This online paper may be cited or briefly quoted in line with the usual academic
                conventions. You may also download it for your own personal use. This paper
                must not be published elsewhere (e.g. mailing lists, bulletin boards etc.) without
                the authors' explicit permission.


Area (1998) 30.3, 000–000

ISSN 0004-0894 © Royal Geographical Society (with The Institute of British Geographers) 1998

Geographies of talk


Geographies of talk:2, 15 ‘Max left a message for you’16, 24

Eric Laurier17

Department of Geography and Topographic Science, University of Glasgow, Glasgow G12 8QQ. Email:

Article37 submitted40 to Area 17 February 1998. Resubmitted41 and revised 19 June 1998.

Summary A little worried about the manner in which transcripts of speech are used in geography, the author telephoned19 a linguist. Unfortunately, she had taken most of her answers elsewhere, leaving only the normal economical opening and delayed listening sequence on her answering machine, to which the author left the following message.20 His lengthy speaking turn, transcribed here from microcassette,21 appears to be an argument about some differences between written texts and interactional talk. Having suggested that talk has specific qualities, the author bemoans the predominance of transcription and coding routines, which lead to a particular kind of representative quotation. Drawing on preceding coffee-time debates with the aforementioned linguist, he wonders aloud22 about using different forms of transcription to shift talk into the scriptural economy of papers, articles and books, and whether more use ought to be made of the analytic methods of conversational analysis. A torn-off piece of computer print-out is included,23 since it was mentioned on the tape as an illustration of the relevance of conversational analysis.

Begin28 (please consult appendix for transcript notation guide)1

Talkproject: Transcript 5

Source: Audiotape

Date: 14 December 1997

Language: English

Participants: Max, investigator;25 Marta, linguist and initial message recipient

Accents: Max, west coast Scottish/Canadian; Marta, non-local English, southern European?

Situation: Leaving message on answerphone/listening to answerphone

Message as follows...

((Light flashing on answerphone))i

Answerphone: ((Speaks in female voice without emphasis)) You have one message. ((Bleeps several times7 and rewinds))ii

Max: ?Hi,10 it’s Max.26 How are you? ((pause)) Listen, it’s just a quickie, hm::: I’ve been thinking about this paper; just a short one for Area or something. I thought about calling it geographies of talk. Anyway, anyway, here’s how it goes: ·hhhh it’s an issue that’s very, been very much on my mind, ’cause I’ve been working on all these transcripts of interviews and conversations from the last couple of projects. The way things are, eh::. You know, it’s a difficulty that’s been building up for me really, that’s very much come out of the research process. This last paper that I was writing on conversational narratives and health was probably the one that triggered it, since I was trying to make a point about talk being different from texts, yeah kinda foolish to take on a problem that’s been occupying philosophers for millennia, I know:::, and, anyways, the practical point for me was that I felt that you shouldn’t analyse a conversation in quite the same way that you would a written biography or a novel or a film or something. Ehm::: the conceptual points were straightforward, that a story constructed in conversation was, in many, many ways, different from a story written into and read out of a typical book, what with the possibilities for interruption, agreement, redirection, attentiveness, receptions and shows of understanding.iii Then::::, then, when I got on to actually analysing how this played itself out in an interview transcript, ((pause)) well, this was where I kept getting bogged down. And you know, the problem is that I, like most geographers, I’d guess, never had any training in the basics of something like conversational analysis. Hhhh. ((pause))

It’s that typical geographer’s problem again, you know, dipping our toes into other disciplines’ ponds, which is good, it’s just that ((imitates the voice of Alan Partridge)),iv sometimes you can’t see the pond for the fish, A: ?ha. ((pause)) Research involves casting nets for words as much as numbers ?I’ve got more fish metaphors than ((pause)) ehm I’ve got fish ((uncertain tone))? Eh:::: ’cause we’re using transcripts of talk more than ever now: from interviews, from participant-observations, from focus groups and the like.v I mean, the transcript has become one of the main forms of non-numerical data to be recorded and archived in geography now. Like, when I hand in my project to the funding body, they want the transcripts set up for archiving so that they can be archived. And in part, that’s why I wanted to talk to you at some point soon, because I remember you saying that in linguistics, ehm, there were a series of different ways of marking up speech, depending on what kind of analysis was going to be done on it: ehm, phonetic, syntactic, dialects and that kind of thing. Maybe transcriptions rather than scriptions; they are moving another register of meaning, which we hear in speaking, but which disappears in writing, apart from a bit of italicization and ehm::: some of those email shorthands. HEY. Your answerphone’s bleeping, I HATE the way it does that. You think either it’s about to cut you off or it’s erasing your message for going on too long.9 >Well. Well I’m not saying that geographers should be doing that kind of linguistic analysis. No. No.< ((pause)) It’s as much, I suppose, ’cause geographers are arriving late in the day to another new field, I mean, sociology of language, and I presume sociolinguists, narratologists, have been working on this field for quite a while, haven’t they? Well, you might not know about the sociologists of language.

((pause)) Ehm, (well there’s/weathers/whether’s) Have you heard of a guy called Harvey Sacksvi? It’s not that important. You might have, just because he was working at Berkeley around the same time as Chomsky and had links with the two big names in North American social interaction studies, Goffman and ?Garfinkel? ((pause)) In the social sciences, you hear lots about Goffman and Garfinkel, which makes sense since they had a fun::damental influence on how social interaction is studied. So, like, Goffman’s work in the, ((pause)) I mean on the performances of self and teamwork, or, ehm, his ethnographic masterpiece Asylums and the like. And I guess Garfinkel remains pivotal for anyone who chooses an ethno-methodological approach in their research. Oh, actually, Sacks and Garfinkel had a working relationship for around twenty years, doing research together on common-sense knowledges and ordinary activities, so you have to see Sacks’ effort as part of that ethno-methodological thing. So::::, so, so the di:::fference with Sacks is, he strayed further into linguistics than either Goffman or Garfinkel. So far into linguistics that Goffman, when he was on the examining board for Sacks’ PhD, blocked the award ’cause it only formed an initial investigation, which means, presumably, that a full investigation required moving out of conversational routines into, quotes, social routines.vii And the thing I want to stress in the article is that, compared with the two Gs, ((pause)) Sacks’ legacy has been, well, less than pervasive, and I don’t think rightly so. I mean, I for one have found his stuff ?so::: useful, since he uses a social interactionist approach, but arising out of conversation. Unlike a lot of linguists’ work I’ve glanced at, ((pause)) who seem always to be going on about or kinda in terms of speech acts and one-off utterances, this guy Harvey Sacks studied the ‘to and fro’ of speech: how one comment projected its possible answer along with several alternatives, while ruling out a whole host of other responses. ((pause)) For example, his whole shift away from treating talk as merely a screen onto which other social practices are projected, to treating it as a practice in its own right, came from listening to a lot of telephone conversations to the Samaritans, examining callers’ strategies for avoiding giving their name by simply saying that they couldn’t hear the Samaritan when they first called (and thus the normal exchange of names on the phone was bypassed). Like: ((pause, change of voice)) ‘Hello, this is Sam Samaritan speaking’ ((different voice)) ‘I CAN’T HEAR YOU’ ((return to first voice)) ‘Sam Samaritan’ ((other voice)) ‘Sam, I’ve got no one else to turn to’ ((pause)). Cool, ?eh. The caller never gave their name. Anyways, that was just the beginning of his work, and ethno-methodologists have used his stuff on conversation to examine how social order is worked out locally in conversations: in courtrooms, classrooms, doctor’s surgeries, air traffic control towers, news interviews and scientific laboratories. All very attractive to those who are interested in spa::ce and pla::ce. Oh, and what’s great is he has one of those kind of mythical academic biographies, like (Saussure/So sure) his work, apart from a tiny quantity of prepared articles, only survives because his lectures were taped and transcribed. Kinda ironic or WHAT? Anyway, he has become one of those founding authors in conversational analyses, because so many of his students went on to develop important careers, ((pause)) well, and of course he was a genius. And he, ·hhhh the poor guy was killed in a car crash on his way to a seminar on the seventies, oops, Freudian slip, I mean in the seventies. Always, helps doesn’t it, if there’s some kind of interesting biography that goes along with an important school of thought.

So, where was I, ((pause)) ehm::: yeah, conversational analysis has a strong focus on how things are told almost more than what is being told. You know, things like, ((pause)) pausing, ·hhhh aspiration and inspiration, turn-taking, emphasis, LOUDNESS, co-telling, correction, overlapping speech, all that really important stuff on language as social action. And dialogue as a social object for study. I’m not saying that geographers haven’t been attentive to language, ?they have. But they have been wedded to the textual metaphor, in, in, in; ((pause)) dear Derri::::da’s influence, for instance, has ended up with writing being privileged over speech, despite his best intentions to avoid opposing the terms, let alone privileging either. Text, text, text, with perhaps a hint of speech here and there. Conversational analysts try to avoid using the term ‘speech’ as much as they can, because of its associations with speech act theory, which is premised on individual utterances rather than dialogues between two or more speakers. Ehm::, so they use talk as an alternative to conversation or dialogue. Talk, talk, talk. But still not all talk, how could it be? ((pause)) Well, okay, that’s kinda broad, brush stroke, I know, since geographers like Nigel Thrift and Marcus Doel have been dealing with language differently.

I KNOW I AM GOING ON FAR TOO LONG FOR YOUR MACHINE and for an acceptable turn. But I’ll forget what I wanted to say to you if I leave it too long.

So:::::: Yes, I know, the other metaphor that has ended up more meta than it should have is voice. Hhhh, so, in geography we carefully include the voice of the other, ((pause)) look and listen for voices that disrupt texts and so on. For all of it, these are, I think, these are still unavoidably scriptural devices. ((pause)) Okay, of course there are plenty of excerpts from transcripts in texts. But they are content-driven quotes.6 Which comes out of the whole coding process. Hhhh, after all, this is the way I’ve worked as well, sifting through transcript after transcript trying to code them, to crunch them down. It’s all for a good reason, since, once they’re coded, then I can extract all the relevant material on the body or the private sphere or anxiety and work or, or … So, coding has made me do some laborious interpretative work and, most painful of all, I’ve summarized hundreds of pages of speech into key codes. But ?then, I just cut and paste a nice quote as a lone representative voice into my text, turning once-spoken, now written, words into what they as representatives (sometimes of discourses or sometimes of social groups) said, not how they said it and in what situation. Quotes that are all too completely shifted into written language, rather than shifting written language toward spoken language, and which hardly ever refer to the dialogue that they were found in. If you think that one of the fundamental principles of conversations is turn-taking (ie one person speaks then another person speaks), this principle disappears in most texts that claim to represent voices that represent discourse. And, so, yes, dialogue needs to be there too, or perhaps talk, since it then loses its associations with a lot of the textual emphasis of dialogical theory.

Talk is a nice term to use, as I said, it has important ramifications in turning us towards the to and fro of conversation rather than singular acts of speech. You probably have some alternatives as a linguist, but I think talk is a friendly uncomplicated term to start out with, even if the theory and methods that follow are just as demanding as those of econometrics or regulation theory. And of course, the still more fascinating thing is that econometrics and regulation theory are not only ideas constructed and communicated through texts, but in and through talk as well. Which, ehh. When:::: it comes down to it, geography is as much talk as it is textual mastery, ehm, I can hardly forget how odd the talk of geographers first sounded when I listened and tried to speak it myself. Conference talk, conference pub talk, seminar talk, post-seminar talk, teaching talk and the like. These were all places of talk, different kinds of talk, and frequently they would be disciplinary talk under the guise of idle chatter.

But you know me, I don’t really want to get too stuck into geographers writing about geographers writing and talking about geography, that’s the way to an island life. No, what I want to suggest in my paper is that there are geographies of talk. Yeah, OBVIOUS POINT. Like Phil Crang was saying about popular geographies, I want to try and emphasize that ((hums theme from popular television series)) the talk is out there.33 That it is other people’s talk that has to be interpreted, because all kinds of actors are geographers, and I do::: include >children, judges, engineers, caravanners, fruit and fruit buyers, cats and catflaps<.viii, 39 So, surprise, surprise, I am going to try and use an example from my research on mobile phones. Just a simple example, though, I’ve emailed it to you.

Excerpt from Max’s email

... where I demonstrated beyond any kind of refutation that the term ‘reality’ should be removed from all academic discourse, apart from in certain situations where it was used ironically or quoted from historical sources. Allowable alternatives are: realistically, realism, unreal, The Real, surreal, though (in reality) I can never see this happening. Next on my list is the term ‘interesting’, and after that, ‘badly written’.

Anyway, enough griping from me. It has been hard choosing the example for the Area article, because I have to make a point in the briefest possible way about the importance of talk to all human activity, and what conversational analysis can offer to geography for analysing talk. I have to do all of that and avoid sounding glib! What offers a possible solution is copying Harvey Sacks’ (1992) format for opening his course on conversational analysis. Rather than making a general categorization, I will begin, as he did, with a particular strip of talk from the research that Chris and I are doing, which will seem familiar to most people:

MYAJ17: Transcript 1

Source: Audiotape

Date: 10 September 1997

Participants: Max, investigator; Sylvia, target phone user

Situation: Phone call between investigator in his office and participant in her car

((the phone rings))

Sylvia: ((looks at phone and then picks it up)) Hi: Max

Max: Hello Sylvia=

Sylvia: =Hel:lo dahling, how are you?

Max: Good, good. ((pause)) How ’bout you?

Sylvia: O:::::::h I’m >fine<.


Max: Where are you?

Sylvia: I’m::::: just passing junction 17.

The conversation is between Sylvia, travelling along the M4 motorway in her car, and myself, sitting in my office in Glasgow. It illustrates turn-taking, and another of the fundamental interests of conversational analysis: adjacency pairs (greeting/greeting and question/answer)—in other words, turn-making, how one half of a conversation begins to select possible responses. Each speaker’s turn will typically project an adjacent response, and often lead ultimately to a non-verbal action; so that saying ‘Max left a message for you’ will either project a query as to what kind of message, what did he say, who is Max, etc and, if it is an answerphone message, will probably result in you striking the play button on the answering machine.ix If one of the pairs were to fail to match up, the conversation or action would falter. However, this a typical and efficient opening to a phone call, which quickly drives the conversation on and, in the process, confirms who each of the speakers are, whether there is any personal business that needs attending to (‘How are you’), and then checks on Sylvia’s location (on which topic more later). Further, this exchange is what is called ‘an opening sequence’: it orientates what will follow, which may be the arrangement of a meeting, a check on some information, a request for assistance and so on.

Turn-taking is frequently described as banal, and yet it is fundamental to the progression of any conversation, and for the initiation of action as, well, being an action itself. Long speeches attributed to one speaker, so often quoted in geographical articles, are distinctly unusual in most conversational situations, whereby a willingness to share the ‘floor’ is a vital ingredient of participation in the situation. What is probably occurring in these long quotations is that their transcriber is dolbying out the unwanted hiss of the interviewer. Yet, this hiss of ‘uh huhs’ and ‘why did you do that?’, long pauses and interruptions, is not just so much unwanted white noise, it is the proverbial rug under the speaker’s feet, which is pulled out for the transformation of turns into statements. The floor is repressed, because the quotation of voices is at its most controllable when it appears as an unsolicited statement, which is then ‘"cited" (as before a court of law)’ (de Certeau 1984, 161). The danger of such smoothed-out evidential quotes is that they shift from being rhetorical-responsive accounts (Shotter 1993) in one time-space to being metonymic statements of undertheorized worlds, which excite and recite the overtheorized time-space of the analyst. In other words, as de Certeau (1984) puts it, on the ‘acts of theory’:

.... we could say that this theorizing operation consists of two movements: the first move cuts out certain practices from an undefined fabric, in such a way as to treat them as a separate population, forming a coherent whole but foreign to the place in which the theory is produced ... The second move turns over the unit thus cut out. At first obscure, silent, and remote, the unit is inverted to become the element that illuminates theory and sustains discourse. (62–3)

Although the answerphone message that I left (or should that be sent?) plays upon the promises of scriptural representation of voices, I do not wish to suggest that such a game leads to a third space or third way that somehow transcends text and voice. It is perhaps another way of writing an other, and yet another reminder of, and worry over, the limits of current procedures for transcribing and inscribing spoken conversation. Conversational analysis shares the cut-and-turn procedures of the theorizing that is carried out elsewhere in human geography. However, it has taken a different flip in its reliance on recorded strips of ‘naturally occurring episodes of interaction’ (Schegloff 1996, 16), which are not (usually) destined for immediate and hurried transcription, in order to claim quick gains in the scriptural economy. As Nigel Thrift (1998) puts it, conversational analysis is a focus on, firstly, the ‘real’30 time of the recorded interaction; secondly, the shifts required to ‘re-mark’ upon the space-timing of events; and thirdly, the production time-space of theory. A primary concern of conversational analysis is with the time-spacing of a conversational event, which leads it into what can seem an almost obsessive concern with who the speaker is, their length of turn, lengths of pauses, closing sequences and opening sequences, allocation of next speaker, speaker’s direction of gaze, listener’s verbal and non-verbal indications of attentiveness, interruptions, overlaps, agreements and even the open-endedness of any ‘conversational event’, which requires work to keep turning, and work to stop turning. These varied space-timings are an intricate, involved and skilful set of procedures, which allow the revolving door31 of conversation to let people in and out of interactions.

Turn-taking, although essential to talk as interaction, is taken for granted, at least, that is, until some small change occurs that requires a reassessment of the ordering involved. Those paying close attention to the earlier strip of talk will have noticed that Sylvia identifies Max before she has even heard his voice on the phone. Sylvia’s car-/mobile phone has a caller identification function that allows her to programme in the phone numbers of her common callers, and the small illumined screen on the phone then displays the name of the caller as it rings. Being greeted by name by the phone responder upsets the typical adjacency pairing of the opening sequence, wherein the person being called responds with a widely oriented greeting. When Sylvia did this to Max for the first time he was thrown by it and spent several minutes questioning Sylvia about how she knew it was him. Sylvia uses the mobile phone function to delegate part of the opening sequence to the phone (ie the identification of caller), and also then creates a telephonic boundary between known callers and unknown callers. For the known callers, Sylvia is able to project familiarity and trust from the moment she picks up the phone, simply by reading who is calling before speaking to them. To understand the importance of rapid recognition for a ‘good’ encounter, one only need consider what happens when the called does not recognize a close-friend caller ‘simply’ through the sound of their voice when they first greet one another on the phone; a great deal of ritual repair work, often disguised as technical repair work, is normally required afterward, such as, ‘It’s a bad line, I didn’t expect to hear from you, have you got a cold?’, ‘You’ve picked up a bit of Scottish accent’, etc. To accept that this may have ramifications beyond the interaction of friends, one has only to consider that Sylvia works in a service industry where one of her main forms of contact with her clients is via the phone. For the multinational company that Sylvia works for, her greetings over the phone are the ‘face’ of that company, and service industries are reliant on a great deal of face-work from their employees (Crang 1998; Hochschild 1983; Lash and Urry 1994; Thrift 1996).

Those who payed close attention earlier may also have noticed the extra opening adjacent pair in this phone call (‘Where are you?’). The telephone has always been a device for the dislocation of conversations, shifting participants out of face-to-face situations and requiring that particular opening sequences be circulated globally so that two localities can be linked up through talk (Boden 1994; Houtkoop-Steenstra 1991). Consider the local acceptance of the global rule, that it is always the person called that will speak first, and the havoc that could be created should a caller try and pre-empt the called. Mobile phones have somewhat paradoxically reinserted the importance of geographical location into opening sequences of talk on the phone. We are now familiar with sections of overheard phone calls in public places: ‘I’m just passing Newcastle, I’ll be there in about an hour’, or ‘I’m in the restaurant, where are you?’ (those of us who use mobile phones are perhaps employing rather than overhearing these re-embedding sequences). This geographical moment of orientation is now a standard part of the opening section of any phone call, since it creates a minimal background in front of which a conversation can take place. This extra adjacency pair was unnecessary before, because the phone number itself was contextually embedded, the caller knew whether they were phoning a work or a home number. What happens now is that the mobile user has to do extra work to relocate their half of the phone call and perhaps also to frame what follows as formal or informal, home or work.

Let us look briefly at Sylvia finishing a call to Ned, another member of the ‘multinational company’, and watch the boundary work going on during the closing sequence of a long conversation, which has been, up until this point, predominantly about arranging meetings and swapping knowledge about difficult clients.

MYAJ17: Transcript 1

Source: Audiotape

Date: September 1997

Present: Car—Sylvia, driver; researcher, passenger seat. On phone—Ned.

Accents: Sylvia—hybrid New Zealand/English accent. Ned—fairly strong Glasgow

Situation: In car, hands free, travelling between house and restaurant in Bristol

N: Awright [well listen] I’ll see you [on Monday].

+ +

S: [(as far as ((s-pause)) is that)]

N: But have a damn nice weekend=

S: =Okay=

N: =And have a nice lunch. ((s-pause))

S: Eh, I, I will, I’m needing it. I need some [food I tell ya, my head is a bit sore today.]


N: [((starts laughing, progressively louder))] You will slap it in ya womin. ((said with stronger Glasgow accent))

S: Ae?

N: ((continues with strong accent)) (Hell) slap it into yee. You shouldnae drink it’s bad for yee.

S: Well I must admit I think sort of leaving the house at five thirty yesterday and getting home last night at four in the morning probably didn’t help.

N: On that note I’m not even gonna ask yee [anymore].


S: [starts laughing] Okay.

N: Have a pleasant weekend.

S: Take care Ned.

N: Aye, you too darlin’.

S: Bye now=

N: =See ya. Bye. ((bleep of call disconnecting))

Ned begins a turn, with which Sylvia overlaps, and it is unclear to the transcriber (and possibly Ned) what she is saying. However, this slight disorder is contained by Ned’s two succeeding turns, which shift the ‘tone’ of the conversation toward informality as he emphasizes first a ‘damn nice weekend’ and then a ‘nice lunch’. Sylvia initially okays the weekend, and then she makes a further but hesitated move to adjust and perhaps to agree, which is overlapped by Ned’s ‘I tell ya’, each of his informal statements being given their extra emphasis through speech. What is worth noting further is the increasing shift into informality and a change in face, or perhaps tone, which Ned is doing by changing his accent. This is a heavily signalled move, which seemingly backfires, since Sylvia suffers a ‘technical failure’ (Goffman 1981), and thus cannot quite respond to what Ned has said, and her brief ‘Ae?’ indication of this is enough to trigger a second go at the same move from Ned. After Ned’s repetition, Sylvia accounts for her sore head via a different line of explanation than the ‘drink being slapped into her’. We might interpret this as Sylvia justifying her actions and thus doing the work of performing her work-identity. She has a headache from long devoted hours on the job, not from too much drink which provides a responsive account to Ned’s joke about slapping drink into her. There is a lot more that could be commented on in this closing sequence, but for the purposes of the article, I hope it demonstrates something of what can be gained in the understanding of talk as social action when it is transcribed and analysed in this way.

Message continues and concludes

Max: E::::::h what was I? Yeahm: an opening sequence and a closing sequence from one of Sylvia’s phone calls would give me the chance to get in another of my worries about interviewing. Ehm its characteristic of being a response to a stranger or to a perceived professional, you know that point, the kinda thing about the strangeness of an interview situation. I mean, it would be very different asking Sylvia how she did her work over the phone compared to simply taping her doing her work over the phone. Both approaches can be used, but by and large, interviewing dominates for constructing research material, and as far as I know, ehm, taping talk-in-action is very rare, ((pause)) the focus group is the closest geographers come to doing it, but funnily enough, they demand there be no cross-talk,38 even though cross-talk is one of the ways in which agreements, disagreements, domination and counter-domination occur. I mean, ((pause)) what topics set cross-talk going, how is the cross-talk resolved without the intervention of the researcher, who in the end gets to take their turn? And, like I was saying, ehm, the interview, let alone the focus group, provides a very curious kind of transcript, which you’ll quickly see if you transcribe an interview or a focus group compared to, say, a business meeting or, ehm, people talking in a pub or a doctor consulting a patient, ((pause)) or whatever. Unless it’s badly done, the interviewer says the bare minimum and usually takes the shortest possible turns. And an interview in its very reasoning implies language as representation rather than language as action.32 A focus group is usually about an iss::ue for group discussion rather than about how focusing operates, just as an interview is about topics rather than how interviewing is achieved.4 HEY YOUR ANSWERPHONE BLEEPED AT ME AGAIN.6 Can you ?hear that?11 It’s a kinda answerphone I’m listening,8 ((pause)) but for how much longer. Answerphones are a conversational turn delegated to a speaking and listening machine, it’s kinda revealing that in Italian they’re called telephone secretaries.

((pause)) Hm:::: that’s disrupted my chain of thought now.12 Is it making any sense to you? Let me think, what have I said? There was that first point that I wanted to make it clear that dealing with talk has been a problem in my own research, since I learnt a lot from textual analysis rather than conversational analysis and, ?hey, texts are not talk.x, 13

((pause)) And then, and then, yeah, the::: ubiquity of transcripts of speech in geography articles, and yet the absence of systems of notation for dealing with speech. After all, if you think about the rigour with which numerical data is dealt. And the fact that there is a great deal of interpretation required as soon as transcription begins, and also endless uncertainty over what is being said. Like, did she just ?say, ((pause)) I can’t say that I’m happy now OR I can say that I’m happy now. The ways of marking up scripts (needn’t be/need to be) totally standardized since there (needn’t be/needs to be) one right way of doing that.

((pause)) Next, I wanted to suggest conversational analysis as developed from the work of Harvey Sacks, in particular, as one way of assisting in the interpretation of talk, and also to echo the claims of ethno-methodologists that ((imitates Alan Partridge)) A: ?ha, talk is everywhere.34

((pause)) Oh, what’s perhaps not clear enough is, ehm, geographies of talk mustn’t be collapsed onto site-specific conversations like corridor talk, office talk, court talk. Well, maybe a little but ehm, how spatiality, location and scale appear in conversation may be just as important. Like with the mobile phone questions, ‘Where are you?’ or ‘Where’s a good place to eat?’ And, more fundamentally, the embedding and re-embedding work that goes on to extend talk out of specific sites, to try and deal with the faceless and placeless moments of disorientation on the phone.

((pause)) So:::: talk is a way of creating local ordering, yet its conversational principles are global. Like turn-taking. It’s similar to Nigel Thrift’s stuff on how the global space of the City of London requires intense conversational and face-to-face interaction to keep it ?organized? enough, or even the scenario used by Marilyn Strathern to visualize what happens when you meet a GLOBAL COMPANY, it turns out to be a face-to-face encounter in Luton with a few executives from IBM. And what then happens, ((pause)) generally a conversation, a meeting, an interview or some other form of talk. And how is it that talk extends itself? Hm:::: now there’s a big question that I think begins to be answered by looking at actually occurring strips of talk and considering what methods are employed within them to shift out in space and time from their local context. Or equally, what conversational procedures are employed to localize what are otherwise decontextualized, uninvolving conversations.

((pause)) I still think that normally what gets pictured when analysing talk is, ehm, socio-spatial relations happening elsewhere and mere chatter happening here, or perhaps brief moments of representation worth collecting for later. Such as thinking about the micro-macro divide; socio-spatial forces are macro and the talk is micro. Except they aren’t, ’cause ((imitates Alan Partridge again)) A: ?ha, a conversation, like the Nissan Micra, it seems small but it’s everywhere and it’s everywhere the same.

((pause)) Okay, you can tell (I’m running out/I’m not running out) of things to say and your machine is running out of tape. ((pause)) I suppose I just have to thank the answerphone for being such a poor talker.

Answerphone: 3.25 pee em.42 You have no more messages. ((tape begins to rewind))



This paper arises out of analytic work on two projects. Firstly, I would like to acknowledge the tireless support of Dr Linda McKie on a health research project undertaken with funding from the Chief Scientist’s Office of the Scottish Office (Grant No K/OPR/17/2). Secondly, I would like to recognize the remarkable road-handling of Professor Chris Philo on a socio-spatial and technical research project on cars and mobile phones, undertaken with funding from the ESRC (Grant No R000222071). The ideas and views expressed in the article are those of the author and are not necessarily those of the funding bodies.18 A greater than usual debt is owed to the co-participants in these two research projects, most notably Ms Maria Cropper, mother of millions. Finally, I have to thank Dr Ludovica Serratrice for creating a critical conversational background to this paper, and for looking after Spunky and Lola whenever I was away overhearing long distant calls.


i Readers will have noted very quickly that this article is written in an unusual style. It is what is known as a reflexive text (Ashmore 1988; Latour 1996; Ronell 1988; Woolgar 1988). Dear readers, please note that writing reflexively is not an integral part of researching talk; I have adopted such a style here in order to place even greater emphasis on the appearance of transcripts, where they appear, how they appear and what they claim to make appear. Although not transcribed from an answerphone, this article is based ‘in fact’ on a conversation that took place between the author and a linguist on the subject of a paper he was thinking of writing. ‘Conversation’ may be overstating the case, since she was busily packing to go on a trip at the time and he basically talked through his paper with only a few short interruptions, which is why it seemed fair if unrealistic to fictionalize his idea for a paper as an uninterrupted answerphone message. However, the transcript of Sylvia and Max (names changed) greeting one another on the phone was recorded during research fieldwork, even though the email excerpt was written for the purposes of this article (rather than for sending to Marta).27

ii See appendix for description of standard conversational analysis (CA) notation.

iii Though we could think about various reading situations, such as a school teacher reading aloud to a class of pupils and the diverse forms of the book itself.

iv Alan Partridge is a comic television character created and performed by Steve Coogan; the character is a chat-show host of excruciating ambition and mediocrity. In a parody of several ‘genuine’ light entertainers on UK television, Alan Partridge has a series of catchphrases and verbal mannerisms—one of which is to end his opening statements with ‘A ha!’ That a speaker, like a discipline, can borrow from other speakers is critical to a speaker’s capacity to extend themselves by shifting out of their local situation. Goffman (1981) puts it thus:

although who speaks is situationally circumscribed, in whose name words are spoken is certainly not. Uttered words have utterers; utterances, however have subjects (implied or explicit), and although these may designate the utterer, there is nothing in the syntax to require this coincidence. (3)

And this is what Goffman calls ‘embedding capacity’ in speaking; the quoting of Goffman being a further example of embedding.

v See, for instance, Area 28(2) (1996), a special edition on focus groups. However, see also, in contrast, the guidelines in Cook and Crang (1995).

vi In keeping with the stylistic requirements of an annotated transcript, references are carried by names (but no dates in brackets which would add confusion to the CA conventions used). A possible (historian’s normal) option for tracing others’ writing in the text (apart from names with years in brackets after them) would be footnoting; however, the constant appearance of numbers in the text and the standard organization of articles for Area suggest that this too is inappropriate. There is nevertheless a bibliography which contains references for all the dateless names cited in the transcript and those cited in a standard fashion in the email.

vii Though Goffman and Sacks rarely commented directly on one another’s work (Schegloff 1988), perhaps because of its proximity. There are, however, several indirect critiques from Goffman (especially ‘Replies and responses’ and ‘Response cries’ in Goffman 1981), who had the advantage in response terms of outliving and outworking Sacks by almost two decades.

viii Do humans and non-humans talk? Some non-humans talk, the answerphone being an obvious example, indeed if we think about conversation as interaction more generally, then there is a constant dialogue between people and things. Sacks (1992) was well aware of the importance of things that humans bring into the conversation or use to keep other things out, though most of his examples drew on local objects, such as people wearing something as a talking point (ie new shoes, a funny hat). However, I am wary of using talk or conversation as another meta-metaphor, claiming that everything is talk and should be treated as such. Or of trying to make a measurable definite object/event that is inescapably talk.

ix The machine themselves normally request playing by flashing a light or bleeping, indicating that they have recorded a message and thus should also be thought of as being involved as part of the social interaction (Latour 1992).

x For the purposes of this article, I have treated text and talk almost as if36 they were utterly separate ways of interacting. Reflecting briefly on the trajectory of this article makes it clear that they are not, since, as I endnoted (Note 1), this article was based on a conversation I had with a linguist, which then was rewritten as this text. Since drafting this article, I have used it as the textual basis for a talk/lecture on conversational spaces, which, for some listeners, was written down and reinterpreted as a text again, thus we have: talkÆtextÆtalkÆtext.35 For further examination of this process, see Goffman (1981), especially Chapter 4, ‘The lecture’. And the rapid to and fro of emails takes on something of the aspects of a conversation (using text rather than talk). Reading groups surround and share out texts with conversation. Derrida’s (1988) ruminations on postcards renders the miscorrespondances of textual ‘to and fro’ into a wise and comic form.

Appendix 114

There are several fairly standard conversational analysis notations used in the text:3

Double brackets are used to enclose a phenomenon that the transcriptionist needs to describe rather than transcribe:

((she let the answerphone take the message because she did not want to speak to him))

And for dealing with pausing, numbers indicate a timed pause, lacking the special machinery required for measuring length of pause I have dealt with them in the telephone conversation transcripts on a rough 3 point scale of short, medium and long:

((s-pause)) ((m-pause)) ((l-pause))

Capital letters are used to indicate words that are said louder than the surrounding speech:

MAYBE I SHOULD PICK UP THE PHONE. No, he’s spoken for too long, he’d guess I was trying to avoid him.

Underlined words indicate emphasis:

I’m not really avoiding him, it’s just that I’m in the middle of watching Brookie.

Audible inhalation is marked by a dot in front of one or several ‘h’s, while exhalation has no dot:

·hhhh the poor guy was killed in a car crash.

Hhhh. He’s still talking to the machine.

Parts of an utterance delivered at a faster pace than surrounding speech are marked with less than and greater than signs:

Still talking. ((angrily)) >Will you stop talking to my machine<

When the transcriptionist has doubt over what was said, this is indicated by brackets around the words, or by ‘/’ slash divided bracketed phrases indicating two or more possible interpretations:

(Why is there so much car advertising/Why does he want me to advise him) at this time of night.

Rising or falling intonation is marked by up or down arrows immediately prior to the speech:

Sometime I feel ?sorry for him, he must be obsessed with his work

Extensions of syllables or sounds are indicated by colons:

This is beyo::::nd obsession. Pu:t the: pho:ne do:wn.

When there is no interval between utterances, they are linked by equals signs:

Max: Let me think what have I said=

Marta: =Quite enough for now

When utterances overlap or speaker makes some form or utterance during an unfinished turn then square brackets are put around the overlapping speech with a plus sign indication which sections overlap:

Max: At which point the guy [we’ve just met falls] over a traffic calming device.


Marta: [Yada, yada, yada]

Sections of speech spoken more quietly than surrounding speech are surrounded by degree signs:

He’s always leaving huge messages on my machine. ?Why am I talking out loud?? I’ll speak to him now, oh, he’s finished.


Agar M (1984) Speaking of ethnography (Sage, London)

Ahearne J (1995) Michel de Certeau: interpretation and its other (Polity Press, Cambridge)

Ashmore M (1988) The reflexive thesis: wrighting the sociology of scientific knowledge (University of Chicago Press, Chicago)

Atkinson J M and Heritage J (1984) Structures of social action (Press Syndicate of the University of Cambridge, New York)

Barnett C (1997) ‘Politics, postcolonialism, and other figures’ Environment and Planning D: Society and Space 15, 137-154

Boden D (1994) The business of talk: organisations in action (Polity Press, Cambridge)

Boden D and Zimmerman D (1991) Talk and social structure (Polity Press, Cambridge)

Chomsky N (1965) Aspects of the theory of syntax (MIT Press, Cambridge, MA)

Cook I and Crang M (1995) Doing ethnographies IBG CATMOG series (Institute of British Geographers, London)

Crang P (1998) Spaces of service (Routledge, London)

de Certeau M (1984) The practice of everyday life (University of California Press, Berkeley)

Derrida J (1979) Writing and difference (Routledge, London)

— (1988) The post card (Chicago University Press, Chicago)

Edwards J A and Lampert M D (1993) Talking data: transcription and coding in discourse research (Lawrence Erlbaum, Hillsdale, NJ)

Garfinkel H (1967) Studies in ethnomethodology (Prentice Hall, Englewood Cliffs, NJ)

Goffman E (1961) Asylums (Penguin, Harmondsworth)

Goffman E (1981) Forms of talk (Philadelphia University Press, Philadelphia and Blackwell, Oxford)

Hochschild A (1983) The managed heart: the commercialisation of human feeling (University of California Press, Berkeley)

Holquist M (1990) Dialogism: Bakhtin and his world (Routledge, London)

Houtkoop-Steenstra H (1991) ‘Opening sequences in Dutch telephone conversations’ in Boden D and Zimmerman D (eds) Talk and social structure (Polity Press, Cambridge), 78-89

Lash S and Urry J (1994) Economies of signs and spaces (Sage, London)

Latour B (1992) ‘On the sociology of a few mundane artefacts’ in Biejker W and Law J (eds) Shaping technology, building society (MIT Press, London), 225-258

— (1996) Aramis: or the love of technology (Routledge, London)

Richardson M (1981) ‘On "The superorganic in American cultural geography"’ Annals of the Association of American Geographers 71(2), 284–7

Riessman C K (1993) Narrative analysis (Sage, London)

Ronell A (1988) The telephone book: technology, schizophrenia and electric speech (University of Nebraska Press, Lincoln)

Sacks H (1992) Lectures on conversation (Blackwell, Oxford)

Saussure F (1931, republished 1974) Course in general linguistics (Fontana, London)

Schegloff E (1988) ‘Goffman and the analysis of conversation’ in Drew P and Wooton T (eds) Erving Goffman: perspectives on the interaction order (Polity Press, Cambridge), 89-135

— (1996) Interaction and grammar (Cambridge University Press, Cambridge)

Shotter J (1993) Conversational realities: constructing life through language (Sage, London)

Strathern M (1995) Shifting contexts: transformations in anthropological knowledge (Routledge, London)

Thrift N (1996) Spatial formations (Sage, London)

Thrift N (1998) personal communication

Whatmore S (1997) ‘Dissecting the autonomous self’ Environment and Planning D: Society and Space 15, 37-53

Woolgar S (1988) Knowledge and reflexivity: new frontiers in the sociology of knowledge (Sage, London)