In this study, the authors explore how classroom tasks that are commonly used in task-based language teaching (TBLT) are achieved as observable aspects of local educational order (Hester & Francis, 2000) through observable and immanently social classroom behaviors. They focus specifically on students' language learning behaviors, which they track through the longitudinal conversation-analytic methodology called learning behavior tracking (LBT) (Markee, 2008). From a theoretical point of view, they situate LBT within the ethnomethodological (EM) perspective on social action pioneered by Garfinkel (1967) and relate it to socially defined ways of understanding planning (Burch, 2014;Markee & Kunitz, 2013). In the empirical part of the article, the researchers analyze TBLT work that was conducted in an English as a foreign language (EFL) classroom in a Swedish junior high school. Specifically, they track the occurrences of a learnable (the spelling of the word disgusting) that was emically oriented to as such by the students as they engaged in planning and accomplishing teacher-assigned tasks. The authors then develop an emic, sequential account of the participants' practical reasoning and dynamically evolving epistemic positions. They argue that this kind of basic empirical research refines our understanding of how TBLT curriculum work is achieved by participants as practical, mundane, and observable activities in language classrooms, and that these insights may feed into more applied research on teacher training, thereby fostering the design of instructional innovations.

In this article, we bring together conversation analysis and usage-based linguistics to investigate the second language (L2) developmental trajectory of a linguistic construction within the complex multimodal ecology of naturally occurring social interaction. We document how, over the course of 15 months, an L2 speaker's use of the French multiword expression comment on dit [how do you say] diversifies in both form and function. Two types of longitudinal change are observed: (a) The expression expands in its context of use: "Literal" uses are observed initially to request a candidate lexical item but are later also found in requests for confirmation, (b) these literal uses become proportionally less frequent, and the expression progressively routinizes as a marker-like element used for indexing cognitive search and floorholding, and eventually also as a preface to self-correction. This routinization entails erosion in form and meaning, in concert with systematic change in co-occurring bodily-visual conduct, in particular gaze and gesture. By documenting change over time in the functional use and the multimodal delivery of the target construction, the findings evidence the longitudinal development of L2 grammar-for-interaction and showcase how linguistic and bodily resources may interface in L2 development. They also have important implications for language teaching and learning.

This paper reviews the research on the development of second language (L2) interactional competence. We first provide an overview of the historical advance-

No abstract