on lexical and null objects. Residual transfer effects were observed among the child learners in the form of insensitivity to the features that serve as the basis for null argument identification and clitic deficits in production. It is also argued that transfer persists despite early and intense exposure to the second language in a natural environment because of the existence of an unmarked argument identification option in the first language.

This book addresses how cross-linguistic interference is represented in the bilingual mind. Examining novel oral production data from older bilingual children representing two Quechua varieties, this research concludes that interference in the feature specification of functional categories leads to language change in a language contact situation, and links convergence, a common set of feature values for the same functional category in both languages to the activation of features related to the informational structure of the sentence. These mechanisms are illustrated in detail by the presence of overt determiners, canonical SVO word order and the absence of accusative marking in bilingual Quechua and by neutralization of case and gender distinctions in direct object pronouns as well as in the emergence of null pronouns with definite antecedents in bilingual Spanish.

In this article, we propose that elements of heritage language grammars—both in the form of axiomatic features and larger combined representational units—are not easily lost over the course of the lifespan. This view contrasts with alternative explanations for the steady-state representation of these grammars that suggest truncated acquisition or erosion are the primary culprits of perceived language loss. In production and comprehension processes for heritage bilinguals, particular elements are more difficult to access than others, leading to differential ways to access representations and feature values. To illustrate and support this hypothesis, we build on previous work by examining the interpretation and use of obligatory mood selection in Spanish desiderative constructions in three groups of heritage speakers with different levels of language proficiency.