Previous studies of variation in interlanguage morphology have led to conflicting and contradictory claims concerning the relation between interlanguage forms and the contexts in which they occur. The present study suggests that such contradictions are due to the descriptively inadequate model of variation used in earlier studies. A multivariate model is proposed and applied to the analysis of variation in the acquisition of the English (s) pluralizaron rule by native speakers of Chinese. Three major groups of factors are shown to influence variation: stage of acquisition, linguistic environment, and communicative redundancy. Findings confirm the hypothesis that there exists a degree of systematicity in interlanguage, but suggest that the interlanguage system is best described in terms of probabilistic rules.

This paper examines language as a multimodal sensory enhancement system, integrating recent neuroanatomical and neurophysiological findings on the ontogenesis of neuronal structures with the generative concept of Universal Grammar (UG) for determination of fundamental differences between primary (PLA) and secondary (SLA) language acquisition. Substantial attention is given to general neurobiological principles such as experience expectant/dependent synaptogenesis, formational/organizational versus associational/reactive plasticity, characteristics of modular cortical organization, and general epigenetic qualities (e.g., intra- and interhemispheric competition, selective neuronal preservation, etc.) of the developing brain. Special emphasis is placed on neurobiological specializations relative to language (e.g., interhemispheric differences in dendritic arborization in Broca's area). The assumed innateness of UG is critically examined, with the neurobiological evidence indicating (a) the first language (L1) does not equal the second language (L2) neurobiologically, and (b) epigenetic factors contributing to PLA are often underestimated. The relevance of these conclusions for SLA is also briefly discussed.

This study examines the use of zero anaphora in the English production of three speaker groups: native speakers, second language learners, and speakers of a non-native institutionalized variety. General discourse function for zero anaphora is found to be similar across speaker groups, although in many cases, ungrammatical by prescriptive standards. In addition, there are important quantitative and structural differences between the native speakers and non-native speakers in how this device is used. The results suggest that the relationship between performance data and second language acquisition needs to be reexamined. In particular, it cannot be assumed that spontaneous production of a given form isa direct indicator of acquisition and conversely, that non-production is necessarily proof of non-acquisition.

This study examines the emergence of complex syntax in interlanguage (IL) development through a functionalist analysis of longitudinal, conversational IL data from two Vietnamese learners of English. The study focuses on discourse-pragmatic factors in the learners′ production and acquisition of complex syntax, specifically on their reliance on collaboration by interlocutors in the production of complex syntactic structures. Its findings offer a test of the utility of Givón's functionalist approach in SLA studies, and of claims by Hatch and others to the effect that “syntax develops out of conversation.” The interdependence of different linguistic levels in IL development is demonstrated, and the use of a variety of analytical units is shown to be critical in characterizing prepositional encoding in IL speech.