本研究は言語教師が教室の中で日常的に行う<教師>としての行為を複数ある行為の定式の中からなぜ特定の定式をどのような手続きで行うのかを明らかにすることを目的に、「授業態度への注意」という1つの教師的行為の定式化手続きを会話分析によって調査した。日本の大学英語授業を録音採集したコーパスデータの中から1つの事例を詳細に分析した結果、学生の授業態度への注意という行為を効果的に行うために、教師は教師と学生の<年齢差>という持ち運び可能アイデンティティを基にした尺度化という手続きで当該行為を効果的に行うということを発見した。今後、<年齢>や他の持ち運びアイデンティティを用いて教師の行為を成す定式化手続きについて成功例だけではなく失敗に至る手続きに関しても記述する研究が行われることが、持ち運びアイデンティティを用いない教師的行為の定式化手続きを明らかにする研究とともに行われることが望まれる。 Language teachers perform a lot of actions in the language classroom associated with their situated role of “teacher” such as asking questions, correcting students’ errors, and explaining grammatical points. Because the classroom is organized by the teacher’s and the students’ enactment of actions tied to their roles of teacher and student (see Aline & Hosoda, 2006; Seedhouse, 2004), how a teacher executes his or her action in the classroom has a great influence on the teacher’s teaching and students’ learning as well as the teacher’s classroom management conducted within the classroom interaction. Thus, the procedures that language teachers employ to practice their everyday classroom actions should be considered a major part of the teacher’s classroom interactional competence (Walsh, 2012). It would be beneficial for both pre- and in-service language teachers to document good and bad methods of undertaking their actions in the language classroom. This study was aimed at describing the ways that language teachers formulate the actions associated with their situated role of teacher in classroom talk. The study employed conversation analysis (CA) as the method of analysis. CA’s objective is “to uncover the tacit reasoning procedures and sociolinguistic competencies underlying the production and interpretation of talk in organized sequences of interaction” (Hutchby & Wooffitt, 2008, p. 12). CA explicates such competencies of the participants through the emic analysis of interaction: CA treats an interaction itself as a system that is organized by sequences of actions, and the participants publicly display their analysis of each other’s action in the next position in the system. So, the analyst’s job is to naturalistically document the competences participants show in their conduct of actions in interaction. The emic analysis of CA offers a way to explicate language teachers’ competence to formulate an action in the language classroom. The data for the analysis was taken from a corpus of 900 minutes of audio-recorded EFL classroom interaction from four different courses at a Japanese university. In this study one perspicuous case was analyzed in which the teacher formulated the action of giving a warning to a student about class participation. The conversation began when the teacher asked a question to a student about the reason why she had been absent a lot. The student answered that she had a ligament injury some months before. Instead of accepting the student’s account, the teacher said that he had the same injury and although his was much worse than hers, he had not missed any class. Here, the teacher’s formulation of the action of giving a warning to a student was practiced through scaling himself and the student in terms of the seriousness of the injury. However, the formulation was undermined by the student’s reluctance to agree that the teacher’s injury was worse than hers. Then, the teacher reformulated the action: He re-scaled the relationship between himself and his student in terms of the transportable identity age and speed of recovery from injury based on the age difference between the teacher and the student. This time, the student accepted the teacher’s warning about her class participation. The difference in the two formulations is their foundation. The initial scaling, seriousness of injury, was based on a verbal account. There was no visible proof to explain the seriousness of the injury such as a supporter, a set of crutches, a cast, or a medical certificate. In other words, the trajectory of the scaling depended on whether or not the student believed the teacher’s verbal account. On the other hand, another scaling, age difference, is visibly available in the interaction between a teacher who is over 60 years old and students who are around 20 years old. The other scaling, speed of recovery, was based on their age differences: It is normatively understood that an older person takes more time to recover from injury than a younger person. Thus, it can be said that the trajectory of a proposed formulation of an action largely depends on the extent to which it is (non)negotiable. It is suggested that, although there are some concerns (see Richards, 2006), to completely ban teachers from orienting to their own or their students’ transportable identities deprives them of a way to conduct their jobs effectively. Therefore, further studies should examine whether teachers’ use of transportable identities is useful for formulating the actions associated with the role of teacher in the language classroom and in other environments as well as whether there are other types of formulation procedures for effectively achieving their work goals.