Skip to main content Skip to navigation

Spoken English: Speech Event

Lecture notes

Course: Spoken English

Week: 3

Lecturer: Keith Richards

Topic: The Speech Event

These notes are extracts only and do not include the arguments developed in the lectures. Neither do they include handouts or workshop activities. They are an additional resource designed primarily for those who attended the relevant lectures.

Important note

The subject of these notes is the speech event rather than the ethnography of communication, which is the broader field of which it is a part. My choice has been determined partly by considerations of time and practicality but also by a consideration of the place that the speech event plays in the ethnography of communication. It is generally accepted as the central concept and in my view represents the best introduction to the field. Duranti sums up the situation succinctly:

For many researchers, the speech event still represents a level of analysis that has the advantage of preserving information about the social system as a whole while at the same time allowing the researcher to get into details of the personal acts

Duranti 1988:219

In view of this, I strongly advise you to follow up the lecture and workshop by reading either Saville-Troike’s excellent brief introduction to the broader field in the McKay collection (1996) or her earlier book (1989) dedicated to the subject. Alternatively, or in addition, you could read Keating (2001).

Review and introduction

In the first week we looked at aspects of communicative competence at the broadest level. This week we’ll move on to attempt to identify appropriate models for describing specific instances of language in use. Just by way of a general reminder and as an indication of the task ahead, here is one attempt to capture the range of relevant components:

Components of communication

1 Linguistic knowledge

(a) Verbal elements

(b) Nonverbal elements

(c) Patterns of possible variants (in all elements and their organization)

(e) Meanings of variants in particular situations

2 Interaction skills

(a) Perception of salient features in communicative situations

(b) Selection and interpretation of forms appropriate to specific situations, roles, and relationships (rules for the use of speech)

(c) Discourse organization and processes

(d) Norms of interaction and interpretation

(e) Strategies for achieving goals

3 Cultural knowledge

(a) Social structure

(b) Values and attitudes

(c) Cognitive maps/schemata

(d) Enculturation processes (transmission of knowledge and skills)

(Saville-Troike 1989: 24)

If we listen to the way people speak, it soon becomes apparent that there are certain activities where interaction seems to be organised in recognisable ways, with rules about what can and cannot be said. We know, for example, that there are accepted ways of issuing and accepting invitations, of making a toast, of making introductions, and so on. This is what lies behind the idea of a speech event, the subject of this week’s sessions.

Speech events in everyday talk

In fact, we use terms which refer to speech events all the time, and these carry a useful semantic load; for example, each of the following utterances refers to a specific speech events. Decide what the relevant interactional rules are:

‘I was late for her lecture.’

‘We had a wonderful chat.’

‘Did you hear the announcement?

Sometimes we take for granted that others know the relevant rules and therefore leave them unstated. If I say, W didn’t have much of a conversation — all he wanted to talk about was his model soldiers,’ I take it as understood that topics of conversation are normally negotiated by participants and that as a result they usually cover different subjects.

The interesting thing about speech events is that they bring together social and linguistic aspects, as Hymes noted:

A general theory of the interaction of language and social life must encompass the multiple relations between linguistic means and social meaning. The relations within a particular community or personal repertoire are an empirical problem, calling for a mode of description that is jointly ethnographic and linguistic.

Hymes 1986:39

Towards a definition

Various definitions of speech events have been offered, and the following discussion is based on those taken from core texts. We begin with three that, taken together, provide a reasonable overview:

The basic unit for the analysis of verbal interaction in speech communities is the speech event ... The speech event is to the analysis of verbal interaction what the sentence is to grammar. When compared with the sentence it represents an extension in size of the basic analytical unit from single utterances to stretches of utterances, as well as a shift in focus from emphasis on text to emphasis on interaction. Speech event analysis focuses on the exchange between speakers

Gumperz 1986: 16-17

At the level of ethnographic description, verbal behavior in all societies can be categorized in terms of speech events: units of verbal behavior bounded in time and space. Events vary in the degree to which they are isolable. They range from ritual situations where behavior is largely predetermined to casual everyday talk. Yet all verbal behavior is governed by social norms specifying participant roles, rights and duties vis-?-vis each other, permissible topics, appropriate ways of speaking and ways of introducing information. Such norms are context and network specific, so that the psycholinguistic notion of individuals relying on their own personal knowledge of the world to make sense of talk in context is an oversimplification which does not account for the very real interactive constraints that govern everyday verbal behavior.

Gumperz 1982: 164-5

A single event is defined by a unified set of components throughout, beginning with the same general purpose of communication, the same general topic, and involving the same participants, generally using the same language variety, maintaining the same tone or key and the same rules for interaction, in the same setting.

Saville-Troike 1989: 27

Gumperz’s comment on the way in which social norms govern verbal behaviour is particularly important, since this is the relationship upon which the concept is based. It is also the reason why the analysis of speech events has played such a central role in the ethnography of communication. Notice, too, his emphasis on the interactive nature of such events within what is a very wide range, from the entirely predictable (ritual) to the more or less open (casual talk). Valuable as these insights are, however, when we look for a definition of the speech event, Gumperz is able to offer nothing more specific than, ‘units of verbal behavior bounded in time and space.’ Taken with the proviso that such events may not be isolable, this leaves the issue of definition pretty much in the air. For a more positive formulation we need to turn to Saville-Troike, whose repetition of the term ‘same’ makes her definition nothing if not consistent. Even so, working from this definition alone, I’m not sure that it would be easy to provide examples of speech events.

Duranti’s comments are worth noting because of their reminder that speech is central to so many of the social activities we engage in, to the extent that some are actually constituted through talk:

The basic assumption of a speech-event analysis of language use is that an understanding of the form and content of everyday talk in its various manifestations implies an understanding of the social activities in which speaking takes place ... Such activities, however, are not simply ‘accompanied’ by verbal interaction they are also shaped by it: there are many ways, that is, in which speech has a role in the constitution of a social event. The most obvious cases are perhaps gossip sessions and telephone conversations, neither of which could take place if talk were not exchanged. But even the most physically oriented activities such as sport events or hunting expeditions rely heavily on verbal communication for the participants' successful coordination around some common task.

Duranti 1988: 218.

Although it does not mention this issue, one of the questions this raises is that of definition because while some of the activities we engage in are very easy to label and to describe, others are more problematic. The same might be said of speech events.

An example

I’ve chosen the following because I found it amusing when I first read it and because it raises the interesting question of how precise our labels need to be. It could just be categorised as ‘small talk’, but in this case it’s clear that certain extra rules are in play that apply to talking to royalty, so I suppose we might characterise it as ‘making small talk with the reigning monarch’. Try to work out the rules yourself and compare your conclusions with mine.

01 Q: Have you been riding today, Mr Greville?

02 G: No, Madam, I have not.

03 Q: It was a fine day.

04 G: Yes, Ma’am, a very fine day.

05 Q: It was rather cold though.

06 G: (Like Polonius) It was rather cold, Madam.

07 Q: Your sister, Lady Francis Egerton, rides, I think, does she not?

08 G: She does ride sometimes, Madam.

09 (A pause, when I took the lead, though adhering to the same

10 same topic.)

11 G: Has your Majesty been riding to-day?

12 Q: (With animation) Oh, yes, a very long ride.

13 G: Has your Majesty got a nice horse?

14 Q: Oh, a very nice horse.

Brett, S. (ed). 1987. The Faber Book of Diaries. London: Faber & Faber.

One way of checking the rules you’ve identified is to find an example where they’re broken and see what the interactional consequences of this are. The following is just such a case, and you might like to see which of the rules you’ve identified are being violated here.

The exchange in question took place in 1960s in England and involved a very popular comedian, Tommy Cooper. Traditionally, the monarch would attend a number of ‘important’ national events, including the Royal Variety Performance and, a few weeks later, the (football) Cup Final. At the end of the former, stars of the show would line up to meet the Queen, having first been informed of the relevant interactional rules (identical to those which applied in the above exchange). Upon being told by the Queen that she had found him very funny, Cooper asked her whether she had really found him funny and, receiving a positive reply, sought further confirmation, to the noticeable discomfort of her attendants and others in the party. At this point, the conversation developed along these lines:

Cooper: May I ask your majesty a personal question?

(Awkward silence)

Queen: (Frostily) So far as I may allow.

Cooper: Do you enjoy football, ma’am?

Queen: No, as a matter of fact I don’t.

Cooper: Well in that case can I have your Cup Final tickets?

(General laughter)

Goffman (1974) has described such behaviour in terms of ‘frames’, which answer the question ‘What is happening here’ and represent the way in which we structure our experience. In the example above Tommy Cooper is ‘breaking frame’. The awkward silence here and the frosty reception are significant both socially and as part of the setting up of the punchline, but the joke has already been prepared for in the earlier establishment of Cooper’s role as a ‘funny man’. We can see a contrast between his behaviour here and that of Greville, whose tortuous efforts to keep to the rules make the earlier example so amusing. It seems to me that the rules in the Queen Victoria example (which don’t seem to have changed) are pretty straightforward: let the Queen take the lead unless a pause in the conversation indicates that a switch is permissible, but make sure you keep to the same topic — and whatever happens, don’t disagree. Tommy Cooper not only takes the lead but selects his own topic (and marks this as a ‘personal’ one).

The SPEAKING model

This is the best known model for analysing speech events. Hymes, who developed it, referred to it as an etic grid and explained the need for such descriptive apparatus as follows:

Even the ethnographies that we have, though almost never focused on speaking, show us that communities differ significantly in ways of speaking, in patterns of repertoire and switching, in the roles and meanings of speech ... Since there is no systematic understanding of the ways in which communities differ in these respects, and of the deeper relationships such differences may disclose, we have it to create. We need taxonomies of speaking, and descriptions adequate to support and test them

Hymes (1986:42-43)

Situation 3 Setting

4 Scene

Participants 5 Speaker, or sender

6 Addressor

7 Hearer, or receiver, or audience

8 Addressee

Ends 9 Purposes — outcomes

10 Purposes — goals

Act sequence 1 Message form

2 Message content

Key 11 Key

Instrumentalities 12 Channels

13 Forms of speech

Norms 14 Norms of interaction

15 Norms of interpretation

Genres 16 Genres

The order here is based on the acronym, but numbers refer to the order in which these components are introduced in Hymes (1986).

Speech event components

One of the things which has always impressed me about Hymes’ SPEAKING model is the subtlety of the distinctions within it: the beauty of the description lies in the divisions within each of the main components. The following comments highlight the main points I made in the lecture:


Hymes’ distinction between setting and scene is an important one. Setting refers to the physical context in which the interaction takes place and may influence the sort of talk that is allowable (e.g. religious building vs bar or caf?). Scene is also part of the situation but its locus is psychological rather than physical. Hymes points out that the same physical setting might be the location for different psychological scenes. Within the same setting participants may move from formal to informal, festive to serious etc.


The distinctions here are along the same lines as Goffman’s ‘production format’ involving the ‘animator’, who produces talk, the ‘author’, who creates talk, and the ‘principal’, who is responsible for talk. The first distinction is between speaker (or sender) and addressor. The former is responsible for the message while the latter is the person who physically delivers it. Normally these are the same, but not always. For example, when in the United States the presidential spokesperson delivers an unpalatable message, nobody points the finger of blame towards the deliverer: it is the President who must bear the brunt of any backlash. The second distinction, which parallels the first, is that between the hearer (or receiver, or audience) and the addressee. The receiver of ‘Now do this in pairs’ in a coursebook may be the students but the main addressee is the teacher.


The distinction which Hymes draws here between outcomes and goals is perhaps less easy to pin down in practice. Outcomes refers to what is conventionally expected or publicly stated as the object of the event from the point of view of the community, whereas the reference to goals recognises that the parties involved may have purposes which are related but not identical to this. Hymes is careful to point out that, for both outcomes and goals, we must be careful to distinguish what is conventionally recognised from what is purely personal or situational.

Act sequence

This refers to the sequence of acts which makes up a speech event. Hymes draws a distinction between message form and message content offering an example of this distinction in terms of ‘He prayed saying “....”’ (where the words appearing between double quotation marks represent the form) and ‘He prayed that he would get well’ which reports the content only. Presumably the difference between the two could be much greater. For example, the message form, ‘Have you seen the time?’ would, in the right context, have a message content which would be represented as, ‘He said it was time they were going.’ This seems to be essentially the same as the locutionary form and illocutionary force distinction originally made by Austin and a key distinction in Speech Act Theory.


The term is used in its conventional sense, to refer to ‘the tone manner, or spirit in which an act is done’ (Hymes 1986:62).


These are, again, conventional. Channel is used in the conventional sense, so it would be important, for example, to distinguish face-to-face communication from talk on the telephone. Forms of speech Hymes identifies as language and dialect, codes, and varieties and registers. It is interesting to note that Saville-Troike includes non-verbal elements under message form, which is surely legitimate. But to represent these (e.g. 1989:166; 169; 173) in general terms as ‘kinesics’ or ‘proxemics’ or ‘eye gaze’ is less acceptable, even when specific examples are provided in the analysis itself. Research in the area of non-verbal communication has demonstrated conclusively that it is an important feature of all face-to-face communication, and detailed studies (e.g. Goodwin 1981 on gaze direction) have revealed that its role is a subtle and complex one. The dilemma for speech event analysis is that this aspect cannot be ignored but there is simply not the space to do it full justice. Selection of ‘relevant’ non-verbal features is therefore inevitably, to some extent, arbitrary.


The two components under this head are particularly important, and many cross-cultural comparisons tend to focus on these areas. Norms of interaction refer to the conventional rules relating to the conduct of the speech event. These will include rules about floor holding, turn-taking, delivery, topic etc. Norms of interpretation are also of crucial importance in speech events and in cross-cultural interaction generally. These refer to the rules which determine the interpretation of particular acts. A failure to understand relevant norms of interpretation was responsible for some very expensive mistakes when the US first engaged in large scale business negotiations in Japan.


The final element is genre, which is not necessarily an element at all. This is not Hymes’ stated position, and he explicitly argues (1986:65) that genres can be invoked within specific events, as when the ‘sermon’ is invoked for humorous purposes within another event. However, it seems to me that this is a special case, and that unless a genre is exploited in this way, it is more likely to be a super-ordinate descriptive category. This is, in fact, how Hymes himself uses it, as the genre (‘Scoring’) within which certain events (shaman’s retribution, girl’s puberty rite, testing of children) are located (1986:67-68).

The model in Perspective

‘The spirit of the model is heuristic, that is, it is designed as an aid to noticing, formulating and organizing materials, and it is designed so as to become itself an object of data- and experience-based critique.’

Philipsen, G. & Coutu, M. 2005. The ethnography of speaking. In K. L. Fitch & R. E. Sanders (eds.), Handbook of Language and Social Interaction. Mahwah, NJ: Lawrence Erlbaum. Pages 355-379. (Page 365)

Basic elements

A number of other writers have developed similar models, all very similar. There’s no need to explore these in depth, but it’s worth noting that the common elements seem to be the first three in Hymes’ list (situation, participants and ends), so you could say these are the core elements, though as far as I know nobody has actually claimed that.

To see how these elements might work when it comes to interpreting an utterance, try working out the purpose (ends) of the following utterance, given the situation and the setting:


Do you know where today’s paper is?


1. Participants: colleague → colleague

Setting: Senior Common Room.

2. Participants: master → servant

Setting: breakfast table

3. Participants: husband → wife

Setting: living room

You should have found the first two very straightforward, but the last one is more problematic, as we saw in the first week:

Husband: Do you know where today’s paper is?

Wife: I’ll get it for you.

Husband: That’s OK Just tell me where it is. I’ll get it.

Wife: No, I’LL get it.

(Gumperz 1982: 135)

This is not an issue we need to explore, but it does serve as a reminder of the points I covered in the first week about the many dimensions of spoken language, which means that it often resists a priori categorisation. As we’ll see next week, this is a position that conversation analysis takes up.

Situation and Event

So far I’ve concentrated on the defining characteristics of the speech event, but the event itself is part of a hierarchy of descriptive terms used by those in the field. Hymes himself offers a fairly extensive list of ‘social units’, but I’ll like to concentrate on the three main elements. Two of these, situation and event, seem to me to be fairly unproblematic, but the speech ‘act’ does throw up a number of problems. For the purpose of discussion we’ll use a slightly adapted example from Saville-Troike (1989:28-29):


Religious service

EVENTS: Call to worship

Reading of scriptures





ACTS: Summons




Closing formula

In this example, the element in italics is carried on to the next level, where it is broken down, so that the situation religious service is represented by the list of events, and from these prayer is chosen as the element to be broken down into acts.

The distinction between (speech/communicative) situation and (speech) event is a straightforward one. The term situation here is used in its conventional sense to stand for the general social context in which communication occurs. A situation is not defined by speaking, although speaking may normally be expected to occur within it. So, for example, speaking would be expected to occur at a party (although it is just about possible to imagine a party where the loudness of the music makes speaking ‘at’ the party impossible), but there will also be other activities, such as dancing, where speaking may not feature. A speech event, however, is defined in terms of speaking, at least in the sense of being governed by the norms relating to this:

The term speech event will be restricted to activities, or aspects of activities, that are directly governed by rules or norms for the use of speech.

Hymes 1986:56

So within the speech situation ‘religious service’, prayer may be a speech event. Obviously, there are forms of prayer which do not involve speech, but the focus here is on ‘spoken’ or ‘public’ prayer. For obvious reasons, Saville-Troike has chosen a clear cut example, and I’ve followed her, but this is not to say that speech events are always so easy to identify, and, as Gumperz notes (1982:164), they may not be easy to isolate. It may also be possible to find one event embedded within another (e.g. an announcement within a lecture) and they may be discontinuous (the lecture may be briefly interrupted as other business is transacted). They can also range from single utterances (‘Fire!’) to extended sequences (e.g. a lecture).

Gumperz’s observation (1982:164) that speech events can range from ritual situations to casual talk seems particularly relevant to TESOL, where coursebooks seem to be to be guilty all too often of ignoring this continuum. We often find them treating as ritual — and therefore subject to precise specification — what properly belongs to the more open end of the range.

Speech Act

Despite these minor reservations, the concept of the speech event is reasonably well-defined, and as an analytical unit it has proved its worth. However, the status of ‘speech act’ is much more problematic. In the end, I think we probably have to accept that the concept of an ‘act’ in EC is far from clear, but it does represent a unit of description below the level of the event. For this reason it seems safest, and perhaps most sensible, to think of acts as part of an ‘act sequence’, which is what Saville-Troike does in her book (e.g. 1989:163). In this way attention is drawn to the fact that they are essentially descriptive units. Having said this, I’d now like to point to what is problematic about the concept, starting with Hymes’ own acknowledgement of the difficulty of labelling acts:

The labelling of the acts is unavoidably somewhat arbitrary.

Hymes 1986:68

The most serious problem associated with the speech act is that its status is by no means clear. Saville-Troike (1996:371) says that it is ‘generally coterminous with a single interactional function, such as a referential statement, a request, or a command, and may be either verbal or nonverbal.’ The problem with this is that the idea of a single interactional function is pretty vague, and ‘generally’ coterminous allows plenty of space for alternatives. The situation isn’t helped by the fact that Hymes states that a joke can be a speech act within a conversation:

The term speech event will be restricted to activities, or aspects of activities, that are directly governed by rules or norms for the use of speech. An event may consist of a single speech act, but will often comprise several. Just as an occurrence of a noun may at the same time be the whole of a noun phrase and the whole of a sentence (e.g. ‘Fire!’), so a speech act may be the whole of a speech event, and of a speech situation (say, a rite consisting of a single prayer, itself a single invocation). More often, however, one will find a difference in magnitude: a party (speech situation), a conversation during the party (speech event), a joke within the conversation (speech act). It is of speech events and speech acts that one writes formal rules for their occurrence and characteristics. Notice that the same type of speech act may recur in different types of speech event, and the same type of speech event in different contexts of situation. Thus, a joke (speech act) may be embedded in a private conversation, a lecture, a formal introduction. A private conversation may occur in the context of a party, a memorial service, a pause in changing sides in a tennis match.

Hymes 1977: 52

I have to admit that I’m not at all comfortable with the idea that a joke is a speech act because jokes can be very extended affairs. There’s also the disturbing possibility that a speech event (e.g. an announcement) might find itself embedded, albeit artificially, within a speech act (it’s easy to imagine it featuring as part of a joke, for example). This is why I feel that the much looser term ‘act sequence’ is useful, since it recognises — perfectly reasonably, it seems to me — that events can sometimes be broken up into a sequence of acts (‘phases’ or ‘stages’ might serve just as well to capture this idea). The precise status of such elements then does not need to be spelt out.

There is another problem with the use of the term ‘speech act’ which arises from its better known use within a related but distinct field. Speech Act Theory developed from the work of the philosopher, Austin, who recognised that certain utterances are used to do more than convey information. He demonstrated that they also perform particular acts. Subsequently, a great deal of work in pragmatics centred on the analysis of such ‘speech acts’, which are always specific utterances. Such work is related indirectly to the concerns of the ethnography of communication, but its treatment of isolated examples, often invented, marks it as coming from a different tradition. Duranti summarises the essential differences between the two fields:

What usually distinguishes the ethnographic approach from pragmatic analysis is a stronger concern for the socio-cultural context of the use of language, with the specific relationship between language and local systems of knowledge and social order, and a lesser commitment to the relevance of logical notation to the strategic use of speech in social interaction.

Duranti 1988:213

Where two traditions use exactly the same term for a central concept, the potential for confusion is considerable, and its not helped when the key figures in each tradition (Austin and Hymes) choose examples which could easily be interchanged. This seems to provide a very good reason for referring to an ‘act sequence’ or following Saville-Troike in using the term ‘communicative act’.

Acts and events

While there is potential for confusion in the use of the term ‘speech act’, it’s also true to say that the connection between event and act is an intimate one. The best way of illustrating this is to ask you to match the following ‘acts’ to the events in which they might occur:


‘Five past ten — is that the time!’

‘It is five minutes past ten, Mr James.’

‘It is 10.05.’





The first thing to notice in these examples is that the form of the utterance is different in each case. It would be very odd, for example, to find someone in a conversation giving the time as 10.05. Perhaps more importantly in view of the issue of acts and their function, it seems clear that a knowledge of the relevant speech event enables us to identify quite clearly the ‘message content’ of these utterances (what Speech Act Theory would call their ‘illocutionary force’). In the first case we know that lectures begin and end at specified times, that the speaker and audience are expected to remain in situ throughout, and that the lecturer is expected, within reasonable limits, to keep to the subject of the lecture. So when Mr James arrives five minutes late and the lecturer reminds him of the time, we know that this is not part of the lecture as such but is an aside which serves as an admonition. In the case of the announcement, which is designed to present information, a bare statement of the precise time achieves the necessary communicative end.

The relationships in the examples are therefore as follows:

Lecture: “It is five minutes past ten, Mr James.”

Chat: “Five past ten — is that the time!”

Announcement: “It is 10.05.”

The ‘chat’ example is perhaps a little more subtle than the others because conversation is the least predictable (in terms of content at least) of all events. However, we know that conversations have to open and close, and if we reflect we will realise that sometimes speakers signal that a conversation has to come to an end (using ‘pre-closers’). In this case the speaker expresses surprise — even alarm — about the time, and from what we know of conversations we can safely assume that it announces that something needs to be done which will either interrupt or end the conversation.

There is, then, a relationship between the interpretation of specific acts within a speech event and the event itself as a representation of shared understandings of the relevant social context; and whatever the shortcomings of the notion of ‘act’ as Hymes represents it, these shouldn’t blind us to the fact that his etic grid offers is a subtle and insightful representation of the components which make up a speech event.

In fact, I think it’s well worth making an effort to get to grips with this concept and with the use of etic grids, but I’m less convinced by some of the more general claims of the ethnography of communication. If you’d like to pursue this (and this very much depends on where your particular interests lie), you’ll need to read Saville-Troike’s 1996 paper in the light of my suggestions in the next section.

Issues in the ethnography of communication

If you’re interested in exploring this field further, Saville-Troike (1996) makes a very good starting point. You might like to read this, noting the following:

  • any apparent contradictions in the statements the author makes;
  • definitions or descriptions which seem to you to be vague or very general in scope;
  • proposals which seem to you to be optimistic;
  • any other problems.

You might also consider how much of what she says here is relevant to TESOL as opposed to TESL or first language teaching.

If you’re interested in my own thoughts on the subject, these can be accessed at:


Duranti, A. 1988. Ethnography of speaking: towards a linguistics of praxis. In F. Newmeyer (ed), The Cambridge Linguistic Survey, Part IV, pp.210-28. Cambridge: Cambridge: Cambridge University Press.

Gumperz, J. 1982. Discourse Strategies. Cambridge: Cambridge University Press.

Gumperz, J. 1986. Introduction. In Gumperz & Hymes (eds), pp.1-25.

Gumperz, J. J. and Hymes, D. (eds). 1986. Directions in Sociolinguistics. Oxford: Blackwell.

Hymes, D. 1977. Foundations in Sociolinguistics. London: Tavistock.

Hymes, D. 1986. Models of the interaction of language and social life. In Gumperz. and Hymes (eds), pp.35-71.

Keating, E. 2001. The ethnography of communication. In P. Atkinson, A, Coffey, S. Delamont, J. Lofland and L. Lofland (eds), Handbook of Ethnography, pp.285-301. London: Sage.

Saville-Troike, M. 1989 The Ethnography of Communication (Second Edition). Oxford: Blackwell.

Saville-Troike, M. 1996. The ethnography of communication. In S.L. McKay (ed), Sociolinguistics and Language Teaching, pp.351-80. Cambridge: Cambridge University Press.

The story

I suggest that you approach this in the following way:

1. Read Labov’s treatment of the subject, either in his book or in the extract reprinted in Joworski and Coupland.

2. Read my brief notes below to check that you have identified the essential elements.

3. Study my analysis of the ‘James Bond’ story.

4. Study the ‘Students as villains’ story and identify any interesting elements in it. Compare your findings with mine in my 1999 paper.


If you plan to write your assignment on this topic, you should explore the papers that appear in the Reading section below (Thornbury & Slade includes a useful chapter). You could also search the Journal of Pragmatics to see whether there have been any studies of narrative in your own language.

Some definitions

We define narrative as one method of recapitulating past experience by matching

a verbal sequence of clauses to the sequence of events which (it is inferred)

actually occurred. Labov 1999: 225


  • A minimal narrative is a sequence of two narrative clauses that are temporally ordered.
  • A narrative clause is confined by temporal juncture. This means that if the order is changed the inferred temporal sequence is changed.
  • A free clause is not confined by any temporal juncture.

Here is an example from Labov 1977/1999. Identify the narrative clauses then check your answer against Labov’s analysis:

a. I know a boy named Harry.

b. Another boy threw a bottle at him right in the head

c. and he had to get seven stitches.

Labov 1999: 227

Story Structure

Labov describes the elements in narratives, then sums these up in terms of the questions to which they respond: 

Abstract What was this about?

Orientation Who, when, what, where?

Complicating Action Then what happened?

Evaluation So what?

Result What finally happened?


Note that the coda is a signal that the narrative has finished and therefore does not respond to a question. Labov says that if you ask the question, ‘And then what happened?’ after a coda, the only possible response is, ‘Nothing; I just told you what happened.’

Types of Story

Narrative As above

Anecdote Notable event and reaction

(resolution not expected)

Exemplum Told to make a moral point

(typically incident and interpretation)

Recount Essentially expository

(record of events)

(Thornbury and Slade 2006: 152-158)

Here is the story on the handout (but note that the alignment on this webpage version is less accurate than on the original):



Annette: Because I just let something happen then that I shouldn’t have let happen.


E:r (.) the:: (.) they’ve been doing a story about-reading about James Bond, and then >we were-< answering some questions on it, and one of the questions said em:: (0.5) er (.) >I can’t remember what the que- >the exactly< question was but it started with Bond, (.) Bond (.) did such and such and they were (.) to say whether it was true or false.=

Keith: =Right.


Annette: And e:m (.) Shafie got out his dictionary and was looking up (.) what I thought was an important word in the question that he didn’t understand, (.) and he was looking up ‘bond’. (.)

Keith: HA! ┌Hahahah

Complicating action 1

Annette: └Because it came at the beginning of the sentence, (.) er (.) ┌he therefore ┐

Keith: └Right right ri┘ght.

Annette: didn’t realise that that capital letter meant that it ┌was ┐ a name, (.)

Keith: └Yes.┘

Embedded orientation 1

Annette: and he’s (.) he s- he showed me in his dictionary

Complicating action 2

Annette: because I- I thought I’d better go and check what he was ┌looking ┐ up.

Keith: └Oh right.┘

Embedded orientation 2

Annette: And then he said ‘It’s this ‘bond’, it says ‘money’ and ‘stocks and shares’ or something. °And° lots of meanings.’

Keith: Hahaha::h=

Complicating action 3

Annette: And I said ‘No no,

Keith: ( )=

Annette: =it’s it’s (.) James Bond,’ >I mean< I pointed to the name on the board and he said ‘O::h yes.’ Heheheh

Keith: Beautiful.


Annette: But I thought I should have picked up on that earlier.

Keith: It’s nice though. Real confusions.

Annette: Yes.

Keith: Yeah. Mmm. (.) Heh




All of the following are available from Warwick library:

Jefferson, G. 1978. Sequential aspects of storytelling in conversation. In J. Schenkein (ed.), Studies in the Organization of Conversational Interaction, pp.219-248. New York: Academic Press.

Johnstone, B. 2003. Discourse analysis and narrative. In D. Schiffrin, D. Tannen and H. E. Hamilton (eds), The Handbook of Discourse Analysis, pp.635-649. Oxford: Blackwell.

Labov, W. 1977. Language in the Inner City. Oxford: Blackwell. Reprinted in A. Jaworski and N. Coupland (eds), The Discourse Reader, London: Routledge, 1999. Pages 221-235.

Norrick, N. R. 2001. Discourse markers in oral narrative. Journal of Pragmatics, 33(6): 849-878

Norrick, N. R. 2005. Interactional remembering in conversational narrative. Journal of Pragmatics, 37(11): 1819-1844.

Schiffrin, D. 1981. Tense variation in narrative. Language, 57 (1): 45-62.

Tannen, D. 1982. Oral and literate strategies in spoken and written narratives. Language, 58(1): 1-21.

Thornbury, S. and Slade, D. 2006. Conversation: From Description to Pedagogy. Cambridge: Cambridge University Press

Toolan, M. J. 2001. Narrative: A Critical Linguistic Introduction (2nd Edition). London: Routledge.

Wolfson, N. 1979. The conversational historical present alternation. Language, 55(1): 168-182.


The following is available from the Applied Linguistics Resources Centre:

Richards, K. 1999. Working towards common understandings: Collaborative interaction in staffroom stories. Text, 19(1): 143-174.

This page has no content yet