The Role of Input and Interaction in Second Language Acquisition Introduction to the Special Issue

SUSAN M. GASS Department o f English Michigan State University A 714 Wells Hall East Lansing, MI 48824-1 027 Email: gassqilot.msu.edu

ALISON MACKEY Department o f Linguistics Gemgztom University Washington, DC 20057-1051 Email: mackeya@gusun.gemgetom.edu

TERESA PICA Grduate School o f Education University o f Pennsylvania 3700 Walnut Street Philadeyhia, PA 19104 Email: teresap@nwf.gse.upenn.edu

This Issue Is Dedicated to the Memory of Charlene J. Sato

I
I

THE THEORETICAL BASIS FOR THE ROLE OF CONVERSATION IN SLA The study of conversational interaction involving second language (L2) learners and their interlocutors has been central to second language acquisition (SLA) research since the early 1980s. A good deal of this work has focused on the ways in which interaction can be influenced by factors of gender, ethnicity, and the role in the social relationship of learners and their interlocutors, and by the nature of the topics, tasks, and activities in which they engage. Considerable attention has also been directed towards the role of interaction with respect to the conditions considered theoretically important for SLA, such as the learners comprehension of input, access to feedback, and production of modified output (cf. Gass, 1997; Long, 1996; Pica, 1994; Wesche, 1994, for recent reviews). The role of interaction has long been central to the study of language acquisition theory. For example, Vygotsky and his colleagues working in Russia in the 1920s conceptualized many constructs that continue to have relevance in inter-

The M o m LanguageJournal, 82, iii, (1998) 0026-7902/98/299-305 $1.50/0 01998 The Modern LanguageJournal

actionist research to date (see Swain & Lapkin, this issue). With the advent of interactionist perspectivesin SLA, emphasis was placed on the empirical study of language learner discourse and social interaction, as SLA researchers gathered data on learners and interlocutors as evidence of language development. Extensive empirical studies of input and interaction explored the ways in which learners manipulated their interlanguage (IL) resources when asked to make their messages more comprehensible. These manipulations, in turn, led learners to restructure their IL toward greater accuracy and complexity. The present special issue focuses on the input and interaction paradigm in SLA, including five empirical studies that are typical of the painstaking data collection that has characterized so much of the interactionist SLA research. Despite the large number of studies dealing with input and interaction in SLA, and, indeed, the wealth of information that such studies have provided, the precise role of interaction in actual development and internalization of L2 knowledge has continued to challenge researchers. As early as 1986, for example, Sato, drawing from her longitudinal research initiated in 1981, proposed that the relationship between learners participation in conversational interaction and their L2 development was one of selectivity and indirectness. This theme has now become central

300 to work on conversational interaction and SLA, and is reflected in the articles included in the present special issue. As such, the articles bring us closer to understanding the ways in which (a) certain dimensions of conversational interaction are more influential than others to the development of a L2 and (b) certain domains of IL develop ment are more responsive than others to the social discourse in which learners and their interlocutors engage. Historically, these articles also represent the latest contributions to a line of research that owes its origins to the interactionist perspective on L2 development put forth by Hatch in a series of seminal articles dating back to the 1970s. Until the 1970s, conversational interaction was believed to serve a reinforcing function in SLA, whereby learners could take grammatical features, structures, and rules that had been presented in classroom lessons and other assignments and apply them to spoken discourse itself-often carefully organized and orchestrated by their teachers and textbooks-to showcase particular grammatical items. This common orthodoxy changed in 1975when Wagner-Cough and Hatch (see also Hatch, 1978; Hatch & Wagner-Cough, 1976) illustrated how learners participation in conversational interaction provided them with opportunities to hear and produce the L2 in ways that went beyond its role as simply a forum for practice. Their analysis of conversations between learners and interlocutors suggested that L2 syntax might develop out of conversation, rather than simply feed into it. From this basic insight stemmed a number of studies that described L2 interaction and attempted to relate it to the linguistic needs of L2 learners, particularly the need for comprehensible input, which, at the time, was considered to be the driving force behind the acquisition process (see Krashen, 1985; Long, 1983a, 1983b). With these studies in the 1980s,Long launched a series of studies that shed light on this relationship. One of Longs foremost contributions was to distinguish between the talk directed toward L2 learners by native speakers (NSs) and the interactions in which they engaged. He showed that such interactions differed from native speakernonnative speaker (NS-NNS) conversation in terms of the conversational structure. Such differences include clarification requests, confirmation of message meaning, and comprehension checks. Longs research revealed that these conversational modifications were not unique to nonnative discourse, and, indeed, were found in NS conversations as well. However, as they were significantlymore abundant in conversations that

The Modern LanguageJournal 82 (1998) involved L2 learners and even more so in learnerto-learner conversations (cf.Varonis & Gas, 1985), Long suggested that they could serve a role in providing the comprehensible input needed for successful L2 learning and proposed a two-step argument concerning the relationship between conversational interaction and acquisition. Longs (1980) proposal went as follows: First, we need to show that comprehension promotes acquisition, and second, we need to show that conversational modifications lead to better comprehension. From this, we would be able to deduce that conversational modifications promote acquisition. Long argued that the first part of his proposal was already supported, albeit indirectly given the lack of evidence that there had ever been a successful language learner in the absence of comprehensible input. To bolster his point, he cited results of research involving hearing children of deaf parents whose language acquisition was delayed or incomplete because comprehensible input was not available through parental interaction (Larsen-Freeman & Long, 1991). The second part of Longs proposal found consistent support through his own research (see Long, 1985) and in studies by Blau (1982),G a s and Varonis (1985a, 1985b),Johnson (1981), and Pica, Young, and Doughty (1987). Thus, throughout the 1980s, important connections between conversational modifications, comprehension, and acquisition were well under way to becoming e s tablished. Despite the promising results of this early research, however, the effect of interaction on acquisition has remained a complex issue. As Long himself pointed out (1985), comprehensible input, in itself, was necessary but not sufficient to promote the acquisition process. Interactional modifications, therefore, cannot be the only mechanism behind the learners L2 develop ment. As noted above, Sato (1986) began to question the claim concerning a direct relationship between interaction and acquisition. She based her argument on findings from her earlier research (1985) on two Vietnamese boys learning English as a second language (ESL), whom she had studied intensively for 10 months and whose primary source of L2 input came from conversational interaction with their teachers and schoolmates, their foster parents, and Sato herself. Sato (1986) found that the boys made little progress toward L2 proficiency, particularly with respect to their application and control of past tense inflections, despite the opportunity to hear and produce these linguistic features in their social discourse on a daily basis. Satos analysis of

Susan M. Gas, Alison Mac@, and TerRta Pica

301

her own conversational interactions with the boys revealed that they were able to establish time reference in ways that were comprehensible but often obviated the need for the application of temporal and aspectual morphosyntax. Relying on the knowledge that they shared with their interlocutor, conversational features (such as their interlocutors use of past tense markings), and their own insertions of adverbial phrases and calendric expressions to indicate time, the boys were able to achieve mutual comprehensibility with Sat0 and other interlocutors without the need for complex grammar. On the basis of her research, Sat0 (1986) did not rule out the importance of conversational interaction in learners access to input for past time marking but suggested that it served more as a source for linguistically salient features such as adverbial phrases and calendric expressions than for verb inflections and other structures that were less perceptible in conversational discourse. She also argued that the role of interaction in acquisition was far more complex than had been heretofore conceived. It was becoming increasingly apparent that researchers would need to look for additional interactional processes that could assist the learners access to L2 forms not readily apparent in the comprehensible input generated by conversational interaction. One such interactional process was advanced by Swain (1985) and led to a line of research that Swain has continued to date (see Swain, 1995;Swain & L a p kin, 1995; and Swain & Lapkin, this issue). Swain (1985) has argued for the utility of what she has called comprehensible output. Her work has expanded and diversified the role of conversation in SLA, as she has suggested that conversation (and production, in general) pushes learners to impose syntactic structure on their utterances. This is in contrast to comprehension, in which it is not always essential for learners to draw on knowledge of L2 syntax. Thus, with respect to the more complex dimensions of L2 syntax, it is the necessity for learners to strive toward comprehensibility in responding to interlocutor feedback, rather than to reach comprehension of interlocutor input, that may play a pivotal and yet somewhat selectiverole in the acquisition process. The importance of feedback, particularly as a source of negative evidence, as a way of elucidating the inadequacy of learners own rule systems, has also been pointed out by White (1987), who suggested that what is necessary for L2 develop ment is not comprehensible input, but incomprehensible input. B y this she means that modifications to language (triggered by something

incomprehensible) become the impetus for learners to recognize the inadequacy of their own rule system. As Gass (1997) argues, it is incomprehensible input that may trigger learners recognition of mismatches between their IL grammar and that of their L2 target. In essence, this is the crux of the current argument as to the possible role that interaction plays in the learning process. As illus trated in a series of articles by Gass and Varonis (see, e.g., Varonis & Gas, 1985), comprehension difficulties or instances of non-understanding are what allow a learner to realize that linguistic modification is necessary. In essence, then, interactional modifications or negotiations (as they have long been referred to by Gass and Varonis and, increasingly, by researchers throughout the field of SLA), can serve to focus learners attention on potentially troublesome parts of their d i s course, providing them with information that can then open the door to IL modification. These modifications may, in turn, lead to subsequent stabilization or language change. Through clarification and elaboration of the message, nonnative speakers (NNSs) can receive more usable input in their quest to understand the L2 and, further, this new or elaborated input can draw attention to IL features that diverge from the L2. It is the realization of divergence between L2 forms and target language (TL) forms that becomes the catalyst for learning. Therefore, negotiation, along with certain classroom activities such as teacher explanation, can bring particular forms to a learners attention-forms that might otherwise be unnoticed-thus enhancing the input (Sharwood Smith, 1985) and making it more salient. This perspective on the relationship between interaction and, more specifically, negotiation and L2 development has, in turn, stimulated research of considerable scope and theoretical import. Long (1996) presents a similar view of the role of interaction/negotiation, noting how it connects input, internal learner capacities, particularly selective attention, and output in productive ways (p. 452) (emphasis ours). Any sort of reformulation of an incorrect utterance (assuming that a learner recognizes it as a reformulation), such as an expansion, recast, or rephrasing, can serve to draw a learners attention to the fact of the incorrectness and can thereby trigger learner-internal mechanisms (e.g., hypothesis testing), which may, in turn, result in immediate output change on the part of the learner. Immediate output change can lead to a quick response to the revised hypothesis and hence a tentative confirmation or rejection of that revised hypothesis.

302 RECENT EMPIRICAL STUDIES

The Modern LanguageJournal 82 (1998) learners internalize and retain these and other linguistic features over the long term. Further studies in this area by LaPierre (1994) and Donato (1994) have suggested that pushed output may result in more permanent IL restructuring (see references in Swain, 1995). Mackey (1995, 1997) also suggested that changes can be maintained, at least for a short period of time. Taken together, these studies of adult and adolescent L2 learners have found that interactional modifications that occur through pushed output can be maintained at least in the short run. Longitudinal data or delayed posttests are obviously a necessary step in order to test this hypothesiswith respect to long-term retention. Other recent studies of the effects of interaction that bring up the topic of long-term change have been reported in the research literature. Gass and Varonis (1994) compared the impact of prescripted modified and unmodified input, with and without the opportunity for interaction, on both comprehension and production, as measured by their NS partners success in following the directions. They found that both the negotiated and modified input positively affected comprehension. In addition, they found that prior negotiation, but not prior modified input, significantly affected subsequent production, leading them to suggest that interaction, with the opportunity for modifications, may impact positively on later language use. Other research has also explored the shortterm effects of interactional modifications on comprehension and acquisition of vocabulary as well as the development of targeted grammatical structures. Loschkys (1994) study focused on these processes with respect to vocabulary items and locative constructions in Japanese as a L2. His results showed a positive effect on comprehension of the vocabulary but no effect on retention or acquisition of the vocabulary and the locatives. With respect to vocabulary, however, a study by Ellis, Tanaka, and Yamazaki (1994) found a greater role for interactional modifications than that of Loschky. This research revealed that, when compared with premodified input, interactionally modified input resulted in both better comprehension and greater vocabulary acquisition. As explained by Ellis et al., these differences in results on acquisition may have been due to the items under study. Loschky investigated locative constructions, whereas Ellis et al. used vocabulary items. The research of Ellis et al., together with that of Gass (1988) and Gass and Varonis (1994), further emphasize the relationship among interaction, comprehension,

OF CONVERSATION AND SLA

The considerable expansion in the body of work on the role of interaction in SLA in the past several years has, in part, been made possible through the use of more innovative and varied methods of data collection-including interactive tasks and computer controlled interactionand through a focus on cognitive processes such as attention and recall. These have long been considered important in L2 development and are now beginning to be operationalized carefully for empirical investigation. From this recent research has emerged a more focused view of the relationship between interaction and L2 development and a reinforcement of the notion that the developmental outcomes of interaction are indeed complex. A number of studies have shown that interactional modifications that are brought about through negotiation for meaning can have a p o s itive effect on the quality of learners immediate production. Studies by Holliday (1995) and Linnell (1995),which have been able to track changes in learner production through their participation on interactive computer tasks, have suggested that the modifications in output that occur as a result of negotiation for meaning are often targetlike in direction. Review papers by Pica (1994, 1996) have extended the claims for negotiation, outlining the ways in which interaction modified through negotiation brings about reformulations, segmentations, and movement of constituents that can provide learners with both lexical and grammatical information about the L2 as well as their own IL system. Such adjustments might also serve as a source of linguistic data for the learner both immediately and possibly also in the longer term. Recent research has also addressed the ques tion of whether the increasinglytarget-like output that learners can obtain through their negotiation for meaning can benefit their L2 develop ment over time. With respect to retention in the short run, Nobuyoshi and Ellis (1993) explored this issue with a small-scale study examining the developmental outcomes of pushed output (i.e., the construct that Swain [1985] had originally labeled comprehensible output). Two of their three experimental learners maintained an improvement in accuracy after 1 week of treatment. This finding indicated that output that is pushed may increase learners control over structures they have already acquired. However, questions remained as to whether it could help

Susan M. Gass, Alison Mackq, and Teresa Pica

303 The article by Polio and Gass in this issue is a partial replication of an earlier study by Gass and Varonis (1994), in which one of the findings was that interaction did not lead to better comprehension of NNSs by NSs. This result was surprising because one would expect that, given the o p portunity to provide clarification, NNS speech would eventually become more comprehensible. Using a design similar to that in the original study, Polio and Gass have attempted to replicate the original results. However, it is important to note that the Polio and Gass design contained subtle modifications as well as a larger sample size. Due to the use of two different scoring procedures and a different statistic, the original results were not confirmed. Comprehension by NSs was better in the interactive condition. In addition to quantitative measures, Polio and Gass (this issue) qualitatively investigated the transcripts of successful versus unsuccessful dyads in order to determine what led to a successful outcome (as defined by success on the particular task used). It is interesting to note that they found that the NS interlocutor was in large part responsible for this difference. In instances of unsuccessful task completion (an information gap task in which the NNS held all of the information), the NS took a leadership role in completing the task. The result was that the NNS was not given an opportunity to complete his or her linguistic agenda. If output is a means of testing hypotheses and receiving feedback, the NNS in the unsuccessful attempts was not able to take advantage of this opportunity. A second matter raised by Polio and Gass relates to broader concerns within the aforementioned research paradigm. In an information gap activity such as that used by Polio and Gass, one must rely on an individual to provide the a p propriate missing information. However, each individual makes his or her own assumptions about what constitutes crucial information and what does not, and there is great individual variation in these assumptions. Therefore, results may depend more on how an information provider perceives the needs of an interlocutor than on linguistic feats or lack thereof. The article by Swain and Lapkin (this issue) investigates the role of conversation within a sociocultural framework, a framework that is complementary to that used in most of the other articles in this issue. The starting point of their investigation is that language production is both a communicative act and a cognitive activity. It is clear that this is an extension of Swains earlier (1985, 1995) work in which she argued that output

and SLA as one in which interaction allows learners to comprehend TL input and in which comprehended input is important for SLA. Additional research has also suggested ways in which both the measure of development and the differences in the interactional conditions might account for their different findings. Mackey (1995, 1997) found that certain types of interaction can have a positive effect on L2 develop ment. She examined the effects of different types of input and interaction on the short-term development of question formation in ESL. Using a pretest, delayed posttest design, Mackey (1995, 1997) found a positive effect for negotiated interaction where there was active participation in the interaction, but no effect for observing interaction without opportunities for production or for receiving modified input with no opportunity for interaction. The interaction in Mackeys study involved NS-NNS dyads carrying out communicative tasks that provided contexts for learners to produce specific morphosyntactic forms. In addition to the research mentioned above, expansions of this line of enquiry continue through a variety of empirical studies in the area of conversational interaction and L2 develop ment. The focus of these studies is increasingly on such issues as the role of negative feedback received through conversational interaction and the role of conversational interaction in promoting noticing and attention to form (e.g., Schmidt, 1994; Tomlin & Villa, 1994). THE CONTRIBUTION OF THE PRESENT SPECIAL ISSUE Many of the recent empirical studies would seem to lead us back to some of the original work on conversation and SLA and to Satos (1986) proposal that conversation is selectively facilitative of L2 development, depending on the structures involved (see also Long, 1996). Satos proposal is becoming the focus of current research. Studies to date suggest that research should focus on (a) the nature of the conversational interaction, (b) whether or not opportunities are present for the conditions and processes that are claimed to facilitate language learning, and (c) the nature of the development that takes place. The aim of the present thematic issue is to address these matters from avariety of perspectives. The issue itself is particularly timely given the burgeoning interest in this topic, which is obvious from the many recent publications, conference papers, colloquia, and doctoral dissertations in this area.

304
serves a learning function. Swain and Lapkins main premise is that through discourse, linguistic knowledge is co-constructed by the interlocutors and, also important, that this represents ongoing learning. The database for Swain and Lapkins present study comes from eighth-grade students in a French immersion classroom engaged in a jigsaw task. The authors present an in-depth analysis of language-related episodes, which they define as episodes where the focus is on language. One pair from Swain and Lapkins database is singled out in their article in order to show how the mental processes (e.g., hypothesis generation, hypothesis testing, and extension of knowledge) involved in dialogic communication mediate actual learning. An interesting aspect of their study and one that strengthens their argument concerning learning as an outcome of collaboration through dialogic interaction can be seen from the results of pretest, posttest comparisons. Adding this component to their research design allows Swain and Lapkin to see ways in which knowledge may be gained through interaction and may in some cases extend to new L2 contexts. As in the Polio and Gass study, these researchers raise the question of individual approaches to task completion and, consequently, the ways in which individuals profit differentially from the interaction. The article by Mackey and Philp (this issue) also suggests that certain aspects of conversational interaction may be particularly beneficial to certain learners. In addition, they demonstrate that the benefits of interaction on IL restructuring may not always be seen immediately. This does not mean that there are no positive effects, but simply that these effects may show up in later language use. Mackey and Philps study addresses the development of question forms in ESL and uses information gap tasks that were designed specifically in order to both promote interaction and to provide contexts in which the targeted forms would be produced. The study focuses on one feature of interaction-recasts-and compares groups of learners who received interactionally modified input while carrying out the tasks with learners who received the same input but containing intensive recasts. Their study explores both the effect of recasts on learners short-term IL development and the nature and content of learners responses to recasts. The results of Mackey and Philps study suggest that, for more advanced learners, interaction with intensive recasts may be more beneficial

Th.e Modern LanguageJournal 82 (1998) than interaction alone in facilitating an increase in production of higher-level morphosyntactic forms. However, for the less advanced learners in their study, recasts were not as effective. The positive developmental effects were found for advanced learners who received recasts even though, while generally acknowledged in the discourse, recasts were usually not repeated and rarely elicited modification by the learners. Mackey and Philps study, therefore, suggests that recasts may be beneficial for short-term IL development even though they are not incorporated into learners immediate responses. As other researchers (Gass, 1988, 1997; Lightbown, 1994) have cautioned, the developmental effects of interaction may not appear immediately. The article by Long, Inagaki, and Ortega (this issue) also focuses on recasts, in an examination of their effects on L2 development. Long (1996) has suggested that negotiated interaction is one way in which implicit negative feedback, in the form of recasts, can be elicited, and that such feedback draws learners attention to mismatches between input and output. Long et al.s article compares models with recasts and addresses the question of the relative contributions of models and recasts to foreign language development of Japanese and Spanish. The treatment in their study was delivered via an information gap communication game, played in either a model or a recast version. The Long et al. article involves two studies focusing on models provided prior to an utterance and recasts provided after an utterance in the foreign languages. The authors found evidence in the Spanish study that recasts were more effective than models in the development of a previously unknown L2 structure, adverb placement. However, on the second structure in the Spanish study-object topicalization-they did not find any (measured) effect on L2 learning for recasts. They suggest that this may be due to that structure being too difficult for students to learn. Long et al. also found limited evidence for improvement in the treatment groups in the related study of Japanese as a L2, although, as they note, those results are difficult to interpret, due to the presence of individuals in the group with prior knowledge of these structures. Here again, we see the matter of individual variation, which emerges from all of the studies in this special issue. The Japanese study focused on adjective ordering and a locative construction, following up on an earlier study conducted by Mito (1993). Long et al. conclude that although possibly aided by a triggering effect of the pretest, recasts in the

Susan M , Gass, Alison Mackq, and Teresa Pica Japanese study produced some learning or resus citation of latent prior knowledge of the structures. Taken together, the results of both the Spanish and Japanese studies provide support for the claim that implicit negative feedback plays a facilitative role in SLA. However, Long et al. do caution against using their results as conclusive evidence and suggest the need for detailed longitudinal case studies in this area. The article by Oliver (this issue) deals with a neglected database in SLA research, that of child L2 learners. She investigates conversational interactions involving child learners of ESL. Because much of SLA research is generally focused on adults or adolescents, Oliver asks whether, like adults, children can negotiate for meaning, what strategies they use to do so, and whether there are differences in the ways adults and children handle such negotiation. Olivers database consists of 192 students, aged 8 to 13years old, who were paired to form 96 ageand gender-matched dyads. The pairs worked together on both a one-way and a two-way communication task. Based on an examination of the transcripts of their conversations, Oliver demonstrated that, like adults, children also negotiate for meaning. However, she also found that the patterns of use demonstrated by children differed from adult patterns of use. The differences were found not in the various types of strategies used but in the proportional use of particular strategies. For example, with respect to comprehension checks, Olivers results suggest that children use this type of negotiation strategy, but that they use it much less often than adults. Oliver claims that the egocentricity of children may account for these differences because this particular conversational adjustment is overtly concerned with helping ones conversational partner to construct meaning rather than just clarifying or confirming meaning for oneself. CONCLUSIONS Despite the emphasis in this special issue on the role of interaction studies in L2 development, it is still advisable to be cautious about the nature of the claims for the role of the environment in SLA. Although interaction may provide a structure that allows input to become salient and hence noticed, interaction should not be seen as a cause of acquisition; it can only set the scene for potential learning. As Long (1996) has pointed out, there are many factors involved in L2 learning; the role of interaction is claimed only to be facilitative. The sources of learning are

305 complex and can be seen as stemming from learner-internal factors, some of which have received extensive treatment in the SLA literature (see Gass, 1997, for a review). However, current research on the role of interaction in L2 developmentcontinues to contribute to our understanding of the relationship between input, interaction, and SLA. The articles in this special issue have brought together a variety of different perspectives and approaches to the study of conversational interaction and L2 development, yet similar themes have also emerged. It seems that interaction can have positive effects on L2 development and that the complex matter of individual differences needs to be considered carefully. Taken as a whole, therefore, this special issue of the MLJreminds us of Satos view that the topic of conversational interaction and SLA is a complex one: The many questions surrounding the study of interaction, development, and L2 learners suggest that this area will continue to provide challenges as well as insights into our understanding of the processes involved in SLA, as it did Sato until her untimely death in 1996. It is to her memory that this special issue of the MLJ is dedicated.

REFERENCES

Blau, E. (1982). The effect of syntax on readability for ESL students in Puerto Rico. TESOL Quarterly, 16,517-528. Donato, R. (1994).Collective scaffolding in second language learning. In J. Lantolf & G. Appel (Eds.), Vygotskian approaches to second language research (pp. 33-56). Norwood, NJ: Ablex. Ellis, R., Tanaka, Y., & Yamazaki, A. (1994). Classroom interaction, comprehension and the acquisition of L2 word meanings. Language Learning, 44, 449-491. Gass, S. M. (1988). Integrating research areas: A framework for second language studies. Applied Linguistics, 9, 198-217. Gass, S. M. (1997). Input, interaction and the second language k a w . Mahwah, NJ: Lawrence Erlbaum. Gass, S. M., &Varonis,E. M. (1985a).Variationin native speaker speech modification to non-native speakers. Studies in Second Language Acquisitiun, 7,37-58. G a s , S. M., & Varonis, E.M. (1985b).Task variation and nonnative/nonnative negotiation of meaning. In S. M. Gass & C. G. Madden (Eds.), I n h t in second language acquisition (pp. 149-161). Rowley, MA: Newbury House. Gass, S. M., &Varonis, E. M. (1994). Input, interaction, and second language production. Studies in Second Language Acquisition, 16,283-302.

306
Hatch, E. (1978). Acquisition of syntax in a second language. In J. Richards (Ed.), Understanding second and fweign language learning (pp. 34-70). Rowley, MA: Newbury House. Hatch, E., & Wagner-Cough, J. (1976). Explaining sequence and variation in second language acquisition. LanguageLearning, 4,39-47. Holliday, L. (1995). NS syntactic modifications in N S N N S negotiations as input data for second language acquif syntax. Unpublished doctoral dissertasition o tion, University of Pennsylvania, Philadelphia. Johnson, P. (1981). Effects on reading comprehension of language complexity and cultural background of a text. TESOL Quarterly, 15, 169-181. Krashen, S. D. (1985). The input hypothesis: Issues and implications. London: Longman. LaF'ierre, D. (1994). Language output i n a cooperative learning setting: Determining its effects on second language learning. Unpublished master's thesis, University of Toronto (OISE),Toronto, Canada. Larsen-Freeman, D., & Long, duction to second language don: Longman. Lightbown, P. (1994). The importance o f timing infocus on fm.Paper presented at the Second Language Research Forum, McGill University, Montreal, Canada. Linnell,J. (1995). Negotiation as a contextfor learningsyntax in a second language. Unpublished doctoral dissertation, University of Pennsylvania, Philadelphia. Input, interaction, and second lann. Unpublished doctoral dissertation, University of California, Los Angeles. Long, M. H. (1983a). Linguistic and conversational adjustments to non-native speakers. Studies in Second Language Acquisition, 5, 177-194. Long, M. H. (198313).Native speaker/non-native speaker conversation and the negotiation of comprehensible input. Applied Linguistics, 4, 126-141. Long, M. H. (1985). Input and second language acquisition theory. In S. M. Gass & C. G. Madden (Eds.), Input i n second language acquisition (pp. 377-393). Rowley, MA: Newbury House. Long, M. H. (1996). The role of the linguistic environment in second language acquisition. In W. C. Ritchie & T. K. Bhatia (Eds.), Handbook o f language acquisition: Vol. 2. Second language acquisition (pp. 413-468). New York: Academic. Long, M. H., Inagaki, S., & Ortega, L. (1998). The role of implicit negative feedback in SLA: Models and recasts in Japanese and Spanish. Modern LanguageJournal, 82,357-371. Loschky, L. C. (1994). Comprehensible input and second language acquisition: What is the relationship? Studies in Second Language Acqu 303-325. Mackey, A. (1995). Stqbping up the pace: Input, interaction f and interlanguage developmat. A n emfi'rical study o questions i n ESL. Unpublished doctoral dissertation, University of Sydney, Australia.

The Modern LanguageJournal 82 (1998)

Mackey, A. (1997). Stqbptng up the pace: Input, interaction and second language develqpment Unpublished manuscript, Michigan State University, East Lansing. Mackey, A., & Philp, J. (1998). Conversational interaction and second language development: Recasts, responses, and red herrings? Modern Language Journal, 82,338-356. Mito, K. (1993). Theeffects o f modellingand recasting on the on o f L2 grammar rules. Unpublished manuscript, University of Hawai'i at Manoa. Nobuyoshi, J., & Ellis, R. (1993). Focused communication tasks and second language acquisition. English Language TeachingJournal, 47,203-210. Oliver, R. (1998). Negotiation of meaning in child interactions. Modern LanguageJournal, 82,372-386. Pica, T. (1994). Research on negotiation: What does it reveal about second-language learning conditions, processes, and outcomes ? Language Learning, 44,493-527. Pica, T. (1996). Second language learning through interaction: Multiple perspectives. Wwking Papers in Educational Linguistics, 12, 1-22. Pica, T., Young, R., & Doughty, C. (1987). The impact of interaction on comprehension. 7ESOL Quarterly, 21, 737-758. Polio, C., & Gass, S. M. (1998). The role of interaction in native speaker comprehension of nonnative speaker speech. Modern Language Journal, 82, 308-319. Sato, C. J. (1985). The syntax o f conversation in interlanguage develupmmt. Unpublished doctoral dissertation, University of California, Los Angeles. Sato, C. J. (1986). Conversation and interlanguage development: Rethinking the connection. In R. R. Day (Ed.), Talking to learn: Conversation in second language acquisition (pp. 5-22), Rowley, MA: Newbury House. Schmidt, R. W. (1994). Deconstructing consciousness in search of useful definitions for applied linguistics. AZLA Review, 1I , 11-26. Sharwood Smith, M. (1985). From input to intake: On argumentation in second language acquisition. In S. M. Gass & C. G. Madden (Eds), Input in second language acquisition (pp. 394-403). Rowley, MA: Newbury House. Swain, M. (1985). Communicative competence: Some roles of comprehensible input and comprehensible output in its development. In S. M. Gass & Madden (Eds.), Input i n second language acon (pp. 235-253). Rowley, MA: Newbury House. Swain, M. (1995). Three functions of output in second language learning. In G. Cook & B. Seidlhofer (Eds.), Principle and practice in applied linguistics Studies in Honour o f H. G. Widdowson (pp. 125144) Oxford: Oxford University Press. Swain, M., & Lapkin, S. (1995). Problems in output and the cognitive processes they generate: A step towards second langage learning. Applied Linguistics, 16, 371-391.

Susan M. Gass, Alison Mackq, and Teresa Pica

307
Wesche, M. (1994). Input and interaction in second language acquisition. In C. Gallaway & B. Richards (Eds.), Input and interaction in language acquisition (pp. 219-249). Cambridge: Cambridge University Press. White, L. (1987). Against comprehensible input: The input hypothesis and the development of L2 competence. Applied Linguistics, 8,95-110.

Swain, M., & Lapkin, S. (1998). Interaction and second language learning: Two adolescent French immersion students working together. M o d a LanguageJournal, 82,320-337. Tomlin, R., &Villa,V. (1994).Attention in cognitive science and second language acquisition. Studies in Sewnd Language Acquisition, 16,183-204. Varonis, E., & Gas, S. M. (1985). Miscommunication in native/non-native conversation. Language in Society, 14,327-343. Wagner-Gough,J., & Hatch, E. (1975).The importance of input data in second language acquisition studies. Language Learning, 25, 297-307.

Tribute to Robert L. Politzer

AS MOST MLJREADERS ARE LIKELYNOW AWARE, ROBERT POLITZER PASSED A W A Y ON MONday,January 26,1998 after a fourteen-month battle with cancer. Based on his long association with the MLJ, it is fitting to find a recognition to his scholarly career in its pages. Born in Vienna, Austria on March 21, 1921, Politzer immigrated to the US. in 1939, fleeing the increasing fascist terrorism of the Nazi regime. When America entered the war, Politzer enlisted and returned to Europe with the US. Army and served with Military Intelligence. After the war, he returned to New York and completed his Ph.D. Degree in linguistics at Columbia University. He subsequently earned a Doctor of Social Sciences degree from the New School for Social Research. During his distinguished scholarly career, he published 14 articles, translations, or notes in the MLJbetween 1951 and 1978. Their titles are as follows: Linguistics and the Elementary Language Course. MLJ, 35 (1951):314-18. On a Linguistic Classification of Teaching Methods. MLJ, 37 (1953):pp. 331-34. A Brief Classification of the Limits of Translatability. MLJ, 40 (1956): 319-22. On the Relation of Linguistics to Language Teaching. MLJ, 42 (1958): 65-68. Assiduity and Achievement. MLJ, 44 (1960): pp. 14-16. Some Reflections on Pattern Practice. MLJ, 48 (1964):24-28. The Impact of Linguistics on Language Teaching: Past, Present, and Future. MLJ, 48 (1964): 146-51. The Macro and Micro Structure of the Foreign Language Curriculum. MLJ 49 (1965): 99-102. Toward a Practice-Centered Program for the Training and Evaluation of Foreign Language Teachers. MLJ, 50 (1966): 251-55. Developmental Aspects of Auditory Discrimination, Echo Response, and Recall (with Louis Weiss). MLJ, 53 (1969):75-85. Practice-Centered Teacher Training: Standard English as a Second Dialect (with Diana E. Bartley). MLJ, 54 (1970):31. Toward Individualization in Foreign Language Teaching. MLJ, 55 (1971):207-12. Developmental Sentence Scoring as a Method of Measuring Second Language Acquisition. MLJ, 58 (1974): 245-50. Errors of English Speakers of German as Perceived and Evaluated by German Natives. MLJ, 62 (1978): 253-61. At the time of his death, Politzer was an Emeritus Professor of Education and Romance Linguistics from Stanford University. The profession recognizes and expresses gratitude for his lasting contributions to foreign language learning.