Although, on first take, the articles in this special issue may look like an indictment of action categorization, a closer examination of the evidence suggests that what is called for is a renewed understanding of action categorization. The keys to this renewed understanding include treating action names as arbitrary labels for identifiable patterns of accountable behavior, recognizing the gradient nature of social behavior in interaction, and acknowledging the cultural and institutional sensitivity of action categorization.
Actions are usually described as generic types by drawing on context-free criteria such as who benefits or performs the actions. Dealing with a specific type of action tied to the setting in which it occurs-that is, actors' enactments in theater rehearsals-we argue that action labels sometimes must be conceptualized in more setting-specific ways, including factors such as participation framework, division of labor, activity types, and practical tasks. Drawing on a corpus of 400 hours of video-recordings of professional theater productions, we show that actors' enactments responding to directors' instructions are not simply implementations of the instruction but propose a possible embodied solution of how a scene could be played. Responsive enactments have a high degree of authorship, as they contain elements that were not part of the original instruction. Enacted proposals do not come in specialized turn-formats and do not formulate future actions. Data are in German.
This study explores the interactional work being done by speakers of American English when they use the format "IWIQ" (I wonder + indirect question; as in, "I wonder how that actually works"). IWIQs in our mundane conversational data are used for making a private, internal state or position publicly available, which in turn allows IWIQs to serve as vehicles for opening an opportunity space for the participants to join in speculating on the matter raised by the IWIQ, without, however, necessarily mobilizing a response. Nonetheless, responses are sometimes produced, having been mobilized through other means, such as gaze. We aim to show that the IWIQ format performs a distinct social action, and thereby to contribute to our understanding of action ascription, categorization, and formation. Data are in American English.
Pointing gestures are ubiquitous in human social interactions. Among pointing gestures are those in which a speaker points at the person being addressed: addressee points. Research has shown that addressee points are frequently used to signal common ground or express agreement. In this article, by contrast, we explore previously unresearched uses of addressee points in conflictual interactions. Through a fine-grained, reductionist analysis of conduct-one involving a reconceptualization of "action" in interaction-we show that addressee points attribute a transgression to the addressee, upgrade the perceptual intensity of the speaker's conduct, and escalate an argument or conflict. We examine the use of addressee points in terminal escalations: close (vs. far) addressee points in extreme conflicts and lax (vs. tense) addressee points in "nonserious" conduct. The findings contribute to research on the kinematic organization of gesture, interactional and gestural space, and the conceptualization of social action. Data are in English.
Recipients' understanding of speakers' actions typically becomes publicly available through their responses, making the next-turn proof procedure central for the analysis of social interaction. However, little is known about how to methodologically warrant the analysis of actions with minimal or no response. This is particularly evident in case of trouble formulations. We examine trouble formulations in the context of practical activities, produced in order to account for bodily conduct that might look problematic to other participants. We show how such trouble reports are designed to preempt any unwanted impression/interpretation and subsequent responses from their recipients. We conclude by discussing methodological implications from this study concerning action categorization in studies of social interaction, and we suggest analytical resources that analysts may draw on to describe an action when there is no or minimal response from recipients. Data are in American English and German.
In conversation analysis, offers have mostly been studied with regard to their linguistic rather than multilingual format. In contrast with requests, offers have been described as benefiting their recipient (vs. doer), even though its nonstraightforward and sometimes ambiguous character has been also highlighted, generating problems of interpretation (is the action a real offer or a token or fake offer?). This article shows how the action of offering something here and now (vs. offering to do a future action) is multimodally produced, involving the body manipulating the offered material objects, and locally categorized as genuine offers vs. as preliminaries inducing other actions. This multi-barreled nature of offers shows how the distinction between benefactor/beneficiary can be questioned and even subverted by the participants. The analyses focus on offers of food samples at the market, categorizable as gifts vs. as conducive to buying. Video data are in French.
This study examines actions that are not directly articulated by speakers but are nonetheless oriented to by participants. Specifically, we examine situations in which a speaker is launching a request project, approaching a request indirectly, with a preliminary, or with further "upstream work," but without ever making an actual request. Prior literature has argued that indirectness and preliminaries afford speakers with deniability and thus allow them to avoid accountability for requesting. We argue that, although these ways of approaching and projecting a request create greater opportunities for alignment and affiliation, they do not allow speakers to evade accountability. Data are in English and Italian.
This introduction to the special issue on action categorization revisits the state of the art, discusses challenges and competing views, and considers future developments in the study of action in social interaction. After reviewing shortly speech act theory, (neo-)Gricean pragmatics, and conversation analysis as approaches to action categorization, it discusses the relationship between action formation and action ascription in interaction vs. the scientific categorization of action by generic labels. It also addresses core issues of labeling actions, such as the relationship between vernacular and technical labels, the issue of granularity of action-types, language- and culture-specificity of actions and their labels, and methodological issues of the empirical categorization of actions based on naturally occurring talk-in-interaction. In a wider perspective, the relevance of action categorization-and, in particular, of generic action labels-is critically discussed and related to other sorts of pragmatic meanings, which equally matter to how participants treat interactional conduct.
Sometimes when people are sharing troubles in their lives with friends and family members, they distinguish problematic behavior from hypothetically more normal, preferable, or even ideal behavior. In so doing, they demonstrate social-moral reasoning and how persuasion is embedded in interactional practices. This article describes two positions in troubles talk where contrasts differentiate and rank categories of people in a corpus of recorded naturally-occurring conversations in U.S. English: in troubles expansion, to recruit affiliation with a complaint about a non-present third party; and wherein the troubles teller responds after encountering possible disaffiliation from the recipient. Both cases show how contrasts may advance the teller's reasonable stance regarding troubles, recruit the recipient into affiliation, and draw moral boundaries around behavior. This research contributes to our understanding of why and how participants use contrasts in conversation, to accomplish what social actions, and with regard to what moral orientations matter to them.
Minimal responses are often seen as limited to fully accepting or rejecting prior talk. This article examines a minimal response practice that is used to withhold or avoid taking such a clear-cut stance. The practice consists of the affirmative particle ken "yes," delivered with a rise-to-high pitch movement. Ken? invites co-participants to review their prior assertions, without targeting an exact part or aspect thereof. Prior speakers assess the relevancy of this invitation-to-review based on the sequential, actional, and relational context. This analysis yields varied responses to ken? including reassertions, retractions, or deferrals of the review invitation. While ken? is often suggesting affective or epistemic incongruence, it may also be used affiliatively, by encouraging the development of a yet unfinished topic. These findings reveal further depths in the use of minimal responses, showing that their lack of specificity has its own affordances. Data are in Hebrew with English glosses and translations.
This study examines how labor and delivery staff in Chinese hospitals announce the gender of newborns during the first encounter with family members. Using conversation analysis of 97 video-recorded birth encounters, the study identifies interactional practices such as guesswork, testing the waters, and gender-specific compliments that delay or soften the gender announcement. These practices function as forms of forecasting, enabling recipients to anticipate and realize the news themselves rather than being told directly. Although the birth of a child is typically treated as good news, announcements of girls are often managed as socially delicate. These findings extend prior work on news delivery and forecasting in clinical settings and contribute to conversation analytic research on gender by showing how participants use interactional practices to make gender relevant and to manage its interactional consequences. Data used in this study are in Mandarin Chinese, with English translation.
This article focuses on J'ai (quelqu'un) statements, "I have (someone)," in French, through which speakers convey that they are attending to a third party requiring their attention. The speaker produces the statement while transitioning from the interaction's opening to the business at hand. The statement initiates the first turn-at-talk of a recruiting sequence oriented to enlisting the recipient in new action. In the same turn-at-talk, the speaker produces the initial recruiting move, for instance, a request. We examine the focal statement in relation to the recruiting move as a preliminary statement in an independent clause, the frame of a recruiting declarative utterance, and the beginning of an extended recruiting telling. The article furthers the study of initial recruiting moves and preliminaries to them-namely, providing background information on contributing circumstances. The data are telephone calls and corridor interactions between coworkers, recorded in French, at two acute-care hospitals in Switzerland.
This article explores the possibility of incorporating coding and the quantitative evidence generated through coding into an investigation of epistemics in social interaction. Specifically, it considers three facets of an epistemic phenomenon as possibly subject to coding and quantifying: i) relationships between a possible practice for managing epistemics and the participants' social identities relevant to the matter in question; ii) relationships between a possible practice and its sequential environment; and iii) the relationships between different practices used within a turn. It is argued that coding can advance a conversation analytic investigation of epistemics when it is adequately informed by and used to inform sequential analyses. Data are in Japanese.
Conversation analysis' inductive qualitative approach consistently relies on contrastive patterns across collections of cases making the data amenable to hypothesis generation about inter-relationships among practices as well as relationships between practices and exogenous variables. But how can this quantification be done in ways that do not return us to a Balesian version of coding that disregards CA principles? CA-grounded coding should be designed to build codes out of rich conversation analytic work, which includes being true to members' orientations to phenomena as well as to incorporating both compositional and positional aspects of phenomena into the coding scheme. In this special issue, we bring together scholars who have conducted CA-grounded coding studies to discuss the methodological issues involved in the development and application of a coding scheme; the practicalities of these endeavors with CA principles; and the ways in which challenges can be overcome.
Coding social interaction has become increasingly attractive for conversation analysts interested in mixed-methods research as a way to demonstrate the robustness of qualitative findings, test relationships between interactional and exogenous variables, and reach a wider audience. However, coding is valuable to conversation analysts only when it is done in a way that attends to participants' orientations to the phenomenon of study. The puzzle then is how to turn the messy richness of conversational data into codes that are interactionally meaningful and valid. In this article, we draw on the existing literature and our own past projects to discuss opportunities and challenges involved in coding social interaction, with an emphasis on three main aspects of the process: constraining a phenomenon by sequential and formal criteria; transforming behavior into variables; and identifying social actions. Data are in English and Italian.
The paper reflects on the coding procedures in a comparative research project on request for confirmation (RfC) sequences in ten languages. The central aim of the project was to compare the linguistic resources used to accomplish RfCs and their responses. We revisit the reflexive process of developing sampling procedures and coding categories that are applicable for a cross-linguistic study of social action. We show that comparative research cannot rely on particular linguistic features in characterizing the action in question. Instead, we developed inclusion criteria that define RfCs in epistemic and sequential terms. Moreover, we illustrate challenges that emerged in coding the RfCs' epistemic profile and different response strategies and argue for two different but related approaches to coding sampled instances of an action sequence in cross-linguistic research as possible solutions: First, a descriptive approach that focuses on linguistic form; second, a step-by-step approach that clearly distinguishes formal and functional criteria. Data are in Spanish and German with English translations.
Repair-the process of detecting and responding to problems with speaking, hearing or understanding in conversation-is the focus of a range of established coding protocols. We discuss the practical process of developing and applying these protocols to a range of verbal and non-verbal repair phenomena. Coding protocols necessarily trade detail for generalization. We consider four payoffs: (i) practical-reduced effort and increased speed and scale of data analysis; ii) empirical-quantitative insights into the distribution of repairs that support comparative analysis and applications such as detecting Alzheimer's Disease; iii) computational -enabling automatic detection of verbal and non-verbal repairs for corpus analysis, dialogue systems development, and to support selective experiments on repair processes; (iv) interdisciplinary-the process of protocol development provides a methodological bridge between qualitative and quantitative disciplines that can foster new insights into how repairs work. Data are in English and German.
This paper deals with one object of coding that is particularly troublesome, the coding of actions. We argue that action-coding within the framework of Conversation Analysis (CA) necessarily leads to problems due to an inherent tension between criteria for good coding schemes and the properties of action in social interaction. While coding schemes require clear definitions of codes, their mutual exclusivity, irreversibility, comprehensiveness and stable coding units, action ascription is characterized by indexicality, multi-layeredness, potentially different understandings of producer vs. recipients, temporality, malleability of units and situatedness. Drawing on several studies involving action-coding, we discuss five recurrent problems of action-coding, which arise from properties of social interaction: the level of abstraction of the action-categorization, multiple and ambiguous actions, actions that do not receive uptake, crossover cases, and problems of data-segmentation. We explain the nature of the problems, present data-examples, and discuss possible solutions. Data are in German.
Coding interactional phenomena for formal quantification poses substantial challenges, especially for conversation analysts committed to the emic validity of analytic categories. Recognizing that each step in the process merits more sustained attention than is possible within a single paper, our aim is to sketch in broad strokes the overall process of developing and implementing coding systems. We touch on key issues and, when possible, suggest guidelines and best practices for four phases of the process: (1) developing the coding system, including the importance of context-sensitive case analysis as a basis for coding categories; (2) determining what will count as a codable case, including the problem of incompletely realized and compound cases; (3) testing and refining the coding system; and (4) coding "for real," addressing various methods for maintaining reliability and promoting a common coding culture. Data are in English.
