In just several decades of its existence, the field of second language research has not only changed the way we think about the teaching and learning of a second language, but has also reshaped our knowledge of human languages in general. The broadly-defined theme for the annual meeting of the Second Language Research Forum held in Madison, Wisconsin, USA in September, 2000 - Second Language Research: Past, Present, and Future - was selected to encourage the discussion of a wide range of issues that comprise this dynamic and rapidly growing field. Under this moniker, we hoped to encourage the participants in the forum to reexamine classic problems associated with second language acquisition (SLA), to demonstrate the application of current advances in the field from a variety of perspectives, and to point to advances in this vast area of inquiry known as "second language research." Keeping in mind what constitutes appropriate data and methods for analysis with which SLA theory is developed (cf. the Modern Language Journal discussion on this issue by Firth & Wagner (1997) and Gass (1998)), we felt that this particular conference should not focus solely on the application of traditional, time-tested theories and methods. We hoped that the forum should also address newer models and approaches that would help scholars reach a deeper understanding of traditional issues, bring together the various domains of SLA, and account for both the emerging and hitherto overlooked communities of language learners.
Obviously, not all of the papers given at the conference could be included in one volume; however, we believe that the 16 peer-reviewed papers included in this year's proceedings will provide the reader with an accurate representation of the major themes of SLA research presented at the conference. The papers included in this volume fall into four broad categories that represent the past, present, and future of SLA research. These are papers addressing phonological and syntactic approaches toward SLA research (I); papers investigating SLA from the perspective of cognitive psychology (II); papers that examine the practices of teachers and students in the second and foreign language classrooms (III); and finally, papers concerned with research methodology and terminology (IV).
I. Phonological and syntactic investigations of SLA
The title of this section refers to the underlying nature of language behavior as it relates to the acquisition of a second (or in one case, a third) language with particular interest in describing and explaining the state of interlanguage at various stages of development or in testing a particular learning model related to acquisition of this underlying knowledge. Although the majority of work in this area is concerned with the acquisition of grammar (including syntax and morphology), we are encouraged by the growing body of literature on the acquisition of second language phonology as evidenced by the first two papers in this volume.
Laura Catharine Smith's paper, L2 acquisition of English liquids: Evidence for production independent from perception, combines models for language processing (Stevens & Halle 1967), language acquisition (Werker & Pegg 1992) and a model known as "kinesthetic feedback loop" to propose that non-auditory feedback plays a larger role in the acquisition of L2 phonology than previously thought. Smith points out that much SLA phonological theory accepts that learners have perceptual mastery of L2 phonemes and phonemic contrasts before they are able to produce these phonemes (Sheldon & Strange 1982; Werker & Pegg 1992). Smith questions this assumption and proposes that, at least for some learners, this process is reversed. Based on data from Japanese and Korean learners acquiring the /l/-/r/ contrast in English, Smith notes that, while learners may be able to produce L2 phonological contrasts at a high level, they may not be able to perceive those contrasts. Smith suggests that production can proceed independently of perception through kinesthetic feedback during articulation. She proposes a study to test the theoretical claims made in the paper and considers possible applications of these findings to second and foreign language pedagogy.
Whereas Smith's paper questions widely held assumptions about the acquisition of second language phonology, Kimberly A.B. Swanson's paper Is L2 learning the same as L1 learning? Learning L2 phonology in Optimality Theory looks to test a newer model of language description and explanation using data from SLA. Swanson's paper tests the learning algorithm within Optimality Theory (OT) for SLA by comparing the phonology of Polish and English. Looking at the phonological processes that differ in these languages, Swanson shows divergent paths within the OT learning algorithm that learners of Polish and English take on their way toward full acquisition. Her results show that in certain instances, L2 learning parallels L1 learning inasmuch as markedness constraints hold for acquisition of L2 as they do for L1. However, the author notes that using a model for acquisition from OT, in some instances, contradicts markedness constraints and Eckman's (1977) Markedness Differential Hypothesis (i.e., learning a more marked process appears to be more difficult than learning a less marked process). Swanson suggests that with modifications, such as re-ordering of higher order faithfulness constraints, the OT learning algorithm can be applied to SLA. Swanson sees the need for more empirical investigation to resolve the contradictions and concludes by discussing the implications of her findings for foreign and second language pedagogy.
The next set of papers address two classic "problems" associated with the acquisition of syntactic knowledge. The first problem, often referred to as the "logical problem of language acquisition" (Bley-Vroman 1990; White 1989) asks how native speakers attain the implicit structural knowledge under the assumption that the linguistic data they are exposed to underdetermines the knowledge they have of their native language. This same situation holds in many cases for L2 grammars as evidenced by Shunji Inagaki, who reports on the acquisition of L2 argument structure in light of the subset principle in a paper titled Japanese learners' acquisition of English motion verbs with locational/ directional PPs. Inagaki attempts to account for the second language acquisition of argument structure as it applies to the learning mechanism of the subset principle. Inagaki argues that argument structure can help to disambiguate two possible scenarios where the L2 is a superset of the L1. In situations where there is a partial fit between the L1 and the L2, learners may work under the misassumption that L1 and L2 are identical and transfer the L1 scenario, or learners may be guided by positive evidence in their construction of an L2 grammar. Inagaki tested the ability of a group of intermediate-level Japanese learners of English to recognize the ambiguity of a sentences such as John swam under a bridge where the object NP can be interpreted either as a goal (the directional reading), or as a location (the locational reading). Verbs that denote manner of motion do not allow the goal (directional) reading in Japanese, and thereby bring about a situation wherein the L2 is a superset of the L1. The results of a picture-matching task suggest that positive evidence plays an important part in argument structure acquisition as long as it is both frequent and unambiguous.
The second problem associated with the acquisition of syntactic knowledge, the "developmental problem" (Gregg 1996; White 1989) is concerned with how a learner forms hypotheses and develops this same system of implicit knowledge. In many cases, this approach attempts to capture acquisitional behavior that is similar in normal, first language acquisition. More recently, the developmental problem has focused on the availability and nature of functional categories in adult L2 acquisition. The papers by Yan-Kit Ingrid Leung and Tania Ionin and Ken Wexler reflect the ongoing debate characterizing the access to, and existence of, functional categories from the developmental point of view.
The initial state of L3A: Full transfer and failed features? by Yan-kit Ingrid Leung examines the extent to which Cantonese-English bilingual learners of French have acquired the [±DEFINITE] feature in the nominal category DP. The main goal of this paper is to characterize the impact that both L1 and L2 have on the L3 initial state. Leung argues that a characterization of the L3 initial state can help to provide support for two competing theories - full transfer or failed functional features - as they relate to the L3 initial state. Using both elicited oral and written production tasks, Leung finds that certain functional features in DP are present in the French initial state of these learners, features that are not a part of their L2 but instead a part of the L3. Although this points toward an L2 influence, it is potentially problematic to determine if the transfer is from L1 or L2. Leung proposes a revised failed functional features approach which predicts that categorial features ([±DEFINITE], for example) can be acquired while non-categorial features ( [±N] or [±V] ) fail.
L1-Russian children learning English: Tense and overgeneration of 'be' by Tania Ionin and Ken Wexler addresses the issue of whether or not frequent lack of tense and agreement morphology is due to an impaired access to those functional categories associated with tense and agreement features. Using both spontaneous production data and a grammaticality judgment task from L1 Russian-speaking children, Ionin and Wexler argue that missing surface inflections are not related to functional category impairment but instead are the result of problems related to incorrect morphological mapping of a somewhat limited English affixal system. They find that the overuse of be as a tense and agreement marker occurs because suppletive agreement in English is often mastered before affixal agreement, a finding that is taken as further support for the existence of those categories being associated with tense and agreement. Ionin and Wexler suggest that one should be careful in making the logical jump from determining a lack of access to those functional categories associated with tense and agreement in English when agreement affixes (such as -ed and -s) are not present.
II. Cognitive accounts of SLA
The next set of papers in this volume reflect cognitive approaches to SLA. Unlike the formal approaches of the previous section, cognitive approaches to SLA view language behavior and development as an indication of a more general property of learning that is not, in principle, an autonomous, language-specific type of knowledge. The somewhat fabricated dichotomy between formal and cognitive accounts is worthy of careful consideration as the two approaches are not necessarily mutually exclusive. Ellen Bialystok's paper Against isolationism: Cognitive perspectives on second language research raises the problems inherent in conflating two often competing perspectives: the cognitive processing approach adopted by psycholinguistics, and the abstract notion of competence associated with the more formal linguistic approach. These competing perspectives have often resulted in a confusing situation where many researchers may conclude that there is no ground where the two may ever meet. To combat this misguided notion, Bialystok proposes that the understanding of traditionally "linguistic" issues such as the nature of language proficiency, the presence or absence of a critical period in SLA, and the effect of bilingualism on language development is in fact in line with the cognitive perspective that she proposes in the paper. Bialystok suggests that a cognitive perspective should not be viewed as a replacement of more formal linguistic approaches to SLA but instead as an enrichment of them.
In addition to Bialystok's paper, three other papers in this section address more conventional cognitive approaches such as the relationship between feature complexity and lexico-semantic acquisition (Agnes Bolonyai and Lida Dutkova-Cope), as well as deeper characterizations of the nature of working memory (WM) in SLA as evidenced by the papers by Mira Goral, Loraine K. Obler, Elaine C. Klein, and Martin R. Gitterman and by Yuki Yoshimura. Bolonyai and Dutkova-Cope implicitly invoke the interdisciplinary goal suggested by Bialystok in their paper L1 attrition of verbal morphology in bilingual children and adults. The authors seek insights into the properties of the mental lexicon by studying the attrition of L1 morphology in a diverse set of learners comprised by Hungarian-English bilingual children and American Czech adults. By looking at L1 attrition, the authors seek to discover whether all grammatical features encoded within verbal inflections (person, number, gender, and definiteness) are equally vulnerable to attrition. Using a lexically-based approach to bilingual production (Levelt, Roelofs, & Meyer 1999; Myers-Scotton & Jake 2000), the authors look at data from naturally occurring conversations as well as more structured interviews and show that these bilinguals used a composite lexico-grammatical system in which the surface morphemes are from the L1 whereas the abstract specifications for agreement (subject-verb and verb-object) are based on the dominant L2 (English). In particular, Bolonyai and Dutkova-Cope report that the least-complex features such as number were more likely to be retained in the L1, while the more complex features that account for definiteness and gender were more likely to be lost. The authors show that attrition in L1 verbal morphology is affected by the language internal (L1) complexity of the structure, by the processing of a particular feature of agreement, as well as by the effects of contact with conflicting abstract lexico-grammatical features in the L2.
The final two studies in this section follow more traditional cognitive topics in SLA as they are concerned with the concept of working memory and lexical and phonemic acquisition in a second language. Translation-equivalent priming and second-language proficiency by Goral, Obler, Klein, and Gitterman investigates the correlation between speakers' level of L2 proficiency and the lexical connections between the first language (L1) and a second language (L2) of speakers of more than one language. A new research platform explores the inter-related strengths of L1 and L2 as evidenced in lexical connections. The authors argue that as learners become more proficient in a L2, they seem to build a more autonomous system independent from their L1 system.
Yoshimura (The role of working memory in language aptitude) looks at the learners' performance on a language aptitude test in its relation to the working memory. Using WM models from Baddeley (1986) and Just and Carpenter (1992), Yoshimura demonstrates that learners with larger WM capacity outperformed those with smaller capacity in both memory and language analytic ability sections in the aptitude test, but not in the section of phonemic coding ability. This finding confirms that WM actually functions to process information in both memory and language analytic ability, although a non-significant correlation between WM and phonemic coding ability implies that WM for phonological information would be somewhat distinct.
III. SLA in the classroom
The five studies in the third section of this volume bring together and challenge long-established perspectives of researchers, teachers, and learners on aspects of classroom discourse practices that are believed to be linked to some concept of "success" in SLA. Researchers commonly investigate psycho-sociological factors at work in the classroom (i.e., motivation) as they relate to success in classroom acquisition of ESL. Two studies in this volume, Bonny Norton's Non-participation, imagined communities, and the language classroom and Paul Russell and Jean Yoo's paper, Learner investment in second language writing, reverse the trend of using ethnographic methods of investigation to examine how socio-psychological factors outside the classroom may determine involvement and success in the ESL classroom.
Norton's paper, Non-participation, imagined communities and the language classroom, draws on the work of Lave and Wenger (1991) and Wenger (1998) by considering how resistance and non-participation in an ESL classroom relates to lack of student success. In a study of two immigrant language learners who, on two separate occasions, withdrew entirely from participation in their ESL classroom, Norton argues that the learners' non-participation is best explained with reference to their investment in their "imagined communities." An imagined community is one with which a given language learner would most like to be affiliated to a community that may well be a reconstruction of past relationships. The paper demonstrates that a language learner's non-participation in a second-language class may result from a disjuncture between the learner's imagined community and the teacher's curriculum goals. The discussion of learners' changing expectations of ESL courses is linked to their shifting identities and their unique desires for the future. The paper concludes that if educators do not acknowledge the imagined communities of the learners in ESL classrooms, they may exacerbate some students' non-participation. At the same time, Norton cautions that it may be problematic to celebrate this imagined community unconditionally. If learners' investments in an imagined community compromise their engagement with the wider target language community, in general, and in particular, in second language classrooms, the value of these investments for language learners raises important questions for teachers, learners, and researchers alike.
Russell and Yoo acknowledge Norton's influence in the paper Learner investment in second language writing as they explore the often-studied issue of motivation in SLA. Like Norton, Russel and Yoo use an ethnographic approach to present a detailed picture of how ESL learners from children to adults reposition themselves as learners based on their investment in their ESL classrooms and in ESL activities. However, the authors believe that learner investment is a more complex issue than simply resistance. The portraits presented in their paper show the complex interweaving of students' many cultural and personal interests and conflicts. The authors are able to show through interviews with the students how these "conflicting desires" are quite transparently reflected in their work as ESL writers.
The final three studies in this section investigate activity inside the classroom and are indicative of the growing momentum of second language acquisition research, as well as the increasing concerns over the transferability of acquisition models used in second language research to the different instructional settings in which adult foreign language acquisition takes place. In particular, all three studies are concerned with corrective feedback practices as perceived by teachers and students in a foreign language classroom. The studies are both similar and complementary in their incorporation of second language research findings, and in the types of theoretical and methodological questions they raise with regard to future research on corrective feedback in light of pre-established categories and existing classroom practices. All three studies were carried out in university settings and include beginning and intermediate foreign language students of Japanese, Spanish, and French. In these foreign language classrooms, all three test Lyster and Ranta's (1997) categorical model of corrective feedback in the context of an L2 immersion classroom. This model identifies five basic types of classroom feedback: (a) recasts, (b) elicitation, (c) clarification requests, (d) metalinguistic feedback, (e) explicit correction, and (f) repetition.
In a replication study, Recasts and learner uptake in Japanese classroom discourse, Mariko Moroishi focuses on the recast in a Japanese as a foreign language classroom. Her findings are consistent with those of Lyster and Ranta in that recasts were the teachers' corrective feedback type of choice, while seemingly being the least effective in leading to uptake and repair. Moroishi suggests ways in which recasts may be perceived by Japanese L2 learners as an ambiguous form of feedback, and concludes that foreign language acquisition may be enhanced if other, less ambiguous, types of feedback are applied more often by teachers of Japanese.
While testing Lyster and Ranta's model in a second-semester Spanish classroom, Ana Oscoz and Judith E. Liskin-Gasparro (Corrective feedback, learner uptake, and teacher beliefs: A pilot study) identify student and teacher content-centered and metalinguistic discourse practices unaccounted for by that model. Oscoz and Liskin-Gasparro conducted a face-to-face interview with the teacher in order to elicit her beliefs about the relative usefulness of corrective feedback practices for various types of classroom activities. They discovered several ways in which that teacher's beliefs contradicted her actual corrective behavior. For instance, the teacher stated in the interview that the best practice was to provide less corrective feedback in communication-centered activities and more error correction in grammar-centered activities. She also expressed preference for the recast as the least intimidating type of corrective feedback. In actual practice, she exhibited identical corrective behavior in both activity types, with recasts being her corrective technique of choice independent of an activity type. Oscoz and Liskin-Gasparro suggest that in her classroom talk, the teacher may have been fostering a discourse practice that accommodates both the instructional goal of communicative competence and a focus on formal accuracy, and that such a practice may be more typical of the foreign language instructional context than of the immersion context.
Tony E. Macheak's study, Corrective feedback in second-language acquisition: Towards an integrated model, represents a shift of focus from teachers' patterns of corrective behaviors to learners' perception of such behaviors. Macheak taps into the learners' interpretations of teachers' corrective behavior in a French as a foreign language classroom using a retrospective protocol. Recordings of classroom interactions were played to students, who were then asked to categorize teacher feedback moves in terms of positive reinforcement vs. negative evidence, and in terms of directness in eliciting uptake. In consultation with teachers involved in these interactions, Macheak conducted his own categorization based on the understanding of the same categories in SLA literature. It was discovered that not only did learners perceive the feedback moves differently from the researcher, but their interpretations often varied from one another. Macheak registers a variety of lexical, phonological, pragmatic, and auditory factors related to instructional content, and discusses teacher and student discourse behaviors that may have caused the gap between the intended purpose of feedback moves and their interpretation by the students.
From three different perspectives, these studies point toward the urgency of further examination of corrective feedback behaviors, and toward the necessity of creating new models that would integrate teachers' and learners' perception of classroom discourse practices. The authors suggest a variety of future directions that, in addition to being beneficial for foreign language acquisition research, may also enrich second language research agendas.
IV. Methodology and terminology
Adopting approaches that not only guide, form, and validate but also advance SLA studies is an important challenge for second language research. Because of the interdisciplinary nature of the field, SLA researchers are faced with the difficult task of determining which tools are best suited to an adequate inquiry in particular issues of language acquisition. Both quantitative and qualitative research models have revealed insightful understandings of the learning processes, the learners' language systems, and the various factors that affect learning and performance. SLA has also been using methodology and terminology from disciplines such as linguistics, psycholinguistics, anthropology, psychology, cognitive studies, and sociology. Thus, the papers concluding this volume question established methods and terminology and propose innovative methodological approaches.
In the paper titled Triangulation in qualitative SLA research on interlanguage pragmatics, Margaret A. DuFon demonstrates on both the theoretical and the empirical levels the advantages of, and even the necessity for, combining multiple research methodologies to reach a sound understanding of the data. Indeed, DuFon shows that a triangulation of method, theory, data, or investigator can increase a study's breadth, validity, and reliability, even though results may show lack of convergence. In her approach to the acquisition of linguistic markers of politeness by learners of Indonesian, DuFon uses both triangulation of data and triangulation of informants. By using three methods of data collection - audio recording mostly by the learner, video recording by the researcher, and learners' dialog journals - DuFon was able to look at a wide variety of instances of language use and learners' self-reflection. In interviews with three groups of informants - the host families, the teaching staff, and the learners themselves - DuFon was able to uncover the insights of participants in the activities she observed. Applying this rich triangulation to the study of the learners' use of terms of address, greetings, and experience exchanges, DuFon found that particular methods (outlined in the paper) are more appropriate for particular data.
While DuFon's paper proposes a broadening of methodological perspectives in SLA research, Kimberly L. Geeslin in her paper entitled Linguistic simplification: Past, present and future links to second language acquisition points to a need for more focus and definition in the linguistic terminology used in SLA research. Geeslin draws attention to the lack of an adequate working definition of the concept of simplification in the research on Spanish SLA and notes the problematic nature of using one catch-all term "simplification" to define phenomena as different as copula deletion as a developmental stage, copula deletion as overgeneralization, and copula deletion as simplification of the rule inventory of the verb. In order to present a more accurate picture of the notion of simplification, Geeslin revisits previous research as well as her own data in an attempt to clarify the different aspects of learner simplification of the Spanish copula. In order to elucidate the nature of simplification, the author shows the variety of types of simplification of copula choice in Spanish SLA and then evaluates definitions of simplification used not only in SLA research but also in language variation studies and creole studies drawing attention to the subtle differences between phenomena such as "simplification," "reduction," and "overgeneralization" in language learning. Geeslin's re-evaluation of the past research in SLA and related fields of linguistics shows that discourse-related strategies used by native speakers also account for some instances of copula deletion in learners of Spanish.
Looking back at the emergence and development of SLA studies in the twentieth century, we think of diversity - not only of research perspectives and approaches, but also of the human languages and communities the field has covered since its inception. We are particularly proud that the "millennial" Second Language Research Forum was held at the University of Wisconsin-Madison. Through the help of both the departments who provided financial support and the numerous individuals who helped to bring this conference to fruition, SLRF 2000 was able to provide an opportunity for researchers to present papers reflecting the myriad of ways of looking at second language phenomena. We hope that the conference proceedings reflect this same wide view of the field, complete with both established and emerging ways of conducting second language research.
As editors, we faced the pleasurable task of editing such a diverse collection of papers in what constitutes one of our first forays into the world of scholarly publishing. We owe a particular debt of gratitude to Dr. Sally Sieloff Magnan and to Dr. Richard Young, who have generously shared their time and offered indispensable guidance while at the same time expecting nothing less than high professional standards in every aspect of scholarly activity. We would also like to thank Hassan Belhiah for help in bringing this volume to completion, and the anonymous reviewers from across the United States and Canada without whose involvement this publication would not have been possible. We hope that this volume contributes to the field of second language research in the true spirit of the SLRF tradition by providing an accessible and widely recognized forum where both young and seasoned scholars can share their visions of and approaches to second language research.
Madison, Wisconsin, Summer 2001
ReferencesBaddeley, Alan D. (1986). Working memory. New York: Oxford University Press.