Tipultech logo

Conversation analysis

Author: Dr Simon Moss


Conversation analysis was developed by Harvey Sacks (e.g., Sacks, 1972), which studies talk during social interactions (see also Have, 1999). Conversation analysis represents an attempt to characterize the patterns and structure of interactions.

To conduct a conversation analysis, researchers scrutinize a short recording of a discussion in depth-perhaps in a therapeutic, legal, business, health, family, or social context. They might, for example, time the pauses between sentences, record rises in pitch and volume, examine interruptions, and consider the precise words and phrases closely. The purpose of this analysis is to understand how individuals fulfill their goals and maintain a sense of order. That is, researchers identify the devices, systems, methods, and procedures individuals use to cultivate a shared understanding with one another.



An example of conversation analysis is presented below. First, the researcher records a conversation, either audio or audio-visual. Second, the researcher recodes this conversation, using some standard nomenclature, developed by Jefferson (1985, 2004;; see also Schenkein, 1978). This nomenclature is intended to represent the words that are expressed as well as other verbal features, such as intonation, timing, and other vocal characteristics. Usually, the researcher listens to the conversation many times to identify subtle vocal features. Sometimes, another research checks whether this transcription is an accurate representation of the conversation.

T1 Fred: Anyway, I was hopin' I could pop by tomorrow (.) if possible

T2 (2.0)

T3 Fred: Probably not, hey

T4 John: .hh. =

T5 Fred: =That's OK?

T6 John: Yeah

T7 Fred: ?I won't stay long?

The nomenclature is stipulated in the following table. For example, the (.) indicates a definite, but short, pause. A number in these brackets, however, specifies the duration oft his pause, and so forth. All para-verbal features that could affect the interpretation of speech are included, whenever possible. Furthermore, no references to the underlying motives or emotions of individuals are included.

(.) Just noticeable pause
(1.3) pause of 1.3 seconds, for example
?word Detectable, obvious rise in pitch
?word Detectable, obvious fall in pitch.
words [words... ]
[words... ]
Square brackets across adjacent lines implies to individuals are speaking simultaneously
.hh Inhalation
hh Exhalation
wo(h)rd Implies laughter while speaking the corresponding word
hehehe Laughter that is separate from the speech
wor- A sharp termination of the word while speaking
wo:rd Implies the sound that precedes the colon has been elongated
(words) Words in brackets are conjectures when the sound or pronunciation is unclear
( ) Unclear talk. Each set of brackets represents one syllable of unclear speech
Implies no pause between two consecutive speakers
word Louder than usual speech
WORD Appreciably louder than usual speech
?word? Quieter than usual speech
>word word< Faster than usual speech
Slower than usual speech
((description)) Double brackets represent descriptions of some verbal behavior that is difficult to write phonetically, such as ((sobbing)) or ((clears throat))

The precise pronunciations are often represented as well, with records such as "dunno" preferred over "don't know" if applicable. Often, researchers diverge from conventional spelling to capture these speech styles. Nevertheless, when such divergence from conventional spelling is overused, readers might struggle to follow the exchange.

Typically, the physical movements, mannerisms, and gestures of individuals, such as holds gaze while scratching cheek, are not included. Such physical movements cannot be described and interpreted as unambiguously as vocal and para-vocal features. The nuances of these movements cannot be represented well.


Third, the researcher then interprets the conversation--by scrutinzing the transcript as well as replaying the recording. First, the researcher offers an intuitive interpretation of the acts and reactions of each person. Second, the researcher considers each response in more detail, attempting to uncover mechanisms and devices the speakers use to maintain a common understanding and to fulfill their goals.

For example, the researcher might maintain the two second pause could imply a negative answer from the perspective of Frank. That is, John has seemed to violate the usual turn taking system-called an adjacency pair in which one comment should evoke a specific response-and hence the pause represents his answer. Furthermore, a pause might reflect an inclination to diverge from a typical, familiar, or preferred response. Typically, a negative response, in a collaborative setting, will reflect an unfamiliar, not preferred, response. The pause, thus, can be inferred to reflect a negative response. Indeed, Frank then explicitly concedes this inference at T3. This example shows how individuals sometimes infer meaning to understand the violation of social rules and norms.

The interpretations also focus on the affective responses and expressions, intended to fulfil their goal to maintain social order. When a common purpose cannot be established, for example, speakers will often experience frustration or confusion, which is manifested in verbal or para-verbal expressions.

Interpretation: Turn taking

Interpretations also focus on how individuals know how to take turns. Individuals, if asked, often maintain that pauses represent cues they use to know when to take turns. Nevertheless, a close examination of conversations challenges this intuition.

Usually, however, individuals shift turns without waiting for a pause. Indeed, pauses often reflect a difficulty in the interaction--a violation of the anticipated pattern (Liddicoat, 2007). Alternatively, pauses during a sentence might indicate a need to reflect or remember something, in which a switch in the speaker could be perceived as discourteous.

No other universal cue or rule guides turn taking. The length of turns does not provide any guide& each turn can range from one word to many paragraphs. Somehow, however, individuals within a conversation develop methods to manage turn taking.

In particular, the syntactic structure, intonation pattern, and gaze direction can all affect the likelihood that a turn might end, called a transition relevant phase. Furthermore, the fulfilment of a goal or action--that is, when the speaker has attained a purpose--also reflects a transition relevant phase.

Interpretation: Adjacency pairs

Researchers often focus on adjacency pairs when they interpret exchanges. An adjacency pair refers to two turns, which are usually consecutive and uttered by different speakers. A question followed by an answer is one example of an adjacency pair. "Hello" followed by "Hi" is another adjacency pair.

These pairs typically correspond to a specific order& for example, a question always precedes an answer. Indeed, some facets of conversation are intended to initiate a specific range of actions by the other person, called first pair parts. A question, for example, is intended to initiate a corresponding answer. The absense of a rejoinder, therefore, violates this pattern and provokes some attempt to redress the breach.

Rather than respond as intended to a first pair part, individuals sometimes communicate counters. For example, in response to a question "Did you enjoy today", the second person might ask "Did you", which in effect becomes an alternative first pair part . The second person might later answer the initial question unprompted.

Some responses to first part pairs, like questions, are immediate--such as accepting an invitation. Other responses to first part pairs might be delayed, perhaps following a pause--such as a declining an invitation. The immediate response is called the preferred action& the delayed response is called the dispreferred action, implying they can evoke problems in social relationships. These dispreferred actions are not only delayed sometimes, but often preceded by a qualification or preface

Typically, according to Sacks (1987), individuals often prefer contiguous responses. For example, in response to two questions, like "Are you tired. Or hungry", the answer might be "I'm not hungry, but I am exhausted". In this instance, the second person ensured some contiguity between the question "Or hungry" and the answer "I'm not hungry". The question "Are you tired" and answer "I am exhausted" could not be contiguous, however.

Interpretation: Expanding

Individuals also often expand adjacency pairs. For example, after an answer, individuals often emit an "Oh" sound and then a question, like "Really"--often intended to elongate the response slightly. Indeed, according to Heritage (1984), the "Oh" in this context often seems to reflect more commitment to desire for more information.

"Oh", in other contexts, however, is often interpreted differently. "Oh" in response to a first part pair often reflects receipt of information--as in the example "Are you going tonight" and "Oh, I'm not sure". That is, the person implies their state has changed as a consequence of the previous remark.

"OK", however, usually represents acceptance of a second part pair--that is, the response to a first part pair like a question. Alternatively, "OK" often provokes closure of a particular sequence of remarks--even in response to a dispreferred action, like a rejection.

Interpretation: Noticing

Interpretations might also refer to other common insights that have been derived from conversation analysis. An example is the concept of noticing in which individuals often communicate inevitable statements like "You didn't go to work" when they see someone at home. Such statements can be very informative. If the following statement by the same person is a specific question, we can infer the noticing seemed to be a probe-but a probe that was not fulfilled.

Interpretation: Teasing

In addition to noticing, individuals often demonstrate a method called teasing when they communicate. For example, in response to "Do I look fat in this", a person might tease with "No fatter than usual". Often, in response to teases, individuals answer frivolously but then, after a few moments, seem to defend themselves. Teases usually demonstrate an element of truth to be effective.

Interpretation: Repair

Furthermore, researchers attempt to uncover repairs-attempts by individuals to redress any miscommunication or misunderstandings. Such repairs offer a key insight into how individuals maintain a sense of social order and shared understanding.

Repairs include any attempts to redress errors and difficulties in the conversation. For example, when individuals cannot retrieve the correct word, they often elongate a previous syllable--as a means to demonstrate they need more time--or deploy non-lexical pertubations such as um or ah. Similarly, after they complete their turn, but recognize an error, they might interject with phrases that begin with "I mean".

On other occasions, the other person might highlight an error, perhaps by emphasizing a word. One person might refer to their desire to lend a book& the other person might repeat this phrase, but emphasize the word lend, to imply that borrow might be the correct term.

Interpretation: Other considerations

The interpretation, however, does not attempt to decipher the feelings or motives of individuals. In addition, the interpretation should not apply generalized preconceptions about society to understand this specific conversation, unless this assertion can be defended from other features of the interaction.


After the initial interpretation, the research might then assess the veracity or validity of their insights. They might assess whether their interpretations or assertions were maintained in a subsequent exchange, for example.

Philosophical underpinnings


Conversation analysis adopts the philosophy of ethnomethodology, formulated by Harold Garfinkel, an American sociologist, in his book "Studies in Ethnomethodology" (Garfinkel, 1967). Ethnomethodology assumes that social order is illusory. The social world appears to be ordered and predictable, but is potentially haphazard and random. Social order itself is a social construction, existing in the minds of individuals in society. Indeed, Garfinkel showed that individuals uncover the patterns of advice they receive even though, unbeknownst to these participants, these suggestions were actually random.

Specifically, a specific utterance cannot be interpreted unambiguously from the words alone (Heritage, 1984). Many factors, such as the preceding discussion, the relationship history between the speakers, and their socio-cultural background shape the interpretation. Accordingly, no individuals can interpret a specific utterance in the same way. Nevertheless, to maintain a sense of order, individuals assume they can understand the meaning of the utterance of another person.

To maintain the illusion or impression of a common, shared understanding, individuals effortlessly and inadvertently apply a variety of methods and procedures while conversing-including both subtle and explicit mannerisms, expressions, and remarks. They might, for example, mimic the emotional expression of the other person. Alternatively, they might show confusion when violations of their expectations arise. As a consequence, when individuals presuppose they share a common perspective, they feel the conversation focuses on objective realizations, uncontaminated by subjective differences.

Ethnomethodological research is conducted to understand the methods and procedures that individuals apply to cultivate this sense of shared order (Taylor & Cameron, 1987). To achieve this goal, scholars might conduct a breaching experiment-in which they infringe upon a social rules, such as driving down the wrong way of road. The results of such breaches can unearth the practices and forces that maintain accountability for social order.

Assumptions that underpin conversation analysis

Several assumptions underpin conversation analysis (see, for example, Liddicoat, 2007). First, conversation is not random or unstructured, but ordered and systematic.

Second, despite this order, no universal structure underpins all conversations. Instead, the participants themselves cultivate this order, structure, and context.

Third, although the participants cultivate the order of their conversation, some of the patterns are repeated across many contexts and cultures.

Fourth, the role of investigators is to understand the skills, devices, and methods that individuals use to generate this order--to guide their own communication, to fulfill their goals, and to understand another person.

Fifth, individuals are not usually aware of how they generate this order and shared understanding& hence, researchers must extract such practices from real conversations not from, for example, introspections about past conversations.

Complications and applications of conversation analysis

Some scholars argue that conversation analysis disregards key issues, such as the contextual and structural forces that impinge on the discourse. That is, the methods that individuals use to maintain social order and to fulfill their goals might, depend, fundamentally on their ideology and social context. For example, according to proponents of critical discourse analysis, conversation analysts should consider these pervasive, but unobservable, issues.

However conversation analysts, like Schegloff (1997), explicitly criticize the consideration of political orientations or theoretical assumptions when conversations are analyzed. These orientations and theories might bias the description and interpretation of conversations. A remark might be interpreted as a means to gain dominance, but might actually function to facilitate the conversation.

Conversation analysis, however, has bee applied to examine the methods and devices individuals apply to maintain and perpetuate inequalities in power. Hutchby (1996), for instance, utilized conversation analysis to show how radio hosts maintain power over their callers. Hutchby showed that radio hosts use various terms or phrases, like "So" or" And", to challenge the pertinence of an argument. Similarly, radio hosts often ascribe and challenge a position to the caller, summarizing a version of their argument they can readily dismiss. In addition, Hutchby demonstrates that radio hosts do not need to offer their own position, which simplifies their role. In this work, conversation analysis provides a unique insight into the attempts of individuals to maintain power, offering an empirical insight into the theoretical mechanisms that proponents of discourse analysis posit.


Arber, A. (2007). "Pain talk" in hospice and palliative care team meetings: An ethnography. International Journal of Nursing Studies, 44, 916-926.

Atkinson, J. M. (1982). Understanding formality: The categorization and production of "formal" interaction'. British Journal of Sociology, 33, 86-117.

Atkinson, J. M. & Heritage, J. (1984). The structure of social action. Cambridge: CUP.

Coates, L., Bavelas, J. B., & Gibson, J. (1994). Anomalous language in sexual assault trial judgements. Discourse & Society, 5, 189-206.

Drew, P (1987). Po-faced receipts of teases. Linguistics, 25, 219-253.

Edwards, D. (1994). Script formulations: an analysis of event descriptions in conversation. Journal of Language and Social Psychology, 13, 211-247.

Edwards, D. (1997). Discourse and Cognition. London: Sage.

Edwards, D., & Potter, J. (1992). Discursive Psychology. London: Sage.

Edwards, D., & Potter, J. (1993). Language and causation: A discursive model of description and attribution. Psychological Review, 100, 23-41.

Fairclough, N. (1992). Discourse and text: linguistic and intertextual analysis within discourse analysis. Discourse & Society, 3, 193-217.

Freeman, S. (1987). Verbal communication in medical encounters: An overview of recent work. Text, 7, 3-17.

Garfinkel, H. (1967). Studies in ethnomethodology. Englewood Cliffs, N.J., Prentice-Hall.

Grice, H.P. (1975). Logic and conversation. In P. Cole & J.L. Morgan (Eds.), Syntax and semantics: Vol. 3. Speech acts (pp. 41-58). New York: Academic Press.

Have, P. T. (1999). Doing conversation analysis. London: Sage

Heath, C. C. (1982). Preserving the consultation: Medical record cards and professional conduct. Journal of the Sociology of Health and Illness, 4, 56-74.

Heritage, J (1984). Garfinkel and ethnomethodology. Cambridge, England: Polity.

Hutchby, I. (1996) Confrontation talk: Arguments, asymmetries and power on talk radio. New Jersey: Lawrence Erlbaum Associates.

Hutchby, I. & Wooffitt, R. (1998) Conversation analysis: Principles, practices and applications. Cambridge: Polity.

Hutchby, I. (2001) Conversation and technology: From the telephone to the internet. Cambridge: Polity.

Hutchby, I. (2002). Resisting the incitement to talk in child counselling: Aspects of the utterance "I don't know". Discourse Studies, 4, 147-168.

Hutchby, I. (2006) Media talk: Conversation analysis and the study of broadcasting. Maidenhead: Open University Press.

Hutchby, I. (2007) The Discourse of child counselling. Amsterdam and Philadelphia: John Benjamins.

Hutchby, I. & Barnett, S. (2005). Aspects of the sequential organisation of mobile phone conversation. Discourse Studies, 7, 147-171.

Hutchby, I. & Tanna, V. (2008) Aspects of sequential organisation in text message exchange. Discourse and Communication, 2, 143-164.

Hutchby, I., & Wooffitt, R. (1998). Conversation analysis. Oxford: Polity.

Hutchby, I. & Wooffitt, R. (2008) Conversation Analysis (2nd Edition). Cambridge: Polity

Jefferson, G. (1972) Side sequences. In D.N. Sudnow (Ed.) Studies in social interaction (pp.294-333). New York, NY: Free Press.

Jefferson, G. (1973). A case of precision timing in ordinary conversation: Overlapped tag-positioned address terms in closing sequences. Semiotica, 9, 47-96.

Jefferson, G. (1974). Error correction as an interactional resource. Language in Society, 3, 181-199.

Jefferson, G. (1985). An exercise in the transcription and analysis of laughter. In T. Van Dijk (Ed.) Handbook of discourse analysis, Vol. 3: Discourse and dialogue (pp.25-34).London, UK: Academic Press.

Jefferson, G. (1996). A case of transcriptional stereotyping. Journal of Pragmatics, 26, 159-170.

Jefferson, G. (2004). Glossary of transcript symbols with an Introduction. In Gene H. Lerner (Ed.) Conversation Analysis: Studies from the first generation (pp. 13-23). Philadelphia: John Benjamins. Liddicoat, A. J. (2007). An introduction to conversation analysis. New York: Continuum.

Long, S. A. (1989). The effects of reader and text characteristics on imagery reponed during and after reading. Reading Research Quarterly, 24, 353-332.

Potter, J. (1996). Representing reality: Discourse, rhetoric and social construction. Thousand Oaks, CA: Sage Publications.

Potter, J., & Wetherell, M. (1987). Discourse and social psychology. London: Sage.

Sacks, H. (1972). Lectures on conversation, Vol. 2. Cambridge: Blackwell.

Sacks, H., Schegloff, H., & Jefferson, G. (1974). A simple systematics for the organization of turn taking in conversation, Language, 50, 696-735.

Schegloff, E. A. (1968). Sequencing in conversational openings. American Anthropologist, 70, 1075-1095.

Schegloff, E. A. (1987). Analyzing single episodes of interaction: An exercise in conversation analysis. Social Psychology Quarterly 50: 101-114.

Schegloff, E. A. (1993). Reflections on Quantification in the Study of Conversation, Research on Language and Social Interaction, 26, 99-128.

Schegloff, E. A., & Sacks, H. (1973). Opening up closings. Semiotica, 8, 289-327.

Schenkein, J. (Ed). (1978). Studies in the organization of conversational interaction. New York: Academic Press.

Taylor, T. J. & Cameron, C. (1987). Analyzing conversation. Oxford: Pergammon.

Wetherell, M., & Potter, J. (1988). Discourse analysis and the identification of interpretative repertoires. In C. Antaki (Ed.), Analysing everyday explanation: A casebook of methods (pp. 168-183). London: Sage.

Wood, L.A., & Rennie, H. (1994). Formulating rape: The discursive construction of victims and villains. Discourse & Society, 5, 125-148. Wooffitt, R. (2005). Conversation analysis and discourse analysis. London & New York: Sage.

Academic Scholar?
Join our team of writers.
Write a new opinion article,
a new Psyhclopedia article review
or update a current article.
Get recognition for it.

Last Update: 6/16/2016