Questions: Conversation Analysis: Order in Interaction

Question 1

Speaker A asks a direct question. Speaker B pauses for three seconds, then says 'Well... I suppose that could work.' According to conversation analysis, what has the pause accomplished before a single word is spoken?

• A. Nothing — CA treats pauses as meaningless noise that falls outside the analytical framework
• B. The pause signals cognitive load and indicates that speaker B needs more processing time
• C. The pause is already a dispreferred response — heard against the background of what the question made relevant, it signals reluctance or resistance before the verbal content arrives
• D. The pause triggers a repair sequence in which speaker A must restate the question more clearly

Answer: (C)

This illustrates CA's core claim that silence is never neutral — it is always heard against the background of what the prior turn made normatively relevant. A question projects an answer as the expected next action. When the expected response is delayed, that delay is itself interactionally meaningful, carrying information (typically dispreference) before any words arrive. Speakers know this and use timing strategically; listeners read timing as part of the response. This is why CA analysts transcribe pauses down to the tenth of a second.

Question 2

How does conversation analysis explain the fact that ordinary conversations produce minimal silence between speakers and minimal overlap, without any central coordinator or explicit negotiation?

• A. People have biologically evolved turn-taking instincts that operate below the level of conscious awareness
• B. Cultural norms about politeness are internalized in childhood and automatically prevent interruption
• C. Speakers and listeners orient to transition relevance places and local speaker-selection rules, producing coordination through decentralized moment-by-moment adjustment
• D. Cognitive models of each other's intentions, built through theory of mind, allow speakers to predict when the other will finish

Answer: (C)

CA's answer is the locally managed turn-taking system. Speakers signal transition relevance places (TRPs) — moments where a turn could legitimately end — through syntax, prosody, and gaze. At a TRP, the current speaker may select the next speaker (obligating them to respond) or leave self-selection open. This system produces orderly coordination without any global schedule. The organization is emergent: it arises from participants following local rules moment by moment, not from internalized politeness norms or theory of mind computations.

Question 3

In conversation analysis, when a speaker asks a question, that question does not merely request information — it normatively obliges the next speaker to respond, such that even silence is heard as a meaningful action.

• T. True
• F. False

Answer: True

This is the adjacency pair concept: a question is the first part of a question-answer pair, and it makes an answer the expected (normatively required) next action. Silence or a non-answer is not neutral — it is heard as a noticeable and accountable absence of the expected response. Participants in interaction are oriented to these normative expectations, which is why a long pause after a question is interactionally meaningful, and why speakers often feel obligated to account for dispreferred responses (e.g., 'I'm sorry, I don't know...').

Question 4

Conversation analysis holds that self-initiated repair and other-initiated repair are equally preferred structures because both serve the same function of restoring shared understanding.

• T. True
• F. False

Answer: False

CA reveals a strong preference ordering: self-initiated repair is preferred over other-initiated repair, and within other-initiated repair, self-correction is preferred over correction by the other. This preference reflects how interactional accountability and social face are organized. When a listener signals trouble (other-initiation), it potentially challenges the speaker's competence or coherence. Self-correction minimizes this threat. Other-correction is dispreferred because it more directly challenges the original speaker. These preferences are not mere politeness conventions — they are systematic features of how repair is organized across languages and cultures.

Question 5

Why does conversation analysis claim that meaning is not a property of words but is produced interactionally?

Model answer: Meaning in CA is always produced in sequence: what any utterance means depends on what it is responding to (the prior turn) and what it makes relevant next (the projected response). The same words mean different things depending on their sequential position. A 'yes' after an invitation is an acceptance; a 'yes' after an accusation is a confession. Meaning is jointly achieved because each speaker produces their turn in orientation to what came before and what it makes conditionally relevant — so meaning is a product of the sequence, not of the words in isolation.

This is CA's deepest theoretical claim, inherited from ethnomethodology: social reality — including meaning — is not pre-given but is continually produced through the practical actions of participants. Applied to language, it means that understanding what someone said requires knowing where it occurs in an unfolding sequence of action, who was obligated to respond to what, and what the response makes relevant in turn. A transcript of words alone is not a record of meaning — the sequential structure is what meaning lives in.