Journal of Pragmatics 39 (2007) 2949 www.elsevier.

com/locate/pragma

Talk in a play frame: More on laughter and intimacy

Jennifer Coates
School of Arts, Roehampton University, Roehampton Lane, London SW15 5PH, United Kingdom Received 16 March 2005; received in revised form 4 January 2006; accepted 5 May 2006

Abstract Conversation is one of the key locuses of humour and it is now widely agreed that shared laughter nurtures group solidarity. This paper will explore the links between laughter and intimacy in everyday conversation. The paper will attempt to clarify the term conversational humour, focussing on informal conversation among friends and on the conversational practices involved in humorous talk. I argue, following Bateson, that conversational humour involves the establishment of a play frame. When a play frame is established, speakers collaborate in the construction of talk in a way that resembles group musical activity, particularly jazz. This way of talking is characterised by, among other things, overlapping speech, the co-construction of utterances, repetition, and a heightened use of metaphorical language. I will argue that play and creativity are linked in signicant ways, and that playful talk is essentially collaborative. # 2006 Elsevier B.V. All rights reserved.
Keywords: Humour; Laughter; Intimacy; Solidarity; Conversation; Collaborative talk; Play

1. Introduction In this paper I shall examine humorous talk occurring in the informal conversation of friends. I shall argue that humorous talk is a form of play, and that talk as play can only be achieved by close collaboration between speakers. Collaboration between speakers constructs solidarity, and thus a key function of playful talk is the creation and maintenance of group solidarity, of intimacy between speakers. In this respect, I shall pursue the line begun by Jefferson et al. (1978) in their paper Notes on laughter in the pursuit of intimacy. I shall examine some of the characteristics of talk as play, drawing on a corpus of informal conversational data involving pairs or groups of

E-mail address: jetcoates@gmail.com. 0378-2166/$ see front matter # 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.pragma.2006.05.003

30

J. Coates / Journal of Pragmatics 39 (2007) 2949

friends, and will argue that talk as play shares features with music, particularly with jazz. The complex, often polyphonic, textual patterns of playful talk index the complex, intricate and intimate links between speakers. 2. Language and humour After many years of relative neglect, humour is now the focus of attention in a range of work being carried out by social psychologists, sociolinguists and conversation analysts, and in a variety of contexts. These include the workplace (Holmes, 2000; Holmes et al., 2001; Holmes and Marra, 2002; Mullaney, 2003); the classroom (Kehily and Nayak, 1997; Davies, 2003); medical settings (DuPre, 1998; Astedt-Kurki et al., 2001; Sullivan et al., 2003); TV discussion groups (Kotthoff, 2003); as well as informal settings such as the home (Norrick, 1993a, 1993b, 2004; Gibbs, 2000; Hay, 2000; Everts, 2003; Coates, in press). In this paper, I shall focus on humour involving conversation among friends in informal settings. Despite growing interest in talk and humour, there does not seem to be general agreement on the meaning of the term conversational humour. Many researchers have used what seems to me a rather narrow interpretation of this term, focusing on specic speech acts such as telling a joke, making a pun, being sarcastic or ironic (see, for example, Chiaro, 1992; Attardo, 1993; Norrick, 1993b; Gibbs, 2000). It could be that this bias in emphasis is gender-related. Recent research exploring gender variation in humour has established a clear pattern of difference among speakers, with men preferring more formulaic joking and women sharing funny stories to create solidarity ` s-Conde, 1997; Hay, 2000; Crawford, (see Crawford and Gressley, 1991; Boxer and Corte 2003). As Crawford (1995:149) remarks, Womens reputation for telling jokes badly (forgetting punch lines, violating story sequencing rules, etc.) may reect a male norm that does not recognise the value of cooperative story-telling. So perhaps the foundational work done by men (e.g. Mulkay, 1988; Attardo, 1993; Norrick, 1993a) grew out of their own orientation to humour. On the other hand, signicant contributions to the literature on humour by female linguists, such as Tannens (1984) chapter (entitled Irony and joking) and Chiaros (1991) book (entitled The Language of Jokes) suggest that a focus on joking rather than humour in conversation is widely accepted as appropriate. Indeed, the rst book-length examination of conversational humour (Norrick, 1993a) is called Conversational Joking, not Conversational Humour. While joking is clearly part of humour, it is surely the case that humour is a much broader, more fuzzy-edged category than the term joking implies. In British English, telling a joke is a very specic speech act, that is, a short formulaic utterance, ending with a punch line, which produces (or is meant to produce) laughter. Telling a joke, moreover, is an activity only rarely associated with friendly conversation. This is not surprising, given that, a joke . . . is likely to disrupt a normal or serious conversation (Chiaro, 1992:114). Moreover, joke-capping sessions (where one speaker tells a joke and then a second speaker tells a joke and so on) are not an everyday occurrence (ibid:113). Chiaros claims are supported by the evidence of my corpus of informal conversation involving friends or family. In this corpus I have found no joke-capping sessions and only one short passage that could be described as a joke-telling. There are some instances in the corpus of the combative style which can be labelled joking around. This is conned to the talk of the youngest speakers (1217 years) and is more frequently used by male speakers. This

J. Coates / Journal of Pragmatics 39 (2007) 2949

31

conforms to the patterns observed by other commentators (see, for example, Back, 1994; Pilkington, 1998; Kehily and Nayak, 1997; Frosh et al. 2002). Frosh et al. (2002) claim that combative humour is central to the construction of masculine identities and hierarchies (op.cit.:103) and that joking seems to be a way of establishing intimacy between men and . . . excluding women (op.cit.:232).1 Joking around and having a laugh are common on the street and in school playgrounds, but not, it seems, so common in friendly informal adult conversation. The conversations Ive collected, then, are not characterised by joke-telling or by jokey repartee, but this does not mean there is no humour. On the contrary, laughter occurs in nearly all the conversations in the corpus. It emerges as the result of humorous stories, or of bantering or teasing among participants, or when speakers pick up a point and play with it creatively. Everyday conversation exhibits spontaneous outbursts of verbal play. These differ from jokes in that a joke is a ready-made unit, learned and repeated by a speaker to amuse an audience. Jokes do not emerge spontaneously in conversation but stand apart from the ow of talk and interrupt its progress. The kind of humorous talk I shall focus on in this paper, by contrast, emerges organically from the ongoing talk and involves the participation of all present. Crucially, it requires conversational participants to adopt an interactive pact (McCarthy and Carter, 2004:172). Unlike a joke, which can be understood away from the context in which it was performed, spontaneous conversational humour relies on shared knowledge and in-group norms, which can make it opaque to outsiders. Everts (2003), for example, discusses a family conversation where the utterance Hes from Virginia is repeated in varying forms throughout the conversation and is unambiguously humorous. A locally emergent expression (Tannen, 1990:45) such as this may acquire a degree of xity or formulaically [sic] as it is repeated throughout the conversation, but does not outlive the conversation out of which it emerges (Everts, 2003:388). Where conversational co-participants collaborate in humorous talk, they can be seen as playing together. Their shared laughter arises from this play and is a manifestation of intimacy, with the voice of the group taking precedence over the voice of the individual speaker. The notion of play is at the heart of what I mean by humour. Everts comments that when she was coding utterances in her conversational data as serious or humorous, she kept in mind the broader interpretation of humour as play (Everts, 2003:379). It is this broader interpretation of humour that will be adopted in this paper. 3. Talk as play The idea of talk as play draws on Batesons (1953) idea of a play frame. Bateson argues that we frame our actions as serious or as play. Conversational participants can frame their talk as humorous by signalling This is play. The notion of a play frame captures an essential feature of humour that it is not serious and at the same time avoids being specic about the kinds of talk that can occur in a play frame: potentially anything can be funny. For a play frame to be established in talk, conversational participants must collaborate with each other. As Holmes and Hay (1997:131) observe: Successful humour is a joint construction
Boys interviewed by Stephen Frosh and his associates saw humour as something exclusively masculine: they implied that girls were not sufciently robust to engage in jokey banter (Frosh et al., 2002:103).
1

32

J. Coates / Journal of Pragmatics 39 (2007) 2949

involving a complex interaction between the person intending a humorous remark and those with the potential of responding. Collaboration is an essential part of playful talk, since conversational participants have to recognise that a play frame has been invoked and then have to choose to maintain it. Because conversational humour is a joint activity, involving all participants at talk, many commentators see its chief function as being the creation and maintenance of solidarity (see, for example, Norrick, 1993a, 1993b; Hay, 1995; Crawford, 1995; Boxer and ` s-Conde, 1997; Holmes and Hay, 1997). The creation of solidarity is an inevitable Corte consequence of the joint construction of a play frame, since interactants who collaborate in humorous talk, necessarily display how nely tuned they are to each other (Davies, 2003:1362). Research to pinpoint the contextual cues which signal a play frame has discussed both speaker intention and audience response. Since textual analysis cannot access speaker intention, analysts have tended to focus on audience response, including verbal responses such as agreement, mirroring, or parody. In particular, analysts have focussed on laughter as a key response. Jefferson (1979, 1984, 1985) carried out meticulous analysis of where laughter occurs in talk, using a Conversational Analytic approach. More recently, Holmes and colleagues working on the Language in the Workplace Project (Victoria University, Wellington, NZ) have paid attention to laughter because of its frequency and salience in their conversational data. However, they do not restrict themselves to laughter. Other contextual clues they consider are the preceding discourse, speakers tone of voice, sudden changes in pitch or rhythm, and paralinguistic clues such as the use of a laughing or smiling voice (Holmes and Hay, 1997:132). Humour often lies in the gap between what is said and what is meant. When a play frame is invoked, we have the choice of joining in the play and responding to what is said, or of reverting to the serious mode. Kotthoff (2003) compared ironic humour in TV discussions with ironic humour in dinner-time conversations and found that, in the TV discussions she analysed, the speakers preferred to return to the serious mode, whereas [i]n informal situations among friends, the preferred strategy is to continue in the humorous key and respond to the said (Kotthoff, 2003:1408). In other words, in relaxed friendly talk, speakers collaborate in talking about one thing while meaning something else, thus maintaining a play frame. As McCarthy and Carter (2004:161) say, in a discussion of hyperbole, it fundamentally depends on a joint acceptance of a distortion of reality. One of the strengths of humour is that it allows us to explore, in new ways, what we know, and even, by using other words, to explore things which are difcult or taboo. There is growing evidence that talk-as-play is qualititatively different from serious talk. Such evidence can be gleaned from the metaphors researchers draw on to represent talk in a play frame. Sully (1902 quoted in Norrick, 1993a:141) talks about the choral nature of playful talk. Davies (2003), working with second language learners, claims that conversational humour is a collaborative activity, and that the verbal art of this specialised joint activity most closely resembles jazz in the world of music (Davies, 2003:1368). Sawyer (2001:19) also draws on jazz as a metaphor, and I have described the collaborative talk of women friends in terms of a jam session (Coates, 1996:117118). These metaphors make parallels between talk and music. Similarly, parallels between spoken language use and poetry have not escaped the scrutiny of linguists (see in particular Tannen, 1990). But in the case of talk-as-play, what seems to be most salient is the collaborative, all-in-together nature of the talk. Humorous talk often involves speakers constructing text as a joint endeavour, just as jazz musicians co-construct music as they improvise on a theme.

J. Coates / Journal of Pragmatics 39 (2007) 2949

33

It is not only in informal contexts involving intimates or close friends that talk-as-play is found. It is always possible for a speaker to introduce a play frame, and research looking at workplace contexts, for example, has uncovered a surprising amount of playful talk (see, for example, Holmes, 2000; Holmes et al., 2001; Mullaney, 2003). All of us, as competent speakers, can switch talk from serious to playful modes. Where talk occurs in a formal context, interactants may switch to a play frame from time to time to defuse tension or to provide light relief from a boring agenda, for example. But in informal contexts where interactants know each other well, talk may switch repeatedly between serious and non-serious frames, and conversational participants will collaborate with each other to bring about the switches. The unpredictability of this kind of talk is part of what makes it fun for participantsanyone can trigger a switch at any time. In this paper, I shall focus on informal talk among friends, because this talk has the potential for frequent and often extended talk in a play frame. 4. Some examples of talk in a play frame In this section, I shall examine three examples of humorous talk. In each example, one of the speakers switches from a serious to a play frame, a switch which is recognised by the speakers co-participants, who proceed to co-construct talk in the play frame. In this section I shall only comment briey on the extracts: detailed analysis of the language features which characterise this humorous talk will be carried out in section 5. The three examples are taken from three very different conversations: rst, a conversation involving three women friends in their 30s, over supper at a house in Surrey, England; second, a conversation involving three young women friends, in their early 20s, all students in Melbourne, Australia; third, a younger group of three young men friends, aged about 19, in South London.2 These extracts have been chosen in part to demonstrate that humorous talk is not the preserve of one particular group but is a normal aspect of friendly everyday talk. The rst example shows how conversational participants can draw on what has been talked about in a serious frame earlier in conversation. Sue tells her two friends that she has brought the school rabbit home for the weekend. They talk briey about the rabbit before the conversation moves on through other topics to a discussion of marriage and relationships. Sue tells a story about a couple she knows where the wife has forbidden the husband to play his guitar, or even to have a guitar in the house. This raises issues about obedience and appropriate behaviour in relationships, and after some more serious talk about the husbands wild youth and nearalcoholism, Sue re-introduces the rabbit theme. The example below represents a very small part of the discussion of the obedient husband. (The extract has been transcribed using stave notation.)3

2 I am extremely grateful to all those who have allowed their conversations to be recorded and analysed as part of my research. Names have been changed to provide anonymity. I would also like to thank Mary-Ellen Jordan who collected the Melbourne data and who has collaborated with me in analysing it (see Coates and Jordan, 1997). 3 This means that all participants contributions are to be read simultaneously, like instruments in a musical stave. Any word, or portion of a word, appearing vertically above or below any other word, is to be read as occurring at the same time as that word. This system allows the reader to see how the utterances of the different participants relate to each other.

34

J. Coates / Journal of Pragmatics 39 (2007) 2949

At the beginning of this extract, the three friends ponder on the obedient husbands life. Lizs utterance oh bless him he doesnt have much of a life triggers Sues laughter as she responds he doesnt really. The switch to a play frame is achieved by the mocking, quasi-maternal tone which

J. Coates / Journal of Pragmatics 39 (2007) 2949

35

Liz adopts in relation to the obedient husband. Sue then introduces a new dimension with her simile: hes like the rabbit, and warms to her theme, continuing I think I should bring him home for weekends. Liz joins in with the suggestion that the bossy wife should get the husband/rabbit a run in the garden, while Anna suggests the two rabbits could meet. Liz fantasises that the husband/rabbit would be happy with a few lettuce leaves and adopts an ingratiating voice to mimic the husband thanking his wife for the lettuce. This is a very good example of Kotthoffs (2003) claim that the co-construction of humour relies on participants responding to what is said (playing with the theme of rabbits, of bringing pets home for the weekend, of making runs in the garden), rather than to what is meant (wives and husbands should have a more equal relationship and should not order each other round). The repetition of the rabbit theme makes the talk of these friends textually cohesive. By reverting to the rabbit theme and using rabbit as a metaphor for obedient husband, these friends are able to play with the parallels that this throws up and to say some pretty devastating things about the couple and their relationship. The next example comes from a conversation between three friends, all students at Melbourne University. At this point in the conversation, Amanda tells her two friends (Jody and Clare) that the mother of a friend of theirs is proposing to marry the man she has been having an affair with for a month. All three friends are horried at the news, but they use humour to good effect to express their critical view of heterosexual marriage, of the particular man talked about and by implication men in general, and to have a laugh about an earlier joke about Clare, sex and the computer.

36

J. Coates / Journal of Pragmatics 39 (2007) 2949

Jodys words whatever they do together in stave 2 are initially received with only a minimal response from Amanda. But Jody chooses to re-focus attention on the idea of whatever they do together by adding I hate to think. This reframes the phrase: whatever they do together is now marked as both humorous and sexual. Clares recognition that a play frame has been introduced is marked by her laughing protest, while Amanda maintains the frame with the joke its probably

J. Coates / Journal of Pragmatics 39 (2007) 2949

37

heterosexual, a joke which inverts the normal pattern of heterosexual unmarked/homosexual marked. Amandas joke is picked up with relish by the other two speakers: Jody launches into a series of utterances which talk about the mans mobile phone, with heavy sexual innuendo. Clare responds to Amandas comment in kind, with the utterance well we KNOW what they do then DONT we in a mock-patronizing reference to the act of sexual penetrationthe implication here is boring! and/or predictable. At the same time, she cohesively ties in Jodys reference to mobile phones by saying, in effect, that what we imagine them doing involves a mobile phone in some unspeakable way. This reading is conrmed by Amanda subsequent teasing remark to Clare: youre the techno sex guru, Clare, you can hardly talk, in which the reference to techno-sex can be understood only if Clares utterance has something to do with techno-sex. The mobile phone joke recurs throughout their talk, and the play frame is maintained throughout the succeeding conversation, with the young women constantly sending up the normative discourse of Romantic Love.4 The last extract comes from a conversation in which three young men are discussing whether miracles are possible (all three are involved in their local church). In this example, Des reinterprets a Bible story with help from Jack and HavJack and Havs contributions are in italics. (This example is a narrative, with one speaker taking the role of narrator; it involves less overlap than the preceding examples. It has been transcribed according to narrative conventions.)5 (3) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21
4

MIRACLES But supposing that he raised someone from the dead? [. . .] It was a little girl [yeah] and she was dead he got to the house too late when she was dead, I cant remember all theAnd. he- he said- he said something ((to her)) Get up, ((xx)) stupid cow and she got up, she was alive. But. when you think about how shit medicine was in those days [mhm] I mean who says she was dead? [yeah I know] she could have been in a coma [yeah] and he could have like triggered something off she could have been lying she could have been lying she could have been really pretending **very very well** **she could have been like-** Right ((Im gonna sort that out)) <CLAPS HANDS> <LAUGHTER>

The entire transcript of this humorous chunk of conversation can be found in the Appendix to Coates and Jordan (1997). 5 The story is presented in numbered lines, each line corresponding to one of the narrators breath-groups or intonation units, typically a grammatical phrase or clause (Chafe, 1980).

38

J. Coates / Journal of Pragmatics 39 (2007) 2949

I cant -I cant keep this up much longer, <SIMULATES GIRLS HIGH VOICE> <LAUGHTER> 23 fucking stupid bearded cunt, <LOUD LAUGHTER> 24 go on, fuck off, <LAUGHTER> 25 for fucks sake, the bastard. <QUIETER> 26 Get up <SERIOUS VOICE> <LAUGHTER> 27 Thank God for that, <REVERTS TO GIRLS VOICE> <LAUGHTER> 28 oh, Jesus Christ <LAUGHTER> that was hard. <LAUGHTER> 29 Right, wheres my ver. <NORMAL VOICE> <LAUGHTER> 30 But I mean lets face it, 31 medicine was crap. [yeah] [words between double asterisks were spoken at the same time] 22 This story is typical of many in the all-male conversations Ive collected, particularly those of younger men. Talk in such conversations switches constantly between serious and non-serious frames, and the men involved collaborate with each other to bring about the switches. This example is in two parts: the rst (serious) part is a re-telling of a Bible story about a miracle; the second (nonserious) part involves a dramatised re-interpretation of the story. Hav attempts to switch to a nonserious frame early on with his contribution Get up, ((xx)) stupid cow (line 9), but Des ignores Hav and completes his initial re-telling of the story. There is then a transitional section (lines 1220) constructed by two speakers, Des and Jack, before the switch to a play frame is fully accomplished. This is an interesting stretch of talk in which the speakers play with ideas which become increasingly fantastic. It is only at line 20, when Hav again uses simulated dialogue: Right ((Im gonna sort that out)) that the humorous version of the story nally takes off. This time, Des accepts the play frame and tells the story again entirely in reported speech. Havs earlier words allow Des the economy of simply saying Get up in a more serious voice (line 26) to bring off the animation of Jesus in the story. Once Des begins his comic re-telling of the story in dialogue, Hav lets him have the oor, and he and Jack become speechless with laughter. Dess joking demand for payment (right wheres my ver?) recognises that he has succeeded in amusing his friends. But note that his switch back to a serious frame in the next line (But I mean lets face it, medicine was crap . . .) is achieved through the co-operation of Jack and Hav: they stop laughing and the minimal response yeah at this point signals their acceptance of the switch back to a serious frame. 5. The linguistic and para-linguistic features of talk in a play frame I have argued that playful talk is qualititatively different from other kinds of talk and that it can often be described in terms of music, particularly jazz. In this section I shall examine ve features of talk in a play frame, features which seem to be intrinsically involved in what it means to play conversationally. The features are: overlapping speech, the co-construction of utterances, repetition, laughter, and metaphor. These ve features are often co-present in a given stretch of talk: for example, overlapping speech often involves repetition, co-constructed utterances often involve two people speaking at the same time and repeating elements from the preceding discourse, and laughter often overlaps with ongoing talk. 5.1. Overlapping speech A play frame can only be established if all conversational participants collaborate in sustaining it. This requires that talk is jointly constructed in a much stronger sense than

J. Coates / Journal of Pragmatics 39 (2007) 2949

39

that intended by Grices notion of cooperation (Grice, 1975). Collaboratively constructed talk, in this strong sense (see Edelsky, 1993; Coates, 1996, 1997a; Coates and Sutton Spence, 2001), starts from the premise that the conversational oor is potentially open to all participants simultaneously. This is very different from a one-at-a-time oor, where (as its name suggests) the oor is inhabited by only one speaker at any one time. Norrick (2004, following Goffman) talks about collaboratively constructed talk of this kind as a team performance. Overlapping speech is the inevitable outcome of joint ownership of the conversational oor. But far from leading to conversational breakdown, overlapping speech in a collaborative oor entails a richer multi-layered texture to talk, where speakers demonstrate their shared perspective on whatever is being talked about and display how nely tuned they are to each other (Davies, 2003:1362). The rst extract is a very good illustration of the kind of overlapping speech that is common in humorous talk among friends. These three friends have been playing for some time with the story of the obedient husband, and when Sue triggers a new play frame, they all collaborate in sustaining it. Their contributions to talk are made simultaneously: each of them develops the rabbit theme in their own way, yet they are clearly attending to each other at the same time, as the repetition of words and meaning demonstrates. The example begins with all three speakers overlapping, and repeating each others words. Liz says he doesnt have much of a life and then Anna and Sue speak simultaneously: [Anna: [Sue: he doesnt by the sounds of it he doesnt really <LAUGHS>

By saying the same thing at the same time, or by echoing what has just been said, interactants bind their utterances together and in this case prepare the ground for Sues hes like the rabbit, which follows straight on. Staves 3, 4 and 5 all consist of more than one speaker speaking at the same time as another speaker. In staves 34, for example, Sue and Liz overlap:

and then Liz and Anna overlap:

The effect of this complex pattern of overlapping is to give the impression of everyone speaking at once, but in a coherent, not chaotic, way. It is this kind of playful talk that has been likened to a jam session (Coates, 1996:117118) or more generally to jazz (Davies, 2003:1368; Sawyer, 2001:19), because speakers voices interweave like instruments improvising on a theme.

40

J. Coates / Journal of Pragmatics 39 (2007) 2949

The transcript of Example 2 shows that there is a great deal of overlapping talk in this extract. A very simple example occurs in stave 8 where Amandas cyber sex overlaps with Jodys virtual sex. Again, we see that where overlap occurs, we also nd repeating patterns (here, of words and ideas). This means that interactants are not having to pay attention to two or three disparate utterances simultaneously, but to two or occasionally three highly cohesive utterances. A more extended example of overlap comes in stave 3: Jody says hes got a bloody mobile phone and then adds he wears it round his waist; simultaneously Clare laughs and comments well we KNOW what they do then DONT we? In overlapping talk like this, involving whole clauses, participants rely on given information (Jody has established that she is focusing on mobile phones) to attend simultaneously to Jodys continued joking and Clares elaboration of Jodys joke. There is no evidence from any of the humorous talk I have collected that participants cannot follow what is going on: on the contrary, the evidence of subsequent talk is that speakers relish choral talk of this kind and are stimulated to make further humorous contributions. In extract 3, we again nd overlapping talk, again coinciding with repetition. Here is the chunk of talk where two speakers talk simultaneously, this time re-organised in stave format:

In stave 2, we can see how Des starts a fourth contribution to the list of she could have been utterances at the same time as Jack extends his utterance with an adverbial, which enriches the syntactic pattern they have established between them. At this moment, Hav seizes the opportunity of Dess ambiguous like to switch into a pretend voice and to use direct speech

J. Coates / Journal of Pragmatics 39 (2007) 2949

41

(be like is now a very common quotative for young speakersbut Des might have been using like as a hedge while he searched for a new verb). Overlapping talk is more common in all-female talk than in all-male talk. The evidence of my database is that men in all-male groups prefer a one-at-a time pattern of turn-taking (for further discussion, see Coates, 1994, 1997b). In most groupings, male speakers avoid overlapping with another speaker, except where the other speaker is a female partner. Where overlap as part of the collaborative construction of talk functions as a display of heterosexual coupledom, men can produce talk as polyphonic as anything shown in the extracts above. But the pressures of hegemonic masculinity mean that male speakers must at all times avoid being perceived as gay. Presumably, conversational duetting, in particular overlapping talk, signals intimacy in a way which risks such a perception, unless the conversational partner is an actual (heterosexual female) partner (Coates, 2006). 5.2. Co-constructed utterances The previous example where one speaker (Hav) continues an utterance begun by another speaker (Des) illustrates how speakers co-construct utterances. Co-construction can involve a second speaker adding just a single word or an entire clause to an utterance, but in all cases of co-constructed utterances, what is achieved is two speakers speaking as if with a single voice. The next two examples are very simple; they are given here to illustrate the concept of co-construction (both examples involve Anna, Sue and Liz, from a different point in their conversation):

In their playful talk about rabbits (and obedient husbands), Sue, Liz and Anna co-construct talk drawing on a rhetorical question structure. Liz, in stave 4, asks a question beginning with the words I wonder why she doesnt . . . and Anna completes this utterance with the words introduce them. At the same time, Liz completes her own utterance with the words get him a run in the garden. This syntactic pattern proves very productive: eventually all three friends use it to develop the fantasy about the husband/rabbit through adding a second part to the utterance (which is not repeated):

42

J. Coates / Journal of Pragmatics 39 (2007) 2949

LIZ: ANNA: ANNA: SUE: SUE: LIZ:

I wonder why she doesnt

get him a run in the garden [stave 4] introduce them [stave 4] introduce them [stave 5] bring him home at weekends [stave 5] (and) let him go out in the garden [stave 6] give him a few lettuce leaves [stave 6]

It is arguably the case that Sue is not adding a second part to Lizs utterance, but is adding a new second part to her own earlier utterance I think I should [bring him home] [stave 3]. In either case, we see speakers constructing utterances where production is shared between speakers, or where production involves cohesive repetition. A further example of co-construction comes from the Melbourne data: in the conversation which follows example (2), Jody, Clare and Amanda continue their playful demolition of the man who is going to marry their friends mother. They ridicule his liking for strudels, in particular, his willingness to drive miles to buy strudels.

Amanda and Clares jointly achieved utterance in stave two here is a very simple example of coconstruction: Amanda begins the clause and Clare completes it. But at the same time, Amanda and Clare are collaborating in reinforcing Jodys initial claim you wouldnt drive from Palm Cove to Bondi, with Amanda repeating the verb drive and Clare repeating the pattern ( from) X to Y (where X and Y are place names). Again, there is evidence that co-constructed utterances are a more normal feature of all-female talk than of all-male talk. As discussed above, only when duetting with a woman partner do men seem happy to deploy collaborative patterns like this which so strongly index mutual knowledge and awareness. 5.3. Repetition As discussion of overlapping talk and co-constructed utterance reveals, repetition is a striking feature of talk in a play frame (see also Norrick, 1993b, 1994). Repetition may occur at many levels: lexical, semantic, syntactic, thematic. In the case of lexical repetition, it seems that once a play frame is in place, we nd locally emergent expressions (Tannen, 1990:45) which become charged with humorous meaning, so rabbits in the rst example becomes charged, as do the words associated with this lexical eld such as run, garden and lettuce. Entire phrases in the rst extract (example 4) are also repeated,

J. Coates / Journal of Pragmatics 39 (2007) 2949

43

with identical, or near-identical words: bring him home at/for weekends (staves 4 and 5); get him a run in the garden/let him go out in a run (staves 4 and 6). It is important to remember that these words and phrases repeat words and phrases used earlier in the conversation during the Rabbit topic. In the second extract (Example 2), the establishment of a play frame means that a term like mobile phone can become highly charged, to the extent that it takes on a multiplicity of meanings, mostly sexual, yet these are never pinned down. The power of the word resides precisely in the lack of clarity as to its meaning in this chunk of talk: innuendo works by forcing the listener to ll the gaps. Amandas question would you want to bloody use this mans mobile phone? at the end of the extract is the climax of a series of rhetorical questions and is funny because the three friends are now tuned to interpreting mobile phone in a non-literal way. The word sex is also repeated in various combinations: heterosexual; techno-sex; cyber-sex; virtual sex. In the last of the three extracts, the use of particular words is crucial to the participants understanding of the second telling of the story; in particular, the phrasal verb get up (lines 9, 10, 26) provides important lexical cohesion between the two parts. In all three examples, the repetition of words and phrases creates lexical cohesion. Syntactic cohesion is also a feature of talk in a play frame. At the end of the second example, the utterance would you want to bloody use this mans mobile phone? only has such an impact because of its positioning in a series of rhetorical questions: would you want to marry this man? would you want to be in the same room as this man? would you want to bloody use this mans mobile phone? Part of the humour here lies in the fact that the speakers produce these questions as if each one presents a worse scenario than the last, whereas each one actually refers to a more trivial situation (on a normal reading of mobile phone). The accumulated force of these three questions is to underline the rstin other words, to revert to the main theme of the conversation, which is their feeling that the friends mother should not marry this man. In the third example, Dess question I mean who says she was dead? sparks a series of utterances, all using the same syntactic pattern: Subject + could have + V: she could have been in a coma and he could have like triggered something off she could have been lying she could have been lying she could have been really pretending very very well The syntactic repetition here signals an increasingly playful mode of talking, with hypotheses becoming more and more fanciful. Repetition allows the talk to move from a serious to a play frame in a very coherent and smooth way; at the same time, the repetition by different speakers of the same syntactic patterns binds the three speakers contributions together. In playful talk, individual voices are less important than the jointly constructed talk. This is why humour is so effective as a means of creating solidarity. Repetition at the semantic level means that speakers say things with the same or similar meanings, but using different words. A good example is the phrase whatever they do together which is played with in various ways in the second example. Because of the semantic emptiness of do, it is possible for every following verb to be infected by the sexual meanings suggested by

44

J. Coates / Journal of Pragmatics 39 (2007) 2949

this phrase. So a supercially innocuous comment like hes got a bloody mobile phone (stave 4) takes on particular semantic overtones, specically sexual ones. This is reinforced by Clares subsequent comment well we KNOW what they do then, which refers back directly to whatever they do together. Another example of semantic repetition comes in the framing set up in Extract (3). In line 12, Des makes the claim But when you think about how shit medicine was in those days. This comes just before his question I mean who says she was dead? which triggers the series of statements which switch the talk into a play frame. This claim is then balanced at the end of the extract by a semantically similar utterance: But I mean lets face it, medicine was crap (line 31). This comes at the end of Dess performance and marks the return to a serious frame. The core of these two utterances is: medicine was shit (line 12) medicine was crap (line 31) These parallelisms make the talk coherent and bind speakers turns together. 5.4. Laughter In their seminal (1978) paper, Jefferson et al. argue that for conversational participants, laughter has the status of an ofcial conversational activity (op.cit.:156). This is an important claim, which moves us away from the idea that laughter is just an accompaniment to talk: it is talk. They also observe that, when conversational participants collaborate in humorous talk, they achieve a display of not merely laughing at the same time, but laughing in the same way (174). These insightful observations (together with the papers title) suggest that laughter and intimacy are signicantly linked.6 Laughter allows participants in playful talk to signal their continued involvement in what is being said, and their continued presence in the collaborative oor. If we assume that a collaborative oor is at all times open to all speakers, then clearly speakers need strategies to signal that they are participating, even when they do not actually produce an utterance. Laughter ts this requirement perfectly. It allows people to signal their presence frequently, while not committing them to speak all the time. At the same time, laughter is the chief culturally recognised way that we acknowledge humour in talk. While analysts cannot rely on laughter as a sign of (successful) humour (laughter may signal surprise or embarrassment, for example), it is certainly an important contextual cue. Not surprisingly, all three extracts show frequent laughter from all participants. In playful talk, laughter may involve the current speaker laughing at their own humour, as in the following examples (from extracts 1 and 2, respectively): (i) (ii) SUE: I wonder why she doesnt get him a RUN in the garden <GIGGLING> AMANDA: I mean the man has a mobile phone <LAUGHING>

Laughter very often occurs when co-participants respond to something funny uttered by the current speaker, as in the following example:

Frustratingly, these observations are applied in the paper only to the laughter associated with improper talk (for example, talk about swimming naked).

J. Coates / Journal of Pragmatics 39 (2007) 2949

45

Note that Annas laughter overlaps with the nal chunk of Lizs turn; Anna responds to Lizs riff about lettuce leaves at a point when Liz might have stopped. But she chooses to embellish her point by acting out the rabbit/husbands thanks to his owner/wife. It is evidence of the coherence of playful talk that Annas laughter works both as a response to the rst part of Lizs utterance and as an accompaniment to the nal part. In the third extract, Hav and Jack laugh in response to every narrative line produced by Des (lines 2127 inclusive), to the extent that he feels justied in (humorously) demanding payment for his bravura performance (Right, wheres my ver). Laughter often occurs at the moment when a play frame is invoked: for example, Sue laughs as she introduces the notion of the obedient husband being like a rabbit, and Jack and Hav laugh when Des begins his dramatised version of the miracle story. Laughter, then, is an important contextual cue in establishing a play frame. Sometimes all participants in talk laugh simultaneously, as when all three young women laugh after Jodys quip this side of Clare hasnt come out yet in the second extract. This pattern (of all participants laughing at once) seems to correlate in many cases with the end of a play frame, as we see at the end of Miracles (extract 3), where group laughter at the end of the dramatized story and again after Des has demanded payment for his performance (lines 2829) signals a recognition that the play frame has potentially come to a close. Alternatively, group laughter can signal the end of a sub-section of humorous talk: in extract (2), there are three occasions when all three speakers laugh at the same time (in staves 7, 9 and 13). In each case, this group laughter coincides with the end of a strand of the extended talk about the man with the mobile phone. These examples suggest that laughter has several roles in playful talk. It signals amusement and appreciation when something humorous is said. It signals the presence in a collaborative oor of co-participants who are not the main speaker but who by laughing can show their involvement in the ongoing talk. It also marks the ongoing talk as solidary in that collaboratively constructed humour relies on in-group knowledge and familiarity. Finally, it plays an important role in structuring playful talk, both in marking speakers recognition of the establishment of a play frame and in marking its close. 5.5. Metaphor Finally, I shall look at the role of metaphor. One of the strategies drawn on by participants in playful talk to create solidarity and to subvert dominant discourses is the use of metaphor (see Gibbs, 2000; McCarthy and Carter, 2004). Metaphor can be dened as a linguistic device whereby one thing is described as if it were another. Its function is to create novel meanings that inspire and disturb by changing our perspective on reality (Eynon, 2001:353). Talk in a play frame frequently involves metaphor; it frequently involves exploitation of the gap between what is said and what is meant (Kotthoff, 2003). It has been suggested that metaphor is the principal device available to us . . . for arriving at a fresh conception of a familiar phenomenon (Hanne, 1999:44). In everyday talk, especially everyday talk among friends, what is talked about

46

J. Coates / Journal of Pragmatics 39 (2007) 2949

is familiar, not erudite. By entering a play frame, speakers can take a fresh look at the everyday and the familiar. Not only does this deepen our understanding of life and of the world around us, it also gives us amusement. While a single pun or a comic aside may amuse briey, here my focus is on sustained playful talk. Examples (1) and (2) demonstrate how metaphorical language functions in extended talk in a play frame. In example (1), the rabbit metaphor is introduced by Sue with her simile: hes like the rabbit (stave 2). From this point until stave 8, Sue and her two friends cooperate in sustaining the rabbit metaphor. During these seven staves, the co-participants never use the word husband, but talk only of the rabbit and of things to do with rabbits such as runs in gardens and lettuce leaves. Their frequent laughter displays their amusement, amusement which arises from their awareness of the gap between their talk about rabbits and the underlying meaning expressed by this metaphor. (Succinctly, a rabbit is perceived as a small, vulnerable, uffy creature while a husband is normatively supposed to be a strong, protective human male. The clash between these two sets of meanings makes these three women laugh, but it also allows them to express their uneasiness with the story of this deviant husband, deviant because he kowtows to his controlling wife.) In example (2), the use of metaphor is more allusive. The topic whatever they do together (stave 1) hints at sexual meanings, so that words used subsequently hint at metaphorical rather than literal meanings. The phrase mobile phone is used with phallic overtones throughout the extract. Even the statement hes an architect (stave 10) acquires sexual overtones, as does Jodys subsequent claim (omitted from transcript above) hes got a spa [jacuzzi] in his ofce (playing on the fact that the main feature of a jacuzzi is that water spurts out in an ejaculatory way). In both examples, the use of metaphorical language leads to an intensication of the humour, with each use of metaphor increasing the humorous impact of the talk. I have looked here at specic uses of metaphor, but in a more general sense, talk in a play frame involves a move away from the literal; as Gibbs (2000:25) says of irony, playful talk can be regarded as a special kind of gurative language. 6. Conclusions: humour and intimacy In this paper, my aim has been to demonstrate what speakers can do with playful talk, what the possibilities of a play frame are. The fact that women in same-sex friendship groups seem more likely to exploit these possibilities than men is a separate issue, which I have addressed in another paper (Coates, in press). The three examples discussed in this paper demonstrate that talk in a play frame can justiably be called a specialised joint activity (Davies, 2003:1368). A play frame can only be sustained if all conversational participants collaborate in sustaining it. This inevitably makes such talk solidary, since co-participants collaborate not only in sustaining a particular topic but also in sustaining a particular way of talking. Successful collaboration arises from shared understandings and shared perspectives, and is a strong demonstration of in-tune-ness. All three examples discussed illustrate this: in every case, all participants are involved, and all share in the maintenance of the play frame and demonstrate how well tuned they are to each other. As Ive shown, this way of talking may involve overlapping speech, the coconstruction of utterances, repetition, and a heightened use of metaphorical language. It is this cluster of linguistic features which has led commentators to describe playful talk in terms of music and particularly jazz. Playful talk is fun: friends meet and talk because, consciously or not, this is a form of play they ` s-Conde (1997:293) have pointed out, we all enjoy a good prize highly. As Boxer and Corte

J. Coates / Journal of Pragmatics 39 (2007) 2949

47

laugh. Laughter makes us feel good and it also demonstrates our togetherness with our fellow speakers. By exploring the linguistic and paralinguistic features of playful talk, my aim has been to improve our understanding of conversational humour, our understanding of talk in a play frame, and our understanding of the links between laughter and intimacy. Acknowledgements I would like to thank those who have read earlier drafts of this paper and whose comments have helped me to improve itJenny Cheshire, Margaret Gottschalk, tope Omoniyi, Frances Rock, Joanna Thornborrow. In particular, I would also like to thank one of the anonymous referees whose comments for the journal led me to make signicant revisions. Finally, I would like to thank Delia Chiaro for her encouragement while I was revising the paper during my time as Visiting Professor at the University of Bologna. References
Astedt-Kurki, Palvi, Isola, Arja, Tammentie, Tarj, Kervinen, Ulla, 2001. Importance of humour to clientnurse relationships and clients well-being. International Journal of Nursing Practice 7 (2), 119129. Attardo, Salvatore, 1993. Violation of conversational maxims and cooperation: the case of jokes. Journal of Pragmatics 19 (6), 537558. Bateson, Gregory, 1953. The position of humor in human communication. In: von Foerster, H. (Ed.), Cybernetics, Ninth Conference. Josiah Macey Jr Foundation, New York, pp. 147. Back, Les, 1994. The white negro revisited. In: Cornwall, A., Lindisfarne, N. (Eds.), Dislocating Masculinity. Routledge, London. ` s-Conde, Florencia, 1997. From bonding to biting: conversational joking and identity display. Journal Boxer, Diana, Corte of Pragmatics 27, 275294. Chafe, Wallace, 1980. The deployment of consciousness in the production of narrative. In: Chafe, Wallace (Ed.), The Pear Stories: Cognitive, Cultural and Linguistic Aspects of Narrative Production. Ablex, Norwood, NJ, pp. 9 50. Chiaro, Delia, 1992. The Language of Jokes. Routledge, London. Coates, Jennifer, 1994. No gap, lots of overlap: turn-taking patterns in the talk of women friends. In: Graddol, David,Maybin, Janet, Stierer, Barry (Eds.), Researching Language and Literacy in Social Context. Multilingual Matters. Avon, Clevedon, pp. 177192. Coates, Jennifer, 1996. Women Talk. Conversations Between Women Friends. Blackwell, Oxford. n, T. (Ed.), Coates, Jennifer, 1997a. The construction of a collaborative oor in womens friendly talk. In: Givo Conversation: Cognitive, Communicative and Social Perspectives. John Benjamins, Philadelphia, pp. 5589. Coates, Jennifer, 1997b. One-at-a-time: the organisation of mens talk. In: Johnson, Sally, Meinhof, Ulrike (Eds.), Language and Masculinity. Blackwell, Oxford, pp. 107129. Coates, Jennifer, 2003. Men Talk: Stories in the Making of Masculinities. Blackwell, Oxford. Coates, Jennifer, 2006. Everyone was convinced that we were closet fags: the role of heterosexuality in the construction of hegemonic masculinity. In: Kyratzis, Sakis, Sauntson, Helen (Eds.), Language, Sexualities and Desires: Cross-cultural Perspectives. Palgrave, London. Coates, Jennifer, in press. It was so funny: humour and the construction of gendered identities. In: Chiaro, D., Baccolini, R. (Eds.), Humour, Language and Gender. Gedit, Bologna. Coates, Jennifer, Jordan, Mary Ellen, 1997. Que(e)rying friendship: discourses of resistance and the construction of gendered subjectivity. In: Livia, Anna, Hall, Kira (Eds.), Queerly Phrased. Oxford University Press, New York, pp. 214232. Coates, Jennifer, Sutton Spence, Rachel, 2001. Turn-taking patterns in Deaf friends talk. Journal of Sociolinguistics 4 (4), 507529. Crawford, Mary, 1995. Talking Difference. Sage, London. Crawford, Mary, 2003. Gender and humor in social context. Journal of Pragmatics 35 (9), 14131430. Crawford, Mary, Gressley, Diane, 1991. Creativity, caring and context: womens and men accounts of humour preferences and practices. Psychology of Women Quarterly 15, 217231.

48

J. Coates / Journal of Pragmatics 39 (2007) 2949

Davies, Catherine Evans, 2003. How English-learners joke with native speakers: an interactional sociolinguistic perspective on humor as collaborative discourse across cultures. Journal of Pragmatics 35, 13611385. DuPre, Athena, 1998. Humor and the Healing Arts. Erlbaum, Mahwah, NJ. Edelsky, Carole, 1993. Whos got the oor? In: Tannen, D. (Ed.), Gender and Conversational Interaction. Oxford University Press, Oxford, pp. 189227. Everts, Elisa, 2003. Identifying a particular family humor style: a sociolinguistic discourse analysis. Humor: International Journal of Humor Studies 16 (4), 369412. Eynon, Terri, 2001. Metaphor: the impossible translation? British Journal of Psychoanalysis 17 (3), 353364. Frosh, Stephen, Phoenix, Ann, Pattman, Rob, 2002. Young Masculinities. Palgrave, London. Gibbs Jr., Raymond W., 2000. Irony in talk among friends. Metaphor and Symbol 15 (1/2), 527. Grice, H.P., 1975. Logic and conversation. In: Cole, P., Morgan, J.L. (Eds.), Syntax and Semantics. Vol. 3: Speech Acts. Academic Press, New York, pp. 4158. Hanne, Michael, 1999. Getting to know the neighbours: when plot meets knot. Canadian Review of Comparative Literature 26 (1), 3550. Hay, Jennifer, 1995. Gender and humour: beyond a joke. Unpublished MA dissertation, Victoria University of Wellington, New Zealand. Hay, Jennifer, 2000. Functions of humour in the conversations of men and women. Journal of Pragmatics 32, 709 742. Holmes, Janet, 2000. Politeness, power and provocation: how humour functions in the workplace. Discourse Studies 2 (2), 159185. Holmes, Janet, Hay, Jennifer, 1997. Humour as an ethnic boundary marker in New Zealand interaction. Journal of Intercultural Studies 18 (2), 127151. Holmes, Janet, Marra, Meredith, 2002. Over the edge? Subversive humour between colleagues and friends. Humor: International Journal of Humor Studies 15 (1), 6587. Holmes, Janet, Marra, Meredith, Burns, Louise, 2001. Womens humour in the workplace: a quantitative analysis. Australian Journal of Communication 28 (1), 83108. Jefferson, Gail, 1979. A technique for inviting laughter and its subsequent acceptance/declination. In: Psathas, G. (Ed.), Everyday Language: Studies in Ethnomethodology. Irvington, New York, pp. 7996. Jefferson, Gail, 1984. On the organization of laughter in talk about troubles. In: Atkinson, J.M., Heritage, J. (Eds.), Structures of Social Action. Cambridge University Press, Cambridge, pp. 346369. Jefferson, Gail, 1985. An exercise in the transcription and analysis of laughter. In: van Dijk, T.A. (Ed.), Handbook of Discourse Analysis. Vol. 3: Discourse and Dialogue. Academic Press, London, pp. 2534. Jefferson, Gail, Sacks, Harvey, Schegloff, Emanuel, 1978. Notes on laughter in the pursuit of intimacy. In: Button, G., Lee, J.R.E. (Eds.), Talk and Social Organisation. Multilingual Matters, Clevedon, pp. 152205. Kehily, Mary Jane, Nayak, Anoop, 1997. Lads and laughter: humour and the production of heterosexual hierarchies. Gender and Education 9 (1), 6987. Kotthoff, Helga, 2003. Responding to irony in different contexts: on cognition in conversation. Journal of Pragmatics 35 (3), 13871411. McCarthy, Michael, Carter, Ronald, 2004. Theres millions of them: hyperbole in everyday conversation. Journal of Pragmatics 36, 149184. Mulkay, Michael, 1988. On Humour: Its Nature and its Place in Modern Society. Polity Press, Oxford. Mullaney, Louise, 2003. Identity and role construction: a sociolinguistic study of gender and discourse in management. Unpublished PhD thesis, Nottingham Trent University. Norrick, Neal, 1993a. Conversational Joking: Humor in Everyday Talk. Indiana University Press, Bloomington. Norrick, Neal, 1993b. Repetition in canned jokes and spontaneous conversational joking. Humor: International Journal of Humor Studies 6 (4), 385402. Norrick, Neal, 1994. Involvement and joking in conversation. Journal of Pragmatics 22, 409430. Norrick, Neal, 2004. Humor, tellability and conarration in conversational storytellng. Text 24 (1), 79111. Pilkington, Jane, 1998. Dont try and make out that Im nice: The different strategies women and men use when gossiping. In: Coates, J. (Ed.), Language and Gender: A Reader. Blackwell, Oxford, pp. 254269. Sawyer, R. Keith, 2001. Creating Conversations: Improvisation in Everyday Discourse. Hampton Press, Cresskill, NJ. Sullivan, T., Weinart, C., Cudney, S., 2003. Management of chronic illness: voices of rural women. Journal of Advanced Nursing 44 (6), 566576. Tannen, Deborah, 1984. Conversational Style. Ablex Publishing, Norwood, NJ. Tannen, Deborah, 1990. Talking Voices. Cambridge University Press, Cambridge.

J. Coates / Journal of Pragmatics 39 (2007) 2949

49

Jennifer Coates is Professor of English Language and Linguistics at Roehampton University. Her published work includes Women, Men and Language (originally published in 1986, third ed., 2004); Women Talk. Conversation Between Women Friends (1996), Language and Gender: A Reader (1998), Men Talk: Stories in the Making of Masculinities (2003) and The Sociolinguistics of Narrative (edited with Joanna Thornborrow, 2005). She has given lectures at universities all over the world and has held Visiting Professorships in Australia, New Zealand, the USA, Germany, Switzerland, Spain and Italy. She was made a Fellow of the English Association in 2002.