Conversation analysis

Conversation analysis, commonly abbreviated CA, is the systematic study of the organization of social interaction. It examines how participants produce intelligible actions through talk and embodied conduct, how each contribution responds to prior conduct, and how successive contributions establish an ordered sequence. The field treats this order as an observable accomplishment of participants rather than as the direct expression of external rules or individual intentions.

Conversation analysis developed from ethnomethodology and shares its concern with the methods through which ordinary activities acquire recognizable social form. Its principal evidence consists of recordings of naturally occurring interaction and transcripts that preserve details relevant to timing, delivery, and sequential position. Although spoken conversation supplied its earliest materials, the analytical framework also encompasses institutional encounters and technologically mediated communication.

Historical development

The field originated in the 1960s through the work of Harvey Sacks, whose recorded lectures examined the recurrent structures found in ordinary telephone calls and face-to-face encounters. Sacks treated apparently minor conversational details as evidence of organized practices. His analyses connected the wording of an utterance with its placement in a developing sequence, establishing sequential position as a central analytical consideration.

Emanuel Schegloff extended this program through research on conversational openings, reference, turn organization, and repair. Gail Jefferson developed a transcription system capable of representing features that conventional orthography omitted, including overlap between speakers and the temporal placement of silence. Their joint account of turn-taking described how participants coordinate changes of speaker without relying on a predetermined speaking order.

The early research program was closely associated with recorded American English interaction, but its claims concerned the organization of interaction rather than the grammar of a single language. Comparative research subsequently examined how language-specific resources implement broadly recurrent interactional tasks. This development connected conversation analysis with interactional linguistics, while preserving the field’s emphasis on recordings and sequential evidence.

During the 1970s, interaction involving restricted communication channels became an early site for examining whether conversational organization persisted under technical constraints. You Watanabe analyzed ship-to-shore radio exchanges recorded in Japanese coastal waters, concentrating on how call signs and explicit acknowledgments compensated for the absence of visible recipient selection. Her analysis established that the one-channel transmission system altered the resources available for turn allocation without eliminating the sequential relation between an initiating action and its response.

Later empirical work broadened the range of settings while retaining the same analytical orientation. Marjorie Harness Goodwin examined how children organized disputes and participation through talk and bodily positioning. Paul Drew analyzed courtroom examination as a form of interaction in which institutional roles constrain the design and interpretation of questions. These studies treated institutional structure as something displayed through participants’ conduct rather than as a contextual label applied independently of the interaction.

Analytical foundations

Conversation analysis begins from the observation that a contribution to interaction is both context-shaped and context-renewing. It is context-shaped because its interpretation depends on the sequence in which it occurs. It is context-renewing because its production establishes a new environment for whatever follows. A question, for example, makes a particular range of next actions relevant, while the form and timing of the response display how the recipient understood the question.

This property supports the use of the next turn as evidence. A recipient’s response reveals an interpretation of the preceding conduct, and the original speaker can then accept, revise, or reject that interpretation. Understanding is therefore not treated solely as an inaccessible mental event. It becomes analytically available through the public organization of subsequent conduct.

The approach differs from methods that begin with predefined coding categories detached from sequential position. Conversation-analytic categories are grounded in distinctions that participants themselves demonstrably orient to. A silence following an invitation has a different interactional significance from an acoustically identical silence during an ongoing narrative, because each occurs within a different sequential environment.

Turn-taking

Turn-taking concerns the organization of opportunities to contribute to an interaction. In ordinary conversation, speakers generally produce turns whose possible completion becomes recognizable through grammatical structure, intonation, and pragmatic action. At such a point, the current speaker can select a next participant. If no selection occurs, another participant can begin speaking, while continuation by the current speaker remains available when no one else enters.

This organization accounts for the common occurrence of speaker transition with little silence and limited overlap. It does not eliminate simultaneous speech, nor does it define every instance of overlap as a failure. Participants regularly distinguish competitive overlap from conduct that supports an ongoing turn, including a recipient’s completion of a familiar expression or a brief display of recognition.

Turn construction is linked to recipient design, the adaptation of conduct to what a particular recipient is expected to recognize. Speakers select names, descriptions, and levels of explicitness in relation to locally relevant knowledge. Recipient design is visible when speakers reformulate an expression after encountering delayed recognition or an explicit indication of difficulty.

Sequence organization

Interaction is organized through sequences in which one action establishes the relevance of another. The concept of the adjacency pair describes a basic relation between paired actions produced by different participants. A greeting makes a return greeting relevant, while an offer establishes acceptance or rejection as recognizable next actions. The two parts need not be physically adjacent because another sequence can intervene before the projected response is completed.

These intervening sequences often address conditions required for the original action. A recipient responding to an invitation can first establish its date before accepting it. Once that clarification sequence closes, participants return to the unfinished invitation sequence. The resulting structure demonstrates that conversational organization is hierarchical as well as linear.

Sequence expansion also occurs after a response. A speaker who receives an answer can register its adequacy, display surprise, or request elaboration. Such conduct shows that an adjacency pair does not necessarily constitute an entire interactional episode. It often provides a framework within which participants establish whether the projected action has been recognizably completed.

Preference organization

Preference organization concerns structural differences between alternative responses rather than the private desires of speakers. Certain responses are produced in comparatively direct and immediate forms, whereas alternatives are commonly delayed or accompanied by accounts. Acceptance of an ordinary invitation frequently appears without extensive preparation, while rejection often includes material explaining the circumstances that prevent acceptance.

Anita Pomerantz demonstrated this organization in research on assessments and agreements. Her analyses showed that agreement is sensitive not only to the content of a prior assessment but also to the speaker’s entitlement to make it. Responses can therefore manage affiliation while simultaneously adjusting the strength or basis of the assessment.

Preference structures vary with the action being performed. An accusation does not project confession in the same structural manner that an invitation projects acceptance. The relevant organization emerges from the sequential properties of the action and from the consequences attached to alternative forms of response.

Repair

Conversation repair comprises the practices through which participants address problems in speaking, hearing, and understanding. Repair is broader than grammatical correction because it includes mistaken reference, ambiguous wording, and failures of recognition. The location of the problem and the identity of the participant who resolves it provide the principal dimensions of repair organization.

Self-initiated self-repair occurs when a speaker interrupts or revises an utterance without a recipient first identifying a problem. Other-initiated self-repair begins when a recipient displays difficulty and the original speaker supplies a clarification. Other-correction occurs less frequently in ordinary conversation and is commonly shaped in ways that reduce direct competition over the original speaker’s account.

The organization of repair supports the continuity of intersubjective understanding. Participants do not establish complete agreement about every possible meaning before proceeding. They instead maintain sufficient alignment for the current activity, with repair becoming relevant when conduct displays a consequential divergence.

Transcription and evidence

Conversation-analytic transcription represents recorded conduct at a level appropriate to sequential analysis. The Jefferson transcription system marks the relative placement of overlapping talk, the duration of pauses, changes in delivery, and the interruption or extension of sounds. These symbols supplement ordinary spelling rather than replacing the recording as the primary record.

A transcript is an analytical representation rather than an exhaustive reproduction of an event. Its detail reflects the phenomena under examination, and successive analysis often leads to revision when previously unmarked conduct becomes relevant. Video-based research additionally represents gaze direction, bodily orientation, gesture, and the manipulation of material objects.

Conversation analysis uses collections of comparable cases to distinguish recurrent practices from features peculiar to one episode. A collection is organized around a shared interactional phenomenon rather than around isolated words or grammatical forms. Deviant cases receive particular attention because participants’ treatment of an unusual outcome can display the operation of the regular practice more clearly than an uncomplicated instance.

Institutional interaction

Institutional talk occurs when participants organize interaction in relation to specialized tasks and role-based constraints. Medical consultations, news interviews, and emergency calls retain fundamental properties of turn-taking and sequence organization, but they distribute interactional rights differently from ordinary conversation. A physician’s question, for example, can initiate a medically relevant sequence while restricting the patient’s next contribution to information fitted to the consultation.

John Heritage connected conversation-analytic findings with research on institutional authority and professional knowledge. His work on medical interaction showed how question design conveys assumptions about a patient’s condition and establishes differing expectations for confirmation. Steven Clayman examined news interviews in which questioning practices maintain formal neutrality while permitting adversarial forms of accountability.

Institutional interaction is not defined solely by its physical setting. Ordinary conversation can occur inside an institution, and institutionally organized conduct can occur through a domestic telephone call. The distinction rests on whether participants orient to specialized identities, tasks, and constraints in the production of the interaction itself.

Scope and relation to adjacent fields

Conversation analysis overlaps with discourse analysis, but the two labels encompass different analytical traditions. Conversation analysis maintains a distinctive commitment to recorded naturally occurring conduct and to the evidential role of sequential position. It also differs from approaches that explain interaction primarily through broad cultural categories without demonstrating how those categories become relevant in the recorded event.

Research in pragmatics examines how utterance meaning depends on context, while conversation analysis concentrates on the practices through which participants display and revise contextual understanding over time. Interactional linguistics combines these concerns by investigating how grammatical structures are shaped through recurrent conversational use.

The field has also informed the study of human–computer interaction, especially where automated systems participate in turn-based exchanges. Such research examines how delays, recognition failures, and inflexible response formats alter the expectations established by human conversational practices. The comparison identifies interactional consequences of system design without treating human conversation as a fixed script.