Second/Foreign Language Learning as a Social Accomplishment: Elaborations on a Reconceptualized SLA

ALAN FIRTH School of Education, Communication and Language Sciences Newcastle University Newcastle Upon Tyne, NE1 7RU United Kingdom Email: alan.rth@ncl.ac.uk JOHANNES WAGNER University of Southern Denmark, Kolding Engstien 1 DK6000 Kolding Denmark Email: jwa@sitkom.sdu.dk

In this article, we begin by delineating the background to and motivations behind Firth and Wagner (1997), wherein we called for a reconceptualization of second language acquisition (SLA) research. We then outline and comment upon some of our critics reactions to the article. Next we review and discuss the conceptual, theoretical, and methodological impact the article has had on the SLA eld. Thereafter, we reengage and develop some of the themes raised but left undeveloped in the 1997 article. These themes cluster around the notions of and interrelationships between language use, language learning, and language acquisition. Although we devote space to forwarding the position that the dichotomy of language use and acquisition cannot defensibly be maintained (and in this we take up a contrary position to that held in mainstream SLA), our treatment of the issues is essentially methodological. We focus on describing a variety of aspects of learning-in-action, captured in transcripts of recordings of naturally occurring foreign, second, or other language interactions. Through transcript analyses, we explore the possibilities of describing learning-in-action devoid of cognitivistic notions of language and learning. In so doing, we advance moves to formulate and establish a reconceptualized SLA.

WITH THE BENEFIT OF 10 YEARS OF HINDsight, it is probably fair to say that our article (Firth & Wagner, 1997) touched a proverbial raw nerve within as well as around the periphery of the second language acquisition (SLA) community. Certainly the articles evident impact on the profession has greatly exceeded our expectations. This is in no small measure thanks to The Modern Language Journals (MLJ ) editor, Sally Sieloff Magnan, who so insightfully showcased the article and its thought-provoking responses, thereby providing a framework for an extraordinarily fruitful and lively debate. For some, the article has
The Modern Language Journal, 91, Focus Issue, (2007) 0026-7902/07/800819 $1.50/0 C 2007 The Modern Language Journal

been a rallying cry for an alternative SLA; for others, it has presaged new lines of inquiry within a rapidly developing and multivaried research eld; for many, it articulated misgivings and dissatisfaction with dominant SLA concepts, theories, and research practices. For others still, it was an unwarranted, misguided, perplexing, and nave cri tique of a well-dened eld of study. We begin by explaining the background to our 1997 article.

BACKGROUND CONTEXT OF FIRTH AND WAGNER (1997) Firth and Wagner (1997) was conceived on the basis of a lingering sense of frustration with SLA as we saw it at that timea sense that, in

Alan Firth and Johannes Wagner large measure, prevailing, dominant SLA theory and concepts were myopic vis-` -vis language a learning as social practice and language as a social phenomenonand a strong sense that the eld had to acknowledge more openly and more consequentially its limited vision, and, if possible, overcome its myopia. The frustrations with SLA emerged gradually, from a bottom-up process while working with our data materials audiorecordings of people at work in a variety of settings, engaging each other in a second/foreign/other language (L2). Our analyses revealed people who were artfully adept at overcoming apparent linguistic hurdles, exquisitely able to work together interactionally, despite having what at rst blush appeared to be an imperfect command of the languages they were using (Firth, 1990, 1991, 1995a, 1995b, 1995c, 1995d, 1996; Wagner 1995a, 1995b, 1996, 1998; Wagner & Firth, 1997). When we turned to apparently relevant areas of SLA for theoretical insight, applicable concepts, and methodologies, something was not right. Rather than depictions of interactional success in an L2, we found an overwhelming emphasis on and preoccupation with the individuals linguistic and pragmatic failure. Rather than talk, we found input. Rather than achievement, we found an abundance of problem-sources. Rather than collaboration, invention, and an extraordinarily creative use of shared resources (which, to us, was learning-in-action), we found references to errors, input modications, interference, and fossilizations. Try as we might at rst, our observations of people using English as a lingua franca (i.e., a mediating language that is not a mother tongue [L1] for any of the interactantssee Firth, 1990, 1996, see also Jenkins, 2006a, 2006b; Seidlhofer, 2001, 2004) just would not t the theory and concepts of SLA. Yet such data materials were, surely, of critical relevance for the SLA program. Here, after all, were nonnative speakers who were clearly acquiring and using (new) forms of language on the yoften language forms that were linguistically and pragmatically marked. Moreover, despite noncollocating noun phrases, verb-concordance, prosodic and morphosyntactic errors, and a host of other linguistic anomalies, they were buying and selling thousands of tons of Danish cheese (as well as sh, owers, and steel) on a daily basis, and maintaining cordial relations with one another. This was real cheese, and they were talking (albeit in linguistically marked ways) real, and big, money. With a background in both SLA and ethnomethodological conversation analysis (CA),1

801 and informed by poststructuralist notions of contingency, uidity, hybridity, and marginality (see, e.g., Lyotard, 1979; Rampton, 1995), we were uneasy about how to approach our data materials. CA, with its emphasis on the socially achieved construction of irredeemably motile, participantdened contextual relevancies, its commitment to the microanalytic explication of naturally occurring (rather than experimental) encounters, and an emic (participant-centered) sensitivity to whats going on, led us to see that our participants were not defensiblythat is, to us, emically identiable as participants, learners, or even nonnative speakersthe standard identity categories of SLA. At the least, such categories were clearly not omnirelevant: These individuals were also, varyingly, sellers, buyers, friends, business acquaintances, customers, and clients. Identity, we had learned, was a motile, liminal, achieved feature of the interaction (Antaki & Widdicombe, 1998; Gumperz, 1982; Rampton, 1995). Andas we will showwe noted that who they are locally and discursively has an impact on how they use and learn language.2 Certainly they were not learners in any formal (or traditional SLA) sense of the term. Yet they were clearly modifying, adapting, and creatively deploying what were to them new forms of language; that is, they were learning, and they would occasionally draw attentionoften in jestto their underdeveloped foreign language abilities when engaging each other. We were hesitant about calling their interlanguage interlanguage, for where and what was the target language here, in this seemingly linguistically lawless lingua franca landscape? Here, it appeared, the spectre of the native speaker, with his or her idealized norms and baseline standards, had evaporated (on this concept, see Jenkins, 2006a).3 SLA tted, but then again, it did not. Something was quite clearly not right. The eld was, we argued, in need of theoretical, methodological, and conceptual refurbishment and expansion. In particular, we (Firth & Wagner, 1997) called for a reconceptualization within SLA in three areas: (a) a signicantly enhanced awareness of the contextual and interactional dimensions of language use, (b) an increased emic (i.e., participant-relevant) sensitivity towards fundamental concepts, and (c) the broadening of the traditional SLA data base (p. 286). From our observations of SLA, it was the individuals disembodied cognitionmore specically, his or her autonomous language processingthat was in the ascendancy, to the detriment of what we might call social cognition, that is, what people do,

802 think, demonstrate, achieve, manipulate, modify, acquire, and learn, together, in concerted social interaction (e.g., Volosinov, 1930/1973; Vygotsky, 1978; Wertsch, 1991). This is not to say, then, that learning was or is uninteresting or somehow incidental to our methodological pursuits. What we did not attempt to show in Firth and Wagner (1997)but will attempt to do in this articleis that learning can be explicated and tracked as it happens and, moreover, that the description of how it happens can be liberated from a cognitivistic position predicated on telementational notions of communication4 and individualistic, monolinguistic, and formalistic perceptions of language.5 These perceptions, we argued, infuse SLA research, in its theorizing on language, learning, and discourse, in its terminologies and fundamental concepts (such as an ascendant native speaker and interlanguage), and in its methodological practices, including the practices of those who focalize interaction in SLA studies. The immediate reactions to the 1997 article were, for the most part, polarized, with rather stern criticism on one side and rm support on the other.6 Because the criticisms go to the heart of the debate, a debate that, as recent publications attest, has in no manner been concluded (see, e.g., the recent exchanges between Gass, 2004; Hall, 2004; Larsen-Freeman, 2004; and Wagner, 2004), and because these criticisms reveal the key counterarguments to ours, while demonstrating how entrenched positionings within SLA can become, let us briey consider the criticisms of Firth and Wagner (1997). CRITICISMS OF FIRTH AND WAGNER (1997) In contradistinction to our argument, Long (1993) had actually argued for SLA theory culling, rather than an expansion or broadening of SLA. In his response to our 1997 article, Long (1997) ventured that the fundamental problem with our position, which among other things emphasized the need to focus on L2 acquisition through L2 use,7 was that SLA is the study of L2 acquisition, [and] not . . . most centrally the language use of second or foreign language speakers (p. 318). Long went on to state that most SLA researchers view the object of inquiry as in large part an internal, mental process (p. 319). However, he conceded that Firth and Wagner are perfectly justied, and probably right, in arguing that a broader, context-sensitive, participant-sensitive, generally sociolinguistic orientation might prove benecial for SLA research (p. 322). Nevertheless, he ended his response with the major

The Modern Language Journal 91 (2007) problem I have with Firth and Wagners polemic remains my skepticism as to whether greater insights into SL use will necessarily have much to say about SL acquisition (p. 322). Kasper (1997) adopted a similar stance. A stands for acquisition, she reminded us, adding, tellingly, that [a] noncognitivist discipline that has learning as its central research object is a contradiction in terms (p. 310; note, however, that we did not, in Firth and Wagner, 1997, call for a noncognitivist SLA). In close agreement with Long (1997), Kasper held that the most nagging problem with the Firth and Wagner paper [is that it] has in fact very little to say about L2 acquisition (p. 310). She went on to state that I am comfortable with an essentially cognitivist denition of SLA (p. 310), though in her self-styled heretical nal note she submitted that
if the excellent microanalytic tools of CA were incorporated into a language socialization approach to SLA, we might be able to reconstruct links between L2 discourse and the acquisition of different aspects of communicative competence that have been largely obscure thus far. (p. 311)8

Poulisse (1997) also offered a defense of the psycholinguistic approach in response to our arguments, maintaining that the task of all researchers [is] to not only describe, but also explain and predict phenomena (p. 325) and it would denitely not do to just look at particular and local phenomena and nd specic explanations for each of them (p. 325). However, Poulisse conceded that Firth and Wagner have a point when they plea for an enhanced awareness of the contextual and interactional dimensions of language use and a more positive view of the NNs [non-native speakers] attempts to interact in the L2 (p. 327). Gass (1998), who found our perspective perplexing (p. 88), maintained that we had got it all very wrong and that our criticisms were nave and misplaced because, among other things, in calling for greater attention to communication, we had not understood that the proper object of attention in SLA is language; thus her correction: the emphasis in input and interaction studies [in SLA] is on the language used and not on the act of communication (p. 84). The goal of my work, she wrote, (and others within the input/interaction framework . . .) has never been to understand language use per se . . . but rather to understand what types of interaction might bring about what types of changes in linguistic knowledge (p. 84). Gass made the erroneous observation that Firth and Wagner portray cognitive

Alan Firth and Johannes Wagner approaches and communication/discourse emphases as mutually incompatible (p. 84). In addition, she claimed that, because we called for an enlargement of the standard SLA database (which we see as being dominated by classroom and experimental settings) to include settings such as the workplace,9 and because we called for greater attention to L2 use from an emically informed viewpoint, the approach we espoused is not actually part of SLA, but part of the broader eld of L2 studies (of which SLA is a subset) (p. 84). She continued, many individuals in a workplace setting . . . are not learners in the sense that is of interest to researchers in SLA (p. 84). We shall return to some of these criticisms presently (see also Firth & Wagner, 1998). What is important to point out at this juncture are the rhetorical methods and arguments deployed both to dismiss our viewpoint and to impose a grand narrative (Lyotard, 1979) or dominant theory of SLA, a theory that is meant to trump ours precisely because it is framed and formulated as dominantfor example, in ex cathedra proclamations about how SLA is properly done, and through references to what most SLA researchers do (see Long, 1997; Gass, 1998). The criticisms also demonstrate rather vividly how readily hegemonic forces can come to the fore when metatheoretical contributions, such as Firth and Wagners (1997), attempt to critique prevailing, established, and thus mainstream, intellectual practices. Admittedly, this is a mainstream that we, in and through our critique, both aid in its construction as mainstream and contribute to its status as mainstream. Long (1997), Kasper (1997), Poulisse (1997), and Gass (1998) drew inwards, and from their self-constructed SLA pantheon they laid down the law by dening SLAs proper intellectual territory (e.g., learners, language, cognition), delineating its key concerns (e.g., acquisition, not use, language, not communication), and by pointing to its borders (e.g., by stipulating what is inside and outside SLA). In so doing, they dened the parameters of legitimate criticism and debate while essentially accusing us of intellectual trespass (see Firth & Wagner, 1998; Thorne, 2000). For some SLA researchers, including some with tangential SLA interests, the Firth and Wagner (1997) article seems to have been viewed as a kind of watershed for SLA in general and for SLA studies of interaction in particular. This reaction to it occurred, at least in part, as the result of the prominence the article was given in the MLJ at the timeas the centerpiece of a focused and extended discussion by prominent scholars contributing to an esteemed journal. However, as we

803 pointed out in our 1997 article, the arguments it contained did not emerge from an intellectual void. Although the article has since been characterized as seminal by a number of scholars, it would be a mistake to see it as some kind of sociocultural or sociointeractional big bang within or without SLA. Our theoretical position vis-` -vis a what we had termed and perceived as mainstream SLA emerged from our own observations of and engagement in an ineluctable sociocultural turn in the study of language that had swept across linguistics in general and discourse studies in particular, as well as across anthropology, social psychology, education, and sociology in the 1970s and 1980s, beginning, we argued, with Gumperz and Hymess (1972), Hallidays (1973), Hymess (1974), and others inuential calls for a socially constituted linguistics. The Firth and Wagner article also developed and was inuenced by a number of extant, socioculturally focused critiques of SLA by scholars such as Block (1996), Kramsch (1993), Lantolf (1996), Rampton (1987), and van Lier (1994). Of course, SLA research of the 1980s and 1990s was itself inuenced by this sociocultural turn as witnessed by the steady increase in studies that acknowledged, thematized, and explored context and interaction. However, we argued in Firth and Wagner (1997) that in the main, SLA work including contextual and interaction-oriented studieswas tenaciously resistant to the full implications of this sociocultural turn (see also similar criticisms made by Breen, 1985; Lantolf & Appel, 1994; Larsen-Freeman, 1983; Rampton, 1987). That is, despite a signicant increase in calls for more socially oriented approaches to L2 learning and acquisition throughout the 1980s and early to mid-1990s, SLA continued to be dominated by an essentially Chomskian mind-set that placed enormous stress on individual cognition to such an extent that, to some, alternative conceptions of SLA were perplexing and just plain wrong. However, on the basis of subsequent publications that reengaged and developed our criticisms of SLA10 (e.g., Atkinson, 2002; Block, 2003; Hall, 2004; Jenkins, 2006a, 2006b; Markee, 2000; Pavlenko, 2002; Rampton, 1997b; Seeley & Carter, 2004; Thorne, 2000), it seems reasonable to conclude that our article articulated a widely held sense of dissatisfaction and unease that was building at the time both within and around the periphery of SLA, an unease that provided impetus for our own call for a reconceptualization of SLA theories, concepts, and methods.11 In 1997, we maintained the position that SLA is part of the nexus of approaches to the wider, interdisciplinary study of languagewhich centrally includes language

804 learningand has the potential to make significant contributions to a wide range of research issues conventionally seen to reside outside its boundariesa potential that is, arguably, yet to be realized. What we called for was an epistemological and methodological broadening of SLA. This broadening does not entail jettisoning the mainstream SLA position, as some have mistakenly interpreted our arguments (e.g., Gass, 1998; Long, 1998; Poulisse, 1997).12 Although we do not subscribe to SLAs fundamental tenets (e.g., an assumption of the natural ascendancy of the native speaker), and we seriously question (a) the validity of such standard SLA dichotomies as acquisition versus use, and language versus communication; (b) problematize the apparently clear-cut separation of the cognitive and the social ; and (c) reject SLAs essentially static view of context and identity, in Firth and Wagner (1997), we nevertheless eschewed dogmatic positioning and pressed instead for the need for greater theoretical, conceptual, and methodological balance within SLA. For let us acknowledge that SLA research has, over the four decades or so of its existence as a discipline, uncovered a wide range of critically important ndings relating to how languages are acquired or learned. In Firth and Wagner (1997), we emphasized the need for a theoretical, methodological, and epistemological broadening of SLA, which included enlarging the standard SLA database to one that reects more accurately the sociolinguistic reality of a vast number of L2 users/learners around the world. We sought an SLA that was more interactionally sensitive, that also made room for an emic stance towards fundamental concepts, and that took seriously the theoretical and methodological consequences of a social view of learning and language. We did so in the belief that the existence of distinct and multiple theoretical traditions may help to explicate the processes of SLA, and subsequently, to develop more accurate heuristics which model these processes and conditions (Thorne, 2000, p. 221). Without specifying in any detailed way what the reconceptualization would entail in methodological terms (this was not, after all, the purpose of our article in 1997), we stressed the need to work towards the evolution of a holistic, bio-social SLA (p. 296). RECONCEPTUALIZATIONS: NEW DIRECTIONS IN SLA How, then, do we assess the eld of SLA, 10 years later? Has the reconceptualization we, and others, called for come about? The most accurate answer to this question is likely to be:

The Modern Language Journal 91 (2007) It depends where you look. In many ways it appears that things are more or less as they were in 1997the mainstream is in full ow (for recent critical expositions, see Block, 2003; Cook, 1999; Jenkins, 2006a, 2006b; Kramsch, 2006; Seeley & Carter, 2004) and the native speaker continues to predominate as the baseline or target that learners should seek to emulate; learning is conceived as a cognitive process that is in essence contextneutral; competence is dened largely in terms of the individuals grammatical competence; etic prevails over emic; and learners in classrooms remains the standard data set. All the while, these learners are viewed as essentially engaged in a continuous, autonomous, cognitive, morphosyntactic struggle to traverse, in linear fashion, along the plane of their interlanguage in pursuit of the target (i.e., native speaker) competence (see e.g., Han, 2003). Nevertheless, our (and others) urgings clearly did not go unheeded. Much SLA research that has been produced over the last decade bears witness to a marked increase in the number of sociocultural and contextual-interactional themes and concepts impacting upon SLAs research agenda, revealing an apparent growing awareness of the need to take seriously the requirement for a more balanced approach to SLA research. It appears that SLA has, over the last decade in particular, undergone a bifurcation between a cognitive SLA (which is being termed mainstream in a number of recent publications)represented perhaps most clearly in work undertaken by, for example, Doughty and Long (2003), who see SLA as a branch of cognitive science (p. 4)and a sociocultural/sociointeractional SLA. An increasing number of researchers are thus displaying a willingness to adopt emic perspectives and explore and attempt to develop cognitive-social approaches to language learning. New sets of metaphors are being deployed, such that allusions to dynamism, interaction, intricacy, and the liminal are nowadays competing with the established SLA metaphors of machinery and computation (e.g., input, output, processing). Kramsch (2002), for example, proposed the development of an ecological approach to SLA and language learning, one that centrally acknowledges the nonlinear, interactional, and contextual characteristics of language use and acquisition. Borrowing from the latest thinking in the natural sciences, Larsen-Freeman (1997, 2007) used the terms chaos and complexity in her attempt to capture more accurately the fact that L2 learning is dynamic, complex, nonlinear, unpredictable, sensitive to initial conditions, sometimes chaotic, open, self-organizing, feedback sensitive, adaptive (Larsen-Freeman, 1997, p. 35). Block (2007)

Alan Firth and Johannes Wagner used the term multimodal to capture the multiplicity of resources (historical, cultural, spatial, temporal, linguistic, etc.) brought to bear upon the processes of language learning and as a framework for more socially oriented SLA analyses. Compared with mainstream cognitive SLA, such work centrally locates and thematizes the contingent, the contextual, the ad hoc-ness, the interrelatedness of linguistic and situational elements, and the unsystematicity inherent in processes underlying L2 learningfactors that have been at worst overlooked and at best downgraded in importance in more traditional SLA research. Research of this chaotic, social, ecological kind is being recognized as representing new directions in the SLA eld (see Cook, 2002; Kramsch 2002; Larsen-Freeman, 2003; Ortega, 2005). Markee and Kasper (2004) thus talked of a recent split between mainstream, cognitive SLA and emergent, sociocultural approaches to SLA (p. 491). This latter type, Markee and Kasper claimed, points the way to future developments in what may prove to be one of the most radical respecications of SLA researchers theoretical priorities and methodological practices in the history of the eld (p. 492). Three topics appear to play a key role in the development of work in this sociocultural, sociointeractional area: (a) whether the research incorporates educational (i.e., classroom) settings or other (e.g., workplace) settings; (b) whether the focus is on language learning or interaction; and (c) whether the research is theory driven or datadriven. Lantolf, with several collaborators (e.g., Lantolf, 2000; Lantolf & Appel, 1994; Lantolf & Thorne, 2006; Pavlenko & Lantolf, 2000), has engaged in SLA theory-building in the Vygotskyan tradition, producing what has become known as the sociocultural approach to SLA.13 At the core of Vygotskyan learning theory, new developmental stages are rst accomplished with the help of others in the social sphere and can then become intrapsychological accomplishments (see Halls, 1997, useful synopsis). The pathway to learning is thus through social practice. The social practice in which learning has been studied thus far has been in L2 classrooms. In sociocultural theory:
Communication, including the instructional conversation of the classroom . . . and the learningdevelopment that emerges from it, arise in the coming-together of people with identities (which entail more than simply whether one is a native speaker), histories and linguistic resources constructed in those histories. (Dunn & Lantolf, 1998, p. 427)

805 room interaction and Vygotskyan ideas of learning are, inter alios, Hall (2002), Hall and Verplaetse (2000), and Ohta (2001). Another distinct research direction is also emerging where the emphasis is on the social, contextual, and interactional. In this case, the work deploys CA methodology. Scholarship that draws together CA and SLA is taking both SLA and CA into new, and largely uncharted, territory, and providing the impetus for a whole new generation of empirically grounded research into how cognitive SLA might be respecied in sociocultural terms14 (Markee & Kasper, 2004, p. 491).15 This work can usefully be divided into two groups. The rst group focuses on the classroom setting and other formal learning environments and is centrally concerned with the theme of L2 learningthough from an interactional perspective (see He, 2004; Hellermann, 2006; Kasper, 2004; Koshik, 2005; Lazaraton, 2002, 2004; Markee, 2000; Markee & Kasper, 2004; Mondada & Pekarek-Doehler, 2004; Mori, 2004a, 2004b; Seedhouse, 2004). A somewhat different emphasis can be found in the CA-based SLA work undertaken by Brouwer (2003, 2004) and Brouwer and Wagner (2004), who explored aspects of L2 learning occurring outside formal educational settings. A second group to have emerged over the last decade also deploys CA methodology and theory. In this case, the focus is not so much on L2 learning, but more on trying to understand and explicate the character of L2 and lingua franca interactions, or L2 use. This work has been undertaken in settings outside the classroomfor example, in government ofces, libraries, and various workplace settings (e.g., Firth, 1996; and contributions to Gardner & Wagner, 2004; Kurhila, 2006; Rasmussen & Wagner, 2002; Schegloff, 2000; Svennevig, 2003, 2004; Wong, 2000a, 2000b). As Wagner & Gardner (2004) explained in the introductory chapter to their Second Language Conversations collection, this research is an ongoing investigation of
whether a micro-analysis of second language conversations can enhance our understanding of what it means to talk in another language, by broadening the focus beyond the sounds, structures and meanings of language to encompass action sequences, timing and interactivity. (p. 14)

With a greater emphasis on data-driven research but still working mainly from the basis of class-

In most of these studieswith Wong (2000a, 2000b) as a notable exceptionexplicating the normality of L2 conversations, in the face of sometimes abnormal (from a native-speaker perspective, at least) linguistic behavior, has underpinned the research, revealing, for example, that due to participants linguistic and interactional alacrity

806 and resourcefulness, interactional success can be achieved (see also Carroll, 2000; Egbert, 2002; contributions to Richards & Seedhouse, 2004; and Seedhouse, 2005). SOCIAL APPROACHES TO L2 LANGUAGE LEARNING Challenging the distinction between language acquisition and language use was not a core issue in Firth and Wagner (1997), although we did address it, albeit briey, in Firth and Wagner (1998). Although our critics pointed out in unison that SLA is about acquisition and not use, we question the utility and theoretical as well as methodological validity of the dichotomy, and see it as being yet another symptom of the cognitivist, reductionist mind-set that prevails in mainstream SLA. To us, acquisition cannot and will not occur without use. Language acquisition, we would argue, is built on language use. Moreover, in order to understand how language acquisition occurs, develops, and is operationalized, we are surely obligated to observe and explicate language in use. At best, making and maintaining a distinction between acquisition and use is highly problematic; at worst, both concepts are so tightly interwoven as to be rendered effectively inseparable. As we noted in Firth and Wagner (1998):
If, as we argue, language competence is a fundamentally transitional, situational, and dynamic entity, then any language users will always be learners [or acquirers] in some respects. New or partly known registers, styles, language-related tasks, lexical items, terminologies, and structures routinely confront language users, calling for the contingent adaptation and transformation of existing knowledge and competence, and the acquisition of new knowledge. (p. 91)

The Modern Language Journal 91 (2007) dened itself as a discipline that produces knowledge about this special phenomenon called acquisition; (b) it cemented its identity as a distinct discipline and secured a foothold in the world of scientic research; and (c) it all but cut off possible links to learning theories residing outside its own (self-constructed) disciplinary boundaries (Rampton, 1997a). This terminological transformation seems to have been critical for SLA. However, if L2 learning is understood differentlyspecically as principally the same type of process as other types of human learningthen SLA is open to extant and potentially insightful learning theories and approaches, and thus placed under immense pressure to reconceptualize, expand, enlarge, or recalibrate its epistemologypressure we see exerted by, for example, Atkinson (2002), Block (2003), Firth and Wagner (1997), Hall (1997, 2004), Jenkins (2006a), (2006b), Lantolf (2000), Liddicoat (1997), Markee and Kasper (2004), Rampton (1997a, 1997b), and Thorne (2000). Admitting different conceptions of learning allows, of course, for nonpsychological theories to become relevant for understanding and explicating L2 learning.16 By way of exemplication, we can identify three potentially relevant avenues of such research. One is sociocultural theory (e.g., Hall, Vitanova, & Marchenova, 2004; Lantolf, 2000), which we have outlined above. A second is constructivism, which is well established outside SLA. A third is a social-interactional approach to learning, which is only just beginning to take root. We will now elaborate briey on these latter two. Constructivism Constructivism (which includes social constructivism, radical constructivism, and cognitive constructivism) is a theory of knowledge acquisition that sees learners constructing their own knowledge and meanings on the basis of personal experiences. Constructivist ideas are readily traced in the work of Dewey (1916, 1980, 1938) and Kant (1781/1946). Drawing on von Glasersfelds work (e.g., von Glasersfeld, 1984, 1995), Doolittle (1999) summarized constructivism as containing:
three essential epistemological tenets of constructivism to which a fourth has been added in light of recent writings: 1. Knowledge is not passively accumulated, but rather, is the result of active cognizing by the individual;

What this acquisitionuse dispute reveals is arguably the most signicant and consequential disagreement between Firth and Wagner (1997) and proponents of a mainstream SLA, namely, the conception of learning. In this regard, some of the implications of our 1997 article were surprising for us as well, for we failed to realize at the time just how fundamentally our views challenged SLAs conception of learning. Indeed, the Chomskian heritage in SLA may be most conspicuous with regard to SLAs concept of learning, for as long as language learning is envisioned as a specic type of learning relating solely to the individuals headthe human language faculty traditional SLA is conceptually safe. In fact, in the early 1970s, when the umbrella term learning was replaced by the technical term acquisition, SLA accomplished three things rather elegantly: (a) it

Alan Firth and Johannes Wagner

2. Cognition is an adaptive process that functions to make an individuals behavior more viable given a particular environment; 3. Cognition organizes and makes sense of ones experience, and is not a process to render an accurate representation of reality; and 4. Knowing has roots in both biological/neurological construction, and social, cultural, and language based interactions (Dewey, 1916, 1980). (p. 1)

807 forms; rather they engage in complex, multimodal, nely tuned co-participation, integrating body posture, gaze, verbal and prosodic activities, rhythm, and pace in their choreography of action (Egbert, Niebecker, & Rezzara, 2004). Research adopting this approach may have the potential to reconceptualize notions of learning, with important consequences for SLA (c.f. Mondada, 2006). SLAs adoption of such theories and approaches would, we submit, have major implications for SLA as a eld of inquiry, as well as for L2 pedagogy, and only hints at possible developments that might have occurred in SLA, modern language teaching, and learning, if behaviorism and, later, information-processing psychology had not been allowed to become established as having primary importance in SLA. LANGUAGE LEARNING AS A SOCIAL ACCOMPLISHMENT How, then, might the analysis and explication of L2 learning be approached from a socialinteractional perspective, one that places contingency, contextuality, dialogue, and liminality at its core? We shall attempt to answer this question by offering cursory analyses of language learning-inaction. Our focus here is not learners in a formal sense of the wordthat is, persons engaged in purposive activities principally in order to develop skills, knowledge, and competences. Nor will we deal with classroom interaction (recent studies of classroom interaction include Hall, 2006; Hall, 2007; Kasper, 2004; Mondada & Pekarek-Doehler, 2004; Mori, 2004a, 2004b; and Seedhouse, 2004). Rather we are concerned to uncover learning as a ubiquitous social activity, as an interactional phenomenon that transcends contexts while being context dependent; as an instance of social cognition in the wild (see Hutchins, 1995), specifically in relation to encounters where an L2 is in use. For although learning may or may not be a drawn-out process, it is certainly a process that takes place in the micromoments of social interaction in communities of practice. It is therefore critically important that we attempt to uncover and understand what goes on, interactionally, in such micromoments. What we want to explicate, albeit extremely cursorily, are aspects of social learning as they are evident in L2 talk-based activities, in each case occurring outside the classroom. Throughout, we emphasize not the failings or deciencies, not the errors or interlanguage of the interactants; rather, we approach the data with a view to explicating and uncovering what L2 users actually do,

Social-Interactional Approach to Learning This approach is in the very earliest stages of development. It is an approach that focalizes learning-in-and-through-social-interaction, though devoid of cognitivist underpinnings. In its current manifestations, it combines insights from three areas: (a) the approach to learning promulgated by Lave and Wenger (1991) and Wenger (1998), who stress a social view of learning, and see learning as occurring in multiplex communities of practice; (b) ethnomethodology, which emphasizes the centrality of dynamically and contingently deployed commonsense reasoning practices in everyday settings (Garnkel, 1967; Heritage, 1984); and (c) CA, which stresses the necessity to attend to and uncover the socially achieved, microstructuring of human activity (Goodwin & Heritage, 1990; Hutchby & Wooftt, 1998; ten Have, 1999). The social-interactional approach can be characterized as follows: Learning is an inseparable part of ongoing activities and therefore situated in social practice and social interaction. In this sense, learning builds on joint actions and as part of a joint action it is publicly displayed and accomplished. Building on the work of Lave and Wenger (1991), Brouwer and Wagner (2004), in their analyses of L2 conversations, proposed such an approach to learningone that shifts the focus from the individuals cognition and grammar, to social praxis in concrete social settings: Instead of studying the acquisition of a grammatical system, we propose to study the texture of communities and the ways in which newcomers get access to them (p. 45). Wagner and PekarekDoehler (2006) demonstrated how participants proceed through their activities by a process of what they call bricolagethat is, complex segments of social activities where interactants collaborate to make sense of their conjoint actions through the situated, contingent deployment of communicative resources. As has been shown in the contributions to Gardner and Wagner (2004), L2 speakers are not interactional dopes. They show perseverance and ingenuity in interacting with others. They produce not so much language

808 in everyday, natural settings. And as we shall see, they deploy, make available, share, adapt, manipulate, and contingently and creatively apply communicative resources in an ongoing attempt to construct meaningful and consequential social interaction. The data excerpts we present and comment on here represent a small window on what might be possible if more in-depth, extensive studies of the same kind are undertaken. An Evidential Argument Consider rst Excerpt 1, where Hansen (H), the Danish dairy-produce salesman, is talking on the telephone with Akkad (A), an Egyptian wholesaler (see the Appendix for transcript conventions): EXCERPT 1 Blowing in the Customs (Firth, 1996, p. 244) 1. A: .. so I told him not to u: :h send the:: cheese after the- () the blowing () in the customs (0.4) A: we dont want the order after the cheese is u: :h () blowing. H: see, yes. A: so I dont know what we can uh do with the order now. () What do you think we should uh do with this is all blo:wing Mister Hansen (0.5) H: Im not uh (0.7) blowing uh what uh, what is this u: :h too big or what? (0.2) A: no the cheese is bad Mister Hansen (0.4) A: it is like () fermenting in the customs cool rooms H: ah its gone off A: yes its gone off H: we: :ll you know you dont have to uh do uh anything because its not ((continues))

The Modern Language Journal 91 (2007) cating uptake, as displaying that he (A) has been understood. Upon hearing Hs uptake at line 4, and now acting on the presumption of a shared common ground, A, in line 6, asks H what do you think we should do with all this blowing? (underlining indicates stress in enunciation). At this juncture, H is evidently compelled to display to make publichis unfamiliarity with the word blowing ; thus Hs Im not uh (0.7) blowing uh what uh, what is this u: :h too big or what? As response is to explain the meaning of blowing; he does this by reformulating: no the cheese is bad, Mr. Hansen (line 10). With no receipt forthcoming (note the 0.4 second pause in line 11), A tries again, in line 12: it is like fermenting in the customs cool rooms. H now displays his updated knowledge by producing his own reformulation (with rising intonation) of blowing , in his ah its gone off? (line 13). A, in the next turn, conrms the appropriateness of the denition of blowing by reusing Hs formulation, hence yes its gone off (line 14). What this excerpt reveals is that the interactants, conjointly, do interactional work to overcome potential or real communicative hurdles in order to establish intersubjectivity and meaning. In so doing, the interactants provide for the availability and utility of interactional and linguistic resources that allow for learning to occur. H has, in this excerpt, quite clearly displayed that he has learned the (for him) new lexical item blowing . What is important is that the excerpt also shows how learning is engaged contingently, as the learned item in question becomes interactionally relevant as the talk unfolds. Learning here, then, is an artifact of interactional exigency and a product of collaboration. Let us now examine how H operationalizes his learning (of what blowing cheese means) in a separate encounter, that is, how his learning is carried over in time and space. Two working days later, H, in Denmark, calls As wholesale company in Cairo. The call is answered by As colleague, B. The call begins as follows: EXCERPT 2 Bad Cheese (Firth, 1991) ((ring)) 1. 2. B: allo H: yes hello Michael Hansen melko dairies Denmark calling () can I please speak to mister Akkad () B: hello mister Michael

2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15.

When this excerpt was initially analyzed (in Firth, 1996, p. 244), the focus of attention was the let it pass procedure and its interactional consequences.17 But we can also use the excerpt to witness an instance of learning-in-interaction. To see this, note, rst, Hs I see, yes (line 4). This turn follows As revelation in lines 13 that the cheese is blowing in the customs (line 1). In line 6, A displays that he has heard Hs I see, yes as indi-

3. 4.

Alan Firth and Johannes Wagner 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. H: is it Barat? B: ye: (h)s, how are you () si::r H: well Im OK, but you ha:d tu- have some uh problems with the: cheese B: uuuuuuhhhhh ((one-second sound stretch)) H: the bad cheese () in the customs (0.5) B: one minute (0.4) mister Akkad will talk () w[ith () you H: [ok yes (1.5) A: YES () mister Hansen H: hello: mister Akkad ().hh we haf some informations for you about the cheese () with the blowing A: yes mister Hansen

809 then, is, potentially at least, a relevant identity category for some of these interactants. In a related way, we may venture that the reason H deploys the item blowing with A, but uses instead cognate formulations with B, is because H is recipient designing (Sacks, Schegloff, & Jefferson, 1974, p. 727) his displays of knowledge so as to achieve communicative (and likely interpersonal) success and to display interactional afnity. To paraphrase Baker, Emmison, and Firth (2005), we can say that H is calibrating his language behavior for his interlocutors competence. We cannot know why H did not use the term blowing with B. Perhaps, if A had not been available or had not been able to engage H, H may subsequently have deployed the item blowing. However, his apparent contingent and selective deployment of the item offers an intriguing angle from which to view language learning. Our argument so far reminds us of the interaction hypothesis (Long, 1996), according to which learning is fueled by the problems that learners encounter. However there are several problems with this explanation for learning: 1. The approach describes a transparent problem-based way of learning, but we nd only few instances of such problem-based learning in our data. 2. It prototypically describes lexical learning. Where does the rest of the language (and communicative behavior more generally) reside? 3. Uptakes are rare. Participants in interactions obviously pick up items and use them in recipient-designed ways, but such actions are only part of the whole language learning story. Although it might partly have to do with the lack of comprehensive corpora, problem-based learning of this kind seems not to be at the core of language learning. 4. The interactive hypothesis constructs some kind of evidential argument: a plain causal chain between use of elements and learning. When we look at the developmental data we have access to, plain causal explanations do not cover many of the complexities we see unfolding in our data. We see learners developing their skills in using resources in more advanced ways for more advanced interactions, but we rarely see a smoking gun, as in Excerpts 1 and 2. There are few, if any, studies showing learning in action (Huth, 2006; Nguyen, 2003). The main obstacle for this type of research has been the lack of longitudinal corpora available for L2 interactions.

16. 17.

Note that in this excerpt it is H who uses the item blowing (line 16)which seems to suggest that not only has H now learned this lexical item in a way that extends beyond the concrete local context where he rst became acquainted with it, but also, critically, that he has learned how to deploy it appropriately in context, which in this case entails knowing whom to use the item with. Consider, rst, how H characterizes the matter while talking to B. First, H uses problems with the cheese (line 7), next, he reformulates the matter as the bad cheese in the customs (line 9), this reformulation likely being a result of an apparent lack of uptake by B (line 8). When A enters the interaction (at line 14), H rst uses the formulation the cheese (line 15), and then, without yet securing uptake from his interlocutor (note the micropause at the end of line 15), formulates the matter as with the blowing (line 16). This formulation succeeds, as we see from As yes Mr. Hansen (line 17). What are the language learning implications of such behavior? They are, rst and foremost, that if learning entails being contingently adaptable as the unfolding context requires, then here, surely, is learning-in-action. Second, here may be evidence of a community of practice (Lave & Wenger, 1991) being constructed before our very eyesnamely, between H and A. The nature of this particular community of practice appears to be consequential for how learning is operationalized, for how learning is displayed, and for how communicative knowledge is rened and tested both within and across interactions. It is arguable that H, in Excerpt 2, orients to A as a co-user, or as a co-member of this microcommunity of practice, of the term blowing . Co-usership,

810 Doing Learning Another way of describing learning does not focus on the evidence but on the process. Our position in this regard is inspired by sociocultural theory, particularly Halls 2006 work. Consider the next excerpt in which Rasmus, a Danish student, coaches Joseph, an American guest student, to learn how to order pizza in Danish. In his own words: Joseph is developing his survival skills.

The Modern Language Journal 91 (2007) core of this standard situationthe Danish sentence would work as the reason for the call, but the opening and closing activities in a phone call are not thematized by Joseph and Rasmus. The sequence is initiated and closed by the L2 speaker, which makes it different from classical teaching activities described, in non-CA research, as variations of an Initiative-Response-Evaluation sequence (Mehan, 1979). The L1 speaker Rasmus appears therefore interactionally not as a teacher but as a language expert who is transferring knowhow to a nonexpert. The participants themselves call the activity learning . Josephs (the learners) main activity is the repeat of the target utterance. We observed that the translation follows the repeat of the Danish expression that shows both participants orienting primarily to the form of the expression. Joseph learns the Danish expression before the translation has made crystal clear what he has learned (see below). What we observe in Excerpt 3 holds for the whole interaction. Joseph provides a prompt, Rasmus delivers a Danish expression that Joseph repeats. However, not all sequences run as smoothly as the one shown in Excerpt 3. In later talk, which we cannot show here due to constraints of space, we observe Rasmus slowing down his speech, isolating troublesome elements in the utterance by prosodic means, and even using the written mode to display a model. The resources for learning used in this interaction are not very different from those of traditional classroom teaching, which include (a) utterance model and repetition, (b) form before content/translation, (c) prosodic formation (speed, rhythm), and (d) written models. In Excerpt 3, the participants show us that they recognize and characterize their activity as learning. They construct conversational routines and patterns of interaction that are typical of language classrooms (Kasper, 2004; Mondada & Pekarek-Doehler, 2004). Although Rasmus and Joseph draw heavily on a classroom format, other studies show how participants in interactions orient in more subtle ways to language learning as one goal in the ongoing activity. Lilja (2006) described instances in interactions between a Finnish host family and guest students where the native speakers in repair sequences do not just repeat trouble sources but strip them of morphological inection and present them in an unmarked form. Theodorsdottr (2007) showed how learners may insist on producing full turns although their coparticipants have indicated their understanding of the ongoing turn before its end. Brouwer

EXCERPT 3 Danish Pizza (Wagner, unpublished raw data) 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. Jo: Ra: Ra: so how d- how do I say say it one time [for] me. [.pt] (0.2) je:g vil gerne bestille en pizza. I would like order a pizza (0.3) jeg vil gerne bestille en pizza. (0.2) jep . (0.5) thats Id like to order pizza. .pt okay,

Jo: Ra: Ra: Jo:

In line 1, Joseph opens a new sequence with so and gives two consecutive prompts for next action: how do I say and say it one time for me. Rasmus provides the appropriate Danish expression (line 4), which Joseph repeats (line 6). Rasmus acknowledges the repeat in line 8 with yep and provides an English translation (line 10). In line 11, Joseph closes the sequence. The sequence runs off at a slow pace without glitches, delays, overlaps, or repairs. Both participants demonstrate, through their actions, that they understand who will produce the next element and which element has been projected. It seems to be clear for both participants what kinds of actions they are engaged in and how the subactivities are distributed. To elaboratethe Danish expression is decontextualized. The I does not refer to the current speaker (Rasmus) but is a generic I for any speaker in a pizza-ordering situation. In this way, Rasmus provides a model as a standard solution for a well-dened activity in a well-dened environment (ordering a pizza by phone). Both Joseph and Rasmus show their knowledge of the script for this activity and treat it as a standard situation where preformulated expressions can be deployed. They refer only to the

Alan Firth and Johannes Wagner and Rasmussen-Hougaard (2007) described how participants in L2 interactions carve out opportunities for doing learning in the ongoing ow of interaction. Compared to the case of Hansen and Akkad, instances of doing learning do not afford an evidential argument for learning. Instances of doing learning show that learning a language and learning to learn are mutually constitutive, as Hall (2006) formulated it for the language classroom:
What our students take away from our classrooms in terms of their target language knowledge and understandings of what it means to learn another language is intimately tied to the kinds of interactional practices that we as teachers create in our talk with students . . . . Through their interactions with us, learners become experienced at guring out the actions that are being implemented in our utterances including how the utterances are constructed and eventually learn to use them to take actions of their own. Our interactional practices are, then, to use Vygotskys term, the mediational means by which we and our students together constitute, represent, and remember what it means to know and do language learning. (p. 27)

811 cess of the company where JG works. This is the second call between BR and JG. EXCERPT 4 Lot of Work 1 (Wagner, unpublished raw data) 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. BR: okay have you a lot of work to do. JG: yes, JG: [we ha a lot a do. BR: [xx BR: a lot of orders. JG: and we e::hm: work also in the weekends BR: ah [(xx the weekend)] JG: [for the next ] e:h fh three four monthth () [I think, BR: [yes (0.8) JG: [so ] BR: [thats] thats good to hear. (0.3) JG: yes, BR: okay. JG: and your company has also. () lot to do, (0.3) BR: oh yes. BR: things are going very well. = JG: yes

Learning is about the object and the ways of learning. Joseph and Rasmus illustrate that they know how to set up a learning situation. They enact the classical activities of language teaching. However, the other studies mentioned here trace ways of learning not only on the genre level but also as practices in ongoing interaction where participants show each other that they are, among other things in the real world, engaged in more than intersubjectivity. They are not content to make themselves understood but clearly demonstrate a desire to do so in ways that are viewed as appropriate and normal in their L2. Other settings, with other kinds of interactional and institutional goals and relationships, will doubtless produce different kinds of talk- and learning-based activities. Doing learning illustrates how participants foreground learning in an interactionally consequential way. This type of learning is different from that of Hansen and Akkad where learning a new item was entirely embedded in the business interaction. But doing learning does not provide evidence that learning is actually happening in these activities. Out of Chaos Comes Order Consider now Excerpt 4, again from a business interaction. Jrgen Gade (JG) in Denmark has been called by a Swiss business acquaintance (BR). At the end of the call, BR enquires about the suc-

BRs question in line 1, have you a lot of work to do is in line 16 mirrored by JGs answer we ha a lot to do (line 3) as well as in his formulation in line 16 and your company has also a lot to do. Regular telephone engagement with known others frequently gives rise to the production of How are you sequences (compare Excerpts 2, 6). As in Excerpt 2, we notice that these how are you sequences are not taken personally but relate instead to the companies in which the protagonists work. Consider now Excerpt 5. EXCERPT 5 Lot of Work 2 1. 2. 3. 4. JG: TT: JG: TT: have you: lot of work to do. yes theyre not too bad [yes] [we-] e-e- picked up another order =eight hundred thousand pound[s [ouw: (0.6) excellent ehhe hhe hhe hhe hhe hhe he

5. 6. 7.

JG: JG:

812 8. 9. 10. 11. 12. 13. 14. 15. 16. TT: JG: TT: TT: G: TT: JG: JG: TT: eight million crowner hhe hh o:ijoijoi. (sxt xx) hh hh hhh so thats e:h thats good yeah we still ned some more though but [it loo]ks looks if its u:h = [mmh.] yes = getting better.

The Modern Language Journal 91 (2007) is, language development in the calls appears to be nonlinear, and the growing complexities in the participants actions are not traceable to simpler preceding forms. This observation is in line with Eskildsens study (Eskildsen, in press; Eskildsen & Cadierno, 2007), which drew on data from a Mexican learner of English over a period of 4 years (20012005). Eskildsen (in press) looked at the development of modal verbs over time and the conclusions he was able to draw on the basis of the data materials do not show a development from a core structure into more complex structures, but rather a patchwork of different uses that appear and disappear over time. In terms of linguistic structure, this study does not show much structure building. We can tentatively conclude that we have been able to point to three aspects of language learning as a social accomplishment. These three aspects, and possibly others, demonstrate in a variety of ways how participants engage in meaningful activities by using an L2. In situated social practices, use and learning are inseparable parts of the interaction. They appear to be afforded by topics and tasks, and they seem to be related to specic people, with particularized identities, with whom new ways of behaving occur as the unfolding talk demands. Studying learning as a social accomplishment shifts our understanding of learning from the construct of a linguistic system or a competence that serves all the speakers purposes. Instead, the development of social relations, the mutual constituency of linguistic resources and tasks, and the specic biography of the language learners come to the foreground. This strand of research has gained momentum over the last 10 years, and quite clearly, much more research into the specics of social interactions in L2 environments is clearly necessary in the years to come.

In Excerpt 5, JG is on the phone with a British colleague with whom he has had regular contact over many years. JGs question in line 1 have you: lot of work to do sounds like an echo of BRs question from Excerpt 4. However, in this case, 2 years have passed between the call in Excerpt 4 and Excerpt 5. It would be a bold move to construe (line 1) as evidence that learning had occurred in the call 2 years earlier. Nevertheless, the resemblance of both formulations over the span of those 2 years is striking. JG has not become more linguistically sophisticated in his inquiry into his interlocutors work commitments. He simply redeploys the utterance formula that he had deployed 2 years previously. We notice comparable development standstill in other recurrent call segments or activities, for example, in call openings, closings, the introduction of new topics, how-are-yous and salutations. In these cases, it appears that comparatively little linguistic development occurs over time. If there is any developmental change at all, we nd it in general interactional uency, such that these activities are produced with fewer delays, smoother turn transitions, and a general increased orderliness; in addition, the content of the calls becomes more personalized over time as the interactants become better acquainted (cf. Brouwer & Wagner, 2004). These observations beg the question: In which environments might we notice language learning as a social accomplishment? If it is not observable in routinized activities and talk episodes, where can it be located? What are the environments where linguistic structures expand and linguistic creativity and development can be observed? They appear to occur in what we might call the body of the calls, that is, in phases following openings and preceding closing routines. Here the talk is more free owing, less planned and less routinized. New expressions and new structures are afforded by new topics and activities. The contingencies of the locally unfolding situation seem to motivate the use of new resources; that

CONCLUSIONS What this entire debate has brought home most vividly is that SLA as a eld of inquiry is today a fertile arena of multifarious approaches, theories, methods, concepts, and, not least, debate. Whether SLA has a mainstream or a dominant approach is a moot point. Cognitivistic SLAthe mainstream that we focused our critique upon in Firth and Wagner (1997)is certainly well established and continues to occupy a prominent place in SLA-related journals, textbooks, doctoral and graduate programs, and publications generally. In

Alan Firth and Johannes Wagner most of its guises, cognitivistic SLA is strikingly different from the SLA of sociocultural theory, the ecological approach to SLA, and the CA-for-SLA movement that we have adumbrated in this article. Thus, although some researchers are exploring the possibilities of developing a holistic, encompassing SLAone that seeks to draw together the social and the cognitiveother researchers position themselves apart from the social, cultural, situational, chaotic, sociolinguistic, ecological, interactional drives, and motives underpinning the work of a growing number of SLA scholars. These differences raise the inevitable question of where SLA is headed, and whether the eld itself is able to withstand the current bifurcations, competing methods, critiques, and internal tensions, and remain generally cohesivein the way that the eld of sociolinguistics, for example, has remained more or less cohesive, despite the discipline being populated with an almost dizzying array of sometimes incommensurable methods, concepts, and theoriesor whether SLA will fracture into cognitive SLA, holistic SLA, sociocultural SLA, conversation-analytic SLA, postmodern SLA, and so on. It is arguable, of course, that such a fracturing either has already taken place or is currently underway. If this is indeed what is happening, or what has occurred, a major issue then becomes one of how the eld or the discipline denes itself. SLA is a relative newcomer to scientic inquiry, and there are inherent risks in allowing such a new eld of research to shift, morph, and fracture, particularly from the viewpoint of those who, through a lengthy professional devotion to the eld or paradigm, conceive of themselves as the intellectual guardians of (in this case) SLA, and see it as their right and obligation to determine, ex cathedra, what is and is not proper SLA. The debates, the arguments, the progress, or the decay is surely an inevitable component of SLAs evolution. We are, then, witness to a natural progression, an intellectual evolution, if one will, where successful paradigms evolve (and sometimes fracture) through both support and critique. If this process is based on sound, creative scholarship, one that leads to advances in knowledge of the many and varied ways in which L2s are learned, acquired, and used (in mutually reinforcing and enlightening ways), then surely SLA will become a more theoretically and methodologically robust and encompassing enterprise. Despite objections from some quarters, the boundaries of SLA are ineluctably being redrawn, and from this particular viewpoint, the future of SLA looks distinctly promising.
ACKNOWLEDGMENTS

813

We are grateful to the following colleagues who generously and insightfully commented on previous versions of this article: Dennis Day, Rod Gardner, Jennifer Jenkins, Chris Jenks, Barbara Seidlhofer, Barbara Lafford, Sally Sieloff Magnan, and three anonymous referees, each of whom provided useful and detailed comments. In most cases, we have incorporated suggestions for improvements into the nal version. Remaining weaknesses and oversights remain our responsibility.

NOTES
1 Conversation analysis (widely known as CA) is a methodology devised by Sacks, Schegloff, and Jefferson in the 1960s (see Sacks, 1992). Building on its ethnomethodological (Garnkel, 1967; Heritage, 1984) foundations, CA endeavors to explicate the microdetails of talk-in-interaction and to uncover the communicative and social competences that structure and render meaningful talk-in-interaction. The materials of CA are video- and audiorecordings of naturally occurring settings where talk is a prime facet of behavior. For descriptions of CAs working methods and theoretical foundations, see ten Have (1999) and Hutchby and Wooftt (1998). 2 Although such observations on our data informed our call for a reconceptualized SLA in Firth and Wagner (1997), they did not feature empirically in that publication. 3 We do not have the space to elaborate the point here, but in essence the argument is that, in the case of English, which is undoubtedly the global lingua franca in an array of domains (e.g., the Internet, diplomacy, science, pop music, tourism), equating target language competence with native speaker competence is inherently problematic, in that such a practice ignores the sociolinguistic reality of the global status and lingua franca uses and functions of English. The implications for SLA are potentially far-reaching, not least in terms of our understanding of interlanguage. This point is well made by Jenkins (2006a, 2006b). 4 The telementational view of communication is the view promulgated by, among others, Saussure (1922), adopted by Chomsky (1957) and, later, by mainstream SLA practitioners. It underpins SLA work in communication strategies (see Firth & Wagner, 1997). According to Harris (1981), the telementational view is a fallacy; it is a thesis about the function of language, namely, that linguistic knowledge is essentially a matter of knowing which words stand for which ideas. For words, according to this view, are symbols devised by man for transferring thoughts from one mind to another. Speech is a form of telementation (p. 9). 5 This monolinguistic notion, we argued (Firth & Wagner, 1997), underpins the prevailing SLA view that sees language users as non-native speakers who are (or ought

814
to be) aiming for a target consisting of the baseline linguistic norms of the native speaker (see, e.g., Konishi & Tarone, 2004). 6 Those who supported our arguments (in their response articles) were Hall (1997), Rampton (1997a), and Liddicoat (1997). 7 To be clear on how we stood in relation to acquisition and use, in Firth and Wagner (1997) we pressed the need for an SLA that can explain how language is used as it is being acquired through interaction, and used resourcefully, contingently, and contextually (p. 296). 8 Kaspers (1997) views were prescient, but note that in Firth and Wagner (1997) we refrained from recommending that CA become part of SLA, or that CAmethodologies be incorporated into SLA. 9 Classrooms are, of course, workplaces, though for the sake of exposition we shall maintain the rudimentary distinction to differentiate classrooms from other social settings in which some kind of occupational or professional activity is taking place. 10 Thorne (2000) is representative of the views held broadly by critics of SLA who, like Firth and Wagner (1997), pressed the need for a reconceptualized SLA along social (to this can be added historical) lines. Thorne wrote: the dominant core of current theories of SLA are for the most part dening a world of a-historical, decontextualized, and disembodied brains. It is my belief that such a theory does not t the evidence (p. 220). 11 Atkinson (2002) pointed out that two prominent SLA researchers, Ellis (1997) and Crookes (1997), seem to have realized the reductionism inherent in [SLA research] (note 11). Ellis (1997) wrote SLA in general has paid little attention to the social context of L2 acquisition, particularly where context is viewed non-deterministically (i.e., as something learners construct for themselves). SLA has been essentially a psycholinguistic enterprise, dominated by the computational metaphor of acquisition (p. 87). Crookes (1997) offered similar sentiments: though cognitive psychology was to be preferred [in SLA] to its dominant predecessor [behaviorism] because it was (purportedly at least) about people (rather than rats), it was a long time before I began to understand that it, like its predecessor, could be seen as a sociocultural construct . . . that reected at least to some extent the presuppositions of the societies in which it developed. That was why it was fundamentally an individualist psychology that treats people as isolates (p. 98). 12 Long (1998), for example, wrote: Instead of dismissing all past work as narrow and awed, and simply asserting that SLA researchers should therefore change their data base to take new elements into account, [critics] should offer at least some evidence that, e.g., a richer understanding of alternate social identities of people currently treated as learners, or a broader view of social context, makes a difference, and a difference not just to the way this to that tiny stretch of discourse is interpretable, but to our understanding of acquisition (p. 92).
13

The Modern Language Journal 91 (2007)

Space limitations preclude a detailed description of the sociocultural approach; see, however, Lantolf (2000); Lantolf and Appel (1994); Lantolf and Thorne (2006); and Pavlenko and Lantolf (2000). Lantolf (2000) described its essence as follows: The most fundamental concept of sociocultural theory is that the human mind is mediated. In opposition to the orthodox view of mind, Vygotsky argued that just as humans do not act directly on the physical world but rely, instead, on tools and labor activity, which allows us to change the world, and with it, the circumstances under which we live in the world, we also use symbolic tools [e.g., language], or signs, to mediate and regulate our relationships with others and with ourselves and thus change the nature of these relationships . . . . The task of psychology, in Vygotskys view, is to understand how human social and mental activity is organized through culturally constructed affairs (p. 1). 14 We question the appropriateness of Markee and Kaspers (2004) characterization of this CA-centered SLA work as sociocultural . The sociocultural approach has already established itself within SLA and is associated with the work of Lantolf (e.g., Lantolf, 2000) and others. According to Hall (1997) and Lantolf (2000), sociocultural SLA research is principally underpinned by Vygotskian theories of mind and learning (e. g. Vygotsky, 1978). CA, however, has its roots quite rmly in Garnkelian sociological theory (Garnkel, 1967; Heritage, 1984). 15 Deploying CAs microanalytic methodology for the analysis of L2 talk-in-interaction poses a number of challenges (which are too numerous and complex for us to reveal and discuss in this article; see Brouwer & Wagner, 2004). It certainly entails much more than familiarizing oneself with CAs terminologies and applying them to L2 data materials. Arguably the greatest challenge for SLA researchers entails coming to terms with CAs ethnomethodological research agenda (Firth, 1995c; Garnkel, 1967; Heritage, 1984), which is radically different from mainstream SLAs cognitivist agenda (see Larsen-Freeman, 2004). The incorporation of L2 data materials into a CA framework also poses challenges for CA. To mention one of several examples: CA has until relatively recently been restricted to the analysis of monolingual, rst language (L1), interactions, where, overwhelmingly, the analyst has relied upon his or her co-membership in the interactants speech community as a resource in the explication of the talk-data. When faced with the analysis of L2 talk, the analyst cannot unproblematically assume linguistic or cultural co-membership; the applicability of an important analytic resource is thereby brought into question. For further information on this topic, see Firth (1996). 16 Let us be clear, however, that the opening up of SLA to noncognitivist theories and approaches to (language) learning is not going to be a trouble-free process; see, for example, Kaspers (1997) view that [a] noncognitivist discipline that has learning as its central research object is a contradiction in terms (p. 310).

Alan Firth and Johannes Wagner

We refer the reader to Firth (1996) for a detailed explication, but in brief, Hs I see (line 5) is quite clearly a let it pass procedure. Such procedures are utilized in interaction when the recipient of a turn lets the unknown or unclear action, word or utterance pass on the (commonsense) assumption that it will either become clear or redundant as talk progresses.
17

815
Dewey, J. (1980). The need for social psychology. In J. A. Boydston (Ed.), John Dewey: The middle works, 18991924 (Vol. 10, pp. 5363). Carbondale, IL: Southern Illinois University. (Original work published 1916) Doolittle, P. E. (1999). Constructivism and online education. Retrieved 12 January, 2007 from Virginia Tech Web site: http://edpsychserver. ed.vt.edu/workshops/tohe1999/tohe2.html Doughty, C. J., & Long, M. (Eds.). (2003). The handbook of second language acquisition. Oxford: Blackwell. Dunn, W., & Lantolf, J. P. (1998). Vygotskys zone of proximal development and Krashens i+1: Incommensurable constructs, incommensurable theories. Language Learning , 48, 411 442. Egbert, M. (2002). Der Reparatur-Mechanismus in deutschen und interkulturellen Gespr chen [The rea pair mechanism in German and intercultural conversation]. Unpublished Habilitationsschrift. Oldenburg, Germany: University of Oldenburg. Egbert, M., Niebecker, L., & Rezzara, S. (2004). Inside rst and second language speakers trouble in understanding. In R. Gardner & J. Wagner (Eds.), Second language conversations (pp. 178200). London: Continuum. Ellis, R. (1997). SLA and language pedagogy: An educational perspective. Studies in Second Language Acquisition, 19 , 6992. Eskildsen, S. W. (in press). Constructing another language: (Beyond) frequency effects in second language acquisition? To appear in Applied Linguistics. Eskildsen, S. W., & Cadierno, T. (2007). Are recurring multi-word expressions really syntactic freezes? Second language acquisition from the perspective of usage-based linguistics. In M. Nenonen & S. Niemi (Eds.), Collocations and idioms 1: Papers from the First Nardic Conference on Syntactic Freezes, Joensuu, May 1920, 2007. Studies in Languages, University of Joensuu, Vol 41. Joensuu: Joensuu University Press. Firth, A. (1990). Lingua franca negotiations: Towards an interactional approach. World Englishes, 9 , 69 80. Firth, A. (1991). Discourse at work: Negotiating by telex, fax, and phone. Aalborg, Denmark: Unpublished doctoral dissertation, Aalborg University. Firth, A. (1995a). Negotiating in the Virtual Marketplace: Making sense of telenegotiating. In K. Ehlich & J. Wagner (Eds.), The discourse of business negotiation. (pp. 127149). Berlin, Germany: Mouton de Gruyter. Firth, A. (1995b). Talking for a change: Negotiating by telephone in commodity trading. In A. Firth (Ed.), The discourse of negotiation: Studies of languages in the workplace (pp. 183222). Oxford: Pergamon Press. Firth, A. (1995c). Accounts in negotiation discourse: A single-case analysis. Journal of Pragmatics, 23, 199 226.

REFERENCES Antaki, C., & Widdicombe, S. (Eds.). (1998). Identities in talk. London: Sage. Atkinson, D. (2002). Toward a sociocognitive approach to second language acquisition. Modern Language Journal , 86 , 525545. Baker, C., Emmison, M., & Firth, A. (2005). Calibrating for competence in calls for technical assistance. In C. Baker, M. Emmison, & A. Firth (Eds.), Calling for help: Language and social interaction in telephone helplines (pp. 3962). Amsterdam: Benjamins. Block, D. (1996). Not so fast: Some thoughts on theory culling, relativism, accepted ndings and the heart and soul of SLA. Applied Linguistics, 17 , 6383. Block, D. (2003). The social turn in SLA. Washington, DC: Georgetown University Press. Block, D. (2007, March). Thinking about SLL research: Conservatism, ghting, toolkits and thinking outside the box. Paper presented at the School of Education, Communication and Language Sciences, Newcastle University, Newcastle Upon Tyne, UK. Breen, M. (1985). The social context for language learningA neglected situation? Studies in Second Language Acquisition, 7 , 136158. Brouwer, C. E. (2003). Word searches in NNSNS interaction: Opportunities for language learning? Modern Language Journal , 87 , 53445. Brouwer, C. E. (2004). On doing pronunciation. In R. Gardner & J. Wagner (Eds.), Second language conversations (pp. 93113). London: Continuum. Brouwer, C. E., & Rasmussen-Hougaard, G. (2007). Conversation for second language learning . Manuscript submitted for publication. Brouwer, C. E., & Wagner, J. (2004). Developmental issues in second language conversation. Journal of Applied Linguistics, 1, 2947. Carroll, D. (2000). Precision timing in novice-to-novice L2 conversations. Issues in Applied Linguistics, 11, 67110. Chomsky, N. (1957). Syntactic structures. The Hague, Netherlands: Mouton. Cook, V. (1999). Going beyond the native speaker in language teaching. TESOL Quarterly, 33, 185209. Cook, V. (Ed.). (2002). Portraits of the L2 user . Clevedon, UK: Multilingual Matters. Crookes, G. (1997). SLA and language pedagogy: A socioeducational perspective. Studies in Second Language Acquisition, 19 , 93116. Dewey, J. (1938). Experience and education. New York: Macmillan.

816
Firth, A. (1995d). Multiple mode, single activity: Telenegotiating as a social accomplishment. In P. ten Have & G. Psathas (Eds.), Situated order: Talk and other embodied activities (pp. 15172). Washington, DC: University of America Press. Firth, A. (1996). The discursive accomplishment of normality: On conversation analysis and lingua franca English. Journal of Pragmatics, 26 , 237 259. Firth, A., & Wagner, J. (1997). On discourse, communication, and (some) fundamental concepts in SLA research. Modern Language Journal , 81, 285300. Firth, A., & Wagner, J. (1998). SLA property: No trespassing! Modern Language Journal , 82, 9194. Gardner, R., & Wagner, J. (Eds.). (2004). Second language conversations. London: Continuum. Garnkel, H. (1967). Studies in ethnomethodology. Englewood Cliffs, NJ: Prentice Hall. Gass, S. M. (1998). Apples and oranges: Or, why apples are not oranges and dont need to be. Modern Language Journal , 82, 8390. Gass, S. M. (2004). Conversation analysis and inputinteraction. Modern Language Journal , 88, 597 602. Goodwin, C., & Heritage, J. (1990). Conversation analysis. Annual Review of Anthropology, 19 , 283207. Gumperz, J. J. (1982). Discourse strategies. Cambridge: Cambridge University Press. Gumperz, J. J., & Hymes, D. (Eds.). (1972). Directions in sociolinguistics. New York: Holt, Rinehart, and Winston. Hall, J. K. (1997). A consideration of SLA as a theory of practice. A response to Firth and Wagner. Modern Language Journal , 81, 301306. Hall, J. K. (2002). Teaching and researching language and culture. London: Pearson Education. Hall, J. K. (2004). Language learning as an interactional achievement. Modern Language Journal , 88, 607 612. Hall, J. K. (2006, June). On the mutually constitutive nature of language teaching and learning . Paper presented at the Summer School 2006 in Odense, Denmark. Hall, J. K. (2007). Redressing the roles of correction and repair in research on second and foreign language learning. Modern Language Journal , 91, 511526 Hall, J. K., & Verplaetse, L. S. (Eds.). (2000). Second and foreign language learning through classroom interaction. Mahwah, NJ: Erlbaum. Hall, J. K., Vitanova, G., & Marchenova, L. (Eds.). (2004). Dialogue with Bakhtin on second and foreign language learning: New perspectives. Mahwah, NJ: Erlbaum. Halliday, M. A. K. (1973). Explorations in the functions of language. London: Edward Arnold. Han, Z. (2003). Fossilization in adult second language acquisition. Clevedon, UK: Multilingual Matters. Harris, R. (1981). The language myth. London: Duckworth. He, A. W. (2004). CA for SLA: Arguments from the Chinese language classroom. Modern Language Journal , 88, 568582.

The Modern Language Journal 91 (2007)

Hellermann, J. (2006). Classroom interactive practices for developing L2 literacy: A microethnographic study of two beginning adult learners of English. Applied Linguistics, 27 , 377404. Heritage, J. (1984). Garnkel and ethnomethodology. Cambridge: Polity Press. Hutchby, I., & Wooftt, R. (1998). Conversation analysis. Cambridge: Polity Press. Hutchins, E. (1995). Cognition in the wild. Cambridge, MA: MIT Press. Huth, T. (2006). Negotiating structure and culture: L2 learners realization of target language pragmatics in complimenting behavior. Journal of Pragmatics, 38, 20252050. Hymes, D. (1974). Foundations in sociolinguistics: An ethnographic approach. University Park: University of Pennsylvania Press. Jenkins, J. (2006a). Points of view and blind spots: ELF and SLA. International Journal of Applied Linguistics, 16 , 137162. Jenkins, J. (2006b). Current perspectives on teaching world Englishes and English as a lingua franca. TESOL Quarterly, 40, 157181. Kant, E. (1946). Critique of pure reason (J. M. D. Meiklejohn, Trans.). New York: Dutton. (Original work published 1781) Kasper, G. (1997). A stands for acquisition: A response to Firth and Wagner. Modern Language Journal , 81, 307312. Kasper, G. (2004). Participant orientations in German conversation-for-learning. Modern Language Journal , 88, 551567. Konishi, K., & Tarone, E. (2004). English constructions used in compensatory strategies: Baseline data for communicative EFL instruction. In D. Boxer & A. D. Cohen (Eds.), Studying speaking to inform second language learning (pp. 174198). Clevedon, UK: Multilingual Matters. Koshik, I. (2005). Beyond rhetorical questions: Assertive questions in everyday interaction. Amsterdam: John Benjamins. Kramsch, C. (1993). Context and culture in language teaching . Oxford: Oxford University Press. Kramsch, C. (Ed.). (2002). Language acquisition and language socialization: Ecological perspectives. London: Continuum. Kramsch, C. (2006). The multilingual subject. International Journal of Applied Linguistics, 16 , 99110. Kurhila, S. (2006). Second language interaction. Amsterdam: John Benjamins. Lantolf, J. P. (1996). SLA theory building: Letting all the owers bloom! Language Learning , 46 , 713749. Lantolf, J. P. (Ed.). (2000). Sociocultural theory and second language learning . Oxford: Oxford University Press. Lantolf, J. P., & Appel, G. (Eds.). (1994). Vygostkian approaches to second language acquisition. Norwood, NJ: Ablex. Lantolf, J. P., & Thorne, S. L. (2006). Sociocultural theory and the genesis of second language development. Oxford: Oxford University Press.

Alan Firth and Johannes Wagner

Larsen-Freeman, D. (1983). The importance of input in second language acquisition. In R. Andersen (Ed.), Pidginization and creolization as language acquisition (pp. 8793). Rowley, MA: Newbury House. Larsen-Freeman, D. (1997). Chaos/complexity science and second language acquisition. Applied Linguistics, 18, 141165. Larsen-Freeman, D. (2003). Teaching language: From grammar to grammaring . Boston: Heinle/ Thomson. Larsen-Freeman, D. (2004). CA for SLA? It all depends. Modern Language Journal , 88, 603607. Larsen-Freeman, D. (2007). On the complementarity of chaos/complexity theory and dynamic systems theory in understanding the second language acquisition process. Bilingualism: Language and cognition, 10, 3537. Lave, J., & Wenger, E. (1991). Situated learning: Legitimate peripheral participation. Cambridge: Cambridge University Press. Lazaraton, A. (2002). A qualitative approach to the validation of oral language tests. Cambridge: Cambridge University Press. Lazaraton, A. (2004). Gesture and speech in the vocabulary explanations of one ESL teacher: A microanalytic inquiry. Language Learning , 54, 79117. Liddicoat, A. (1997). Interaction, social structure, and second language use. Modern Language Journal , 81, 313317. Lilja, N. (2006, November). Language learning opportunities in everyday conversation. Paper presented at the 2006 AFinLA conference, Jyv skyl , Finland. a a Long, M. H. (1993). Assessment strategies for SLA theories. Applied Linguistics, 14, 225249. Long, M. H. (1996). The role of the linguistic environment in second language acquisition. In W. C. Ritchie & T. K. Bhatia (Eds.), Handbook of second language acquisition (pp. 413468). San Diego, CA: Academic Press. Long, M. H. (1997). Construct validity in SLA research. Modern Language Journal , 81, 318323. Long, M. H. (1998). SLA: Breaking the siege. University of Hawaii Working Papers in ESL, 12, 79129. Lyotard, J. F. (1979). La condition postmoderne: Rapport sur le savoir (The post-modern condition: A report on knowledge). Paris: Minuit. Markee, N. (2000). Conversation analysis. Mahwah, NJ: Erlbaum. Markee, N., & Kasper, G. (2004). Classroom talks: An introduction. Modern Language Journal , 88, 491 500. Mehan, H. (1979). Language lessons. Cambridge, MA: Harvard University Press. Mondada, L. (2006, June). Contingent opportunities for learning: Sequential organization of participation changes (some thoughts about CAs contributions to the study of learning). Paper presented at the Summer School 2006 in Odense, Denmark. Mondada, L., & Pekarek-Doehler, S. (2004). Second language acquisition as situated practice: Task accom-

817
plishment in the French second language classroom. Modern Language Journal , 88, 501518. Mori, J. (2004a). Negotiating sequential boundaries and learning opportunities: A case from a Japanese language classroom. Modern Language Journal , 88, 536550. Mori, J. (2004b). Pursuit of understanding: Rethinking negotiation of meaning in view of projected action. In R. Gardner & J. Wagner (Eds.), Second language conversations (pp. 157177). London: Continuum. Nguyen, H. (2003). The development of communication skills in the practice of patient consultations among pharmacy students. Unpublished doctoral dissertation, University of Wisconsin, Madison. Ohta, A. (2001). Second language acquisition processes in the classroom. Mahwah, NJ: Erlbaum. Ortega, L. (2005). Methodology, epistemology, and ethics in instructed SLA research: An introduction. Modern Language Journal , 89 , 317327. Pavlenko, A. (2002). Poststructuralist approaches to the study of social factors in second language learning and use. In V. Cook (Ed.), Portraits of the L2 user (pp. 257302). Clevedon, UK: Multilingual Matters. Pavlenko, A., & Lantolf, J. P. (2000). Second language learning as participation and the (re)construction of selves. In J. P. Lantolf (Ed.), Sociocultural theory and second language learning (pp. 155177). Oxford: Oxford University Press. Poulisse, N. (1997). Some words in defense of the psycholinguistic approach: A response to Firth and Wagner. Modern Language Journal , 81, 324328. Rampton, B. (1987). Stylistic variability and not speaking normal English: Some post-labovian approaches and their implications for the study of interlanguage. In R. Ellis (Ed.), Second language acquisition in context (pp. 4758). London: Prentice Hall. Rampton, B. (1995). Crossing . London: Longman. Rampton, B. (1997a). Second language research in late modernity: A response to Firth and Wagner. Modern Language Journal , 81, 329333. Rampton, B. (1997b). A sociolinguistic perspective on L2 communication strategies. In E. Kellerman & G. Kasper (Eds.), Advances in research on communication strategies (pp. 279303). Oxford: Oxford University Press. Rasmussen, G., & Wagner, J. (2002). Language choice in international telephone conversations. In K. K. Luke & T.-S. Pavlidou (Eds.), Telephone calls (pp. 111131). Amsterdam: John Benjamins. Richards, K., & Seedhouse, P. (Eds.). (2004). Applying conversation analysis. London: Palgrave Macmillan. Sacks, H. (1992). Lectures on conversation (Vols.1 & 2). Oxford: Blackwell. Sacks, H., Schegloff, E. A., & Jefferson, G. (1974). A simplest systematics for the organization of turntaking for conversation. Language, 50, 696735. Saussure, F. de (1922). Cours de linguistique g n rale (A e e course in general linguistics). Paris: Payot.

818
Schegloff, E. A. (2000). When others initiate repair. Applied Linguistics, 21, 205243. Seedhouse, P. (2004). The interactional architecture of the language classroom: A conversation analysis perspective. Oxford: Blackwell. Seedhouse, P. (2005). Conversation analysis and language learning. Language Teaching , 38, 165187. Seeley, A., & Carter, B. (2004). Applied linguistics as social science. London: Continuum. Seidlhofer, B. (2001). Closing a conceptual gap: The case for a description of English as a lingua franca. International Journal of Applied Linguistics, 11, 133 158. Seidlhofer, B. (2004). Research perspectives on teaching English as a lingua franca. Annual Review of Applied Linguistics, 24, 209239. Svennevig, J. (2003). Echo answers in native/non-native interaction. Pragmatics, 13, 285309. Svennevig, J. (2004). Other-repetition as display of hearing, understanding and emotional stance. Discourse Studies, 6 , 489516. ten Have, P. (1999). Doing conversation analysis. Thousand Oaks, CA: Sage. Theodorsdottir, G. (2007, April). Insisting on your turn Second language learners struggle for turn completion. Paper presented at the 17th International Conference on Pragmatics and Language Learning, Honolulu, HI. Thorne, S. L. (2000). Second language acquisition theory and some truth(s) about relativity. In J. Lantolf (Ed.), Sociocultural theory and second language learning (pp. 219243). Oxford: Oxford University Press. van Lier, L. (1994). Forks and hope: Pursuing understanding in different ways. Applied Linguistics, 15, 328346. Volosinov, V. N. (1973). Marxism and the philosophy of language. New York: Seminal Press. (Translated by Ladislav Matejka and I. R. Titunik. Original work in Russian published in 1930). von Glasersfeld, E. (1984). An introduction to radical constructivism. In P. Watzlawick (Ed.), The invented reality (pp. 1740). New York: Norton. von Glasersfeld, E. (1995). A constructivist approach to teaching. In L. P. Steffe & J. Gale (Eds.), Con-

The Modern Language Journal 91 (2007)

structivism in education (pp. 316). Hillsdale, NJ: Erlbaum. Vygotsky, L. S. (1978). Mind and society: The development of higher mental processes. Cambridge, MA: Harvard University Press. Wagner, J. (1995a). Negotiating activity in technical problem solving. In A. Firth (Ed.), The discourse of negotiation. Studies of language in the workplace (pp. 223246). Oxford: Pergamon. Wagner, J. (1995b). What makes a discourse a negotiation? In K. Ehlich & J. Wagner (Eds.), The discourse of business negotiation (pp. 936). Berlin, Germany: Mouton de Gruyter. Wagner, J. (1996). Language acquisition through foreign language interactionA critical review of studies on second language acquisition. Journal of Pragmatics, 23, 215235. Wagner, J. (1998). On doing being a guinea pigA response to Seedhouse. Journal of Pragmatics, 30, 103113. Wagner, J. (2004). The classroom and beyond. Modern Language Journal , 88, 612616. Wagner, J., & Firth, A. (1997). Communication strategies at work. In E. Kellerman & G. Kasper (Eds.), Advances in research on communication strategies (pp. 323344). Oxford: Oxford University Press. Wagner, J., & Gardner, R. (2004). Introduction. In R. Gardner & J. Wagner (Eds.), Second language conversations (pp. 117). London: Continuum. Wagner, J., & Pekarek-Doehler, S. (2006, May). Progressive bricolage of turn constructional units in second language talk. Paper presented at the International Conference on Conversation Analysis, ICCA06, Helsinki, Finland. Wenger, E. (1998). Communities of practice. Cambridge: Cambridge University Press. Wertsch, J. V. (1991). Voices of the mind. Hemel Hempstead, UK: Harvester-Wheatsheaf. Wong, J. (2000a). Delayed next turn repair initiation in native/non-native speaker English conversation. Applied Linguistics, 21, 244267. Wong, J. (2000b). Repetition in conversation: A look at rst and second sayings. Research on Language and Social Interaction, 33, 407424.

Alan Firth and Johannes Wagner APPENDIX Transcript Conventions

[] (0.0) (.) :: .,?

819

WORD word hhh word xxx =

Left and right brackets indicate beginning and end of overlap. Numbers in parentheses indicate silence by tenths of seconds. Micropause. Colons indicate prolongation of the immediately prior sound. The longer the colon row, the longer the prolongation. Arrows indicate shifts into especially high or low pitch. Punctuation markers are used to indicate intonation: , level intonation ; slightly falling intonation . falling intonation to low slightly rising intonation ? rising intonation to high Upper case indicates especially loud sounds relative to the surrounding talk. Degree signs bracketing a sound, word, phrase, and so on, indicate especially soft sounds relative to the surrounding talk. A raised dot-prexed row of hs indicates an inbreath. Without the dot, the hs indicate an outbreath. Underlining indicates stressed syllables unintelligible speech latching between turns or parts of turns creaky voice