The underlying orderliness in turn-taking Examples from Australian talk Rod Gardner, Richard Fitzgerald, and Ilana Mushin ABSTRACT: The 1974 paper ‘Simplest systematics for the organization of turn taking in conversation’, by Sacks, Schegloff, and Jefferson, is widely regarded as groundbreaking for its detailed and methodical understanding of the routine methods of turn taking in conversation. However, these findings also raise questions of what role, if any, a broader sociocultural context of the talk may play in organising social behaviour, and whether different kinds of orderliness, or even a different turntaking machinery, may be managed and attended to according to different social or cultural conventions. In this paper, we provide examples from a range of Australian face-to-face conversations that show that, even in talk involving extended overlap or extended gaps, the same foundational principles of order in turn-taking apply. From this evidence, we suggest that variations in length and proliferation of gaps and overlaps are not symptomatic of different turn-taking machinery, but rather are contingent on contextual and situational factors. Introduction: Turn-taking and social order Rod Gardner, School of Education & Professional Studies, Griffith University, Australia; Richard Fitzgerald, School of Journalism & Communication, University of Queensland, Australia; Ilana Mushin, School of English, Media Studies, & Art History, University of Queensland, Australia. This paper presents an exploration of the ways in which orderliness is achieved in social interaction through turn-taking. Schegloff (2000) states that ‘the orderly distribution of opportunities to participate in social interaction is one of the most fundamental preconditions for viable social organization’ (p. 1). In echoing the claims underpinning the linguistic turn in the social sciences, Schegloff places language and language use as central to any understanding of The underlying orderliness in turn-taking: Examples from Australian talk 65 36(3)-CS3-20091116.indd 65 17/11/2009 2:29:26 PM contemporary society. While the interest in language use within and as part of the social order has taken many forms, an attention to the detailed methods that participants use in order to conduct their talkin-interaction has offered a corrective to the traditional sociological focuses of race, class, and gender. Through paying attention to the ways in which people enact social identity through talk-in-interaction, social identity is revealed as a lived members’ phenomenon irredeemably embedded in social action locally produced through language use (Benwell & Stokoe, 2006; Button, 1991; Garfinkel, 1967, 2002; Hester & Housley, 2002; Sacks, 1974, 1995). For Harvey Sacks, one the founders of this approach, the sociologically defined ‘problem’ of social order was not to be seen as a problem for members but endemic to academic sociology in its routine description and research of society. As Sacks (1995) points out in one of his early lectures: I am trying to develop a sociology where the reader has as much information as the author and can reproduce the analysis…The trouble with their work [Hymes’s Ethnography of Speaking and Sociology] is that they are using informants; that is they’re asking questions of their subjects. That means that they’re studying the categories that members use, to be sure, except at this point they are not investigating their categories by attempting to find them in the activities they’re employed. And that of course is what I am attempting to do. (Vol. 1 Lecture 4, p. 27) For Sacks, the new approach called ‘conversation analysis’ should start with the phenomenon (in this case language and categories of identity) and remain with this phenomenon, rather than imposing further analysts’ conceptions on top of, and in the process of, explaining what, for the participants, was adequately understandable. Indeed, such was the power of this approach that two of Sacks’s earliest pieces of work, ‘An initial investigation of the usability of conversational data for doing sociology’ (later incorporated into ‘On the analysability of stories by children’, 1974) and the ‘Simplest systematics for the organization of turn taking in conversation’ (Sacks, Schegloff, & Jefferson, 1974, henceforth SSJ) remain at the heart of conversation analysis. As Schegloff’s recent discussions (Schegloff, 2006, 2007a, 2007b) reiterate in respect to the central and enduring findings of this original work: One of the most fundamental organizations of the practice for talk-in-interaction is the organization of turn taking. For there ���������� ������� �� ������������� � ��� �� ��� ���� 66 36(3)-CS3-20091116.indd 66 17/11/2009 2:29:27 PM to be the possibility of responsiveness—of one participant being able to show that what they are saying and doing is responsive to what another has said and done—one party needs to talk after the other, and, it turns out, they have to talk singly. It is the organization of the practices of turn-taking that is the resource relied upon by parties to talk-in-interaction to achieve these outcomes routinely. (Schegloff, 2007b, p. 1) Sacks’s work, then, emerges directly out of a concern with what became coined as the ‘missing what’ in academic sociology, which passes over the actual lived detail of how members make sense of and go about in their society. This respecification of sociological social order as a members’ phenomenon was not confined to sociology but became part of a wider critique of the social and human sciences through emphasising the use of language as central to social action. As Button points out: As sociologists have discussed ‘action’ and ‘actors’ without reference to the fact that it is people who engage in embodied action in ‘real time’, so too has linguistics often discussed ‘language’ without reference to its use by speaking people… Anthropology has often glossed over details of the circumstantial action through having the occasioned account of the native informant stand proxy for a society. (Button, 1991, p. 6) The idea that talk and action are inseparable from, and actively construct, the social world thus underpins the work of ethnomethodology and conversation analysis. At the heart of this is the recognition that one of the basic building blocks of society is the requirement that its members interact with each other, and in a way that is mutually practically purposeful (whatever that is), and reflexively constitutive of the common sense understandings of the unfolding event and the participants thereof (Schegloff, 2007b). In this way, the ethnomethodological focus on language use as revealing and enacting social order underpins our discussion as we now turn to the methodical organisation of that interaction as discussed in Sacks et al. (1974). The seminal findings of SSJ have become well established for proposing norms for the organisation of turn-taking in ‘ordinary’ conversation1. This groundbreaking paper identified a number of recurrent features within interaction by and through which the smooth transition of turns between speakers occurs and recurs. Underlying these observations is The underlying orderliness in turn-taking: Examples from Australian talk 67 36(3)-CS3-20091116.indd 67 17/11/2009 2:29:27 PM an understanding that such effective floor management is necessary in order for people to enact their social lives—that is, managing turntaking is a necessary achievement through which to conduct effective social interaction. As Schegloff (2000, p. 1) puts it, ‘the orderly distribution of opportunities to participate in social interaction is one of the most fundamental preconditions for viable social organization’. One of the fundamental organisations of turn-taking is the orientation of participants to ‘one speaker at a time’ (Schegloff, 2006, p. 71). Corresponding to this is that participants have the ‘problem of coordination if the talk is to be without recurrent substantial silences and overlaps’ (Schegloff, 2006, p. 72). While departures from this can certainly be found, for example, in greetings sequences or choral co-production (Lerner, 2002), there remains a general consensus among conversation analysts that speakers do manage their talk so as to minimise gaps and overlaps, thus maintaining optimal conditions for one speaker at a time to talk. Schegloff (2006, p. 72) thus concludes that ‘so far [the SSJ] account works across quite a wide range of settings, languages, and cultures’. The finding that conversation participants orient to ‘one speaker at a time’ emerges from the ‘gross observations’ (SSJ) or ‘basic features’ (Sacks & Schegloff, 1979) of conversation. These observations were built on a large corpus of (largely) American and British conversations. The two relevant gross observations for the analysis presented here are that ‘overwhelmingly one party talks at a time’ and that ‘transitions from one turn to the next with no gap and no overlap between them are common. Together with transitions characterised by slight gap or slight overlap, they make up the majority of transitions’ (SSJ, p. 700-1). These have been interpreted as claiming that conversation in general is ordered for the smooth transitions of turns with as minimal a gap or overlap as possible. Indeed, talk characterised by ‘smooth transitions’ has been taken as the benchmark of conversational behaviour in recent work summarising the findings of conversation analysis for linguistic enquiry (Fox, 2007; Kärkkäinen, Sorjonen, & Halasvuo, 2007). However, as many have observed, talk frequently does not consist of smooth transitions (e.g., Reisman, 1974; Tannen, 1984; Coates, 1986 for talk involving extended overlaps; Basso, 1990; Scollon & Scollon, 1981; Lehtonen & Sajavaara, 1985 for extended gaps between turns), and there needs to be a way of accounting for this2. In this paper, we present a number of cases that appear to represent rather extreme cases of deviation from these gross observations in ���������� ������� �� ������������� � ��� �� ��� ���� 68 36(3)-CS3-20091116.indd 68 17/11/2009 2:29:27 PM terms of both extended silence and extended overlap. For readers who may not be familiar with SSJ, we begin by presenting one extract of conversation that illustrates how very orderly and precision-timed conversation can be. This is followed by two extracts that seem to reflect a relaxation of the orientation to gap minimisation. One of these extracts is from a corpus of conversations recorded between elderly Indigenous Australian women from the Garrwa language group (spoken on the west side of the Gulf of Carpentaria in northern Australia), and the other is taken from a corpus of Anglo-Australian couples. The silences found in these two extracts (and elsewhere in these and other conversations in our collection) appear not to be associated with continuing states of incipient talk (Schegloff & Sacks, 1973), as they occur regularly after a current speaker has selected a next speaker, or during storytelling or complaint sequences, for example. Finally, we present an extract that illustrates a relaxation of the orientation to overlap minimisation. The data is from a televised political debate between the leaders of the two main political parties in Australia before the 1996 federal election. There are several very extended spates of simultaneous talk—up to 30 seconds. The two debaters recurrently depart from the debate format to the extent that the talk becomes a free-for-all for a while. In this debate, there appears to be regular and sustained talk with more than one speaker at a time3. In all of these examples, and in many others we have examined, despite the seeming departure from the ‘one speaker at a time’ principle, what we have found is that the speakers still orient to the underlying rules of turn-taking as proposed by SSJ4. In terms of the gross observations we focus on here, we find that, regardless of the degree of overlap or gap, there still appears a general orientation to turn transition (SSJ, 1974, p. 700) and to speaker change transition relevance places. This may well reflect a basic orientation to orderliness in social interaction, one that may hold across all social contexts, in support of Sidnell’s (2001) claim that ‘the orderliness of conversation…is grounded in a species-specific adaptation to the contingencies of human social intercourse’ (p. 1263). Thus we note that, while the basic patterns of turn construction and turn allocation outlined in SSJ are supported, variations in the timing of turns (gaps and overlaps) according to local contingencies should also be considered as normative. To preview our findings: The first extract exhibits all of the features of the turn-taking rules and gross observations noted in SSJ. The second The underlying orderliness in turn-taking: Examples from Australian talk 69 36(3)-CS3-20091116.indd 69 17/11/2009 2:29:27 PM and third extracts clearly deviate from gross observation 4: We find that gaps are very common, but lack of overlaps is common, rather than ‘lack of gaps or overlaps are common’. The fourth extract clearly deviates from gross observation 3: We find that occurrences of more than one speaker at a time are very common in that talk, and often extended, rather than ‘occurrences of more than one speaker at a time are common, but brief’5. The evidence for these claims and an account for why they happen and why they do not undermine SSJ in any significant way is presented in section 2 below. �� ��� ���� ����������� ����� Out of the gross observations and the analysis of thousands of naturally occurring turn transitions, SSJ proposed a set of turn-taking rules for the ways in which people manage both turn transition and turn allocation. The rules themselves distinguish between next-speaker selection by current speaker, and self-selection. The rules are ordered hierarchically: first, if the current speaker selects the next speaker, then the selected speaker must speak at the next transition relevance place (i.e., the next place at which the utterance underway is possibly complete—a transition relevance place (TRP)). If no next speaker is selected, then any other participant in the conversation may self-select—at the next TRP. Only if no other speaker self-selects may the current speaker continue. If the current speaker continues speaking under the third of these options, then the rules are recycled, in the same order, until speaker transition occurs. Gaps and overlaps do of course occur, even with such rules, but they are not considered part of ‘smooth transitions’. Gaps may be treated as signs of trouble, for example, that the upcoming turn will be dispreferred or disagreeing (Levinson, 1983). Overlaps are observed to be brief, with one speaker dropping out quickly, to preserve the progressivity of the talk (SSJ, 1974). It is these aspects of turn-taking that we will examine closely in our data extracts to demonstrate an overall preference to preserve the social order through orientation to these turn-taking rules and practices. In the rest of this section, we present four contrasting extracts from Australian talk-in-interaction to demonstrate the extraordinary coordination and precision timing that participants can achieve, and their orientation to shared rules and procedures of turn-taking. ���������� ������� �� ������������� � ��� �� ��� ���� 70 36(3)-CS3-20091116.indd 70 17/11/2009 2:29:27 PM 2.1. Lyn and Mal The first example is a snippet of mundane conversation between a husband and wife. Lyn and Mal are talking about a mutual acquaintance who has lost his job. This conversation occurred during an economic recession in Australia in the early 1990s, and the talk up to this point had been about relatives and acquaintances who had been affected by the recession. We first describe what is happening in this extract before turning to a detailed description of how the turn-taking facilitates the action sequence. (1a) L&MH3b-1-40-54 1 2 3 4 5 6 7 8 9 Lyn: ·hhh So- (.) whad about- that gu:y:-, who’d bin retren:ched befo:re,= ‘n was so te:hrrifie:d,= �es he lost the jo:b? (0.5) Mal: °Ah,°= �which �gu:y. Lyn: hhh ya know,= Dick sai:d,= thet he::- (.) hi:red someone [a mon]th ago:,= [>wh’d bin<]= Mal: [O:h. ] [�Y e s :�,] Lyn: =retre:nched.= This fragment of talk marks a topical shift from the immediately prior talk, during which they had been talking about a company that had been downsizing and retrenching many of its employees. Lyn asks a question in the first three lines of the extract, but the question is embedded in what turns out to be a problematic person reference. Lyn presents this first reference to ‘that guy’ (line 1) as someone recognisable to Mal. This is a first reference to this person, at least in the several minutes of the conversation that have been transcribed prior to this point. As Lyn does not know the person’s name, she provides a description of the person as someone who had ‘been retrenched before’ and who ‘was so terrified’. After her turn there is half a second of silence, before Mal initiates repair by requesting further information on the ‘guy’, thereby claiming non-recognition6. Lyn’s third description of the ‘guy’ reports that her brother-in-law, Dick, had told them that he had hired someone a month previously. This attempt is successful, with Mal producing, first, a ‘change-of-state’ token, ‘oh’ (Heritage, 1984), an affirming ‘yes’, and finally an answer to Lyn’s question, ‘Yes, he’s gone’. We have so far described the progression (or lack thereof) of the talk in extract (1), and the types of interactional activities in which Lyn and Mal are engaged. However, the purpose of this analysis is to draw The underlying orderliness in turn-taking: Examples from Australian talk 71 36(3)-CS3-20091116.indd 71 17/11/2009 2:29:27 PM attention to how the activities in which they are engaged are finely tuned to the turn-taking behaviours of the participants. As her first turn emerges, Lyn comes to four points at which her utterance could have been complete, marked by the superscripted down-arrows in (1b). Note that, as this is a first mention of the person, the turn is not possibly complete after ‘guy’ in line 1, as there has not been sufficient information provided to identify him. (1b) L&MH3b-1-40-54 1 2 3 4 5 Lyn: hhh So- (.) whad about- that gu:y:-, who’d bin retren:ched� befo:re,=� 'n was so te:hrrifie:d,=� �es he lost the jo:b?� (0.5) Mal: °Ah,°= �which �gu:y. At none of these four points of possible completion does Mal attempt to take a turn. Indeed, even after the turn is finally and actually complete, he says nothing for half a second. When he does say something, it is to initiate repair (line 5). Lyn’s second attempt at making the ‘guy’ recognisable, in lines 6-7, has more success. This turn passes three points at which the turn is possibly complete, marked with the down-arrows in (1c). (1c) L&MH3b-1-40-54 6 7 8 9 10 11 Lyn: hhh ya know,= Dick sai:d,= thet he::- (.) hi:red someone �[a mon]th ago:,= Mal: [O:h. ] Lyn: =�[>wh'd bin<] retre:nched.�= Mal: [�Y e s :�,] Mal: =Yes,= he’s go:ne, Person reference has now surfaced as the main activity focus, thereby making some claim of recognition by Mal highly salient. In contrast to Lyn’s first attempt in lines 1-3, in this second attempt Mal responds three times, and each of these three responses is at a point of possible completion: the ‘Oh’ after ‘You know, Dick said that he hired someone’ in line 7, the ‘Yes’ after the ‘a month ago’ in line 7, and the ‘Yes, he’s gone’ after ‘who’d been retrenched’ in line 9. The point to note here is that Mal does not come in anywhere during the term but each time at precisely the points at which Lyn’s turn could have been complete. ���������� ������� �� ������������� � ��� �� ��� ���� 72 36(3)-CS3-20091116.indd 72 17/11/2009 2:29:27 PM The precision timing of Mal’s three short utterances is, though, not merely a case of deft manipulation of the turn-taking machinery. At points at which a response from Mal becomes a relevant action in the first three lines of the sequence, he produces no response. Only when the pressure builds up as a gap emerges after line 3 does he talk, with the repair initiation. At some point during the second attempt, Mal achieves recognition (or at least claims it). The ‘Oh’ marks an announcement of a change of mental state (Heritage, 1984), a move from not knowing something to knowing something: an epistemic shift. This shift, however, may not have occurred at exactly the point of completion of ‘someone’, but some moments before this; it is much more likely that he withheld his response until the point of possible utterance completion before producing his ‘Oh’. Note also that the ‘Oh’ does not answer Lyn’s question. That probably happens with his next utterance, ‘Yes’, in line 10, though there is something neatly equivocal about this token: It could be linked more to the ‘Oh’, affirming his recognition, or it could be a first answer to Lyn’s question, thus making it a pivotal transition from claiming recognition to answering. Again, the token is not placed just anywhere, but at precisely the next point at which Lyn’s turn could have been complete, after ‘ago’. Thus precision timing has occurred twice in succession. This, however, is not the end of the story. Lyn goes on to finish her turn with ‘who’d been retrenched’, and for a third time Mal produces something at precisely the next point at which Lyn’s turn could have been complete, this time an unequivocal answer, ‘Yes, he’s gone’. Either lightning has struck three times in the same place, or Mal has demonstrated remarkable precision timing in his application of conversational turn-taking rules. 2.2. Katelin, Daphne, and Hilda We have so far examined an episode of talk in which there are precisely timed orientations to points of possible completion of TRPs, even though there is some trouble with progressivity in the talk due to a lack of recognition of person reference. For the next extract, we use data from a remote Aboriginal community to focus on talk in which longer gaps between turns regularly appear. It is not our purpose to characterise Aboriginal cultural influences on ways of talking (or argue that this does or does not happen). Three elderly Indigenous women have been sitting for more than 2 hours on the porch of a house, some of the time telling stories for the The underlying orderliness in turn-taking: Examples from Australian talk 73 36(3)-CS3-20091116.indd 73 17/11/2009 2:29:27 PM language research of the third author of this paper. At the point in the conversation at which this extract occurs, they are sitting around doing not very much. The conversation is continuous, in the sense that there are no lapses in the talk, but it continues in a languid fashion, with only sporadic maintained topical focus. There are some quite extended silences during this extract. We noted in our discussion of extract (1) above that the only silence of any length (0.5 seconds) was associated with a problem in the talk: Mal was having trouble understanding to whom Lyn was referring. It has been claimed that if silences do occur in conversation, there is a standard tolerance for about one second of silence before another speaker regularly starts to speak again (Jefferson, 1989). When silences longer than one second occur, they have usually been found to be associated with non-talk activities (such as eating, writing, or grooming), or with some problem in the talk (Goodwin, 1981; Goodwin & Goodwin, 1987; Jefferson, 1989; Pomerantz, 1984b). In contrast with these analyses of (mostly) Anglo-American talk, lengthy silences have been described as normal features of Australian Aboriginal talk (e.g., Eades, 2000; Mushin & Gardner, 2009; Walsh, 1991). Mushin and Gardner (2009) have also found a greater prevalence of silence in Indigenous Australian conversation than has generally been reported for Western conversation. Our Garrwa corpus thus allows us numbers of sequences of talk where longer silences abound, and yet there appears to be no trouble in the talk, nor attendance to talkhindering activities. Note, in particular, in extract 2, the silence at line 795. (2) Porch2.8:785:IR-3:2’07” 785 HG: Fra:zh one >kuna [nayi<;= ( ).] Fresh one kuna nayi Q here Here are fresh ones, (aren’t there)? 786 DG: [>Gi’ me dat bru:sh,= there] Give me that brush there, 787 >bardibard’ ba’ nga’;= mamanumba.< bardibardi baki ngayu mamanumba old.woman and 1Sg lose old woman, I lost it 788 (0.6) 789 HG: Wh:at. ���������� ������� �� ������������� � ��� �� ��� ���� 74 36(3)-CS3-20091116.indd 74 17/11/2009 2:29:27 PM 790 -> (2.3) 791 DG: *˘Uh br:ush,= nga:ki.˘* 1SgDAT My brush 792 (1.8) 793 Hil: Jeki fr:esh one;= >barri nanda< w:aliji;= That fresh one barri that meat 794 ^ngarri? TAG (That one), it’s fresh, that beef, isn’t it? 795 -> (5.5) 796 Kat: °Mm hm:¿° 797 (1.6) ((HG glances up and looks east)) 798 HG: ‘ay’marri [wanyi Wanymarriwanyi White.woman-ERG The white woman (is coming) 799 KS: [Nayi waynmarriw’ ngana;= Nayi waynmarriwanyi ngana Here white.woman-ERG 1Sg-ACC 800 b:un:malimba yih know,= ˘~all:hh da:y) tire.out 801 wulana. (.all: da:y:.~˘ wulani day.before That white woman is tiring me out, you know. All day, yesterday, all day. 802 (2.0) 803 KS: ˘~From hheight to s:even:-n; We focus on the 5.5-second silence in line 795. The meandering topical focus is evident in this extract. Daphne requests Katelin to pass her a hair brush that is on the ground in front of Katelin, which she does. Meanwhile Hilda is talking about some fresh meat. The three women are not very mobile, and they are hungry. They have already requested passers-by to fetch them some food, without success. Hilda repeats her question in lines 792-3, ‘It’s fresh, that beef, isn’t it?’. According to the rules in the SSJ paper, Katelin is obliged to start her response immediately after the question is asked. However, Katelin is drinking from a cup at this point. This is what Goodwin (1981) has called activity-occupied withdrawal. Katelin could, of course, have stopped drinking to answer the question. However, even when she does stop drinking, after 2.1 seconds, there is a further delay of 3.4 The underlying orderliness in turn-taking: Examples from Australian talk 75 36(3)-CS3-20091116.indd 75 17/11/2009 2:29:27 PM seconds before the ‘Mm hm’ is produced as an answer. First, she sets the cup down on the ground, which takes another 2.3 seconds. She then returns her hands to her lap, which takes another 0.8 seconds, and it is not until 0.3 seconds after her hand has settled that she produces her ‘Mm hm’. There is no hurrying to finish the drinking, nor any hurry to respond after she has finished. Not even after the cup is settled on the ground does she answer. Instead she waits until her hand has returned to neutral position in her lap before producing the answer. Thus while drinking can be seen as an inhibitor to an immediate answer, this ‘problem’ cannot account for the whole of the delay, as Katelin had opportunities to answer earlier, and she could have created opportunities to answer even earlier. Note that there is no disagreement, nor is there any other problem in the talk associated with this silence. In fact, the speakers are fully aligned with each other. Neither are they engaged in any activities that would interfere with talk: Daphne is waving away flies and then she picks up a small object and shakes it. Hilda is stroking a coolamon (a vessel made of bark or wood for carrying water, babies, etc.) throughout this sequence, but she still produces trouble-free talk in lines 785, 789, 793, and 798. We found this slow pace with regular lengthy silences between turns very widely throughout this and other Garrwa conversations. On the other hand, with very few exceptions, we do not find in our conversational data very long silences of, for example, more than 10 seconds. It could, then, perhaps be construed that we may have a cultural practice of turn-taking here, specifically relating to how silences are treated and the obligation to speak immediately when selected. This practice is widely observed among researchers of Aboriginal social interaction, and is substantiated in our own data. However, this does not mean that the women we recorded do not recognisably turn-take in ways consistent with the general characterisation of SSJ. Indeed, much of their talk does proceed with little or no gap, and with little or no overlap7. Additionally, we find in our Anglo-Australian data extended sequences in which longer untroublesome gaps are also found. Extract 3 is from a conversation between an Anglo-British wife and her Anglo-Australian husband. (3) L&MC2ai:239 1 2 3 Mel: Tom Barry got a:ll that- (0.2) stuff off to (0.3) Liz: Did �ee?= ���������� ������� �� ������������� � ��� �� ��� ���� 76 36(3)-CS3-20091116.indd 76 17/11/2009 2:29:27 PM 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 Mel: =th’ schoo:l. Mel: °Yea:h. ( ).°= Liz: =An: so he wor- managed ta make it wo:rk. (0.8) Mel: Yeah-. (0.5) Liz: °W’ll° �there you �go:. (2.1) Mel: °Mm- There ya go°.= Liz: =Ho:w long d’t ta:ke ‘im. (2.9) Mel: A::h;= °nod ‘ll tha-t lo:ng°, (7.3) Mel: Very clever ma:n;= is To:m¿ (1.3) Liz: �How o:ld �is he. (2.5) Mel: Don’know;= ee’d be: late �twenties:¿ �early �thirdies¿ (1.0) Liz: He needs a hai:rcut. (2.2) Mel: He:’s a (0.7) hippy:;= �fr’m the sevendies. (0.4) Liz: Doesn’ he kno:w¿= th’t it’s ni:nedee:n ni:nedy o:ne?= Mel: =Hehh (0.3) ·HHUH (4.4) Liz: �’x��tra:wrdinry. Mel: °°Yehh°°. (3.9) While there are some differences in this extract from the Indigenous conversation extract that was discussed above, we do find some long silences between the turns, such as the 2.9-second silence between Liz’s question in line 14 and Mel’s answer in line 16. Liz’s asking of the question selects Mel as next speaker and he is ‘obliged to take the next turn to speak’ (SSJ, p. 704). Transfer of speakership should occur at the transition relevance place. There is, though, a delay of nearly three seconds before Mel responds. After Mel’s answer, there is an even longer silence of 7.3 seconds in line 17. Mel’s answer, of course, does The underlying orderliness in turn-taking: Examples from Australian talk 77 36(3)-CS3-20091116.indd 77 17/11/2009 2:29:28 PM not select next speaker, so Liz may, but need not, speak. She does not, but neither does Mel continue his turn until he self-selects after this very long silence, to make an observation about Tom, about whom they had been talking. This observation about Tom being very clever is followed by another silence of 1.3 seconds before Liz self-selects with another question, thereby selecting Mel, and again there is a long gap of 2.5 seconds before he answers. As with the Garrwa conversations above, there is no evidence of any problem here. We have no video of this conversation, but on the audio recording we can hear that they are at the dinner table, which means that the activity of eating might account for some of these silences. This notwithstanding, there appears to be no orientation to a minimisation of gaps. What may be significant is that Mel and Liz are intimates; they are at home alone, in the evening; there are no pressures to get things done, no deadlines. If situations that promote longer gaps between turns are more regular, or occur more conventionally in the lives of Australian Aboriginal people, then this may account for its association with the social life of a particular culture—a culturally specific variation on the notion of what constitutes the ‘right’ length of space between turns. However, there is one striking similarity between the three old women in the Garrwa data and Mel and Liz in the couples data: they are all intimates engaged in desultory conversation with no pressure to ‘get things done’. In general, what our data, as exemplified in extracts 2 and 3, tells us is not that the gross observations 3 and 4 in SSJ concerning orientation to gaps and overlaps were wrong, but that they are perhaps insufficient to capture some variation that would appear to occur widely in some (ordinary) conversations or talk in some situations: for example, where there is a lack of pressure to take a turn at the first opportunity. 2.3. Paul and John The next example presents a very different and contrasting case to the orderliness of minimal gaps (as seen in extract 1), and indeed the preservation of orderliness across extended gaps between turns we find in extracts 2 and 3. In extract 4, we see turn-taking that appears at first hearing to be very disorderly to the point of breakdown. It is from a televised debate before the 1996 Australian federal election between the incumbent Labor Prime Minister Paul Keating and his soon-to-be successor, the conservative Liberal John Howard, with television personality Ray Martin as moderator. A full analysis of this extract is beyond the scope of this paper, so two fragments from this extract ���������� ������� �� ������������� � ��� �� ��� ���� 78 36(3)-CS3-20091116.indd 78 17/11/2009 2:29:28 PM will be discussed below. At the beginning of this extract, Howard is attacking a trade union leader, Bill Kelty, a supporter of Keating’s Labor Party. Peter Reith was a shadow government minister. (4a) KHDeb:928:20-30:4’22”Getting like Parliament 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 JH: PK: JH: PK: JH: PK: JH: PK: JH: PK JH: PK: JH: PK: JH: PK: JH: PK: JH: PK: JH: PK: JH: PK: JH: PK: RM: PK: [E:h he - eh whaddee ↓did w’z: to: ·hhh e↑sayda thee Estraylien people,= if ↑you ↓have the ↑ne:rve ↓or the ga:ll¿ ·hh to even ↑contem↓plate; ·hh electing a coalition gover’men’;= we’re genna ↑rip the place apa:rt. (.) ↑You know-; (.) en’ ↓I: kno:w, (0.5) th’t if ↑we win thee election? (0.5) the tra:de union movement;= will ↑wo:rk with u:s? (0.2) °°O:[h yea:h°°. [Eh- ↑Kelty to:ld Peter Reith; (0.2) ‘ee said- (.) ‘ee said;= if ↑you: win thee election;= ↑we’ll hafta work with you¿ [·hh [we ↑might- ] we-we [might-]= [‘ee de[nie:d ↓that.] [that- ]= =[we might-] =[he denied] that. (.) we-we mightn’ [like it. [‘ee denied it. Well ↓well;=↑in otha words;= you thing it’s pehrfickly [ligid]imate- (.) ↑Ah-; (.) w’ll you= [↑No; ] =↑do:[n’. [W e l l, w-]why did you;= [↑I’m juss- [↑No; (b’t-) =[in↑do:rse et] then. =[well (juss) ] ↑le’ me jess say this. I mean look- look[i- [i- i[f ↑you win [ thee election John,]= [e- [e- [Kel[↑Kelty is: th’ one;]= =[‘f ↑you win thee-](0.2)[well ↑don’ ↓t]a:lk a:ll= =[<↑K e l t y r]: u n[s ↓y o: u ,>] =night,= [if you:-] i f : e- y~o u-](.) [if ↑you:-]= [nuh- ↑no] no:;↑ <no ↑you ↓h]ad a[: v e: r y]= =[if you win]: thee el]: e c t i o n]:, ·h h h ]= =[good run:;] P a u: l,] >l’k c’m on;] you ‘ad a]= =[--(0.2)--] w a :]ges will:= =[very good] run<.] =[ get hi:gher¿ ] =[>’sa bit li’ pa]:rl’men’;=izn’ et<.= =[Wages [ will-] The underlying orderliness in turn-taking: Examples from Australian talk 79 36(3)-CS3-20091116.indd 79 17/11/2009 2:29:28 PM 39 JH: =[Ih[ noh- ] w e:ll-] [well- ] 40 RM: [>↓Bit-] like pa]:rl’men’;= [rilly.] 41 JH: ·phhhn yih k[n o w ; ] 42 |(0.5)| 43 RM: [>W’d you] like j’s finish the li:ne;= en’ 44 then go da- (0.2) tsk= 45 JH: =yeah;= 46 RM: =[Paul.] 47 JH: =[·hh ] b’t- (.) you ↑ca:n’ (.) ↓deny: the fact; hh 48 ·hhh what ↑Bill ↓Keldy did la:st week,= w~a:s to;= 49 in↑timida:te¿ ·hh ↓and ↑bully¿ ·hh ¯the Estraylien 50 ↑people¿ As we noted in the introduction, in spates of talk like this one—and there were many—any formal debate structure has broken down, so that while this talk is not found within ‘ordinary’ conversation (cf. extracts 1-3), it is like conversation at this time in the recording. Here the participants revert to vigorous argument and heckling of each other, albeit with an overhearing audience of millions. In line 12, Keating interjects with ‘he denied that’ to challenge Howard’s claim that the trade union leader, Kelty, was duplicitously telling the Australian people that he would ‘rip the place apart’, while at the same time telling the conservative coalition parties that he would work with them. Keating’s interjection is timed to begin precisely at a point when Howard’s turn could be possibly complete. Keating’s interjection, however, is not a one off. It is twice repeated (with a further bit of talk, a standalone ‘that’, in line 12). The denial thus turns into a heckle. We might conjecture that the high stakes being played for here—contention for the highest office in the land—lead Keating to attempt to drown out potentially damaging accusations from his opponent. Be that as it may, orderliness of the talk, such as we saw in the first extract above, appears to have broken down, as the effect of this heckling is to disrupt Howard’s talk so that the initial elements of his turn are recycled with four ‘might’s and six ‘we’s: ‘we might- we-we migh- we might- we-we mightn’t like it’. As Schegloff (2000) says, ‘many overlaps are the site of hitches and perturbations in the production of the talk’ (p. 10). Such hitches and perturbations are ‘deflections in the production of the talk from the trajectory which it had been projected to follow’ (p. 11), that is, the smooth production of a turn. The turn’s progressivity is disrupted. What most commonly occurs in such simultaneous talk is an orientation to resolving the overlap, by the use of what Schegloff (2000) calls an overlap resolution device, namely the deployment ���������� ������� �� ������������� � ��� �� ��� ���� 80 36(3)-CS3-20091116.indd 80 17/11/2009 2:29:28 PM of hitches and perturbations (such as sound stretches, cut-offs, or repetitions) in the talk to orient its speaker towards a point at which the other speaker’s talk may be coming to an end, thus opening the floor to a resumption of single participant speakership (‘one at a time’). We may note here that Howard, in line 11, cuts off his first ‘we might—’, then starts again immediately after the first of Keating’s challenges, ‘He denied that’, is complete. However, Keating very quickly comes in again with a ‘that’, which again coincides with Howard breaking off his talk after ‘might’, only to start again just as Keating starts his second heckling ‘he denied that’ in line 14. Once again, finding he is not sole speaker, Howard cuts off his third ‘might’, and one beat (a micropause) after the end of Keating’s second challenge, he starts again—for the fourth time. This time he remains in the clear for longer, which provides him with enough time to be able to finish the whole utterance, despite Keating coming in for a second repeat of ‘he denied that’ in overlap with the last two words of Howard’s turn. Note that the effect of Keating’s repetitions is to stop Howard making a potentially damaging (to Keating) claim in full hearing of the audience. The point here is that, despite apparent breakdown and disorderliness, there is still orientation on the part of both speakers to possible transition relevance places, to ‘one speaker at a time’, and to completion of utterances, even if it means not initially completing the utterance. If we turn our attention now to lines 25 to 37, reproduced as extract 4b here, we can see what can happen when simultaneous talk is extended to the point at which neither speaker backs down. (4b) KHDeb:928:20-30:4’55”Getting like Parliament 25 26 27 28 29 30 31 32 33 34 35 36 37 38 JH: PK: JH: PK: JH: PK: JH: PK: JH: PK: JH: PK: RM: PK: I mean look- look[i- [i- i[f ↑you win [ thee election John,]= [e- [e- [Kel[↑Kelty is: th’ one;]= =[‘f ↑you win thee-](0.2)[well ↑don’ ↓t]a:lk a:ll= =[<↑K e l t y r]: u n[s ↓y o: u ,>] =night,= [if you:-] i f : e- y~o u-](.) [if �you:-]= [Nuh- �no] no:;� <no ↑you ↓h]ad a[: v e: r y]= =[if you win]: thee el]: e c t i o n]:, ·h h h ]= =[good run:;] P a u: l,] >l’k c’m on;] you ‘ad a]= =[--(0.2)--] w a :]ges will:= =[very good] run<.] =[ get hi:gher¿ ] =[>’sa bit li’ pa]:rl’men’;=izn’ et<.= =[Wages [ will-] The underlying orderliness in turn-taking: Examples from Australian talk 81 36(3)-CS3-20091116.indd 81 17/11/2009 2:29:28 PM 39 40 41 42 43 44 45 46 JH: =[Ih[ noh- ] w e:ll-] [well- ] RM: [>↓Bit-] like pa]:rl’men’;= [rilly.] JH: ·phhhn yih k[n o w ; ] |(0.5)| RM: [>W’d you] like j’s finish the li:ne;= en’ then go da- (0.2) tsk= JH: =yeah;= RM: =[Paul.] In line 25, Howard begins a turn haltingly, with hitches, which is an aftermath of the competitive overlapping talk that had preceded this fragment (see 4a above). In line 26, Keating comes in to compete for the floor with ‘If you win the election, John’, and what follows is similar to what was described above: recycling of parts of turns. Howard, though, after a shaky start, manages to produce the remainder of his turn fluently, despite the overlapping talk. As Schegloff claims: [s]ome speakers, on some occasions, continue talking in solo production, with no hitches or perturbations, as if no one else were talking at all. Such speakers embody…an apparently exclusive orientation to producing their own talk to completion, whatever else may be going on. (Schegloff, 2000, p. 30). It must be said that Howard here may well be orienting his talk not to Keating, nor even to Ray Martin, the moderator, but to the millions of voters watching the debate. However, on this occasion, when it becomes apparent that neither is backing down from the floor, and indeed Howard is producing fluent talk, Keating changes tack, and, instead of trying to make a critical point about what would happen if Howard won the election, he addresses the turn-taking system, challenging Howard’s right to continuing to talk: ‘well don’t talk all night’ (lines 28/30). This appeal to ‘one-at-atime’, and to a reasonable turn size (even in debate) is unsuccessful, for it is now Howard who addresses turn-taking and rights to speakership, with his ‘nuh no no no, you had a very good run Paul, look come on, you had a very good run’ (lines 31/33/35). With remarkable precision timing in his change of stance from debater to turn-regulator, Howard has latched his riposte at precisely the point of completion of Keating’s admonishing ‘Well don’t talk all night’ in lines 30-31. Keating then returns for a third time to articulate the consequences of a Howard election victory (from line 30), and once again, up ���������� ������� �� ������������� � ��� �� ��� ���� 82 36(3)-CS3-20091116.indd 82 17/11/2009 2:29:28 PM against Howard’s hitch-and-perturbation-free claim that Keating has had a generous allocation of speaking time, Keating recycles his turn beginning, this time with another four ‘if you’s (lines 30/32), before struggling to completion in line 36. Note that he holds back with the final part of his accusation, the part that had not been repeated (‘if you’ six times and ‘win the election John’ twice), and the part that packs the punch (‘wages will rise’) until just before Howard completes his riposte, so that most of it will be out in the open without having to compete against another voice—only to find Ray Martin coming in to overlap the punch line with a comparison of their interaction with the rowdiness of (the Australian) parliament. The underlying orderliness of turn-taking is displayed in these almost entropic spates of talk by the very fact of their apparent entropy: the talk disintegrates in various ways because the speakers are striving to return to the orderliness of turn-by-turn, recycling elements of their turns in an effort to pinpoint a place where they will be the sole speaker. Chaos is not the default, but their very political survival is at stake. This is a fight: each is looking to damage the other, even deliver a ‘knock-out blow’, but at the same time they must defend and parry, and one means of defence is to try to obliterate the talk of the other by talking over them. Nevertheless, even in this extreme case scenario, the speakers seek out points that would be legitimate loci for speaker transition. Even within such apparently chaotic sequences of talk, it is out of the ordinary for such completion points to be totally ignored (as when Keating heckles ‘he denied that’ (lines 12, 14 and 17), or Howard ‘continue(s) talking in solo production, with no hitches or perturbations’ (Schegloff, 2000, p. 30)), and they carry on with little regard for the other speaker—and even here, they are doing what they are doing to talk over the talk of the other. Thus even in disorder and disarray, they seek order. In this high-stakes encounter, the two main speakers are in a fight for political life. Unlike Mal and Lyn in extract 1, they are using their talk to drown out the other speaker. They even address turn-taking protocols. Nevertheless, even here they are responding to each other, tracking each other’s talk for points of completion. �� ���������� In our analysis and discussion above, we have identified a number of the turn allocation rules and gross observations of social interaction discussed in the SSJ paper as part and parcel of interaction within and The underlying orderliness in turn-taking: Examples from Australian talk 83 36(3)-CS3-20091116.indd 83 17/11/2009 2:29:28 PM across cultures in our data set. Moreover, we argue that, rather than regarding any differences identified in their use as social or cultural deviations or tied to any social category, instances of variations be regarded as locally contingent task- and situation-based members’ methods of doing interaction in situ. From this we suggest that members’ orientation to situation- and task-based methods of turn allocation is as much apparent within the urgency and high stakes of an argument such as displayed in the Keating-Howard television debate, as through the unhurried, silence-punctuated talk of the old Garrwa friends or the Anglo couple home alone. In this sense, we strongly argue against equating an instance of turn allocation that differs from that identified in the SSJ as a deviation tied to a specific culture, as this would entail arguing that the culture of politics is represented in its entirety through the Keating-Howard debate and similar examples, and similarly that long pauses and gaps identified in the Garrwa data are culturally specific to the Garrwa (or Indigenous Australian) people. Clearly ‘arguing’ is something found across many cultures, as is a tolerance for long pauses and gaps. Rather, our argument is that there would seem to be a cross-cultural conformity with many of the gross observations discussed in the SSJ, which indicates that these may not be tied to and so define specific cultures, but that they may be tailored in any instance through an attention to the interactional task at hand, in whatever culture and by whomever. In this way the occurrence of long pauses, which are treated as non-problematic by the Anglo couple, are also treated as non-problematic by the Garrwa women. Thus, what has been observed and described elsewhere as a cultural difference can be reframed as an adaptation of human practices of interacting to recurring local interactional contingencies. Over time, such practices may become conventional and normalised for certain sociocultural groups. We are not claiming that people do not experience cross-cultural communication breakdowns or problems that may be traceable to variation in some variables in conversational style. Such breakdowns can no doubt sometimes be attributed to turn-taking. For example, Eades (2007) shows that in courtrooms under questioning from lawyers, Aboriginal silence is often misinterpreted. The lawyers did not recognise ‘that Aboriginal answers to questions often begin with considerable silence’ (p. 287). What we believe may be contributing to these situations is a clash not of fundamental cultural turn-taking practices, but of tolerance for silence in a particular institutional setting. Clearly, some institutional tasks use a more formal or specialised turn���������� ������� �� ������������� � ��� �� ��� ���� 84 36(3)-CS3-20091116.indd 84 17/11/2009 2:29:28 PM taking organisation that differs from routine conversational interaction, but that draws upon routine cultural turn-taking practices familiar to members. However, Aboriginal people may not be familiar with the characteristics of particular institutions, such as in legal settings, and may transfer their conversational tolerance for silence, for example after questions, to this setting. On the other side, lawyers (and perhaps also teachers and health workers) may not accommodate to these Aboriginal silences. We suggest that cross-cultural communication breakdowns occur not because the turn-taking systems of various cultures are fundamentally different, but because one or more parties are not accommodating to the local social and institutional variables of the turn-taking systems. On the other hand, if we do not have the normative orderliness in turn-taking in conversation that we saw in the first example (‘What about that guy…’), or in the extracts we looked at from the Garrwa conversations, the social work of interaction and the tasks we carry out relying upon our understanding of the methods of interaction would create a different society. Finally, our discussion suggests a move away from treating the turntaking methods and rules found in the SSJ paper as forming some kind of base line of ‘ordinary conversation’ in Anglo culture from which all other (institutional) interactional tasks within the culture are modifications. Indeed, the original data examined in the SSJ paper included examples from institutional and non-institutional interactions and was largely unreflective as to its ‘ordinary’ or otherwise status, except in pondering the notion of a continuum towards the end (p. 730). It would seem that, while this provided impetus to examine other types of interaction, the reification of the turn allocation techniques as ‘ordinary conversation’ may detract from the flexibility and power of the original rules and methods identified in the paper. That is to say that interaction is necessarily task based, and all interaction has some form of turn allocation methods within and as part of it. These are not necessarily derived or modified from any base line but irredeemably locally task-based methods oriented to by participants in situ. As we have hopefully made clear in our discussion, this is not to deny the original strength of the turn-taking methods identified in the SSJ, as we place them at the core of social action, but rather to broaden these in a way that reconnects with the original revolutionary drive of ethnomethodology of understanding the sociological ‘problem’ of social order (see McHoul, this issue). In that first and foremost people largely interact with a purpose, that purpose is what shapes the methods of turn allocation and broader interactional goals as social interaction. The underlying orderliness in turn-taking: Examples from Australian talk 85 36(3)-CS3-20091116.indd 85 17/11/2009 2:29:28 PM Notes 1. While not relevant to the original paper, the term ‘ordinary’ conversation has become the routine way of describing talk-ininteraction that is not ‘institutional’. Within the 1974 discussion, all interaction was described as ‘conversation’. Moreover, we leave aside the turn-taking requirements of institutions, which may indeed reflect different heuristics for orderliness. 2. Conversation analysts have indeed described extended overlap (e.g., Jefferson, 1983; Jefferson & Schegloff, 1975; Lerner, 1989; Schegloff, 2000, 2002) and gaps between turns (e.g., Davidson, 1984; Goodwin, 1981, 1994, 1995; Hayashi, 2003; Jefferson, 1986, 1989; Pomerantz, 1984a, 1984b) as phenomena in talk. However, what we feel is still lacking in this literature is a confrontation with the gross observations 3 (‘Occurrences of more than one speaker at a time are common, but brief’) and 4 (‘Lack of gaps or overlaps are common’) in the face of extreme talk that appears to deviate strongly from these observations. 3. While this debate is not quotidian conversation, we argue that the participants have in these sequences departed massively from a formal debate format, and, perhaps despite the vast overhearing studio and television audience, reverted to something more akin to face-to-face argumentative conversation, displaying a major departure from ‘one-at-a-time’. 4. We acknowledge that in presenting only four extracts of talk we are not in a position to make large claims. However, the point of the paper is to demonstrate that speakers i) are capable of remarkable precision timing (not a new claim); ii) in some situations and circumstances will expand the silences between turns way beyond the TRP (not necessarily associated with trouble); and iii) on some occasions can speak simultaneously for very extended spates of talk (not documented in the literature in this way). 5. We do not claim that these gross observations in SSJ are false; what we are claiming is that we have found that in certain conversations or talk in our corpus there are regular deviations from these gross observations. 6. Mal clearly has a problem identifying the person to whom Lyn is referring. It has been established that, where possible, speakers show a preference to make the person reference recognisable for ���������� ������� �� ������������� � ��� �� ��� ���� 86 36(3)-CS3-20091116.indd 86 17/11/2009 2:29:28 PM the recipients; and if possible, speakers also prefer to use just one reference expression (Sacks & Schegloff, 1979; Stivers, Enfield, & Levinson, 2007). Most commonly, reference in English is done by name or personal pronoun, as these are both minimal, and—in the appropriate context of the talk with the appropriate recipient design—recognisable. On some occasions, though, a name is not available, so some other tack is taken to identify the person. This is what happens in the above extract. 7. Gardner & Mushin (2007) found that overlaps were mostly brief and similar to overlaps described for Anglo-English conversations, although there was at least one pattern of overlapping (‘post start up overlap’) which appears to differ from overlaps described elsewhere. References Basso, K. (1990). To give up on words: Silence in Western Apache. Western Apache Language and Culture. Tuscon: University of Arizona Press. Benwell, B., & Stokoe, E. (2006). Discourse and identity. Edinburgh: Edinburgh University Press. Berry, A. (1994). Spanish and American turn-taking styles: A comparative study. Pragmatics and Language Learning, 5, 180-190. Button, G. (1991). Ethnomethodology and the human sciences. Cambridge: Cambridge University Press. Coates, J. (1986). Women, men and language. London: Longman. Coates, J., & Cameron, D. (Eds.). (1988). Women in their speech communities. London: Longman. Davidson, J. (1984). Subsequent versions of invitations, offers, requests, and proposals dealing with potential or actual rejection. In J. Maxwell Atkinson & John Heritage (Eds.), Structures of social action (pp. 102-128). Cambridge: Cambridge University Press. Eades, D. (2000). I don’t think it’s an answer to the question: Silencing Aboriginal witnesses in court. Language in Society, 29(2), 161-195. Eades, D. (2007). Understanding Aboriginal silence in legal contexts. In Helga Kotthoff & Hilda Spencer-Oatey (Eds.), Handbook of Intercultural Communication (pp. 285-301). Berlin: Mouton de Gruyter. Fox, B. (2007). Principles shaping grammatical practices: An exploration. Discourse Studies, 9(3), 299-318. Gardner, R., & Mushin, I. (2007). Post-start-up overlap and disattentiveness in talk in a Garrwa Community. Australian Review of Applied Linguistics, 30, 35:1-35:15. Garfinkel, H. (1967). Studies in ethnomethodology. Englewood Cliffs: PrenticeHall. Garfinkel, H. (2002). Ethnomethodology’s program: Working out Durkheim’s aphorism. Lanham: Rowman and Littlefield. The underlying orderliness in turn-taking: Examples from Australian talk 87 36(3)-CS3-20091116.indd 87 17/11/2009 2:29:29 PM Goodwin, C. (1981). Conversational organization. New York: Academic Press. Goodwin, C. (1994). Professional vision. American Anthropologist, 96(3), 606-633. Goodwin, C. (1995). Seeing in depth. Social Studies of Science, 25, 237-274. Goodwin, C., & Goodwin, M. (1987). Concurrent operations on talk: Notes on the interactive organization of assessments. IPRA Papers in Pragmatics, 1(1), 1-54. Hayashi, M. (2003). Language and the body as resources for collaborative action: A study of word searches in Japanese conversation. Research on Language and Social Interaction, 36(2), 109-141. Heritage, J. (1984). A change-of-state token and aspects of its sequential placement. In J. Maxwell Atkinson & John Heritage (Eds.), Structures of social action (pp. 299-347). Cambridge: Cambridge University Press. Hester, S., & Housley, W. (Eds.). (2002). Language, interaction and national identity: Studies in the social organisation of national identity in talk-ininteraction. Aldershot: Ashgate. Jefferson, G. (1983). Two explorations of the organization of overlapping talk in conversation: Notes on some orderliness in overlap onset. Tilburg papers in language and literature, no. 28� Tilburg, NL: Tilburg University. Jefferson, G. (1986). Notes on ‘latency’ in overlap onset. Human Studies, 9, 153-183. Jefferson, G. (1989). Preliminary notes on a possible metric which provides for a ‘standard maximum’ silence of approximately one second in conversation. In D. Roger & Peter Bull (Eds.), Conversation: An interdisciplinary perspective (pp. 166-196). Clevedon: Multilingual Matters. Jefferson, G., & Schegloff, E. (1975, December). Sketch: Some orderly aspects of overlap in natural conversation. Paper presented at the meeting of the American Anthropological Association. Kärkkäinen, E., Sorjonen, M., & Helasvuo, M. (2007). Discourse stucture. In Timothy Shopen (Ed.), Language typology and syntactic description (pp. 301-371). Cambridge: Cambridge University Press. Lehtonen, J., & Sajavaara, K. (1985). The silent Finn. In Deborah Tannen & Muriel Saville-Troike (Eds.), Perspectives on silence (pp. 193-201). Norwood, N.J.: Ablex Publishing Corporation. Lerner, G. (1989). Notes on overlap management in conversation: The case of delayed completion. Western Journal of Speech Communication, 53, 167-177. Lerner, G. (2002). Turn-sharing: The choral co-production of talk-in-interaction. In Cecilia E. Ford, Barbara A. Fox, & Sandra A. Thompson (Eds.), The language of turn and sequence (pp. 225-256). Oxford: Oxford University Press. Levinson, S. (1983). Pragmatics. Cambridge: Cambridge University Press. Mushin, I., & Gardner, R. (2009). Silence is talk: Conversational silence in Australian Aboriginal talk-in-interaction. Journal of Pragmatics, 41(10), 2033-2052. Pomerantz, A. (1984a). Agreeing and disagreeing with assessments: Some features of preferred/dispreferred turn shapes. In J. Maxwell Atkinson & John Heritage (Eds.), Structures of social action (pp. 57-101). Cambridge: Cambridge University Press. ���������� ������� �� ������������� � ��� �� ��� ���� 88 36(3)-CS3-20091116.indd 88 17/11/2009 2:29:29 PM Pomerantz, A. (1984b). Pursuing a response. In J. Maxwell Atkinson & John Heritage (Eds.), Structures of social action (pp. 152-163). Cambridge: Cambridge University Press. Reisman, K. (1974). Contrapuntal conversations in an Antiguan village. In Richard Bauman & Joel Sherzer (Eds.), Explorations in the ethnography of speaking (pp. 110-124). Cambridge: Cambridge University Press. Sacks, H. (1974). On the analysability of stories by children. In Roy Turner (Ed.), Ethnomethodology (pp. 216-232). Harmondsworth: Penguin. Sacks, H. (1995). Lectures in conversation. Oxford: Blackwell. Sacks, H., & Schegloff, E. A. (1979). Two preferences in the organization of reference to persons in conversation and their interaction. In George Psathas (Ed.), Everyday language: Studies in ethnomethodology �pp. 15-21). New York: Irvington. Sacks, H., Schegloff, E. & Jefferson, G. (1974). A simplest systematics for the organization of turn-taking for conversation. Language, 50, 696-735. Schegloff, E. (2000). Overlapping talk and the organization of turn-taking for conversation. Language in Society, 1, 1-63. Schegloff, E. (2001). Accounts of conduct in interaction: Interruption, overlap, and turn-taking. In Jonathan H. Turner (Ed.), Handbook of sociological theory (pp. 287-321). New York: Kluwer Academic/Plenum Publishers. Schegloff, E. A. (2006). Interaction: The infrastructure for social institutions, the natural ecological niche for language, and the arena in which culture is enacted. In N. J. Enfield & S. C. Levinson (Eds.), Roots of human sociality: Culture, cognition and interaction (pp. 70-96). London: Berg. Schegloff, E. A. (2007a). Categories in action: Person-reference and membership categorization. Discourse Studies, 9(4), 433-461. Schegloff, E. A. (2007b). Sequence organization in interaction: A primer in conversation analysis. Cambridge: Cambridge University Press. Schegloff, E. & Sacks, H. (1974). Opening Up Closings. In R. Turner (Ed). Ethnomethodology. Harmondsworth: Penguin. Scollon, R., & Scollon, S. (1981). Narrative, literacy and face in interethnic communication. Advances in discourse processes. Norwood, N.J.: Ablex. Sidnell, J. (2001). Conversational turn-taking in a Caribbean English Creole. Journal of Pragmatics, 33, 1263-1290. Stivers, T., Enfield, N. J., &. Levinson, S. C. (2007). Person reference in interaction. In N. J. Enfield & Tanya Stivers (Eds.), Person reference in interaction: Linguistic, cultural, and social perspectives (pp. 1-20). Cambridge: Cambridge University Press. Tannen, D. (1984). Conversational style: Analysing talk among friends. Norwood N.J.: Ablex. Walsh, M. (1991). Conversational styles and intercultural communication: An example from northern Australia. Australian Journal of Communication, 18(1), 1-12. The underlying orderliness in turn-taking: Examples from Australian talk 89 36(3)-CS3-20091116.indd 89 17/11/2009 2:29:29 PM ���������� ������� �� ������������� � ��� �� ��� ���� 90 36(3)-CS3-20091116.indd 90 17/11/2009 2:29:29 PM