J. Dean Mellow

Using Time-Series Research Designs to Investigate the Effects of Instruction on SLA.

This paper argues that the study of second language acquisition theory and pedagogy can be enhanced through the use of time-series research designs. As quasi-experiments, time-series designs have features that improve internal validity. In addition, because these designs only require a small number of subjects, they are very practical, encouraging a greater number of empirical investigations of the many claims within the field and permitting the use of authentic measures that have high construct validity. The longitudinal nature of the designs also enhances construct validity, potentially yielding new insights into the effects of instruction on SLA. The designs utilized in two time-series studies (Kennedy, E., 1988, The Oral Interaction of Native Speakers and Non-Native Speakers in a Multicultural Preschool: A Comparison between Freeplay and Contrived NS/NSS Dyads , unpublished masters thesis, University of British Columbia, Vancouver; Mellow, J. D., 1996, April, A Longitudinal Study of the Effects of Instruction on the Development of Article Use by Adult Japanese ESL Learners , unpublished manuscript, University of British Columbia, Vancouver) are reviewed in order to illustrate the design features, the questions that may be investigated, and the issues that are raised in interpreting data.

Connectionism, HPSG signs and SLA representations: specifying principles of mapping between form and function

A current limitation of the connectionist approach to second language acquisition (SLA) research is that it does not, to my knowledge, include complex linguistic representations. This article proposes a partial solution to this limitation by motivating and illustrating specific analyses that utilize the sign-based representations developed within Head-driven Phrase Structure Grammar (HPSG). To motivate the proposed representations, the article applies them to an analysis of four types of mappings between form and function: one-to-one, primed redundancy, nonprimed redundancy and polyfunctional. The paper summarizes representative SLA data that indicate how these mappings may appear in second language (L2) production. Key properties of HPSG analyses are discussed, indicating how they are consistent with connectionist assumptions. Sign-based representations of the four types of mappings are then provided, including several modifications to HPSG formalisms. The article concludes with a discussion of future directions.

On the Primacy of Theory in Applied Studies: A Critique of Pienemann and Johnston (1987).

This article critiques Pienemann and Johnston (1987), an influential model of the acquisition of ESL morphology. While Pienemann and Johnston claimed that their model is confirmed and supported by the generative approach to morphology developed in Selkirk (1982), this article indicates that Selkirks lexicalist theory actually opposes their proposals. In addition, this article demonstrates that their proposals are incompatible with syntactic analyses of word formation. The widespread influence of the Pienemann and Johnston model suggests that L2 researchers must ensure that models of SLA are solidly grounded in foundational theories.