for a social reconceptualization of central tenets of second language acquisition (SLA) research, this special issue of The Modern Language Journal focuses on practices for teaching and learning a second language (L2) with special attention to the details of sociointeractional contexts of teaching and learning behaviors/activities. Its goal is to unveil learning processes and practices as socially observable phenomena in situ and in vivo and to discuss pedagogical implications of the findings. As such, the issue focuses on some well-established concepts from the SLA field, including noticing, attention, and corrective feedback, but aims to explore and reconceptualize them in terms of social displays of behavior and social practices as seen through the lens of conversation analysis. This Introduction sets the stage for the articles in the special issue by tracing SLA's interest in socio-interactional aspects of learning before moving on to a brief discussion of the epistemology of CA. We then outline the ways in which the individual articles empirically contribute to a social understanding of learning and cognition in SLA, before summarizing the main points addressed in the special issue.

This study aims to show how multimodality, that is, the mobilization of various communicative resources in social actions (Mondada, 2016), can be used to teach grammar. Drawing on ethnomethodological conversation analysis (Sacks, 1992), the article provides a detailed analysis of 2 corrective feedback sequences in a Swedish-as-a-second-language classroom. It shows that teaching grammar using corrective feedback sequences is a collaborative activity between teachers and students, which requires both verbal and other embodied practices. Specifically, it demonstrates how the teachers made grammatical constructs visible, noticeable, and thus learnable through the use of multiple resources such as annotating and illustrating on a whiteboard or projection screen, using concrete meta-talk (Storch, 2008), together with nonverbal actions such as gesturing. The article argues that the practice of marking a linguistic structure through multiple resources creates 'landmarks' for teaching purposes. These landmarks were used (a) for an instructed vision through which the intelligibility of abstract grammatical concepts and relations as cognitive phenomena is constituted by a concrete set of observable and reportable actions, and (b) as prompts in organizing knowledge not only for the purpose of the current activity of teaching but also for future occasions.MULTIMODALITY HAS BEEN A TOPIC OF inquiry for some time in the domain of textual analysis, quite simply because most contemporary texts are no longer just straight word-based textual artifacts but show an intricate interlacing of multimodal resources in building texts. In contrast with a textual-analytical stance, the present study adopts an explicitly pedagogical orientation from an ethnomethodological conversation analytic (EM/CA) point of view (Mondada, 2014(Mondada, , 2016. The study aims to show how multimodality, that is, the use of multiple communicative resources, such as talk, body, and available artifacts, is indispensable to action-building in human activities including language teaching, here