1
The Development of Second and Foreign Language Learning Through Classroom Interaction
Joan Kelly Hall
Lorrie Stoops Verplaetse
In the field of second language acquisition (SLA), interaction has long been considered important in language learning. However, much of this research has been based on a traditional psycholinguistic perspective of language and learning. In this view, language is assumed to be a discrete set of linguistic systems external to the learner, whereas learning is viewed as the process of assimilating the structural components of these systems into preexisting mental structures. Although the specific goals of research carried out from this perspective may vary from finding the most effective way of facilitating the assimilation of new systemic knowledge to discovering and constructing the ideal linguistic system, it is generally agreed that language learning and language use are two distinct phenomena (Firth & Wagner, 1997, 1998; Gass, 1998; Hall, 1993, 1995, 1997).
By and large, the authors of the studies presented in this volume on classroom interaction and language learning began their explorations into second and foreign language learning from this more traditional perspective. However, in the investigations they report on here, they cross disciplinary borders and draw on theoretical insights and empirical evidence found in areas outside of what has generally been considered the main purview of the SLA field. Their findings join our interests in reconceptualizing second and foreign language learning using a broader, sociocultural perspective of language and learning with practical concerns for nurturing classroom communities of successful second and foreign language learners. In addition to helping build a foundation for the development of a more comprehensive model of second and foreign language learning, we believe that what we learn from these studies will help us in the creation of a set of principles for identifying and sustaining classroom interactional practices that foster additional language development. In what follows, we first provide a brief review of the findings on interaction and additional language learning that have emerged from the more traditional strands of research in SLA and discuss the assumptions about language and learning on which these studies are based. We then provide an overview of the theoretical assumptions on language and learning on which the studies presented here are based. Concluding the chapter is an overview of each of the 12 studies comprising the volume.
INTERACTION AND SLA: AN HISTORICAL PERSPECTIVE
Interaction as an Outgrowth of Foreigner Talk Literature
Research on interaction and second language learning grew out of early studies on foreigner talk (FT). Once it was determined that native speakers do indeed modify their input when talking with non-native speakers (NNSs; Ferguson, 1975) in a manner similar to the way caretakers alter their talk to babies (BT), the FT research expanded from simply describing the linguistic features of FT to exploring FTâs role in interaction. Given the achievement differences between first and second languages, Freed (1980) sought to determine the differences between BT and FT and found that although both were syntactically similar, they differed functionally. The primary function of FT was to convey information, whereas the function of BT was to elicit interaction.
Shortly thereafter, Longâs (1981) study of modifications in native speaker (NS) input to NNSs made a distinction between linguistic modifications and interactive modifications, claiming that the interactive modifications were facilitative and necessary for second language acquisition. In this study, he credited the NS input with the following interactive modifications âpresumably intended to facilitate comprehension and participation by the NNSâ (p. 263):
1. Topics are treated simply and briefly as compared to NS talk to NSs (NT).
2. Topics are dropped unexpectedly and shifted to accommodate miscommunications by the NNS.
3. Topics initiated in FT are signaled by additional stress, left dislocation, intrasentential pauses, question forms as topic initiators, and frames (i.e., âok,â âwellâ).
Long (1981) further credited FT with the following techniques âto sustain conversation and to lighten other aspects of the NNSâs interactional burdenâ (p. 264):
1. NSs engage in cooperative dialogue by supplying information to the NNSâs utterances to help the latter express an idea.
2. NSs answer their own questions and ask rhetorical questions.
3. NSs frequently use an interrogative style, thereby requiring answers and, hence, sustaining the conversation.
4. NSs use many and frequent clarification devices to avoid conversation problems and to repair miscommunications.
Long (1981) claimed that these modifications are necessary and facilitative to SLA, arguing that âcurrent knowledge suggests they [the interactive modifications] are found in all cases of the successful acquisition of a full version of SLâ (p. 275). His claim was supported by other studies (Arthur, Weiner, Culver, Lee, & Thomas, 1980; Hatch, 1983).
Although Longâs arguments are convincing, he made no distinction between different aspects of SLA that may be differentially affected by these modifications. Nor did he account for the fact that such FT modifications are also present in cases of unsuccessful SLA; that is, these modifications are found in nearly all cases of NS interactions with beginning NNSs. Moreover, although such modifications may well be necessary for comprehension and hence facilitative in early stages of SLA, and although they may be intended to facilitate participation by the NNS, their full impact on the NNSâs opportunities for interaction has not been satisfactorily analyzed. In fact, in a recent study, Verplaetse (1993) found NS modifications to have patently detrimental effects on NNS participation in conversations, thus challenging the general premise that NS modifications are primarily beneficial to NNSs. She suggested that at some developmental point in the second language learnerâs acquisition process, certain NS modifications become a hindrance to the NNSâs opportunities for production.
Importance of NNSâs Role in Interaction
A second group of SLA researchers followed this interest in interaction and second language learning, focusing on the importance of the NNSâs role in interaction. Research such as that by Scarcella and Higa (1982), Gass and Varonis (1985), Pica, Doughty, and Young (1986), and Pica (1988) echoed the importance of interaction in SLA, focusing in particular on the NNSâs role in the negotiation of meaning. These studies acknowledged that comprehensible input is necessary for SLA, but they claimed that it is the NNSâs work in the negotiation of meaning that increases and ensures that the input is maximally comprehensible. Thus the NNSâs role in the process of negotiating meaning became crucial to the acquisition process.
Importance of Output in Interaction
In the early to mid-1980s the research on interaction had been focused primarily on the learnerâs input. Beginning in the mid-1980s, however, investigatory concerns were enlarged to include the importance of the learnerâs output in interaction. According to Swain (1985), output provided three functions: noticing, hypothesis testing, and reflection. During the time that a learner is required to process input only, he or she may not need to attend to all features of language to comprehend the content. However, at the point when he or she must produce output, the learner may first notice that a gap of linguistic knowledge exists between what he or she wants to convey and his or her ability to convey it. When the learner attempts production, using what linguistic knowledge is currently available in his or her interlanguage, the learner tests out hypotheses about the organization of the language system. Finally, through the learnerâs output and the interlocutorâs response to that output, the learner can reflect on and ultimately modify his or her language use. In a more recent study, Swain (1995) provided empirical evidence to support her claim concerning the role of output in SLA.
Role of Interaction in the Classroom Event. Because so much of language learning occurs in the classroom, SLA researchers have also focused on the role of interaction in the classroom event. One primary concern has been with the role that teacher talk plays in SLA. Krashenâs (1980, 1989) theoretical work on comprehensible input has perhaps given most attention to this talk. According to Krashen, meaningful teacher talk is central to the process of language learning. Although in his own work Krashen provided little empirical evidence for this claim, it motivated much interest among others.1 For example, there has been quite a bit of work attempting to define the features of teacher talk considered crucial to its role in making the message comprehensible. These investigations have included the examination of such features as the degree of syntactic complexity, the rate of speech, and the length of utterance. Discourse features of teacher talk such as feedback, error correction, and use of questions have also been examined for the roles they play in making talk comprehensible.2 Because second language learning also occurs in the content classroom, additional studies have described and compared the nature of teacher talk to native speaking and NNS students in content classrooms (Early, 1985; Green, 1992; Schinke-Llano, 1983; Slaughter & Bennett, 1982;Verplaetse, 1998).
In a related direction, there has been interest in investigating input enhancement and form-focused instructional talk. Of specific concern has been the examination of the roles these special genres of teacher talk play in the raising of students' consciousness about the syntactic aspects of the target language and their subsequent learning of these forms. In a summary of this research, Schmidt (1994) concluded that such teacher talk helps to explicitly focus the learnersâ attention on the syntactic forms in question, and thus facilitates their development of the knowledge and use of these forms in the second or foreign language.
Finally, there has been research concerned with the description of task conditions from which negotiated speechâmorphosyntactically simplified speech that has been claimed to facilitate second language acquisitionâemerges (Crookes & Gass, 1993a, 1993b).3 The data category of interest here is talk produced between NS-NNS and NNS-NNS pairs in particular classroom activities, which is examined for indications of speech modifications such as clarification requests and comprehension checks. For the most part, the interest has been in examining the conditions influencing the emergence of such speech, such as one-way versus two-way information flow, and collaborative versus competitive goals, and less so in the impact such negotiation has had on second language learning in general.
To summarize, over the last 20 years or so, the role of interaction in second language learning has been examined from different angles, moving from early FT studies to the study of the role of the NNS in interaction and on to the study of teacher- and task-based talk in the foreign and second language classroom. Although these studies on interaction and SLA have taken somewhat different routes in investigating language development, most share some fundamental assumptions about the nature of language and learning that differ from those on which the studies in this volume are based. Perhaps most striking is the way that additional language learning has been operationalized: as distinct increases in the comprehension or use of correct grammar. There has been little consideration of other areas considered essential to competent language use such as discourse or sociolinguistic competence. Likewise, although there has been some acknowledgment of the socially constructed nature of language learning, few studies have been placed in the larger social context of real communicative interaction.
Voicing some concern with this narrow construction of language and learning, researchers in the field of SLA have begun to look into other fields for theoretical and methodological insights into lan...