firth and wagner 2007

Second/Foreign Language Learningas a Social Accomplishment:Elaborations on a ReconceptualizedSLAALAN FIRTHSchool of Education, Communication and LanguageSciencesNewcastle UniversityNewcastle Upon Tyne, NE1 7RUUnited KingdomEmail: [email protected]

JOHANNES WAGNERUniversity of Southern Denmark, KoldingEngstien 1DK–6000 KoldingDenmarkEmail: [email protected]

In this article, we begin by delineating the background to and motivations behind Firth andWagner (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 thearticle. Next we review and discuss the conceptual, theoretical, and methodological impact thearticle has had on the SLA field. Thereafter, we reengage and develop some of the themesraised but left undeveloped in the 1997 article. These themes cluster around the notions ofand interrelationships between language use, language learning, and language acquisition.Although we devote space to forwarding the position that the dichotomy of language use andacquisition cannot defensibly be maintained (and in this we take up a contrary position to thatheld in mainstream SLA), our treatment of the issues is essentially methodological. We focuson describing a variety of aspects of learning-in-action, captured in transcripts of recordingsof naturally occurring foreign, second, or other language interactions. Through transcriptanalyses, we explore the possibilities of describing learning-in-action devoid of cognitivisticnotions of language and learning. In so doing, we advance moves to formulate and establish areconceptualized SLA.

WITH THE BENEFIT OF 10 YEARS OF HIND-sight, it is probably fair to say that our article(Firth & Wagner, 1997) touched a proverbial rawnerve within as well as around the periphery ofthe second language acquisition (SLA) commu-nity. Certainly the article’s evident impact on theprofession has greatly exceeded our expectations.This is in no small measure thanks to The Mod-ern Language Journal’s (MLJ ) editor, Sally SieloffMagnan, who so insightfully showcased the articleand its thought-provoking responses, thereby pro-viding a framework for an extraordinarily fruit-ful and lively debate. For some, the article has

The Modern Language Journal, 91, Focus Issue, (2007)0026-7902/07/800–819 $1.50/0C©2007 The Modern Language Journal

been a rallying cry for an alternative SLA; for oth-ers, it has presaged new lines of inquiry within arapidly developing and multivaried research field;for many, it articulated misgivings and dissatisfac-tion with dominant SLA concepts, theories, andresearch practices. For others still, it was an un-warranted, misguided, perplexing, and naıve cri-tique of a well-defined field of study. We begin byexplaining the background to our 1997 article.

BACKGROUND CONTEXT OF FIRTH ANDWAGNER (1997)

Firth and Wagner (1997) was conceived onthe basis of a lingering sense of frustration withSLA as we saw it at that time—a sense that, in

Gaby Casillas

Highlight

Alan Firth and Johannes Wagner 801

large measure, prevailing, dominant SLA the-ory and concepts were myopic vis-a-vis languagelearning as social practice and language as a so-cial phenomenon—and a strong sense that thefield had to acknowledge more openly and moreconsequentially its limited vision, and, if possi-ble, overcome its myopia. The frustrations withSLA emerged gradually, from a bottom-up pro-cess while working with our data materials—audiorecordings of people at work in a vari-ety of settings, engaging each other in a sec-ond/foreign/other language (L2). Our analysesrevealed people who were artfully adept at over-coming apparent linguistic hurdles, exquisitelyable to work together interactionally, despite hav-ing what at first blush appeared to be an imper-fect 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 apparentlyrelevant areas of SLA for theoretical insight, ap-plicable concepts, and methodologies, somethingwas not right. Rather than depictions of interac-tional success in an L2, we found an overwhelmingemphasis on and preoccupation with the individ-ual’s linguistic and pragmatic failure. Rather thantalk, we found input . Rather than achievement , wefound an abundance of problem-sources. Ratherthan collaboration, invention, and an extraordi-narily creative use of shared resources (which, tous, was learning-in-action), we found references toerrors, input modifications, interference , and fossiliza-tions. Try as we might at first, our observations ofpeople using English as a lingua franca (i.e., a me-diating language that is not a mother tongue [L1]for any of the interactants—see Firth, 1990, 1996,see also Jenkins, 2006a, 2006b; Seidlhofer, 2001,2004) just would not fit the theory and conceptsof SLA.

Yet such data materials were, surely, of criticalrelevance for the SLA program. Here, after all,were nonnative speakers who were clearly acquir-ing and using (new) forms of language on thefly—often language forms that were linguisticallyand pragmatically marked. Moreover, despitenoncollocating noun phrases, verb-concordance,prosodic and morphosyntactic errors, and a hostof other linguistic anomalies, they were buyingand selling thousands of tons of Danish cheese(as well as fish, flowers, and steel) on a daily basis,and maintaining cordial relations with one an-other. This was real cheese, and they were talking(albeit in linguistically marked ways) real, and big,money.

With a background in both SLA and eth-nomethodological conversation analysis (CA),1

and informed by poststructuralist notions of con-tingency, fluidity, hybridity, and marginality (see,e.g., Lyotard, 1979; Rampton, 1995), we were un-easy about how to approach our data materials.CA, with its emphasis on the socially achievedconstruction of irredeemably motile, participant-defined contextual relevancies, its commitmentto the microanalytic explication of naturally oc-curring (rather than experimental) encounters,and an emic (participant-centered) sensitivity to“what’s going on,” led us to see that our partici-pants were not defensibly—that is, to us, emically—identifiable as participants, learners, or even nonna-tive speakers—the standard identity categories ofSLA. At the least, such categories were clearly notomnirelevant: These individuals were also, vary-ingly, sellers, buyers, friends, business acquain-tances, customers, and clients. Identity, we hadlearned, was a motile, liminal, achieved featureof the interaction (Antaki & Widdicombe, 1998;Gumperz, 1982; Rampton, 1995). And—as we willshow—we noted that who they are locally and dis-cursively has an impact on how they use and learnlanguage.2

Certainly they were not learners in any for-mal (or traditional SLA) sense of the term. Yetthey were clearly modifying, adapting, and cre-atively deploying what were to them new formsof language; that is, they were learning, andthey would occasionally draw attention—often injest—to their underdeveloped foreign languageabilities when engaging each other. We were hes-itant about calling their interlanguage interlan-guage , for where and what was the target languagehere, in this seemingly linguistically lawless linguafranca landscape? Here, it appeared, the spectreof the native speaker, with his or her idealizednorms and baseline standards, had evaporated(on this concept, see Jenkins, 2006a).3 SLA fit-ted, but then again, it did not. Something wasquite clearly not right. The field was, we argued, inneed of theoretical, methodological, and concep-tual refurbishment and expansion. In particular,we (Firth & Wagner, 1997) called for a reconcep-tualization within SLA in three areas: “(a) a sig-nificantly enhanced awareness of the contextualand 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 individ-ual’s disembodied cognition—more specifically,his or her autonomous language processing—thatwas in the ascendancy, to the detriment of what wemight call social cognition, that is, what people do,

802 The Modern Language Journal 91 (2007)

think, demonstrate, achieve, manipulate, modify,acquire, and learn, together, in concerted socialinteraction (e.g., Volosinov, 1930/1973; Vygotsky,1978; Wertsch, 1991). This is not to say, then,that learning was or is uninteresting or some-how incidental to our methodological pursuits.What we did not attempt to show in Firth andWagner (1997)—but will attempt to do in thisarticle—is that learning can be explicated andtracked as it happens and, moreover, that the de-scription of how it happens can be liberated froma cognitivistic position predicated on telementa-tional notions of communication4 and individual-istic, monolinguistic, and formalistic perceptionsof language.5 These perceptions, we argued, in-fuse SLA research, in its theorizing on language,learning, and discourse, in its terminologies andfundamental concepts (such as an ascendant na-tive speaker and interlanguage), and in its method-ological practices, including the practices of thosewho focalize interaction in SLA studies.

The immediate reactions to the 1997 articlewere, for the most part, polarized, with ratherstern criticism on one side and firm support onthe other.6 Because the criticisms go to the heartof the debate, a debate that, as recent publicationsattest, 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 keycounterarguments to ours, while demonstratinghow entrenched positionings within SLA can be-come, let us briefly consider the criticisms of Firthand 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 ourposition, which among other things emphasizedthe need to focus on L2 acquisition through L2use,7 was that SLA is “the study of L2 acquisi-tion, [and] not . . . ‘most centrally the languageuse of second or foreign language speakers’”(p. 318). Long went on to state that “most SLAresearchers view the object of inquiry as in largepart an internal, mental process” (p. 319). How-ever, he conceded that “Firth and Wagner are per-fectly justified, and probably right, in arguing thata broader, context-sensitive, participant-sensitive,generally sociolinguistic orientation might provebeneficial for SLA research” (p. 322). Never-theless, he ended his response with “the major

problem I have with Firth and Wagner’s polemicremains my skepticism as to whether greater in-sights into SL use will necessarily have much to sayabout SL acquisition” (p. 322).

Kasper (1997) adopted a similar stance. “‘A’stands for acquisition,” she reminded us, adding,tellingly, that “[a] noncognitivist discipline thathas learning as its central research object is a con-tradiction in terms” (p. 310; note, however, thatwe did not, in Firth and Wagner, 1997, call fora noncognitivist SLA). In close agreement withLong (1997), Kasper held that “the most naggingproblem with the Firth and Wagner paper [is thatit] has in fact very little to say about L2 acquisition”(p. 310). She went on to state that “I am comfort-able with an essentially cognitivist definition ofSLA” (p. 310), though in her self-styled hereticalfinal note she submitted that

if the excellent microanalytic tools of CA were in-corporated into a language socialization approach toSLA, we might be able to reconstruct links betweenL2 discourse and the acquisition of different aspectsof communicative competence that have been largelyobscure thus far. (p. 311)8

Poulisse (1997) also offered a defense of thepsycholinguistic approach in response to ourarguments, maintaining that “the task of allresearchers [is] to not only describe, but also ex-plain and predict phenomena” (p. 325) and “itwould definitely not do to just look at particu-lar and local phenomena and find specific ex-planations for each of them” (p. 325). However,Poulisse conceded that “Firth and Wagner have apoint when they plea for ‘an enhanced awarenessof the contextual and interactional dimensions oflanguage use’ and a more positive view of the NN’s[non-native speaker’s] attempts to interact in theL2” (p. 327).

Gass (1998), who found our perspective “per-plexing” (p. 88), maintained that we had got itall very wrong and that our criticisms were naıveand misplaced because, among other things, incalling for greater attention to communication,we had not understood that the “proper” objectof attention in SLA is language; thus her correc-tion: “the emphasis in input and interaction stud-ies [in SLA] is on the language used and not onthe act of communication” (p. 84). “The goal ofmy work,” she wrote, “(and others within the in-put/interaction framework . . .) has never been tounderstand language use per se . . . but rather tounderstand what types of interaction might bringabout what types of changes in linguistic knowl-edge” (p. 84). Gass made the erroneous obser-vation that “Firth and Wagner portray cognitive


approaches and communication/discourse em-phases as mutually incompatible” (p. 84). In ad-dition, she claimed that, because we called for anenlargement of the standard SLA database (whichwe see as being dominated by classroom and ex-perimental settings) to include settings such asthe workplace,9 and because we called for greaterattention to L2 use from an emically informedviewpoint, the approach we espoused “is not ac-tually part of SLA, but part of the broader fieldof L2 studies (of which SLA is a subset)” (p. 84).She continued, “many individuals in a workplacesetting . . . are not learners in the sense that is ofinterest to researchers in SLA” (p. 84).

We shall return to some of these criticismspresently (see also Firth & Wagner, 1998). Whatis important to point out at this juncture arethe rhetorical methods and arguments deployedboth to dismiss our viewpoint and to impose agrand narrative (Lyotard, 1979) or dominant the-ory of SLA, a theory that is meant to trumpours precisely because it is framed and formu-lated as dominant—for example, in ex cathedraproclamations about how SLA is properly done,and through references to what most SLA re-searchers do (see Long, 1997; Gass, 1998). Thecriticisms also demonstrate rather vividly howreadily hegemonic forces can come to the forewhen metatheoretical contributions, such as Firthand Wagner’s (1997), attempt to critique prevail-ing, established, and thus mainstream, intellec-tual practices. Admittedly, this is a mainstreamthat we, in and through our critique, both aidin its construction as mainstream and contributeto its status as mainstream. Long (1997), Kasper(1997), Poulisse (1997), and Gass (1998) drewinwards, and from their self-constructed SLA pan-theon they laid down the law by defining SLA’sproper intellectual territory (e.g., learners, lan-guage, cognition), delineating its key concerns(e.g., acquisition, not use, language, not commu-nication), and by pointing to its borders (e.g., bystipulating what is inside and outside SLA). In sodoing, they defined the parameters of legitimatecriticism and debate while essentially accusing usof intellectual trespass (see Firth & Wagner, 1998;Thorne, 2000).

For some SLA researchers, including some withtangential SLA interests, the Firth and Wagner(1997) article seems to have been viewed as a kindof watershed for SLA in general and for SLA stud-ies of interaction in particular. This reaction toit occurred, at least in part, as the result of theprominence the article was given in the MLJ atthe time—as the centerpiece of a focused andextended discussion by prominent scholars con-tributing to an esteemed journal. However, as we

pointed out in our 1997 article, the argumentsit contained did not emerge from an intellectualvoid. Although the article has since been char-acterized as seminal by a number of scholars, itwould be a mistake to see it as some kind of so-ciocultural or sociointeractional big bang withinor without SLA. Our theoretical position vis-a-viswhat we had termed and perceived as mainstreamSLA emerged from our own observations of andengagement in an ineluctable sociocultural turnin the study of language that had swept acrosslinguistics in general and discourse studies in par-ticular, as well as across anthropology, social psy-chology, education, and sociology in the 1970sand 1980s, beginning, we argued, with Gumperzand Hymes’s (1972), Halliday’s (1973), Hymes’s(1974), and others’ influential calls for a sociallyconstituted linguistics. The Firth and Wagner ar-ticle also developed and was influenced by a num-ber of extant, socioculturally focused critiques ofSLA by scholars such as Block (1996), Kramsch(1993), Lantolf (1996), Rampton (1987), and vanLier (1994).

Of course, SLA research of the 1980s and 1990swas itself influenced by this sociocultural turn—as witnessed by the steady increase in studies thatacknowledged, thematized, and explored contextand interaction. However, we argued in Firth andWagner (1997) that in the main, SLA work—including contextual and interaction-orientedstudies—was tenaciously resistant to the full im-plications of this sociocultural turn (see also sim-ilar criticisms made by Breen, 1985; Lantolf &Appel, 1994; Larsen-Freeman, 1983; Rampton,1987). That is, despite a significant increase incalls for more socially oriented approaches to L2learning and acquisition throughout the 1980sand early to mid-1990s, SLA continued to be dom-inated by an essentially Chomskian mind-set thatplaced enormous stress on individual cognition—to such an extent that, to some, alternative con-ceptions of SLA were perplexing and just plainwrong. However, on the basis of subsequent pub-lications that reengaged and developed our criti-cisms 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 con-clude that our article articulated a widely heldsense of dissatisfaction and unease that was build-ing at the time both within and around the periph-ery of SLA, an unease that provided impetus forour own call for a reconceptualization of SLA the-ories, concepts, and methods.11 In 1997, we main-tained the position that SLA is part of the nexusof approaches to the wider, interdisciplinary studyof language—which centrally includes language


learning—and has the potential to make signif-icant contributions to a wide range of researchissues conventionally seen to reside outside itsboundaries—a potential that is, arguably, yet tobe realized.

What we called for was an epistemological andmethodological broadening of SLA. This broad-ening does not entail jettisoning the mainstreamSLA position, as some have mistakenly interpretedour arguments (e.g., Gass, 1998; Long, 1998;Poulisse, 1997).12 Although we do not subscribe toSLA’s fundamental tenets (e.g., an assumption ofthe natural ascendancy of the native speaker), andwe seriously question (a) the validity of such stan-dard SLA dichotomies as acquisition versus use ,and language versus communication; (b) problema-tize the apparently clear-cut separation of the cog-nitive and the social ; and (c) reject SLA’s essentiallystatic view of context and identity, in Firth and Wag-ner (1997), we nevertheless eschewed dogmaticpositioning and pressed instead for the need forgreater theoretical, conceptual, and methodolog-ical balance within SLA. For let us acknowledgethat SLA research has, over the four decades or soof its existence as a discipline, uncovered a widerange of critically important findings relating tohow languages are acquired or learned.

In Firth and Wagner (1997), we emphasized theneed for a theoretical, methodological, and epis-temological broadening of SLA, which includedenlarging the standard SLA database to one thatreflects more accurately the sociolinguistic real-ity of a vast number of L2 users/learners aroundthe world. We sought an SLA that was more in-teractionally sensitive, that also made room for anemic stance towards fundamental concepts, andthat took seriously the theoretical and method-ological consequences of a social view of learn-ing and language. We did so in the belief that“the existence of distinct and multiple theoreticaltraditions may help to explicate the processes ofSLA, and subsequently, to develop more accurateheuristics which model these processes and condi-tions” (Thorne, 2000, p. 221). Without specifyingin any detailed way what the reconceptualizationwould entail in methodological terms (this wasnot, after all, the purpose of our article in 1997),we stressed “the need to work towards the evolu-tion of a holistic, bio-social SLA” (p. 296).

RECONCEPTUALIZATIONS: NEWDIRECTIONS IN SLA

How, then, do we assess the field of SLA,10 years later? Has the reconceptualization we,and others, called for come about? The mostaccurate answer to this question is likely to be:

It depends where you look. In many ways it ap-pears that things are more or less as they were in1997—the mainstream is in full flow (for recentcritical expositions, see Block, 2003; Cook, 1999;Jenkins, 2006a, 2006b; Kramsch, 2006; Seeley &Carter, 2004) and the native speaker continues topredominate as the baseline or target that learn-ers should seek to emulate; learning is conceivedas a cognitive process that is in essence context-neutral; competence is defined largely in terms ofthe individual’s grammatical competence; etic pre-vails over emic; and learners in classrooms remainsthe standard data set. All the while, these learnersare viewed as essentially engaged in a continuous,autonomous, cognitive, morphosyntactic struggleto traverse, in linear fashion, along the plane oftheir interlanguage in pursuit of the target (i.e.,native speaker) competence (see e.g., Han, 2003).

Nevertheless, our (and others’) urgings clearlydid not go unheeded. Much SLA research thathas been produced over the last decade bears wit-ness to a marked increase in the number of socio-cultural and contextual-interactional themes andconcepts impacting upon SLA’s research agenda,revealing an apparent growing awareness of theneed to take seriously the requirement for a morebalanced approach to SLA research. It appearsthat SLA has, over the last decade in particu-lar, undergone a bifurcation between a cogni-tive SLA (which is being termed mainstream ina number of recent publications)—representedperhaps most clearly in work undertaken by, forexample, Doughty and Long (2003), who see SLAas “a branch of cognitive science” (p. 4)—and asociocultural/sociointeractional SLA. An increas-ing number of researchers are thus displaying awillingness to adopt emic perspectives and ex-plore and attempt to develop cognitive-social ap-proaches to language learning.

New sets of metaphors are being deployed,such that allusions to dynamism, interaction, in-tricacy, and the liminal are nowadays competingwith the established SLA metaphors of machineryand computation (e.g., input, output, process-ing). Kramsch (2002), for example, proposed thedevelopment of an ecological approach to SLA andlanguage learning, one that centrally acknowl-edges the nonlinear, interactional, and contextualcharacteristics of language use and acquisition.Borrowing from the latest thinking in the natu-ral sciences, Larsen-Freeman (1997, 2007) usedthe terms chaos and complexity in her attempt tocapture more accurately the fact that L2 learningis “dynamic, complex, nonlinear, unpredictable,sensitive to initial conditions, sometimes chaotic,open, self-organizing, feedback sensitive, adap-tive” (Larsen-Freeman, 1997, p. 35). Block (2007)


used the term multimodal to capture the multiplic-ity of resources (historical, cultural, spatial, tem-poral, linguistic, etc.) brought to bear upon theprocesses of language learning and as a frame-work for more socially oriented SLA analyses.Compared with mainstream cognitive SLA, suchwork centrally locates and thematizes the contin-gent, the contextual, the ad hoc-ness, the interre-latedness of linguistic and situational elements,and the unsystematicity inherent in processes un-derlying L2 learning—factors that have been atworst overlooked and at best downgraded in im-portance in more traditional SLA research.

Research of this chaotic, social, ecological kindis being recognized as representing new direc-tions in the SLA field (see Cook, 2002; Kram-sch 2002; Larsen-Freeman, 2003; Ortega, 2005).Markee and Kasper (2004) thus talked of a re-cent “split between mainstream, cognitive SLAand emergent, sociocultural approaches to SLA”(p. 491). This latter type, Markee and Kasperclaimed, “points the way to future developmentsin what may prove to be one of the most radi-cal respecifications of SLA researchers’ theoreti-cal priorities and methodological practices in thehistory of the field” (p. 492).

Three topics appear to play a key role in the de-velopment of work in this sociocultural, socioin-teractional area: (a) whether the research incor-porates educational (i.e., classroom) settings orother (e.g., workplace) settings; (b) whether thefocus is on language learning or interaction; and(c) whether the research is theory driven or data-driven. Lantolf, with several collaborators (e.g.,Lantolf, 2000; Lantolf & Appel, 1994; Lantolf &Thorne, 2006; Pavlenko & Lantolf, 2000), has en-gaged in SLA theory-building in the Vygotskyantradition, producing what has become known asthe sociocultural approach to SLA.13 At the coreof Vygotskyan learning theory, new developmen-tal stages are first accomplished with the help ofothers in the social sphere and can then becomeintrapsychological accomplishments (see Hall’s,1997, useful synopsis). The pathway to learning isthus through social practice. The social practicein which learning has been studied thus far hasbeen in L2 classrooms. In sociocultural theory:

Communication, including the instructional con-versation of the classroom . . . and the learning-development that emerges from it, arise in thecoming-together of people with identities (which en-tail more than simply whether one is a native speaker),histories and linguistic resources constructed in thosehistories. (Dunn & Lantolf, 1998, p. 427)

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

room interaction and Vygotskyan ideas of learn-ing are, inter alios, Hall (2002), Hall and Ver-plaetse (2000), and Ohta (2001).

Another distinct research direction is alsoemerging where the emphasis is on the social,contextual, and interactional. In this case, thework deploys CA methodology. Scholarship thatdraws together CA and SLA is taking both SLAand CA into new, and largely uncharted, ter-ritory, and providing “the impetus for a wholenew generation of empirically grounded researchinto how cognitive SLA might be respecified insociocultural terms”14 (Markee & Kasper, 2004,p. 491).15 This work can usefully be divided intotwo groups. The first group focuses on the class-room setting and other formal learning environ-ments and is centrally concerned with the themeof L2 learning—though from an interactional per-spective (see He, 2004; Hellermann, 2006; Kasper,2004; Koshik, 2005; Lazaraton, 2002, 2004; Mar-kee, 2000; Markee & Kasper, 2004; Mondada &Pekarek-Doehler, 2004; Mori, 2004a, 2004b; Seed-house, 2004). A somewhat different emphasis canbe found in the CA-based SLA work undertakenby Brouwer (2003, 2004) and Brouwer and Wag-ner (2004), who explored aspects of L2 learningoccurring outside formal educational settings.

A second group to have emerged over the lastdecade also deploys CA methodology and the-ory. In this case, the focus is not so much on L2learning, but more on trying to understand andexplicate the character of L2 and lingua francainteractions, or L2 use. This work has been under-taken in settings outside the classroom—for exam-ple, in government offices, libraries, and variousworkplace settings (e.g., Firth, 1996; and contribu-tions to Gardner & Wagner, 2004; Kurhila, 2006;Rasmussen & Wagner, 2002; Schegloff, 2000;Svennevig, 2003, 2004; Wong, 2000a, 2000b).As Wagner & Gardner (2004) explained in theintroductory chapter to their Second LanguageConversations collection, this research is an on-going investigation of

whether a micro-analysis of second language conversa-tions can enhance our understanding of what it meansto talk in another language, by broadening the fo-cus beyond the sounds, structures and meanings oflanguage to encompass action sequences, timing andinteractivity. (p. 14)

In most of these studies—with Wong (2000a,2000b) as a notable exception—explicating thenormality of L2 conversations, in the face of some-times abnormal (from a native-speaker perspec-tive, at least) linguistic behavior, has underpinnedthe research, revealing, for example, that due toparticipants’ linguistic and interactional alacrity


and resourcefulness, interactional success can beachieved (see also Carroll, 2000; Egbert, 2002;contributions to Richards & Seedhouse, 2004; andSeedhouse, 2005).

SOCIAL APPROACHES TO L2LANGUAGE LEARNING

Challenging the distinction between languageacquisition and language use was not a core issuein Firth and Wagner (1997), although we did ad-dress it, albeit briefly, in Firth and Wagner (1998).Although our critics pointed out in unison thatSLA is about acquisition and not use, we ques-tion the utility and theoretical as well as method-ological validity of the dichotomy, and see it asbeing yet another symptom of the cognitivist, re-ductionist mind-set that prevails in mainstreamSLA. To us, acquisition cannot and will not occurwithout use. Language acquisition, we would ar-gue, is built on language use. Moreover, in orderto understand how language acquisition occurs,develops, and is operationalized, we are surely ob-ligated to observe and explicate language in use.At best, making and maintaining a distinction be-tween acquisition and use is highly problematic;at worst, both concepts are so tightly interwovenas to be rendered effectively inseparable. As wenoted in Firth and Wagner (1998):

If, as we argue, language competence is a fundamen-tally transitional, situational, and dynamic entity, thenany language users will always be “learners” [or “ac-quirers”] in some respects. New or partly known regis-ters, styles, language-related tasks, lexical items, termi-nologies, and structures routinely confront languageusers, calling for the contingent adaptation and trans-formation of existing knowledge and competence,and the acquisition of new knowledge. (p. 91)

What this acquisition–use dispute reveals is ar-guably the most significant and consequentialdisagreement between Firth and Wagner (1997)and proponents of a mainstream SLA, namely,the conception of learning. In this regard, someof the implications of our 1997 article were sur-prising for us as well, for we failed to realize atthe time just how fundamentally our views chal-lenged SLA’s conception of learning. Indeed, theChomskian heritage in SLA may be most conspic-uous with regard to SLA’s concept of learning,for as long as language learning is envisioned asa specific type of learning relating solely to theindividual’s head—the human language faculty—traditional SLA is conceptually safe. In fact, in theearly 1970s, when the umbrella term learning wasreplaced by the technical term acquisition, SLAaccomplished three things rather elegantly: (a) it

defined itself as a discipline that produces knowl-edge about this special phenomenon called ac-quisition; (b) it cemented its identity as a distinctdiscipline and secured a foothold in the world ofscientific research; and (c) it all but cut off pos-sible links to learning theories residing outsideits own (self-constructed) disciplinary boundaries(Rampton, 1997a).

This terminological transformation seems tohave been critical for SLA. However, if L2 learn-ing is understood differently—specifically as prin-cipally the same type of process as other typesof human learning—then SLA is open to extantand potentially insightful learning theories andapproaches, and thus placed under immense pres-sure to reconceptualize, expand, enlarge, or recal-ibrate its epistemology—pressure we see exertedby, for example, Atkinson (2002), Block (2003),Firth and Wagner (1997), Hall (1997, 2004),Jenkins (2006a), (2006b), Lantolf (2000), Liddi-coat (1997), Markee and Kasper (2004), Rampton(1997a, 1997b), and Thorne (2000).

Admitting different conceptions of learningallows, of course, for nonpsychological theoriesto become relevant for understanding and ex-plicating L2 learning.16 By way of exemplifica-tion, we can identify three potentially relevantavenues of such research. One is socioculturaltheory (e.g., Hall, Vitanova, & Marchenova, 2004;Lantolf, 2000), which we have outlined above. Asecond is constructivism, which is well establishedoutside SLA. A third is a social-interactional ap-proach to learning, which is only just beginningto take root. We will now elaborate briefly on theselatter two.

Constructivism

Constructivism (which includes social con-structivism, radical constructivism, and cognitiveconstructivism) is a theory of knowledge acqui-sition that sees learners constructing their ownknowledge and meanings on the basis of per-sonal experiences. Constructivist ideas are readilytraced in the work of Dewey (1916, 1980, 1938)and Kant (1781/1946). Drawing on von Glasers-feld’s work (e.g., von Glasersfeld, 1984, 1995),Doolittle (1999) summarized constructivism ascontaining:

three essential epistemological tenets of construc-tivism to which a fourth has been added in light ofrecent writings:

1. Knowledge is not passively accumulated, butrather, is the result of active cognizing by the indi-vidual;


2. Cognition is an adaptive process that functionsto make an individual’s behavior more viable given aparticular environment;

3. Cognition organizes and makes sense of one’sexperience, and is not a process to render an accuraterepresentation of reality; and

4. Knowing has roots in both biologi-cal/neurological construction, and social, cultural,and language based interactions (Dewey, 1916,1980). (p. 1)

Social-Interactional Approach to Learning

This approach is in the very earliest stagesof development. It is an approach that focal-izes learning-in-and-through-social-interaction,though devoid of cognitivist underpinnings. Inits current manifestations, it combines insightsfrom three areas: (a) the approach to learningpromulgated by Lave and Wenger (1991) andWenger (1998), who stress a social view of learn-ing, and see learning as occurring in multiplexcommunities of practice; (b) ethnomethodology,which emphasizes the centrality of dynamicallyand contingently deployed commonsense rea-soning practices in everyday settings (Garfinkel,1967; Heritage, 1984); and (c) CA, which stressesthe necessity to attend to and uncover the sociallyachieved, microstructuring of human activity(Goodwin & Heritage, 1990; Hutchby & Wooffitt,1998; ten Have, 1999).

The social-interactional approach can be char-acterized as follows: Learning is an inseparablepart of ongoing activities and therefore situatedin social practice and social interaction. In thissense, learning builds on joint actions and as partof a joint action it is publicly displayed and ac-complished. Building on the work of Lave andWenger (1991), Brouwer and Wagner (2004), intheir analyses of L2 conversations, proposed suchan approach to learning—one that shifts the fo-cus from the individual’s cognition and grammar,to social praxis in concrete social settings: “In-stead of studying the acquisition of a grammat-ical system, we propose to study the texture ofcommunities and the ways in which newcomersget access to them” (p. 45). Wagner and Pekarek-Doehler (2006) demonstrated how participantsproceed through their activities by a process ofwhat they call bricolage—that is, complex segmentsof social activities where interactants collaborateto make sense of their conjoint actions throughthe situated, contingent deployment of commu-nicative resources. As has been shown in the con-tributions to Gardner and Wagner (2004), L2speakers are not interactional dopes. They showperseverance and ingenuity in interacting withothers. They produce not so much “language

forms”; rather they engage in complex, multi-modal, finely tuned co-participation, integratingbody posture, gaze, verbal and prosodic activities,rhythm, and pace in their choreography of action(Egbert, Niebecker, & Rezzara, 2004). Researchadopting this approach may have the potentialto reconceptualize notions of learning, with im-portant consequences for SLA (c.f. Mondada,2006).

SLA’s adoption of such theories and ap-proaches would, we submit, have major implica-tions for SLA as a field of inquiry, as well as forL2 pedagogy, and only hints at possible develop-ments that might have occurred in SLA, modernlanguage teaching, and learning, if behaviorismand, later, information-processing psychology hadnot been allowed to become established as havingprimary importance in SLA.

LANGUAGE LEARNING AS A SOCIALACCOMPLISHMENT

How, then, might the analysis and explicationof L2 learning be approached from a social-interactional perspective, one that places contin-gency, contextuality, dialogue, and liminality at itscore? We shall attempt to answer this question byoffering cursory analyses of language learning-in-action. Our focus here is not learners in a formalsense of the word—that is, persons engaged inpurposive activities principally in order to developskills, knowledge, and competences. Nor will wedeal with classroom interaction (recent studiesof 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 asa ubiquitous social activity, as an interactionalphenomenon that transcends contexts while be-ing context dependent; as an instance of socialcognition in the wild (see Hutchins, 1995), specif-ically in relation to encounters where an L2 is inuse. For although learning may or may not be adrawn-out process, it is certainly a process thattakes place in the micromoments of social inter-action in communities of practice. It is thereforecritically important that we attempt to uncoverand understand what goes on, interactionally, insuch micromoments.

What we want to explicate, albeit extremely cur-sorily, are aspects of social learning as they areevident in L2 talk-based activities, in each case oc-curring outside the classroom. Throughout, weemphasize not the failings or deficiencies, notthe errors or interlanguage of the interactants;rather, we approach the data with a view to expli-cating and uncovering what L2 users actually do,


in everyday, natural settings. And as we shall see,they deploy, make available, share, adapt, manip-ulate, and contingently and creatively apply com-municative resources in an ongoing attempt toconstruct meaningful and consequential social in-teraction. The data excerpts we present and com-ment on here represent a small window on whatmight be possible if more in-depth, extensive stud-ies of the same kind are undertaken.

An Evidential Argument

Consider first Excerpt 1, where Hansen (H),the Danish dairy-produce salesman, is talking onthe telephone with Akkad (A), an Egyptian whole-saler (see the Appendix for transcript conven-tions):

EXCERPT 1Blowing 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

2. (0.4)3. A: we don’t want the order after the

cheese is u: :h (·) blowing.4. H: see, yes.5. A: so I don’t know what we can uh do with

the order now. (·) What do you6. think we should uh do with this is all

↑blo:wing Mister Hansen7. (0.5)8. H: I’m not uh (0.7) blowing uh what uh,

what is this u: :h too big or what?9. (0.2)

10. A: no the cheese is ↑bad Mister Hansen11. (0.4)12. A: it is like (·) fermenting in the customs’

cool rooms13. H: ah it’s gone off↑14. A: yes it’s gone off↓15. H: we: :ll you know you don’t have to uh

do uh anything because it’s not((continues))

When this excerpt was initially analyzed (inFirth, 1996, p. 244), the focus of attention wasthe let it pass procedure and its interactional con-sequences.17 But we can also use the excerpt towitness an instance of learning-in-interaction. Tosee this, note, first, H’s “I see, yes” (line 4). Thisturn follows A’s revelation in lines 1–3 that the“cheese is blowing in the customs” (line 1). In line6, A displays that he has heard H’s I see, yes as indi-

cating uptake, as displaying that he (A) has beenunderstood. Upon hearing H’s uptake at line 4,and now acting on the presumption of a sharedcommon ground, A, in line 6, asks H “what do youthink we should do with all this blowing?” (un-derlining indicates stress in enunciation). At thisjuncture, H is evidently compelled to display—to make public—his unfamiliarity with the wordblowing ; thus H’s “I’m not uh (0.7) blowing uhwhat uh, what is this u: :h too big or what?” A’sresponse is to explain the meaning of blowing; hedoes this by reformulating: “no the cheese is bad,Mr. Hansen” (line 10). With no receipt forthcom-ing (note the 0.4 second pause in line 11), A triesagain, in line 12: “it is like fermenting in the cus-toms’ cool rooms.” H now displays his updatedknowledge by producing his own reformulation(with rising intonation) of blowing , in his “ah it’sgone off?” (line 13). A, in the next turn, confirmsthe appropriateness of the definition of blowingby reusing H’s formulation, hence “yes it’s goneoff” (line 14).

What this excerpt reveals is that the interac-tants, conjointly, do interactional work to over-come potential or real communicative hurdlesin order to establish intersubjectivity and mean-ing. In so doing, the interactants provide for theavailability and utility of interactional and linguis-tic resources that allow for learning to occur. Hhas, in this excerpt, quite clearly displayed that hehas learned the (for him) new lexical item blow-ing . What is important is that the excerpt alsoshows how learning is engaged contingently, asthe learned item in question becomes interaction-ally relevant as the talk unfolds. Learning here,then, is an artifact of interactional exigency and aproduct of collaboration.

Let us now examine how H operationalizes hislearning (of what blowing cheese means) in a sepa-rate encounter, that is, how his learning is carriedover in time and space. Two working days later, H,in Denmark, calls A’s wholesale company in Cairo.The call is answered by A’s colleague, B. The callbegins as follows:

EXCERPT 2Bad Cheese(Firth, 1991)((ring))

1. B: allo2. H: yes hello Michael Hansen melko

dairies Denmark ↑calling (·) can Iplease speak to mister Akkad

3. (·)4. B: ↑hello mister Michael


5. H: is it Barat?6. B: ye: (h)s, how are you (·) si::r7. H: well I’m OK, but you ha:d tu- have some

uh problems with the: cheese8. B: uuuuuuhhhhh ((one-second sound

stretch))9. H: the bad cheese (·) in the ↑customs

10. (0.5)11. B: ↑one minute (0.4) mister Akkad will

talk (·) w[ith (·) you12. H: [ok ↑yes13. (1.5)14. A: YES (·) mister Hansen↓15. H: hello: mister Akkad (·).hh we haf some

informations for you about thecheese (·)

16. with the ↑blowing17. A: ↑yes mister Hansen

Note that in this excerpt it is H who uses theitem blowing (line 16)—which seems to suggestthat not only has H now learned this lexical itemin a way that extends beyond the concrete localcontext where he first became acquainted withit, but also, critically, that he has learned how todeploy it appropriately in context, which in thiscase entails knowing whom to use the item with.Consider, first, how H characterizes the matterwhile talking to B. First, H uses “problems with thecheese” (line 7), next, he reformulates the matteras “the bad cheese in the customs” (line 9), thisreformulation likely being a result of an apparentlack of uptake by B (line 8). When A enters theinteraction (at line 14), H first uses the formu-lation “the cheese” (line 15), and then, withoutyet securing uptake from his interlocutor (notethe micropause at the end of line 15), formulatesthe matter as “with the blowing” (line 16). Thisformulation succeeds, as we see from A’s “yes Mr.Hansen” (line 17).

What are the language learning implicationsof such behavior? They are, first and foremost,that if learning entails being contingently adapt-able as the unfolding context requires, then here,surely, is learning-in-action. Second, here may beevidence of a community of practice (Lave &Wenger, 1991) being constructed before our veryeyes—namely, between H and A. The nature ofthis particular community of practice appears tobe consequential for how learning is operational-ized, for how learning is displayed, and for howcommunicative knowledge is refined and testedboth within and across interactions.

It is arguable that H, in Excerpt 2, orients to A asa co-user, or as a co-member of this microcommu-nity of practice, of the term blowing . Co-usership,

then, is, potentially at least, a relevant identity cat-egory for some of these interactants. In a relatedway, we may venture that the reason H deploysthe item blowing with A, but uses instead cognateformulations with B, is because H is “recipient de-signing” (Sacks, Schegloff, & Jefferson, 1974, p.727) his displays of knowledge so as to achievecommunicative (and likely interpersonal) successand to display interactional affinity. To paraphraseBaker, Emmison, and Firth (2005), we can say thatH is calibrating his language behavior for his in-terlocutors’ competence. We cannot know why Hdid not use the term blowing with B. Perhaps, ifA had not been available or had not been ableto engage H, H may subsequently have deployedthe item blowing. However, his apparent contin-gent and selective deployment of the item offersan intriguing angle from which to view languagelearning.

Our argument so far reminds us of the interac-tion hypothesis (Long, 1996), according to whichlearning is fueled by the problems that learnersencounter. However there are several problemswith this explanation for learning:

1. The approach describes a transparentproblem-based way of learning, but we find onlyfew instances of such problem-based learning inour data.

2. It prototypically describes lexical learning.Where does the rest of the language (and com-municative behavior more generally) reside?

3. Uptakes are rare. Participants in interac-tions obviously pick up items and use them inrecipient-designed ways, but such actions are onlypart of the whole language learning story. Al-though it might partly have to do with the lackof comprehensive corpora, problem-based learn-ing of this kind seems not to be at the core oflanguage learning.

4. The interactive hypothesis constructs somekind of evidential argument: a plain causal chainbetween use of elements and learning. When welook at the developmental data we have access to,plain causal explanations do not cover many of thecomplexities we see unfolding in our data. We seelearners developing their skills in using resourcesin more advanced ways for more advanced inter-actions, but we rarely see a “smoking gun,” as inExcerpts 1 and 2.

There are few, if any, studies showing learning inaction (Huth, 2006; Nguyen, 2003). The main ob-stacle for this type of research has been the lackof longitudinal corpora available for L2 interac-tions.


Doing Learning

Another way of describing learning does notfocus on the evidence but on the process. Ourposition in this regard is inspired by socioculturaltheory, particularly Hall’s 2006 work. Consider thenext excerpt in which Rasmus, a Danish student,coaches Joseph, an American guest student, tolearn how to order pizza in Danish. In his ownwords: Joseph is developing his survival skills.

EXCERPT 3Danish Pizza(Wagner, unpublished raw data)

1. Jo: so how d- how do I ↑say say it one time[for] me.

2. Ra: [.pt]3. (0.2)4. Ra: je:g vil gerne bestille en pizza.

I would like order a pizza5. (0.3)6. Jo: jeg vil gerne bestille en pizza.7. (0.2)8. Ra: ◦jep◦.9. (0.5)

10. Ra: that’s I’d like to order pizza.11. Jo: .pt okay,

In line 1, Joseph opens a new sequence with“so” and gives two consecutive prompts for nextaction: “how do I say” and “say it one time forme.” Rasmus provides the appropriate Danish ex-pression (line 4), which Joseph repeats (line 6).Rasmus acknowledges the repeat in line 8 with“yep” and provides an English translation (line10). In line 11, Joseph closes the sequence.

The sequence runs off at a slow pace withoutglitches, delays, overlaps, or repairs. Both partic-ipants demonstrate, through their actions, thatthey understand who will produce the next ele-ment and which element has been projected. Itseems to be clear for both participants what kindsof actions they are engaged in and how the sub-activities are distributed. To elaborate—the Dan-ish expression is decontextualized. The I doesnot refer to the current speaker (Rasmus) butis a generic I for any speaker in a pizza-orderingsituation. In this way, Rasmus provides a modelas a standard solution for a well-defined activityin a well-defined environment (ordering a pizzaby phone). Both Joseph and Rasmus show theirknowledge of the script for this activity and treatit as a standard situation where preformulated ex-pressions can be deployed. They refer only to the

core of this standard situation—the Danish sen-tence would work as the reason for the call, butthe opening and closing activities in a phone callare not thematized by Joseph and Rasmus.

The sequence is initiated and closed by the L2speaker, which makes it different from classicalteaching activities described, in non-CA research,as variations of an Initiative-Response-Evaluation se-quence (Mehan, 1979). The L1 speaker Rasmusappears therefore interactionally not as a teacherbut as a language expert who is transferring know-how to a nonexpert.

The participants themselves call the activitylearning . Joseph’s (the learner’s) main activity isthe repeat of the target utterance. We observedthat the translation follows the repeat of the Dan-ish expression that shows both participants orient-ing primarily to the form of the expression. Josephlearns the Danish expression before the transla-tion has made crystal clear what he has learned(see below).

What we observe in Excerpt 3 holds for thewhole interaction. Joseph provides a prompt, Ras-mus delivers a Danish expression that Joseph re-peats. However, not all sequences run as smoothlyas the one shown in Excerpt 3. In later talk, whichwe cannot show here due to constraints of space,we observe Rasmus slowing down his speech, iso-lating troublesome elements in the utterance byprosodic means, and even using the written modeto display a model. The resources for learningused in this interaction are not very different fromthose of traditional classroom teaching, which in-clude (a) utterance model and repetition, (b)form before content/translation, (c) prosodic for-mation (speed, rhythm), and (d) written mod-els. In Excerpt 3, the participants show us thatthey recognize and characterize their activity aslearning. They construct conversational routinesand patterns of interaction that are typical oflanguage classrooms (Kasper, 2004; Mondada &Pekarek-Doehler, 2004).

Although Rasmus and Joseph draw heavily ona classroom format, other studies show how par-ticipants in interactions orient in more subtleways to language learning as one goal in the on-going activity. Lilja (2006) described instancesin interactions between a Finnish host familyand guest students where the native speakersin repair sequences do not just repeat troublesources but strip them of morphological inflec-tion and present them in an unmarked form.Theodorsdottır (2007) showed how learners mayinsist on producing full turns although their co-participants have indicated their understandingof the ongoing turn before its end. Brouwer


and Rasmussen-Hougaard (2007) described howparticipants in L2 interactions carve out oppor-tunities for doing learning in the ongoing flow ofinteraction.

Compared to the case of Hansen and Akkad,instances of doing learning do not afford an ev-idential argument for learning. Instances of do-ing learning show that learning a language andlearning to learn are mutually constitutive, as Hall(2006) formulated it for the language classroom:

What our students take away from our classrooms interms of their target language knowledge and under-standings of what it means to learn another languageis intimately tied to the kinds of interactional practicesthat we as teachers create in our talk with students . . . .Through their interactions with us, learners becomeexperienced at figuring out the actions that are beingimplemented in our utterances including how the ut-terances are constructed and eventually learn to usethem to take actions of their own. Our interactionalpractices are, then, to use Vygotsky’s term, the medi-ational means by which we and our students togetherconstitute, represent, and remember what it means toknow and do language learning. (p. 27)

Learning is about the object and the ways oflearning. Joseph and Rasmus illustrate that theyknow how to set up a learning situation. They en-act the classical activities of language teaching.However, the other studies mentioned here traceways of learning not only on the genre level butalso as practices in ongoing interaction where par-ticipants show each other that they are, amongother things in the real world, engaged in morethan intersubjectivity. They are not content tomake themselves understood but clearly demon-strate a desire to do so in ways that are viewedas appropriate and normal in their L2. Other set-tings, with other kinds of interactional and institu-tional goals and relationships, will doubtless pro-duce different kinds of talk- and learning-basedactivities.

Doing learning illustrates how participants fore-ground learning in an interactionally consequen-tial way. This type of learning is different fromthat of Hansen and Akkad where learning a newitem was entirely embedded in the business inter-action. But doing learning does not provide evi-dence that learning is actually happening in theseactivities.

Out of Chaos Comes Order

Consider now Excerpt 4, again from a businessinteraction. Jørgen Gade (JG) in Denmark hasbeen called by a Swiss business acquaintance (BR).At the end of the call, BR enquires about the suc-

cess of the company where JG works. This is thesecond call between BR and JG.

EXCERPT 4Lot of Work 1(Wagner, unpublished raw data)

1. BR: okay have you a lot of work to do.2. JG: yes,3. JG: [we ha’ a lot a do.4. BR: [xx5. BR: a lot of orders.6. JG: and we e::hm: work also in the week-

ends7. BR: ah [(xx the weekend)]8. JG: [for the next ] e:h fh three four

monthth (·) [I think,9. BR: [yes

10. (0.8)11. JG: [so ]12. BR: [that’s] that’s good to ↓hear.13. (0.3)14. JG: yes,15. BR: ↑okay.16. JG: and your company has also. (·) lot to

do,17. (0.3)18. BR: oh yes.19. BR: things are going very well. =20. JG: yes¿

BR’s question in line 1, “have you a lot of work todo” is in line 16 mirrored by JG’s answer “we haa lot to do” (line 3) as well as in his formulationin line 16 “and your company has also a lot todo.” Regular telephone engagement with knownothers frequently gives rise to the production ofHow are you sequences (compare Excerpts 2, 6).As in Excerpt 2, we notice that these how are yousequences are not taken personally but relate in-stead to the companies in which the protagonistswork. Consider now Excerpt 5.

EXCERPT 5Lot of Work 2

1. JG: have you: lot of work to do.2. TT: yes¿ they’re not too bad¿3. JG: [yes]4. TT: [we-] e-e- picked up another order¿

=eight hundred thousandpound[s

5. JG: [↓ouw:¿6. (0.6)7. JG: excellent ehhe hhe hhe hhe hhe hhe

he


8. TT: eight million cr↓owner¿ hhe hh9. JG: ↓↓∗o:ijoijoi.∗

10. TT: (sxt xx) hh hh ·hhh11. TT: so thats e:h thats g↓ood¿12. G: ◦yeah¿◦

13. TT: we still ned some m↓ore though but[it loo]ks looks if its u:h =

14. JG: [mmh.]15. JG: yes¿16. TT: = getting better.

In Excerpt 5, JG is on the phone with a Britishcolleague with whom he has had regular contactover many years. JG’s question in line 1 “have you:lot of work to do” sounds like an echo of BR’squestion from Excerpt 4. However, in this case, 2years have passed between the call in Excerpt 4and Excerpt 5. It would be a bold move to con-strue (line 1) as evidence that learning had oc-curred in the call 2 years earlier. Nevertheless,the resemblance of both formulations over thespan of those 2 years is striking. JG has not be-come more linguistically sophisticated in his in-quiry into his interlocutor’s work commitments.He simply redeploys the utterance formula thathe had deployed 2 years previously. We noticecomparable development standstill in other re-current call segments or activities, for example,in call openings, closings, the introduction ofnew topics, how-are-you’s and salutations. In thesecases, it appears that comparatively little linguis-tic development occurs over time. If there is anydevelopmental change at all, we find it in gen-eral interactional fluency, such that these activ-ities are produced with fewer delays, smootherturn transitions, and a general increased orderli-ness; in addition, the content of the calls becomesmore personalized over time as the interactantsbecome better acquainted (cf. Brouwer & Wagner,2004).

These observations beg the question: In whichenvironments might we notice language learningas a social accomplishment? If it is not observablein routinized activities and talk episodes, wherecan it be located? What are the environmentswhere linguistic structures expand and linguisticcreativity and development can be observed?

They appear to occur in what we might callthe body of the calls, that is, in phases followingopenings and preceding closing routines. Herethe talk is more free flowing, less planned andless routinized. New expressions and new struc-tures are afforded by new topics and activities. Thecontingencies of the locally unfolding situationseem to motivate the use of new resources; that

is, language development in the calls appears tobe nonlinear, and the growing complexities in theparticipants’ actions are not traceable to simplerpreceding forms.

This observation is in line with Eskildsen’s study(Eskildsen, in press; Eskildsen & Cadierno, 2007),which drew on data from a Mexican learner of En-glish over a period of 4 years (2001–2005). Eskild-sen (in press) looked at the development of modalverbs over time and the conclusions he was ableto draw on the basis of the data materials do notshow a development from a core structure intomore complex structures, but rather a patchworkof different uses that appear and disappear overtime. In terms of linguistic structure, this studydoes not show much structure building.

We can tentatively conclude that we have beenable to point to three aspects of language learningas a social accomplishment. These three aspects,and possibly others, demonstrate in a variety ofways how participants engage in meaningful activ-ities by using an L2. In situated social practices,use and learning are inseparable parts of the in-teraction. They appear to be afforded by topicsand tasks, and they seem to be related to specificpeople, with particularized identities, with whomnew ways of behaving occur as the unfolding talkdemands.

Studying learning as a social accomplishmentshifts our understanding of learning from theconstruct of a linguistic system or a compe-tence that serves all the speaker’s purposes. In-stead, the development of social relations, themutual constituency of linguistic resources andtasks, and the specific biography of the languagelearners come to the foreground. This strandof research has gained momentum over the last10 years, and quite clearly, much more researchinto the specifics of social interactions in L2 en-vironments is clearly necessary in the years tocome.

CONCLUSIONS

What this entire debate has brought home mostvividly is that SLA as a field of inquiry is todaya fertile arena of multifarious approaches, theo-ries, methods, concepts, and, not least, debate.Whether SLA has a mainstream or a dominantapproach is a moot point. Cognitivistic SLA—themainstream that we focused our critique upon inFirth and Wagner (1997)—is certainly well estab-lished and continues to occupy a prominent placein SLA-related journals, textbooks, doctoral andgraduate programs, and publications generally. In


most of its guises, cognitivistic SLA is strikingly dif-ferent from the SLA of sociocultural theory, theecological approach to SLA, and the CA-for-SLAmovement that we have adumbrated in this arti-cle. Thus, although some researchers are explor-ing the possibilities of developing a holistic, en-compassing SLA—one that seeks to draw togetherthe social and the cognitive—other researchersposition themselves apart from the social, cul-tural, situational, chaotic, sociolinguistic, ecolog-ical, interactional drives, and motives underpin-ning the work of a growing number of SLA schol-ars. These differences raise the inevitable questionof where SLA is headed, and whether the field it-self is able to withstand the current bifurcations,competing methods, critiques, and internal ten-sions, and remain generally cohesive—in the waythat the field of sociolinguistics, for example, hasremained more or less cohesive, despite the dis-cipline being populated with an almost dizzyingarray of sometimes incommensurable methods,concepts, and theories—or whether SLA will frac-ture into cognitive SLA, holistic SLA, socioculturalSLA, conversation-analytic SLA, postmodern SLA,and so on.

It is arguable, of course, that such a fracturingeither has already taken place or is currently un-derway. If this is indeed what is happening, or whathas occurred, a major issue then becomes oneof how the field or the discipline defines itself.SLA is a relative newcomer to scientific inquiry,and there are inherent risks in allowing such anew field of research to shift, morph, and frac-ture, particularly from the viewpoint of those who,through a lengthy professional devotion to thefield or paradigm, conceive of themselves as theintellectual guardians of (in this case) SLA, andsee it as their right and obligation to determine,ex cathedra, what is and is not proper SLA. Thedebates, the arguments, the progress, or the decayis surely an inevitable component of SLA’s evolu-tion. We are, then, witness to a natural progres-sion, an intellectual evolution, if one will, wheresuccessful paradigms evolve (and sometimes frac-ture) through both support and critique. If thisprocess is based on sound, creative scholarship,one that leads to advances in knowledge of themany and varied ways in which L2s are learned,acquired, and used (in mutually reinforcing andenlightening ways), then surely SLA will becomea more theoretically and methodologically robustand encompassing enterprise. Despite objectionsfrom some quarters, the boundaries of SLA areineluctably being redrawn, and from this partic-ular viewpoint, the future of SLA looks distinctlypromising.

ACKNOWLEDGMENTS

We are grateful to the following colleagues who gener-ously and insightfully commented on previous versionsof this article: Dennis Day, Rod Gardner, Jennifer Jenk-ins, 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 forimprovements into the final version. Remaining weak-nesses and oversights remain our responsibility.

NOTES

1 Conversation analysis (widely known as CA) is amethodology devised by Sacks, Schegloff, and Jeffer-son in the 1960s (see Sacks, 1992). Building on itsethnomethodological (Garfinkel, 1967; Heritage, 1984)foundations, CA endeavors to explicate the microdetailsof talk-in-interaction and to uncover the communica-tive and social competences that structure and rendermeaningful talk-in-interaction. The materials of CA arevideo- and audiorecordings of naturally occurring set-tings where talk is a prime facet of behavior. For descrip-tions of CA’s working methods and theoretical founda-tions, see ten Have (1999) and Hutchby and Wooffitt(1998).

2 Although such observations on our data informedour call for a reconceptualized SLA in Firth and Wagner(1997), they did not feature empirically in that publica-tion.

3 We do not have the space to elaborate the pointhere, but in essence the argument is that, in the case ofEnglish, which is undoubtedly the global lingua francain an array of domains (e.g., the Internet, diplomacy,science, pop music, tourism), equating target languagecompetence with native speaker competence is inher-ently problematic, in that such a practice ignores the so-ciolinguistic reality of the global status and lingua francauses and functions of English. The implications for SLAare potentially far-reaching, not least in terms of our un-derstanding of interlanguage. This point is well madeby Jenkins (2006a, 2006b).

4 The telementational view of communication is theview promulgated by, among others, Saussure (1922),adopted by Chomsky (1957) and, later, by mainstreamSLA practitioners. It underpins SLA work in communi-cation strategies (see Firth & Wagner, 1997). Accordingto Harris (1981), the telementational view is a fallacy; itis a thesis about the function of language, namely, that“linguistic knowledge is essentially a matter of knowingwhich words stand for which ideas. For words, accordingto this view, are symbols devised by man for transferringthoughts from one mind to another. Speech is a form oftelementation” (p. 9).

5 This monolinguistic notion, we argued (Firth & Wag-ner, 1997), underpins the prevailing SLA view that seeslanguage users as non-native speakers who are (or ought


to be) aiming for a target consisting of the baseline lin-guistic norms of the native speaker (see, e.g., Konishi &Tarone, 2004).

6 Those who supported our arguments (in their re-sponse articles) were Hall (1997), Rampton (1997a),and Liddicoat (1997).

7 To be clear on how we stood in relation to acquisitionand use , in Firth and Wagner (1997) we pressed the needfor an SLA that can explain “how language is used as it isbeing acquired through interaction, and used resourcefully,contingently, and contextually” (p. 296).

8 Kasper’s (1997) views were prescient, but note thatin Firth and Wagner (1997) we refrained from rec-ommending that CA become part of SLA, or that CA-methodologies be incorporated into SLA.

9 Classrooms are, of course, workplaces, though forthe sake of exposition we shall maintain the rudimen-tary distinction to differentiate classrooms from othersocial settings in which some kind of occupational orprofessional activity is taking place.

10 Thorne (2000) is representative of the views heldbroadly by critics of SLA who, like Firth and Wagner(1997), pressed the need for a reconceptualized SLAalong social (to this can be added historical) lines.Thorne wrote: “the dominant core of current theories ofSLA are for the most part defining a world of a-historical,decontextualized, and disembodied brains. It is my be-lief that such a theory does not fit the evidence” (p.220).

11 Atkinson (2002) pointed out that two prominentSLA researchers, Ellis (1997) and Crookes (1997), “seemto have realized the reductionism inherent in [SLAresearch]” (note 11). Ellis (1997) wrote “SLA in gen-eral has paid little attention to the social context ofL2 acquisition, particularly where context is viewednon-deterministically (i.e., as something learners con-struct for themselves). SLA has been essentially a psy-cholinguistic enterprise, dominated by the computa-tional metaphor of acquisition” (p. 87). Crookes (1997)offered similar sentiments: “though cognitive psychol-ogy was to be preferred [in SLA] to its dominant pre-decessor [behaviorism] because it was (purportedly atleast) about people (rather than rats), it was a long timebefore I began to understand that it, like its predeces-sor, could be seen as a sociocultural construct . . . thatreflected at least to some extent the presuppositionsof the societies in which it developed. That was why itwas fundamentally an individualist psychology that treatspeople as isolates” (p. 98).

12 Long (1998), for example, wrote: “Instead of dis-missing all past work as ‘narrow’ and ‘flawed’, and simplyasserting that SLA researchers should therefore changetheir data base to take new elements into account, [crit-ics] should offer at least some evidence that, e.g., a richerunderstanding of alternate social identities of peoplecurrently treated as ‘learners’, or a broader view of so-cial context, makes a difference, and a difference notjust to the way this to that tiny stretch of discourse isinterpretable, but to our understanding of acquisition”(p. 92).

13 Space limitations preclude a detailed descriptionof 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 funda-mental concept of sociocultural theory is that the humanmind is mediated . In opposition to the orthodox view ofmind, Vygotsky argued that just as humans do not actdirectly on the physical world but rely, instead, on toolsand labor activity, which allows us to change the world,and with it, the circumstances under which we live inthe world, we also use symbolic tools [e.g., language],or signs, to mediate and regulate our relationships withothers and with ourselves and thus change the nature ofthese relationships . . . . The task of psychology, in Vygot-sky’s view, is to understand how human social and men-tal activity is organized through culturally constructedaffairs” (p. 1).

14 We question the appropriateness of Markee andKasper’s (2004) characterization of this CA-centeredSLA work as sociocultural . The sociocultural approachhas already established itself within SLA and is associ-ated with the work of Lantolf (e.g., Lantolf, 2000) andothers. According to Hall (1997) and Lantolf (2000),sociocultural SLA research is principally underpinnedby Vygotskian theories of mind and learning (e. g.Vygotsky, 1978). CA, however, has its roots quite firmly inGarfinkelian sociological theory (Garfinkel, 1967; Her-itage, 1984).

15 Deploying CA’s microanalytic methodology for theanalysis of L2 talk-in-interaction poses a number of chal-lenges (which are too numerous and complex for usto reveal and discuss in this article; see Brouwer &Wagner, 2004). It certainly entails much more thanfamiliarizing oneself with CA’s terminologies and ap-plying them to L2 data materials. Arguably the great-est challenge for SLA researchers entails coming toterms with CA’s ethnomethodological research agenda(Firth, 1995c; Garfinkel, 1967; Heritage, 1984), whichis radically different from mainstream SLA’s cognitivistagenda (see Larsen-Freeman, 2004). The incorporationof L2 data materials into a CA framework also poseschallenges for CA. To mention one of several exam-ples: CA has until relatively recently been restricted tothe analysis of monolingual, first language (L1), inter-actions, where, overwhelmingly, the analyst has reliedupon his or her co-membership in the interactants’speech community as a resource in the explication ofthe talk-data. When faced with the analysis of L2 talk,the analyst cannot unproblematically assume linguisticor cultural co-membership; the applicability of an im-portant analytic resource is thereby brought into ques-tion. For further information on this topic, see Firth(1996).

16 Let us be clear, however, that the opening up of SLAto noncognitivist theories and approaches to (language)learning is not going to be a trouble-free process; see, forexample, Kasper’s (1997) view that “[a] noncognitivistdiscipline that has learning as its central research objectis a contradiction in terms” (p. 310).


17 We refer the reader to Firth (1996) for a detailedexplication, but in brief, H’s I see (line 5) is quite clearlya let it pass procedure. Such procedures are utilized ininteraction when the recipient of a turn lets the un-known or unclear action, word or utterance pass on the(commonsense) assumption that it will either becomeclear or redundant as talk progresses.

REFERENCES

Antaki, C., & Widdicombe, S. (Eds.). (1998). Identitiesin talk. London: Sage.

Atkinson, D. (2002). Toward a sociocognitive approachto second language acquisition. Modern LanguageJournal , 86 , 525–545.

Baker, C., Emmison, M., & Firth, A. (2005). Calibratingfor competence in calls for technical assistance. InC. Baker, M. Emmison, & A. Firth (Eds.), Callingfor help: Language and social interaction in telephonehelplines (pp. 39–62). Amsterdam: Benjamins.

Block, D. (1996). Not so fast: Some thoughts on theoryculling, relativism, accepted findings and the heartand soul of SLA. Applied Linguistics, 17 , 63–83.

Block, D. (2003). The social turn in SLA. Washington,DC: Georgetown University Press.

Block, D. (2007, March). Thinking about SLL research:Conservatism, fighting, toolkits and thinking outsidethe box. Paper presented at the School of Edu-cation, Communication and Language Sciences,Newcastle University, Newcastle Upon Tyne, UK.

Breen, M. (1985). The social context for languagelearning—A neglected situation? Studies in SecondLanguage Acquisition, 7 , 136–158.

Brouwer, C. E. (2003). Word searches in NNS–NS inter-action: Opportunities for language learning? Mod-ern Language Journal , 87 , 534–45.

Brouwer, C. E. (2004). On doing pronunciation. In R.Gardner & J. Wagner (Eds.), Second language con-versations (pp. 93–113). London: Continuum.

Brouwer, C. E., & Rasmussen-Hougaard, G. (2007). Con-versation for second language learning . Manuscriptsubmitted for publication.

Brouwer, C. E., & Wagner, J. (2004). Developmentalissues in second language conversation. Journal ofApplied Linguistics, 1, 29–47.

Carroll, D. (2000). Precision timing in novice-to-noviceL2 conversations. Issues in Applied Linguistics, 11,67–110.

Chomsky, N. (1957). Syntactic structures. The Hague,Netherlands: Mouton.

Cook, V. (1999). Going beyond the native speaker inlanguage teaching. TESOL Quarterly, 33, 185–209.

Cook, V. (Ed.). (2002). Portraits of the L2 user . Clevedon,UK: Multilingual Matters.

Crookes, G. (1997). SLA and language pedagogy: A so-cioeducational perspective. Studies in Second Lan-guage Acquisition, 19 , 93–116.

Dewey, J. (1938). Experience and education. New York:Macmillan.

Dewey, J. (1980). The need for social psychology. In J.A. Boydston (Ed.), John Dewey: The middle works,1899–1924 (Vol. 10, pp. 53–63). Carbondale, IL:Southern Illinois University. (Original work pub-lished 1916)

Doolittle, P. E. (1999). Constructivism and onlineeducation. Retrieved 12 January, 2007 fromVirginia Tech Web site: http://edpsychserver.ed.vt.edu/workshops/tohe1999/tohe2.html

Doughty, C. J., & Long, M. (Eds.). (2003). The handbookof second language acquisition. Oxford: Blackwell.

Dunn, W., & Lantolf, J. P. (1998). Vygotsky’szone of proximal development and Krashen’si+1: Incommensurable constructs, incommen-surable theories. Language Learning , 48, 411–442.

Egbert, M. (2002). Der Reparatur-Mechanismus indeutschen und interkulturellen Gesprachen [The re-pair mechanism in German and intercultural con-versation]. Unpublished Habilitationsschrift. Old-enburg, Germany: University of Oldenburg.

Egbert, M., Niebecker, L., & Rezzara, S. (2004). Insidefirst and second language speakers’ trouble in un-derstanding. In R. Gardner & J. Wagner (Eds.),Second language conversations (pp. 178–200). Lon-don: Continuum.

Ellis, R. (1997). SLA and language pedagogy: An ed-ucational perspective. Studies in Second LanguageAcquisition, 19 , 69–92.

Eskildsen, S. W. (in press). Constructing another lan-guage: (Beyond) frequency effects in second lan-guage acquisition? To appear in Applied Linguis-tics.

Eskildsen, S. W., & Cadierno, T. (2007). Are recurringmulti-word expressions really syntactic freezes?Second language acquisition from the perspec-tive of usage-based linguistics. In M. Nenonen &S. Niemi (Eds.), Collocations and idioms 1: Papersfrom the First Nardic Conference on Syntactic Freezes,Joensuu, May 19–20, 2007. Studies in Languages,University of Joensuu, Vol 41. Joensuu: JoensuuUniversity Press.

Firth, A. (1990). ‘Lingua franca’ negotiations: Towardsan interactional approach. World Englishes, 9 , 69–80.

Firth, A. (1991). Discourse at work: Negotiating by telex,fax, and phone . Aalborg, Denmark: Unpublisheddoctoral dissertation, Aalborg University.

Firth, A. (1995a). Negotiating in the ‘Virtual Market-place’: Making sense of telenegotiating. In K.Ehlich & J. Wagner (Eds.), The discourse of businessnegotiation. (pp. 127–149). Berlin, Germany: Mou-ton de Gruyter.

Firth, A. (1995b). Talking for a change: Negotiating bytelephone in commodity trading. In A. Firth (Ed.),The discourse of negotiation: Studies of languages inthe workplace (pp. 183–222). Oxford: PergamonPress.

Firth, A. (1995c). Accounts in negotiation discourse: Asingle-case analysis. Journal of Pragmatics, 23, 199–226.


Firth, A. (1995d). Multiple mode, single activity: Telene-gotiating as a social accomplishment. In P. tenHave & G. Psathas (Eds.), Situated order: Talk andother embodied activities (pp. 151–72). Washington,DC: University of America Press.

Firth, A. (1996). The discursive accomplishment ofnormality: On conversation analysis and ‘linguafranca’ English. Journal of Pragmatics, 26 , 237–259.

Firth, A., & Wagner, J. (1997). On discourse, communi-cation, and (some) fundamental concepts in SLAresearch. Modern Language Journal , 81, 285–300.

Firth, A., & Wagner, J. (1998). SLA property: No tres-passing! Modern Language Journal , 82, 91–94.

Gardner, R., & Wagner, J. (Eds.). (2004). Second languageconversations. London: Continuum.

Garfinkel, H. (1967). Studies in ethnomethodology. Engle-wood Cliffs, NJ: Prentice Hall.

Gass, S. M. (1998). Apples and oranges: Or, why applesare not oranges and don’t need to be. Modern Lan-guage Journal , 82, 83–90.

Gass, S. M. (2004). Conversation analysis and input-interaction. Modern Language Journal , 88, 597–602.

Goodwin, C., & Heritage, J. (1990). Conversation analy-sis. Annual Review of Anthropology, 19 , 283–207.

Gumperz, J. J. (1982). Discourse strategies. Cambridge:Cambridge University Press.

Gumperz, J. J., & Hymes, D. (Eds.). (1972). Directionsin sociolinguistics. New York: Holt, Rinehart, andWinston.

Hall, J. K. (1997). A consideration of SLA as a theory ofpractice. A response to Firth and Wagner. ModernLanguage Journal , 81, 301–306.

Hall, J. K. (2002). Teaching and researching language andculture . London: Pearson Education.

Hall, J. K. (2004). Language learning as an interactionalachievement. Modern Language Journal , 88, 607–612.

Hall, J. K. (2006, June). On the mutually constitutive natureof language teaching and learning . Paper presentedat the Summer School 2006 in Odense, Denmark.

Hall, J. K. (2007). Redressing the roles of correction andrepair in research on second and foreign languagelearning. Modern Language Journal , 91, 511–526

Hall, J. K., & Verplaetse, L. S. (Eds.). (2000). Second andforeign language learning through classroom interac-tion. Mahwah, NJ: Erlbaum.

Hall, J. K., Vitanova, G., & Marchenova, L. (Eds.).(2004). Dialogue with Bakhtin on second and for-eign language learning: New perspectives. Mahwah,NJ: Erlbaum.

Halliday, M. A. K. (1973). Explorations in the functions oflanguage . London: Edward Arnold.

Han, Z. (2003). Fossilization in adult second language ac-quisition. Clevedon, UK: Multilingual Matters.

Harris, R. (1981). The language myth. London: Duck-worth.

He, A. W. (2004). CA for SLA: Arguments from the Chi-nese language classroom. Modern Language Jour-nal , 88, 568–582.

Hellermann, J. (2006). Classroom interactive practicesfor developing L2 literacy: A microethnographicstudy of two beginning adult learners of English.Applied Linguistics, 27 , 377–404.

Heritage, J. (1984). Garfinkel and ethnomethodology. Cam-bridge: Polity Press.

Hutchby, I., & Wooffitt, 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: L2learners’ realization of target language pragmaticsin complimenting behavior. Journal of Pragmatics,38, 2025–2050.

Hymes, D. (1974). Foundations in sociolinguistics: Anethnographic approach. University Park: Universityof Pennsylvania Press.

Jenkins, J. (2006a). Points of view and blind spots: ELFand SLA. International Journal of Applied Linguis-tics, 16 , 137–162.

Jenkins, J. (2006b). Current perspectives on teachingworld Englishes and English as a lingua franca.TESOL Quarterly, 40 , 157–181.

Kant, E. (1946). Critique of pure reason (J. M. D. Meikle-john, Trans.). New York: Dutton. (Original workpublished 1781)

Kasper, G. (1997). A stands for acquisition: A responseto Firth and Wagner. Modern Language Journal , 81,307–312.

Kasper, G. (2004). Participant orientations in Germanconversation-for-learning. Modern Language Jour-nal , 88, 551–567.

Konishi, K., & Tarone, E. (2004). English constructionsused in compensatory strategies: Baseline data forcommunicative EFL instruction. In D. Boxer & A.D. Cohen (Eds.), Studying speaking to inform secondlanguage learning (pp. 174–198). Clevedon, UK:Multilingual Matters.

Koshik, I. (2005). Beyond rhetorical questions: Assertivequestions in everyday interaction. Amsterdam: JohnBenjamins.

Kramsch, C. (1993). Context and culture in language teach-ing . Oxford: Oxford University Press.

Kramsch, C. (Ed.). (2002). Language acquisition and lan-guage socialization: Ecological perspectives. London:Continuum.

Kramsch, C. (2006). The multilingual subject. Interna-tional Journal of Applied Linguistics, 16 , 99–110.

Kurhila, S. (2006). Second language interaction. Amster-dam: John Benjamins.

Lantolf, J. P. (1996). SLA theory building: Letting all theflowers bloom! Language Learning , 46 , 713–749.

Lantolf, J. P. (Ed.). (2000). Sociocultural theory and sec-ond language learning . Oxford: Oxford UniversityPress.

Lantolf, J. P., & Appel, G. (Eds.). (1994). Vygostkian ap-proaches to second language acquisition. Norwood,NJ: Ablex.

Lantolf, J. P., & Thorne, S. L. (2006). Sociocultural the-ory and the genesis of second language development .Oxford: Oxford University Press.


Larsen-Freeman, D. (1983). The importance of inputin second language acquisition. In R. Andersen(Ed.), Pidginization and creolization as languageacquisition (pp. 87–93). Rowley, MA: NewburyHouse.

Larsen-Freeman, D. (1997). Chaos/complexity scienceand second language acquisition. Applied Linguis-tics, 18, 141–165.

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, 603–607.

Larsen-Freeman, D. (2007). On the complementarityof chaos/complexity theory and dynamic systemstheory in understanding the second language ac-quisition process. Bilingualism: Language and cog-nition, 10 , 35–37.

Lave, J., & Wenger, E. (1991). Situated learning: Le-gitimate peripheral participation. Cambridge: Cam-bridge University Press.

Lazaraton, A. (2002). A qualitative approach to the valida-tion of oral language tests. Cambridge: CambridgeUniversity Press.

Lazaraton, A. (2004). Gesture and speech in the vocab-ulary explanations of one ESL teacher: A micro-analytic inquiry. Language Learning , 54, 79–117.

Liddicoat, A. (1997). Interaction, social structure, andsecond language use. Modern Language Journal ,81, 313–317.

Lilja, N. (2006, November). Language learning opportu-nities in everyday conversation. Paper presented atthe 2006 AFinLA conference, Jyvaskyla, Finland.

Long, M. H. (1993). Assessment strategies for SLA the-ories. Applied Linguistics, 14, 225–249.

Long, M. H. (1996). The role of the linguistic envi-ronment in second language acquisition. In W.C. Ritchie & T. K. Bhatia (Eds.), Handbook of sec-ond language acquisition (pp. 413–468). San Diego,CA: Academic Press.

Long, M. H. (1997). Construct validity in SLA research.Modern Language Journal , 81, 318–323.

Long, M. H. (1998). SLA: Breaking the siege. Universityof Hawai’i Working Papers in ESL, 12, 79–129.

Lyotard, J. F. (1979). La condition postmoderne: Rapportsur le savoir (The post-modern condition: A reporton knowledge). Paris: Minuit.

Markee, N. (2000). Conversation analysis. Mahwah, NJ:Erlbaum.

Markee, N., & Kasper, G. (2004). Classroom talks: Anintroduction. Modern Language Journal , 88, 491–500.

Mehan, H. (1979). Language lessons. Cambridge, MA:Harvard University Press.

Mondada, L. (2006, June). Contingent opportunities forlearning: Sequential organization of participationchanges (some thoughts about CA’s contributions tothe study of learning). Paper presented at the Sum-mer School 2006 in Odense, Denmark.

Mondada, L., & Pekarek-Doehler, S. (2004). Second lan-guage acquisition as situated practice: Task accom-

plishment in the French second language class-room. Modern Language Journal , 88, 501–518.

Mori, J. (2004a). Negotiating sequential boundaries andlearning opportunities: A case from a Japaneselanguage classroom. Modern Language Journal , 88,536–550.

Mori, J. (2004b). Pursuit of understanding: Rethinking‘negotiation of meaning’ in view of projected ac-tion. In R. Gardner & J. Wagner (Eds.), Second lan-guage conversations (pp. 157–177). London: Con-tinuum.

Nguyen, H. (2003). The development of communicationskills in the practice of patient consultations amongpharmacy students. Unpublished doctoral disserta-tion, University of Wisconsin, Madison.

Ohta, A. (2001). Second language acquisition processes inthe classroom. Mahwah, NJ: Erlbaum.

Ortega, L. (2005). Methodology, epistemology, andethics in instructed SLA research: An introduc-tion. Modern Language Journal , 89 , 317–327.

Pavlenko, A. (2002). Poststructuralist approaches to thestudy of social factors in second language learn-ing and use. In V. Cook (Ed.), Portraits of the L2user (pp. 257–302). Clevedon, UK: MultilingualMatters.

Pavlenko, A., & Lantolf, J. P. (2000). Second languagelearning as participation and the (re)constructionof selves. In J. P. Lantolf (Ed.), Sociocultural theoryand second language learning (pp. 155–177). Ox-ford: Oxford University Press.

Poulisse, N. (1997). Some words in defense of the psy-cholinguistic approach: A response to Firth andWagner. Modern Language Journal , 81, 324–328.

Rampton, B. (1987). Stylistic variability and not speakingnormal English: Some post-labovian approachesand their implications for the study of interlan-guage. In R. Ellis (Ed.), Second language acquisitionin context (pp. 47–58). London: Prentice Hall.

Rampton, B. (1995). Crossing . London: Longman.Rampton, B. (1997a). Second language research in late

modernity: A response to Firth and Wagner. Mod-ern Language Journal , 81, 329–333.

Rampton, B. (1997b). A sociolinguistic perspective onL2 communication strategies. In E. Kellerman &G. Kasper (Eds.), Advances in research on commu-nication strategies (pp. 279–303). Oxford: OxfordUniversity Press.

Rasmussen, G., & Wagner, J. (2002). Language choicein international telephone conversations. In K. K.Luke & T.-S. Pavlidou (Eds.), Telephone calls (pp.111–131). Amsterdam: John Benjamins.

Richards, K., & Seedhouse, P. (Eds.). (2004). Applyingconversation analysis. London: Palgrave Macmil-lan.

Sacks, H. (1992). Lectures on conversation (Vols.1 & 2).Oxford: Blackwell.

Sacks, H., Schegloff, E. A., & Jefferson, G. (1974). Asimplest systematics for the organization of turn-taking for conversation. Language , 50 , 696–735.

Saussure, F. de (1922). Cours de linguistique generale (Acourse in general linguistics). Paris: Payot.


Schegloff, E. A. (2000). When ‘others’ initiate repair.Applied Linguistics, 21, 205–243.

Seedhouse, P. (2004). The interactional architecture of thelanguage classroom: A conversation analysis perspec-tive . Oxford: Blackwell.

Seedhouse, P. (2005). Conversation analysis and lan-guage learning. Language Teaching , 38, 165–187.

Seeley, A., & Carter, B. (2004). Applied linguistics as socialscience . London: Continuum.

Seidlhofer, B. (2001). Closing a conceptual gap: Thecase for a description of English as a lingua franca.International Journal of Applied Linguistics, 11, 133–158.

Seidlhofer, B. (2004). Research perspectives on teach-ing English as a lingua franca. Annual Review ofApplied Linguistics, 24, 209–239.

Svennevig, J. (2003). Echo answers in native/non-nativeinteraction. Pragmatics, 13, 285–309.

Svennevig, J. (2004). Other-repetition as display of hear-ing, understanding and emotional stance. Dis-course Studies, 6 , 489–516.

ten Have, P. (1999). Doing conversation analysis. Thou-sand Oaks, CA: Sage.

Theodorsdottir, G. (2007, April). Insisting on your turn—Second language learner’s struggle for turn completion.Paper presented at the 17th International Con-ference on Pragmatics and Language Learning,Honolulu, HI.

Thorne, S. L. (2000). Second language acquisition the-ory and some truth(s) about relativity. In J. Lan-tolf (Ed.), Sociocultural theory and second languagelearning (pp. 219–243). Oxford: Oxford Univer-sity Press.

van Lier, L. (1994). Forks and hope: Pursuing under-standing in different ways. Applied Linguistics, 15,328–346.

Volosinov, V. N. (1973). Marxism and the philosophy oflanguage . New York: Seminal Press. (Translated byLadislav Matejka and I. R. Titunik. Original workin Russian published in 1930).

von Glasersfeld, E. (1984). An introduction to radicalconstructivism. In P. Watzlawick (Ed.), The inventedreality (pp. 17–40). New York: Norton.

von Glasersfeld, E. (1995). A constructivist approachto teaching. In L. P. Steffe & J. Gale (Eds.), Con-

structivism in education (pp. 3–16). Hillsdale, NJ:Erlbaum.

Vygotsky, L. S. (1978). Mind and society: The developmentof higher mental processes. Cambridge, MA: HarvardUniversity Press.

Wagner, J. (1995a). Negotiating activity in technicalproblem solving. In A. Firth (Ed.), The discourseof negotiation. Studies of language in the workplace(pp. 223–246). Oxford: Pergamon.

Wagner, J. (1995b). What makes a discourse a negotia-tion? In K. Ehlich & J. Wagner (Eds.), The discourseof business negotiation (pp. 9–36). Berlin, Germany:Mouton de Gruyter.

Wagner, J. (1996). Language acquisition through for-eign language interaction—A critical review ofstudies on second language acquisition. Journal ofPragmatics, 23, 215–235.

Wagner, J. (1998). On doing being a guinea pig—Aresponse to Seedhouse. Journal of Pragmatics, 30 ,103–113.

Wagner, J. (2004). The classroom and beyond. ModernLanguage Journal , 88, 612–616.

Wagner, J., & Firth, A. (1997). Communication strate-gies at work. In E. Kellerman & G. Kasper(Eds.), Advances in research on communication strate-gies (pp. 323–344). Oxford: Oxford UniversityPress.

Wagner, J., & Gardner, R. (2004). Introduction. In R.Gardner & J. Wagner (Eds.), Second language con-versations (pp. 1–17). London: Continuum.

Wagner, J., & Pekarek-Doehler, S. (2006, May). Progres-sive bricolage of turn constructional units in secondlanguage talk. Paper presented at the InternationalConference 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 Hemp-stead, UK: Harvester-Wheatsheaf.

Wong, J. (2000a). Delayed next turn repair initiation innative/non-native speaker English conversation.Applied Linguistics, 21, 244–267.

Wong, J. (2000b). Repetition in conversation: A lookat ‘first and second sayings’. Research on Languageand Social Interaction, 33, 407–424.


APPENDIXTranscript Conventions

[] Left and right brackets indicate beginning and end of overlap.(0.0) 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

WORD Upper case indicates especially loud sounds relative to the surrounding talk.◦word◦ Degree signs bracketing a sound, word, phrase, and so on, indicate especially soft sounds relative to the

surrounding talk.·hhh A raised dot-prefixed row of h’s indicates an inbreath. Without the dot, the h’s indicate an outbreath.word Underlining indicates stressed syllablesxxx unintelligible speech= latching between turns or parts of turns∗ creaky voice

firth and wagner 2007

Documents