Keywords

1 A (Historical) Prelude: The Demands of the Social World and the Advancement of Research

Let me start with a prelude. The importance of interactional competence (IC) for people’s participating in the social world—be it in their L1 or in their L2, L3, etc.—cannot be overestimated in the twenty-first century. The emergence of a knowledge- and service-based economy as well as the growing diversification/globalization of our economic and cultural landscapes highlight in unprecedented ways the importance of people’s adaptive capacities and mastery of communicative tools (see the papers collected in Pekarek Doehler et al. 2017). IC in different languages is a central component of the wider social abilities by which people gain access to multiple institutional and social worlds, are recognized as members in the related communities of practice, learn, construct their identities, pass through processes of educational or professional selection, socialize in the workplace, and much more: IC, including in an L2, is instrumental in people’s being in and moving through the social world.

Yet, it is exactly this competence that represents a central stumbling stone when it comes to teaching and testing languages around the world, across settings, methods or cultures. We know from experience how, after 6 or 8 years of L2 learning in the classroom, we (or others) can find our(them)selves helpless when it comes to engaging in spontaneous L2 interaction. One may reasonably argue that the problem lies in the very nature of the object at stake, i.e., the intricate abilities it takes to manage the situated dynamics of social interaction. Yet, one may also reasonably argue that at least part of the issue is due to the relatively limited knowledge we currently have about the nature and, in particular, the development of these abilities. Today, we look back on more than a century of research on language structure in modern linguistics, of which more than half includes research on the development of L2 grammar, linguistic forms or form-function mappings. By contrast, we have so far witnessed merely a decade of empirical research into the development of L2 IC—albeit backed up by some 50 years of research in conversation analysis (CA) concerned with “the competences that ordinary speakers use and rely on in participating in intelligible, socially organized interactions” (Heritage and Atkinson 1984, p. 1), which, however, has not been developmental in nature.

Importantly, the emergence of the notion of IC in the field of SLA cannot be reduced to highlighting one subcomponent of language learning. Rather, it is symptomatic for a shift in our very understanding of what language learning is. Throughout the past two decades, L2 learning has been increasingly understood as the development of linguistic means for engaging in the social world, as a fundamentally socio-cognitive process, not enclosed inside the individual’s cognition, but driven through language use, the prototypical site of which is social interaction (cf. Firth and Wagner 1997, 2007). The construct of interactional competence can therefore be seen as spearheading a new perspective on language learning in the field of SLA—a perspective that has important implications for language teaching, as evidenced in the contributions to this volume (see e.g., Eskildsen, this volume; Huth, this volume; and Walters, this volume). Though I will not go into this here, it is important to note that the construct itself has been applied to language teaching, stressing the importance of the detailed ways in which teachers interact with students in the classroom (see Walsh 2011, 2013 on the notion of Classroom Interactional Competence).

Historically, calls for a better understanding of L2 development in light of the dynamic nature of language use in interaction go back to the 1980s (e.g., Kramsch 1986) and initial conceptualizations of the notion of IC have seen the light in the 1990s (Hall 1999; He and Young 1998). Yet, it is only within the past decade (see especially Hellermann’s 2008 book-length study and the papers in Hall et al. 2011), that the development of L2 IC over time has gained systematic attention in empirical SLA research (for state of the art discussions see Skogmyr Marian and Balaman 2018; Pekarek Doehler and Pochon-Berger 2015). The lesson to draw from this, despite the increasing amount of empirical findings that we have available today, is one of modesty: There is still a long way to go.

2 Understanding Interactional Competence and Its Development

2.1 Conceptual Challenges

Arguably, it is exactly in the relative recent nature of empirical research into L2 IC that lies an opportunity—but also a challenge. We know from experience that, when the picture available is not yet fully rendered, there is promising space for discussion, adaptation, and mutual influence between different vantage points, not only as to the conceptual implications of the diverse evidence they offer, but also as to the very questions they raise. When it comes to issues of L2 acquisition and L2 teaching, in order to be productive, such a dialogue needs to be grounded in a mutually compatible understanding of language, learning, and ultimately (interactional) competence.

The fact, however, is that the concept of IC and the related understandings of L2 learning are not solidly ‘out there’, that is, are not substantially addressed within L2 policies or curriculum design, teacher training or classroom practice. Once we leave the field of CA-SLA research, IC often remains only vaguely circumscribed or tends to be conflated with the notion of ‘communicative competence’ (see below). More generally, the socio-cognitive nature of L2 learning as anchored in language use in interaction is often overshadowed by the dominant focus on individual learners and their cognitive processing. There is hence an urgent need for spelling out, based on empirical research on IC and its development, a coherent understanding of the notion and of its implications.

Within current SLA research, the conceptualization of IC is grounded in a socio-constructivist understanding of cognition, competence, and learning as fundamentally situated, distributed (Lave 1988; Hutchins 1995), locally accomplished in and through social interaction (Garfinkel 1967): IC is viewed as an ability for joint contextually contingent action (see below). Such an understanding, however, fits quite uneasily with cognitivist views of language and of competence as properties of the individual, which have historically provided the theoretical backbone for frameworks for L2 teaching at several levels of granularity. Such contrasting conceptualizations represent a central challenge for developing implications for L2 education based on results from L2 research. This is so because views of language and of learning have a structuring effect on curricula design as well as classroom practices, and therefore contribute to shaping local affordances for language development within instructional settings.

2.2 An Epistemologically Coherent and Empirically Validated Definition of the Target Object: From Communicative Competence to Interactional Competence

The notion of IC cannot be reduced to an expansion of the target object of L2 learning to include interactional abilities in addition to linguistic, pragmatic or socio-cultural ones; rather, as mentioned above, the notion of IC reflects a shift, within the field of SLA, in our very understanding of what language learning is. The prominent lines of SLA research have for long been grounded in a fundamentally monologic and individualistic language ontology, concerned with linguistic form, form-function mappings, and individual cognitive (input) processing (for earlier critiques of such a view see Markee 1994; McNamara 1997; Firth and Wagner 1997). As a consequence, contextual communicative practices and the organization of social interaction have not been a concern for mainstream SLA, and social interaction tended to be either left out of the picture, or treated as a mere setting (among others) allowing for the acquisition of linguistic forms (see e.g., the Interaction Hypothesis, Long 1996, and ensuing work).

With Hymes’ (1972) conceptualization of communicative competence, the field of SLA saw a groundbreaking shift toward a more holistic understanding of language use, yet without embracing the dynamic nature of language use in and for social interaction. Ensuing Hymes’ work, sociolinguistic abilities have been foregrounded, relating to culturally specific norms of conduct (e.g., politeness), as well as pragmatic abilities, pertaining to the realization of speech acts (e.g., requests) or to issues of discourse coherence (e.g., discourse markers). The distinctive feature of these developments—which differentiates them from current concerns with IC—is their focus on social conventions rather than on locally situated procedures for action. Furthermore, while research on communicative competence has substantially advanced our understanding of the spoken modality, it has largely remained attached to a monologic perspective. In Canale and Swain’s (1980) and Canale’s (1983) work, for instance, communicative competence has been subdivided into linguistic, sociolinguistic, strategic, and discursive competence, with a focus on the individual production of the learner, rather than on the learner’s participation in social interaction and the related process of mutual adaptation. In this context, the notion of competence has furthermore tended to be conceptualized as a decontextualized cognitive property of the individual, that is, a competence that is put to use within language practice independently from the situational context of such use and from the co-participants’ actions.

Now, such an understanding strongly contrasts with more recent socio-cognitivist and socio-constructivist conceptualizations of human cognitive functioning (e.g., Hutchins 1995; Lave 1988; Rogoff 1990; Wertsch 1991) and of language learning (Firth and Wagner 1997) as profoundly contextual, i.e., contingent upon the local circumstances of use. From this perspective, competence is not an abstract property enclosed in the brain of the individual, but is situated and hence continually adapted to the local circumstantial details of its use within people’s acting in the social world. As Wertsch (1991) put it: “Human mental functioning is inherently situated in social interactional, cultural, institutional and historical contexts” (p. 86). These developments have also radically put into question classical dichotomies regarding cognition—such as the distinction between individual and social processes, abstract capabilities and contextualized ones, and ultimately competence and performance (for SLA see Firth and Wagner 1997). For instance, in a study on arithmetic tasks, Lave (1988) documented 30 years ago already that participants tend to perform better in practical real-life situations (such as calculating prices on the market) than when solving tasks of the same degree of difficulty in formal tests. This provides a speaking example of how competencies (even those relating to such ‘hard-core’ issues as mathematics) are situated in context, and hence cannot be understood as context-independent cognitive properties or abilities of the individual.

Within the field of SLA, socio-cognitivist and socio-constructivist understandings of L2 learning have been increasingly foregrounded within the past two decades, and it is in this context that the nature and the development of L2 IC has become a central concern. In a pioneering statement, Kramsch argued already in 1986 against what she referred to as an “oversimplified view of human interaction” (p. 367) in SLA, and in the 1990s, researchers started to offer more dynamic and dialogic and contextualized conceptualizations of competence, focusing on social interaction (e.g., Hall 1999; He and Young 1998; Firth and Wagner 1997). Yet, it is only within the past decade that social interaction has started to be empirically investigated as the very object of L2 learning.

To date, the most important advancements in understanding L2 IC and its development have been provided by longitudinal (and in some cases cross-sectional) conversation analytic studies on SLA (CA-SLA; for recent discussions see Pekarek Doehler and Pochon-Berger 2015; Pekarek Doehler 2019; Skogmyr Marian and Balaman 2018). Following CA’s epistemological roots in ethnomethodology, a line of research in sociology, IC has been defined in terms of members’ ‘methods’ (cf. Garfinkel 1967) for organizing social interaction (Hellermann 2008; Pekarek Doehler 2010, 2019; Nguyen 2017). ‘Methods’ are systematic procedures (of turn-taking, opening or closing a story-telling, repairing interactional trouble, etc.) through which participants in an interaction coordinate their actions, accomplish roles and relationships, establish mutual comprehension, and maintain intersubjectivity. These procedures include verbal resources—but also prosodic and embodied resources such as gesture, posture, gaze—that contribute to situated meaning-making and the coherent coordination of mutual actions within social encounters. As part of participants’ public action in conjunction with others, these procedures are observable in the details of participants’ conduct; by virtue of that fact, they are inspectable by the researcher, both for their local deployment and their development over time.

Importantly, the above conceptual grasp of IC is based on a long tradition of empirical CA research that has amply documented that ‘competent’ members (typically L1 speakers) have at their disposal alternative methods for getting the same interactional jobs accomplished. For instance, they may have different ways of showing disagreement (use of polarity marker of the type ‘no’, or more subtle turn-constructional formats such as ‘yes…. but…’; Pomerantz 1984). They choose between these alternative methods according to the local circumstances of their interaction, which allows them to deploy conduct that is adapted to the situation at hand and to their precise interlocutors, i.e., conduct that is context sensitive and recipient-designed (Sacks et al. 1974). The availability of alternative methods is exactly what L2 speakers often lack, which entails limited adaptive abilities on their part (see below).

In a nutshell, then, IC consists of the ability to deploy procedures for the management of social interaction (turn-taking, opening or closing a conversation, disagreeing, initiating a story-telling, and so forth) in ways that are relevant, i.e., adapted, to the local circumstances of the interaction and to the specific others who participate therein. IC includes both the ability to understand the interactional context and the expected practices therein, and to deploy locally relevant conduct based on verbal and non-verbal resources. This understanding hence highlights the socially situated and distributed nature of IC as an ability to act conjointly with others.

2.3 An Empirically Grounded Understanding of the Developmental Trajectories of L2 Interactional Competence

CA-SLA studies on IC have investigated several of the abovementioned types of interaction-organizational procedures. The cumulative evidence stemming from investigations on such diverse objects as turn-taking, disagreeing, opening tasks and story-tellings shows that when interacting in their L2, speakers build on interactional abilities they had developed since infancy, yet they also re-calibrate, re-adapt these as part of their developing IC in the L2. Beginner L2 speakers may for instance employ only basic methods for turn-taking (such as soliciting someone by name, or raising one’s voice; Cekaite 2007), for disagreeing (such as using the polarity marker ‘no’; Pekarek Doehler and Pochon-Berger 2011), or for opening tasks (Hellermann 2008), but then diversify these over time in the process of becoming more efficient L2 speakers. This process of course centrally involves linguistic resources; over time, these become invested with new, specifically interactional, functions. For instance, in Korean L2 the use of the connective kutney (roughly corresponding to English ‘but’) as a disagreement marker has been shown to emerge only over time (Kim 2009), although the form itself was available to the L2 speaker earlier on. For English L2, the expression what do you say has been shown to expand in use, first occurring in the sense of ‘how do you say X’, and later on being also used as a request for repetition (in the sense of ‘what did you just say’) and for eliciting co-participant’s opinion (in the sense of ‘what do YOU say/think’) (Eskildsen 2011). And for French L2, comment on dit‚ ‘how do you say’, has been shown to progress in use from doing a request for translation to working additionally as a marker of cognitive search and a floor-holding device (Pekarek Doehler and Berger 2019). These findings testify to the development of an L2 grammar-for-interaction as an integral part of L2 IC (Pekarek Doehler 2018).

Diversification of speakers’ procedures for dealing with practical interactional issues as well as expansion of the interaction-functional realm of precise grammatical resources are hence key characteristics of the developmental trajectory of L2 IC over time. And this has been documented both in classroom studies and in studies on interactions outside of the classroom. It is exactly this diversification/expansion that allows speakers to use language for the purpose of coordinating social interaction, and to adapt their conduct to the local situational constraints and to the precise others they are interacting with, i.e., to deploy conduct that is increasingly context-sensitive and recipient-designed (cf. Sacks et al. 1974). This is what makes L2 speakers increasingly ‘competent’ as members of the L2 community in which they act and interact.

In sum, the conceptualization of IC in terms of members’ ‘methods’ is in line with a conception of learning and of competence as situated and mutually adaptive: Learning a language is defined as a social practice (learning-in-action, Firth and Wagner 2007), and IC as an ability for joint action, that is co-constructable, i.e., shaped through the participants’ mutual actions, and contingent upon the details of the social interactions L2 speakers participate in (competence-in-action, Pekarek Doehler 2010). This means that IC is understood to emerge from members’ cumulative experience of social interactions while continuously being adapted in the course of such interactions: IC is not simply brought along by individuals to new situations, but is brought about, in interaction with others, by the local circumstantial details of the social interaction.

3 Longitudinal Studies on the Development of L2 IC and Their Implications: The In-Principle Affordances of the Classroom and Beyond

The conceptual developments and empirical findings in the field of CA-SLA research as outlined above boil down to a deconstruction of the competence-performance dichotomy: Competence is understood as a competence for interaction, and as a competence that grows out of interaction. While CA work in the field specifically focuses on practices (or: ‘methods’) and linguistic resources for interaction, it also fundamentally resonates with larger usage-based approaches that evidence how linguistic constructions emerge from language use (Ellis and Larsen-Freeman 2006)—and often from language use in interaction (e.g., Eskildsen 2015). As a consequence, participation in social interaction is seen as key to learning.

While research on L2 IC stresses the need for adaptation and diversification of resources, based on the conceptual apparatus of CA and an ethnomethodologically grounded understanding of IC, it also focuses on generic principles of interaction: turn-taking organization, repair organization, sequence organization (i.e., the organization of turns into ‘pairs’ such as question-answer), and the larger structural organization (i.e., conversational openings and closings). This is where the conceptual and epistemological foundations of current work on CA-SLA come to play a key role in view of identifying the opportunities offered by classroom interaction for IC development: As generic principles of social interaction are at work in any situation—institutional or not—they can in principle be ‘practiced’ in any social interaction. It is the ways that these principles are managed—i.e., the methods and resources speakers deploy for organizing interaction—that vary in context-sensitive ways. This has important implications for how we see the classroom as an opportunity space for interaction, and for the development of L2 IC.

We know from ample research on classroom interaction that the L2 classroom is a diversified interactional arena (e.g., Markee 2000; Sert 2015; Seedhouse 2004; Walsh 2006; Waring 2015; see also many of the papers collected in this volume and in Markee 2015), offering in principle a plethora of opportunities for L2 interactional development. One of the key issues for developing sound measures in view of favoring the development of L2 IC in and through classroom interaction is to tease apart what can reasonably be taught or practiced within the classroom and what cannot effectively be addressed inside the classroom, and to identify how out-of-classroom experiences can be made profitable within the classroom.

To give just a couple of examples, from our own research: There is evidence, for instance, that practices for doing L2 disagreements in classroom interaction diversify across time within the classroom, in ways that bring the L2 students closer to what we know from L1 speakers. Some years ago we conducted a cross-linguistic study on disagreements, comparing intermediate and advanced (9th and 12th grade) French L2 students in a German-speaking environment (Pekarek Doehler and Pochon-Berger 2011). While disagreements were not an explicit target of L2 instruction, debates on topical and potentially controversial issues (abortion, the military, environmental policies, etc.) provided ample opportunities for disagreeing with others, and such debates were implemented at both levels of schooling. The comparison between the two levels showed that the L2 students developed their abilities for doing disagreements through the very fact of interacting in the L2 within the classroom, and without disagreement having been the target of instruction or structured classroom practices. At lower level of proficiency students tended to uniformly do disagreements through the use of turn-initial polarity markers, such as non ‘no’, while at upper levels they diversified their practices, using for instance disagreement prefaces within a ‘yes-but’ structuring of their disagreeing turns. These findings suggest that the development of ‘methods’ for doing disagreement and a range of other dispreferred actions (rejections of requests or invitations, for instance) may be favored by specific types of classroom interaction, such as debates. These very ‘methods’ may, however, also lend themselves to explicit instruction and structured practice in the classroom (Barraja-Rohan 2011; Huth and Taleghani-Nikazm 2006; Wong and Waring 2010).

The above results resonate with findings on L1 development. For instance, in a range of studies on young people’s L1 IC on the transition between lower-secondary and upper secondary schooling (and the workplace), we identified a strong continuity between lower and upper secondary school regarding issues of interactional engagement and participation (see the papers in Pekarek Doehler et al. 2017): Interactional processes within the lower secondary classroom, and in particular teachers’ implicit or explicit encouragements for students to act in precise ways that furthered issues such as interactional engagement, assertiveness, and local adaptation of one’s conduct to ongoing activities, tended to become appropriated by students as patterns of reference guiding their conduct at upper secondary levels. It is exactly these patterns that were shown to be called for in work-related situations, such as job interviews or actual workplaces: The cooperative participatory classroom culture based on students’ initiative, the diversification of turn-taking practices, and the negotiation of knowledge observed at the upper end of the school trajectory reverberates with the increased demands for interactional flexibility encountered in diverse work-related situations. This is a strong argument showing how classroom practice without overt instruction, combined of course with out-of-classroom socialization processes, profited IC development (in an L1).

The above examples—along many others (see recently Watanabe 2017 on the development of turn-taking and participation in the L2 classroom; see also Eskildsen this volume, on the development of embodied interactional resources)—stress the fact that we need to learn much more about how social interaction within the classroom favors IC development over time (see Pekarek Doehler and Fasel Lauzon 2015 for an overview of longitudinal studies of L2 classroom interaction). This is so because ‘simple’ L2 interaction, that is, interaction that does not specifically target a given learning object, can easily be underestimated as a mere site of putting to use what one has already acquired, yet exactly this same type of interaction can be a key site of mutual adaptation, experimentation, informal instruction (or ‘informal assessment’, Can Daşkın, this volume), and ultimately interactional development.

Other research results draw a less promising picture. We conducted a set of studies on au pairs who had had years of L2 instruction before immersing into a stay of several months in an L2 environment. Results showed that some aspects of their IC developed relatively late in their overall learning trajectories, but change in these occurred relatively fast once the L2 speakers were immersed in everyday L2 use: Such fast L2 development was observed for instance with practices for opening story-telling in recipient-designed ways (Pekarek Doehler and Berger 2018) or for soliciting recipient’s help during word-searches (Pekarek Doehler and Berger 2019) in ways that minimize the disruptiveness of these searches, as well as with the use of grammatical resources for the social coordination of interaction (Pekarek Doehler 2018). Results here suggest that short-term total immersion through a stay in an L2-speaking environment (even 2 or 3 months) out-weighs long-term classroom instruction with regard to selective aspects of IC. This, of course, calls for more extended research on the multiple facets of IC and how their trajectories develop selectively within precise settings.

Given the massive time-limitation for practicing interaction within the classroom, a central question is how the classroom and ‘the wild’ (i.e., out of classroom language experiences) can be combined to create opportunity spaces for interactional development. This issue is addressed in great detail in the papers collected in Hellermann et al. (2019), which stress the need for a reflexive relationship between the classroom and ‘the wild’ (Wagner 2015): They argue for integrating into school curricula language-learning experiences in out-of-school social interactions, for instance through student-exchanges (as currently practiced throughout many European countries), the assignment of out-of-classroom on-line interactional tasks (Balaman and Sert 2017) or more local integration of opportunities for naturalistic interactions (e.g., Piirainen-Marsh and Lilja 2019). On the one hand, such enterprises can capitalize on the classroom’s power to transform language experience into learning, for instance when participants’ self-recorded out-of-classroom experiences are brought back into the classroom for reflection and teaching purposes (see e.g., Thorne 2013 and the papers in Hellermann et al. 2019); on the other hand, this may take advantage of the complementary opportunities for learning that out-of-classroom naturalistic interactions offer compared to classroom instruction. The importance of such endeavors cannot be overestimated in light of what we know, today, about L2 IC and its development.

4 Conclusion

In a recent paper discussing the interaction between L2 speakers in the classroom and out-of-classroom L2 experiences, Wagner (2019) argues for an ethnomethodologically and sociologically grounded understanding of learning as the keystone for a new experiential pedagogy that is able to prepare L2 speakers for participation in the social world. Such an understanding focuses on the situated and contextualized nature of learning and of competence and sees language use as the driving force for learning, rather than seeing linguistic knowledge as the prerequisite for use. Yet, this understanding, while it is in line with current conceptualizations of IC as outlined above, stands in sharp contrast to cognitive-individualistic views of learning as the internalization of knowledge. It is exactly in such contrasting epistemologies of learning (or teaching) that a key challenge emerges when it comes to formulating SLA research-based implications for L2 education (see also Pekarek Doehler, this volume), and to bridging the gap between research and practice (see the contributions in Salaberry and Kunitz 2019).

5 Implications—In a Nutshell

The conceptual and empirical developments described in this paper have a range of implications for second language education, which are addressed in several contributions to this volume. Some consequences for the classroom have been mentioned above, and I have spelled out other consequences later on in this volume (Pekarek Doehler this volume). In a nutshell:

  • Integrating classroom and out-of-classroom language practices wherever possible. In order to prepare L2 learners for their participation in real-world L2 encounters, classroom practice needs to be more consequentially completed with opportunities for out-of-classroom language experiences (see above), and these experiences should be brought back to the classroom as objects of reflection and of teaching. This means capitalizing on the learning potential of the classroom in ways that are nourished by a wider range of interactional practices than the classroom alone can offer.

  • Integrating IC and the related understanding of L2 learning into teacher training and curriculum design. A convergent epistemology of language learning represents an indispensible basis for bridging the gap between research and practice. The socio-constructivist conceptualization of language learning that emanates from several lines of current research is in need of clarification in the field of language education, and so is the notion of IC, and how it differs from communicative competence. Furthermore, there is a parallel need to raise teachers’ awareness for IC and how it can be observed in social interaction. The arenas for such endeavors are teacher training (see Pekarek Doehler this volume; Sert this volume; Waring this volume) and curriculum design (see Markee’s 1997 classic work on managing curricular innovations).

  • Designing assessment models and practices that recognize the social, i.e., mutually adaptive, nature of language use in interaction (see already McNamara 1997). We need operational criteria for assessing IC (e.g., Kley 2019; Walters this volume), practicable testing designs and situations (e.g., Huth and Betz 2019), and ultimately recognition of the fact that IC is rooted in jointly acting with others and that—consequently—the testee’s acting is inevitably contingent upon the tester’s acting.

  • Adapting current reference frameworks for L2 teaching. Existing reference frameworks such as the CEFR call for a specification of the (often vague) descriptors for ‘interactional competence’, and for a moving away from its treatments as just one sub-component of ‘speaking’, adjoined in an additive manner to other components such as accuracy, fluency of coherence. Reference frameworks need to be better aligned with the current state of research which offers a more encompassing understanding of IC that highlights social interaction as the typical (and ontogenetically as well as phylogenetically primary) site of language use, and understands language, learning, and competence as fundamentally situated, emerging in and through social interaction. Such an understanding, however, stands in sharp contrast to the very epistemological foundations of existing reference frameworks that continue to be indebted to a monologic and individualistic view of L2 learning and use (see Huth this volume).