Abstract
In conversation analysis (CA), through Sacks, Schegloff, Jefferson, and others, the conceptual architecture is joined at the hip to a technical architecture of transcripts, sequence, and turn productions. That the conceptual was to be found and demonstrated in the material detail of temporal productions was central to CA’s extraordinary innovations. As with CA, an Epistemic CA has the task of giving evidence of its conceptual order in actual materials, and thus animating the materials to show them. The task and relationship are emblematically reflexive: we shall find the expression ‘Oh’ indexing ‘changes of state’ or ‘inapposite inquiries’, for example, as of the account-able animations of turns and sequences of turns. Our shared attachments to sequential analysis deliver the expectation that we shall see how Epistemic order is achieved on actual occasions, through actual materials, rendered as transcript. The discussion turns to how the Epistemic Program (EP) engages and acquits this analytic expectation.
Introduction
From the articles in this Special Issue, we now have a collection of Epistemic conceptualizations, noticings, objects, and organizational things drawn from its publications. It is a hybrid collection of conversation analytic things like turns, positions, and sequences; and Epistemic things like changes of state, particles that ‘index’, and tensions about access, propriety, and the authority to speak of whatever the parties may be speaking of. Perhaps central to the Epistemic Program’s (EP) distinctive programmatic identity is the introduction of contested fields of greater and lesser epistemic agency (Heritage, 2002, passim), revealing asymmetric claims and endowments operating in the background, shaping our ways of speaking (and hearing) by shaping our authorizations to do so. No less central for the EP are the seemingly unavoidable intersections of ‘knowledge’, ‘experience’, information and their possessions, and also different forms of knowledge and experience, routinely parsed as greater and lesser forms.
As the EP develops in its more contemporary publications, natural conversation comes to be about ‘monitoring epistemic status’, giving evidence of imbalance, achieving balance (see Lindwall et al., this issue), and thereby, in the course of these operations and their preoccupations, producing the evident order of talk-in-interaction (Heritage, 2012a, 2012b). As the parties act as epistemic claimants, pressing and/or deflecting authority, access, and ownership, often ‘indexed’ by ‘Oh’ productions and modulated by morphosyntax to yield the orderly action formations and turn constructions we then find, we can expect that these orientations, tasks, and tensions can be – and will be – revealed in the actions and constructions they shape. Or at least this is the expectation of our attachments to conversation analysis (CA), as it was and has been CA’s achievement to wed its sociology of natural language use to actual exhibits in their constitutive detail. CA has produced not only descriptions of conversation’s organizations in multiple domains, but also descriptions of social actions – grammars of action – shown in the temporal-material detail of transcript on actual occasions, an expectation that seems to anchor the EP to material worlds and material study as well.
Cases in their constitutive detail
Ethnomethodology and conversation analysis (EMCA) were leveraged from a deep dispute with the normative analytic cultures of their day (and to this day). There were many disputes, but this one may have been central: by conventional academic wisdom, ‘everyday life’ – in the profusion of its expressions, occasions, unremarkable engagements, and recurrence – was a dense quotidian mask laid over where and how the engines of social order might actually be found and disclosed. If anything, everyday life was a mis-direction, its orderliness was hidden, and social science was the corrective (see Lynch and Wong, this issue). As Sacks (1984) observed, the consensus modus was that overwhelmingly the products of everyday life were the dross of social order, and the aim of formal analysis was to recover the ‘golden chain’ that would show itself only here and there to the credentialed devices of formal-analytic reasoning.
To these presumptive reckonings, EM and CA proposed radically alternate understandings: that everyday life was already in possession of the terms and resources for understanding its relentlessly achieved order. This was part of the shock of CA for language study. It had been assumed that vernacular order owned neither provenance nor authorization. It did not run deeply enough. Worse, were vernacular grammars admitted, the disciplines might confront a horizon of uncountable grammatical forms and competencies, outstripping all best efforts at systematic reduction. 1 EMCA’s understandings of vernacular worlds put these long-standing formal-analytic ambitions at serious risk, and still do. Like CA’s ‘proof procedure’ wherein next turns are consulted for understanding the sense of prior turns, our interests, descriptions, and measures do not stray far from vernacular reckonings, or from our transcripts. The aim is not to take our leave of the vernacular order, at least until we understand it. And once understood, there may be no reason to take leave of it.
Transcript has thus been the ‘portal’ for CA studies onto an unexamined orderliness running in full and public view. It is through transcript that social action is rendered as concertedly on-going productions in finely crafted durations. These productions are what hold our interest, and the sense of the phrase ‘animating transcript’ refers to the work of how we leverage these production accounts, getting our transcripts to speak of the objects, descriptions, and orientations we claim for them. 2 Often enough the task is naturalistically descriptive, as when noting an overlapping turn, a greeting, or even a disagreement. Routinely, we hear them as any competent speaker would hear them. Elsewhere, however, descriptions are achieved by the accounts that organize and deliver them. For CA, and presumably the EP, transcript is the generative, constitutive field of findings, and those findings include the accounts that reflexively leverage them into view (as in accounts of ‘doing indirection’ (Sacks, 1992, vol. II: 101), the repair space (Schegloff, Sacks & Jefferson, 1977), every discussion of ‘possibles’ (Schegloff, 2006: 145), or even a ‘question’ (Schegloff, 1984: 29)).
Talk of EM reflexivity is seldom easy, and we want to pursue it with a small collection of materials that may begin to get at this work. By turning to treatments of actual materials, we are returning to the expectation within CA that sequential analysis will treat actual cases ‘formally’ and ‘informatively’, and further that we might return to the mat-erials themselves to assess analytic disputes or differences (Sacks, 1984). As Schegloff (1991) observed, the ‘test of the adequacy of a description of some practice [is] its capacity to yield convincing analysis of singular episodes of conversation’ (p. 153). In part, this ‘test’ is a way to assess the cogency of our accounts and whether they are revealing of what the participants are doing and also, and especially, whether – and how – our accounts can have use for the profuse, unremarkable, and yet constitutive detail registered in our transcripts.
Implicit to these proposals was not only a penetrating critique of formal analysis. There was also the proposal that natural language study could not be disengaged or uncoupled from the mastery of natural language. One could not do this work and not be a competent over-hearer (see Wong and Olsher, 2000). For this reason, we invite the reader to engage the discussions that follow as sequential analysts, on the one hand, and also as competent over-hearers. It is by both competencies that we may have something interesting to say about ordinary talk, its orderly productions, and their description. 3
The corpus
The material presented in the following appears in the early publications discussed in Macbeth et al. (this issue), and also subsequent publications. As readers of the EP may have noticed, many of the transcribed materials appear multiple times across publications, and also multiple times within single publications.
The first sequence appears at least in Heritage and Raymond (2005) and in Heritage (2011), and we are using the narrative from the later publication. It takes up assessments and ‘… cases in which one party evaluates some state of affairs to which the other has no access at all …’ (Heritage, 2011: 160). In the ‘no access’ we have a foundational separation that underwrites the epistemic organization of interaction as between those who have access, knowledge, or information, and those who do not. As one might then expect, a recurrent project for the EP is the parsing of kinds or ways or measures of knowing.
Lottie and Emma: First assessments
Heritage (2011: 160) introduces the following sequence (his exhibit 3) by noting that ‘Emma’s sister Lottie has returned from an apparently exhilarating trip to visit friends in Palm Springs. Her method of representing the house she stayed in centers on its inaccessibility to her sister’: (1) 1 LOT: h h Jeeziz Chr 2 ↓n 3 EMM: [I b
In the discussion following the transcript, we read Patently lacking the resources to enter into a direct appreciation of the house by the very terms of Lottie’s assessment, Emma aligns with Lottie’s evaluation by means of a subjunctive expression of her likely evaluation, thereby achieving a simulacrum of agreement … (Heritage, 2011: 160)
This is a dense account of a three-line transcript of two turns. Note first the framing of what is interesting about the sequence: it is an item to a collection of assessments about matters ‘to which the other party has no access at all …’ (Heritage, 2011: 160). But there are two potential difficulties here, beginning with how first assessments are recipient designed. We don’t normally offer knowing assessments of an operatic performance, for example, to others who know nothing about opera (unless we are teaching them). Both Sacks (1992) and Pomerantz (1984: 61, 63) remind us that first assessments are produced for persons who can perfectly well offer a second, and thus agree or not in next turn, and we think that is clearly so in this case as well. In the particulars here, we can also ask, is Lottie’s first turn an assessment calling for a second, or an enthusiastic preface to a telling of her trip experience? By either account a response is called for in next turn, and CA’s proof procedure recommends that we consult next turn for understanding its prior.
Second, the expression ‘you have no idea’ figures centrally in the treatment of Emma’s ‘access’. But if one examines the audio record and transcript as it was prepared by Jefferson, you will find multiple uses of the idiom ‘you have no idea’ as part of Lottie’s breathless accounting, and this provides a very different impression of Emma’s ‘access’. In the Jefferson transcript, the target sequence is found in lines 81–84. But in the prior talk, they discuss the house in terms we shall hear again, along with its location – its neighborhood and landmarks – a location that Emma evidently knows well enough (see Schegloff, 1972 on formulations of place): (1a) [CABank/Jefferson/NB:IV:10:R/21swimnude.cha]
4
43 Emm: = 44 Lot: °O 45 (.) 46 Lot: °J 47 (.) 48 Emm: M 49 Lot: 50 Torrero. 51 Emm: Oh:::. 52 Lot: 53 Emm: Oh n 54 (1.0) 55 Lot: °ihYe:: 56 57 Emm: ⌊·hhhh ⌊ 58 Lot: = it yihkn⌈ow tuh P⌉ 59 Emm: ⌊ Ye
Emma knows these formulations of place, and as their conversation continues we encounter the sequence we are discussing and discover a different sequence, and specifically, a very different second turn by Emma: (1b) 79 Lot: [But] 80 Emm: [Oh 81 Lot: ⌊h h⌋J 82 E(h)mma 83 84 85 [u- 86 Lot: 87 Sundee 88 Emm: ⌊ehh h
As can be seen, Emma’s actual turn at lines 83–84 is Emm: I bet it’s a dre
We cannot tell how Emma knows of the pool from the prior talk. And in subsequent talk, Lottie corrects the account: the pool isn’t enclosed, it is ‘outside the big glass doors’, though Lottie treats the mis-description as her own (lines 529–534). But what is quite clear is that the transcript, once seen, rather deeply qualifies the account of what Emma ‘knows’ as ‘no access’, a formulation that seems to be operating from a kind of ‘sense-impression empiricism’, whereby if one party ‘saw’ something that the other party didn’t see, the other party would have ‘no access’ and the first would have sole access. But again, this re-writes our understanding of assessment sequences. Moreover, one may know of things not seen ‘this time’ (rush hour traffic, dreary weather, magnificent houses, etc.) well enough to join a knowing conversation about them. An insistence on ‘sense impressions’ is not quite recognizable in a literature that has taken keen interest in accounts, tellings, and the account-ability of ordinary worlds. Via the accounts of others, we know far more of the world – and its warranted assessments – than what we have ‘actually seen’. 5
So, it is not that Emma would mistake this house for a boat, or that she has never heard Lottie’s expression about ‘having no idea’ before (it is first heard in line 49–50). Lottie is not speaking, nor is Emma hearing quite so literally. And Emma’s reply shows she does indeed ‘have an idea’, of how Lottie is speaking: a first idiomatic expression receives a next; just as she ‘has no idea’, then it ‘must be dream’, for them both. Neither is literally so, and notwithstanding subjunctive expressions, ‘alignments’ do more than mark absences.
One last point on the issue of ‘access’ and the notion that Emma ‘… has no access at all …’. Briefly, in Lottie’s first turn, she remarks, 1 h h J 2 |↓n
We want to note the turn production ‘…
So by various moves and framings, the Epistemic treatment animates the transcript to leverage its findings. There is a formality and stiffness to the account. There is nothing of the evident excitement the parties share in the telling, and it is hard to find evidence of their demonstrable orientations. The turns seem to be characterized rather than analyzed in their production features. Furthermore, to find a ‘simulacrum of agreement’ here is to write a metric of ‘kinds of agreements’ and then declare what kind this one is. It is an over-hearer’s measure entirely.
Trixie: Epistemic assessments and Oh-prefaced agreements
The second sequence has to do with a conversation between two dog breeders discussed in Heritage (2002), Heritage and Raymond (2005), and Heritage (2012a). A recurrent theme across many EP analyses of transcript is one of ‘proprietary relations’, or how one may ‘own’ things like knowledge, experience, spouses, children, grandchildren, cats and dogs. This ownership can be central to epistemic status and authority. In the particulars here (from Heritage, 2002: 204–205, exhibit 11), [T]wo dog breeders – Norman and Ilene – have been talking about the readiness of one of Norman’s younger dogs to have a first litter. . . . And at line 9, Ilene mentions one of Norman’s other dogs (Trixie), who apparently began breeding at a young age: (2) [Heritage 1:11:4] 1 Ile: No well 2 Nor: [She:: ] 3 Ile: = uh[: 4 Nor: [She wz a 5 Ile: 6 Nor: [ 7 Ile: [Y 8 Nor: = But 9 Ile: 10 Nor: 11 Ile: = °Ye:h° =
The narrative resumes, Here Norman’s oh-prefaced agreement (line 10), in conveying the independence of his assessment from Ilene’s, also alludes to his epistemic priority with respect to the information in question … At the same moment, Ilene’s tag question (line 9) downgrades the epistemic strength of what would otherwise be a flat assertion.
And further, … the epistemic priority of the second, oh-prefacing speaker is available from the topic and context of the interaction and inexplicitly indexed in the talk. (Heritage, 2002: 205)
There are several things to tease out here. The sequence begins with Ilene suggesting that the pup in question is ‘still a bit young’. Norman cites her age in line 4, and Ilene seems to receive it as settling the matter. But she returns with the recollection about Trixie in line 9, and her re-completion, ‘didn’t she’, seems less a downgrade than a continuation of her reminder about the case of Trixie with a call for agreement. Ilene is not ‘asserting information’; she seems to be calling for confirmation, and she gets it. Her turn seems to be the touch off for Norman in line 10, in his softly spoken ‘°Oh:: ye:s°’. But note that it is in overlap with Ilene’s re-completion. So the re-completion (‘didn’t she’) is not the turn completion that Norman addresses; he addresses Ilene’s reminder of line 9, where soft speaking can be a mark of uncertainty, and discovery.
And given that they are both dog breeders with evident access to a shared history, it is difficult to hear Norman conveying either ‘epistemic priority’ or ‘independence’ in line 10, whether ‘inexplicitly indexed in the talk’ or not (the sense of the phrase is not developed). It is difficult to see how the transcript delivers the account the analysis suggests.
Instead, it seems that Norman has belatedly discovered Ilene’s first allusion to ‘troubles’ produced in line 1. Ilene artfully pursues it in her reference to Trixie in line 9 (and how Norman may be proceeding with a decision that did not work well the last time). And that he agrees with it in line 10, once he discovers it, then receives Ilene’s softly latched confirmation in line 11 of what Norman has now found as the thing she had been pointing to from the beginning. But this is a very different account of the sequence. It attaches to the detail of the transcript rather than to priorities or statuses ‘inexplicitly indexed’. It treats the record in evidence. Of the difference, we think that Schegloff’s (1991) measure of ‘convincing analysis of single episodes of conversation’ (p. 153) continues to be very useful. More directly, One of the key tasks of researchers in developing claims for a phenomenon is not to sacrifice the detailed examination of single cases on the altar of broad claims – especially when the cases are meant as evidence for the broad claim; one of the key tasks of readers is to examine the detailed analysis of single cases as episodes with their own reality, deserving of their own rigorous analysis without respect to their bearing on the larger argument for which they are being put forward. (Schegloff, 2010: 42, italics in original)
Jan is sick: Inapposite inquiries
The next sequence is about ‘Oh-prefaced responses to inquiries’ and is placed under the heading ‘Questions rendered inapposite by context’. In this instance, … the respondent treats the object of an inquiry as self-evident by virtue of its physical context, or of persons’ cultural or personal knowledge. In exhibit 15, Jan comes to the phone and gives a cough (1–>). Ivy’s subsequent inference (2–>) gets an oh-prefaced confirmation (3–>). (Heritage, 1998: 301) (3) [Heritage, 1998: 301–302, exhibit 15] [Rah:A:1(2):1] 1 Jan: 2 Ivy: [Oh 3 Jan: 4 Ivy: 5 Jan: [khh 6 Jan: 7 Ivy: [O 8 Ivy: Yih
This narrative has Ivy displaying her inference that Jan is still sick in line 4, and Jan producing her ‘Oh-prefaced’ reply in line 6 to mark it as ‘inapposite’, as Oh-prefaced replies in such places are said to do, because Jan’s state of health was already evident in her cough-interrupted greeting.
We can note first an equivocation in the description of Ivy’s line 4: it is framed as an ‘inquiry’ but then characterized as an ‘inference’, and perhaps there is good cause for it. We are not sure of the logical relations between inquiries and inferences, but to complicate things, line 4 also appears to be a conclusion, and an immediate conclusion from within its local production environment. Note also how Ivy’s first inference/conclusion is evidenced in line 2 – her ‘Oh dear me’. Note further how the production of line 4, slightly simplified, seems to be Ivy: Ahr you still’v go i:t.
Absent the audio record, we can’t settle the matter, but a change in its production – a repair – seems to be in play. It seems to begin as ‘Are you still …?’, and becomes ‘you still’v got i:t’. That is, what may have been launched as an inquiry is completed as a conclusion. And there is good sense for the hearing when we consult the production of the turn in its course: 3 Jan: khh= 4 Ivy: = A[h(r) you still’v got i:t. 5 Jan: [khh 6 Jan: Oh yes. I wz pretty bad y(h)estiday. [.hhhh-
Line 3 shows a cough in the clear, to which Ivy latches her line 4 with ‘Ahr …’. But her turn beginning is then overlapped with a second cough and now we have Ivy speaking in the context of consecutive coughing, and insofar as coughing evidences illness, she (seemingly) concludes as much, and repairs the course of her turn.
The premise of the discussion of ‘Oh-prefaced replies to inquiries’ is that an inquiry is inapt if its answer is already in evidence, and we do seem to have the second part of that formulation in play. That ‘you still’v got i:t’ is indeed in evidence: the coughing brackets the turn’s initiation in line 4. But it is Ivy who is showing an orientation to ‘what’s evident’ in how her turn takes the trajectory it does. This raises further questions: is her conclusion in line 4 – ‘you still’v got i:t’ – ‘inapposite’ in its orientation to what is already in evidence, and if so, would every such conclusion from what is in evidence be inapposite? 6 And if Ivy’s conclusion of line 4 is not inapposite for these reasons, then in what sense is Jan marking it so with her ‘Oh-preface’ in line 6? 7 And is ‘inappositeness’ a judgment that falls to the recipient whenever a current turn remarks on affairs ‘in evidence’? But of course, the domain in question isn’t just any remark on affairs in evidence. The relevant domain is ‘inquiries’, and this returns us to the equivocation in the treatment of Ivy’s turn as an ‘inquiry’, an ‘inference’, or now a ‘conclusion’ (our term). And in the equivocation, there is the suggestion of a potentially enormous organizational domain – remarks on things in evidence – for which the epistemic treatment offers no guidance as to where ‘inquiries about things in evidence’ stand with respect to the larger domain. There is no discussion of what things in evidence are sensibly found inapposite when heard as inquiries, inferences, or conclusions, and when, much less a consideration of what other work an ‘Oh-prefaced’ response to an inquiry might do. 8
In this way, and for this sequence, we can find and affirm an orientation to ‘what’s in evidence’. But it is relieved of any sense of inappositeness. That sense, for the EP, seems to rely entirely on the assertion that ‘Oh-prefaced replies’ are so motivated; the assertion seems to be determinative of the finding, and perhaps for this reason the finding has an uncertain relationship to the materials. While Jan’s second turn in line 6 is certainly ‘Oh-prefaced’, and while the discussion writes a motivated account for it, it is not then the case that the turn itself gives much evidence for the account. The ‘Oh’ here could well be an appreciation of Ivy’s conclusion about Jan’s circumstances, a conclusion leveraged from her (Ivy’s) close attention to their exchange and the opening it affords for Jan’s telling that follows. There is lots of coughing, and Ivy misses none of it.
As sick people do, we often answer the phone with evidence that we’re sick. And with such evidence, all inquiries about ‘how are you?’ are virtually questions with known answers. This kind of redundancy seems far from inapposite. Instead, it gives the call receiver the resources with which to show the measure not of an inapposite question or inference, but of just how sick she is, or has been, as Jan does in the same turn. As an impression, commiseration, sympathy, and condolence sequences are routinely made of redundancy (as are greetings, closings, and deliveries of news). Measures surely are at play here, measures of how sick Jan was and is now, but for these parties they are not about inapposite inquiries. And in the measure, we gain some sense for how well ‘information redundancy’ is serving our understanding of actual cases in their turn-by-turn production.
Hyla’s boyfriend: Free-standing ‘Ohs’ and their opacity
The fourth sequence appears in multiple publications: in Heritage (1984, 2005, 2012a, 2012b, 2013a, 2013b) and in multiple places within single publications. We are conjoining two of them presented in 1984, and collating discussions about ‘free-standing Ohs’ and their opacity. As was discussed in Macbeth et al. (this issue), the recipient of a ‘free-standing Oh’ will await further instruction or invitation before taking a next turn because the opaque expression defeats an understanding of what is to be done with it, next. In this sequence, Nancy is talking to her friend Hyla about Hyla’s new romantic interest in San Francisco: (4) [Heritage, 1984: 310, exhibit 26] [HG:II:25] 1 Nan: .hhh Dz he ‘av ’iz own apa:rt[mint?] 2 Hyl: [.hhh] Yeah, = 3 Nan: =Oh:, 4 5 Nan: How didju git ’iz number, 6 (.) 7 Hyl: I(h) (.) c(h)alled infermation’n San Fr’ncissc(h) [uh. 8 Nan: 9 (.) 10 Nan: (Heritage, 1984: 325)
The narrative tells us that ‘[In line 4], the informative party withholds further on-topic talk after an “oh” receipt until receiving a request to do so …’. In this context, it may further be noted that, whereas ‘oh’ may propose a change of state in response to an informing, it is entirely opaque as to the quality or character of the change of state proposedly undergone by its producer. Thus an informant/‘oh’ recipient may withhold from further talk with a view to permitting/inviting the ‘oh’ producer to elaborate what lay behind the production of the particle. (Heritage, 1984: 325)
So, Nancy is quizzing Hyla about her new romantic interest, and the narrative tells us that it is the ‘opacity’ of Nancy’s ‘Ohs’ in response to Hyla’s disclosures that presses the sequence forward, as Hyla withholds to permit or invite Nancy to ‘elaborate what lies behind the particle’. Thus, what change lies behind the particle, in its opaque and generic silence, substantially shapes the progression of the sequence.
Note further that the opacity and the withholding are expressions of uncertainties of information, and thus uncertainty organizes sequence development. And no doubt there are times and occasions where this is so. Imagine interrogations, whether by police of suspects, or parents of their children, or children of their parents (see Heritage, 2012a; Terasaki, 2004[1976]). And in fact, we have something very much like that here, a practical interrogation. But it doesn’t seem to be an expression of a hidden dynamic indexed only here and there by opaque tokens. Rather than expressing something that is ‘prior to’ or ‘underneath’ the grammars of conversation, what Hyla and Nancy are doing here seems instead to be a familiar way of speaking.
That Hyla and Nancy are engaged in a practical and knowing course of inquiry is evidenced when we consult the larger sequence found in Schegloff (1988b). It begins with a question from Hyla:
9
(4a) [Schegloff, 1988b: 451–452, exhibit 7] [HG:22-23] 01 Hyla: Y’ know w’t I did las’ ni: [ght? 02 Nancy: [ Wha:t, = 03 Hyla: = Did a te:rrible thi:::: [ng 04 Nancy: [ You called Si:m, 05 (0.4) 06 Hyla: No:, 07 (.) 08 Nancy: What, 09 (.) 10 Hyla: t’ hhhh [Well I hed-] 11 Nancy: [You called ] Richard, = 12 ( ): = hh-hh = 13 Hyla: = (h)y(h)Yea (h)h en I h(h)ung up w(h)un ‘e 14 a(h)ns [wer 15 Nancy: [ Oh: Hyla why:::::
The last turn may remind the reader of the modest purchase of the informational interrogative as a characterization of what work such turns can do. But the larger point is that we see how Nancy and Hyla became engaged in a guessing game called ‘Do you know what I did last night?’. It is a perfectly ordinary thing, and a lovely candidate for talk about ‘who knows what’. And indeed it unfolds as sequences of Nancy pursuing what Hyla knows but isn’t saying, yet (as in the game ‘20 questions’). It shows us how we may well find exchanges in which one party knows things the other does not, and the other proceeds, with the cooperation of the knowing party, to find out. And whenever one finds such an occasion, it will be found as the occasioned features of ‘how we’re speaking now’.
We therefore certainly have in this sequence something like a distribution of K+ and K− relations. But depending on whether we treat them as occasioned relations or formal structures, we will be led to very different animating descriptions. In the particulars of the sequence, as of Hyla and Nancy knowing perfectly well how they are speaking, and as it becomes a series of Nancy’s paced inquiries, each next answer to which receives her ‘Oh’ and then a duration of her thinking about it and then a next question, there is no puzzle or mystery or opacity as to ‘what lies behind the production of the particle’. None. The sense of these ‘Ohs’ and the durations that follow them lies right on the surface of the sequences whose productions they join as an emerging account. We have here a collaborative course of inquiry. The parties know who knows what, as of this ‘language game’ they are playing. The play of Nancy’s ‘Ohs’ is evident, and whatever they may be ‘indexing’ of an epistemic asymmetry, the parties have been knowingly and openly speaking in those ways from the beginning. Perhaps it is the nature and achievement of sequential analysis and Jeffersonian transcripts that they leave little space or need for conjectures of hiddenness. The relentlessly familiar work of common understanding seems written on the surfaces of interaction, and for the parties first. These ‘Ohs’ show us the parties’ demonstrable orientations, and what they bring to our attention is first shown and found by them. Formal accounts are late comers to these practical tasks and achievements.
Discussion
It may be an unexpected feature of the CA corpus, meaning that perhaps no one had reason to imagine it 40 years ago, that readers of the literature will find and note the reappearance of transcripts we have seen before. Same sequences have been treated many times to same and different issues, and we know many of them by their ‘handles’, like ‘Two girls’, or ‘Chicken dinner’, or ‘Therapy talk’. Returning to them is like renewing an acquaintance, and the yield is a richly familiar corpus filled with practical demonstrations of what topics ‘same materials’ can sustain, and also a sense of the emergence of topics, interests, and conceptual distinctions across CA’s formative history. They give evidence of a distinctively disciplined form of inquiry. Yet when we turn to Epistemic re-analyses of these same and other materials, we can find a schism.
A central and recurrent puzzle for our reading of the EP is how occasioned production features of turn and sequence, like those we have reviewed and many others, are rendered expressions of a durable formal structure operating in the background. Transcript is animated on behalf of an omnipresent engine of ‘epistemic order’ (Heritage, 2008: 309). These renderings are programmatic achievements, generic in their terms and operations, and treated as causative of what order may be found on any actual occasion. This would seem to be the central aim and achievement of the EP’s animations, notwithstanding that generalizations such as these are routinely crafted at the expense of actual, produced, and constitutive detail, and what that detail may show us (cf. Sacks, 1992, vol. 2: 430; Schegloff, 2010). 10
Our interests from the outset have taken up the continuities and departures of the EP from the conceptual excavations of EMCA. On the one hand the debt is unmistakable. On the other, departures are too. These are of course conceptual matters, and as Winch (1990) shows us, every task, distinction, and puzzle of social science – everything to be gained and lost – is a conceptual matter.
We have addressed several such conceptual departures through the early publications of what has become a recognizable literature and program of Epistemic studies of natural conversation. We could variously sum those departures, as in, for example, a departure from the demonstrable orientations of the parties as both the aim and map-works for understanding their sequential productions. In their place, we find the formal structures of generic particles, gradients, and statuses operating prior to any actual particularization. Or – and it is an allied move – a departure from production accounts of temporal-sequential order in preference for more nearly ordinal accounts of position. Or, how Heritage (2012c) and Levinson (2013) write critiques of CA’s ‘proof procedure’, whereby we consult what a first turn yields in next turn, for understanding what work the first is doing. These descriptions are of course expressed in terms of the vernacular reckonings of natural language, about things like questions, complaints, and compliments. And for this reason there is for Levinson (2013) a hermeneutic ‘soft underbelly’ of vernacular practice verging on ‘the occult’ (p. 105) that organizes them (see Lindwall et al., this issue). But the next turn ‘proof procedure’ is not really about ‘proving’ (nor is every clarifying analytic description a proof; see Lynch and Wong, this issue). It is rather a way of making use of vernacular reckonings to discipline our professional reckonings, by consulting how the parties discipline their own. As noted above, EMCA does not take its leave of vernacular grammars of action, and Schegloff has written clear replies to those who propose we should (Schegloff, 1987, 1988a, 1988c, 1991, 1996, 2009, 2010). Those grammars are indeed what EMCA aims to understand. Nor does it seem quite newsworthy to suggest that vernacular order is no more than an occult production. The assessment is the familiar self-appointment of formal analysis; to it, everything else is an occult production.
But perhaps more telling for the EP, having little use for CA’s ‘proof procedure’, it is in no way clear what the EP offers in its stead; that is, what analytic disposition or instruction it offers whereby we align our inquiries and descriptions with what the parties are demonstrably doing, as both a resource for our inquiries and a measure for assessing the adequacy of our descriptions. If the CA ‘proof procedure’ is to be set aside, what will do its work within the EP?
The question seems quite open. But we can see what may be a kind of procedure in discussions of action formation wherein someone initiating the action of ‘requesting information’, regarded by the EP as the ‘the ultimate paradigm of an adjacency-pair first action’ (Heritage, 2012a: 3), returns to it in third turn, to confirm that that was indeed what she was doing in first turn. Thus, in the following sequence Jen’s ‘Oh receipt’ of a ‘first-person’ informational account works to confirm that indeed, her prior report of a ‘hearsay account’ was a request for information and/or confirmation from the first-hand knower she is speaking to now. Admittedly, the set-up is not easy, but hopefully the mat-erial shows it clearly: (5) [Heritage, 2012a: 10, exhibit 10] [Rah:12:4:ST] [arrows are in original text] 1 Jen: 2 3 Ida: Yes I’m s’pose to be tihmorrow yes, 4 Jen: 5 Ida: [Yeh,
The commentary observes, Jenny’s declaratively framed utterance references information that is in her recipient’s epistemic domain and is treated as a request for confirmation (line 3). Here it can again be noted that Jenny’s change of state (K− → K+) oh-receipt confirms by implication that her original declarative was indeed a question in search of information. (Heritage, 2012a: 10)
The suggestion of a ‘proof procedure’ is that for having produced her ‘Oh-receipt’ in line 4, we have proof ‘by implication’ that Jen’s first turn ‘was indeed a question in search of information’, or alternatively a request for confirmation of the prior account. We now have grounds for identifying the action formed in Jen’s first turn. But if so, and by either account of the first turn’s action – a search for information or a request for confirmation – it would be a curious procedure. The relevant ‘proof’ of a ‘question in search of information/confirmation’ in Jen’s first turn would seem to be Ida’s next turn in line 3. This is the ‘next turn procedure’ produced by and for the parties, such that if something else were being solicited by Jen in first turn, the misunderstanding could be revealed.
But we seem to have here instead a third-turn procedure. After Ida’s next turn, Jen confirms what her first was with her ‘Oh’ receipt in line 4. We have then, it seems, a ‘second proof’. But what need has Jen for a ‘confirmation’ at all; it was her question. And if the confirmation were for Ida – a confirmation in third turn of the hearing first evidenced in her next-turn reply – what need has Ida? Or, if Ida has need for a confirmation of the hearing evidenced in her next turn, why would this third-turn procedure be provincial to the action formation of ‘questions in search of information’? That is, why would not every question, compliment, complaint, assessment, agreement, and the full panoply of action formations not benefit from third-turn confirmations too? But then, absent evidence of the parties’ need for this procedure, would not a more quotidian understanding of this expression on this occasion (‘Oh’) save us the difficulties that follow? 11 There is no evidence that natural conversation could actually ever go on this way, in a regime of third-turn confirmations. But perhaps, alternatively, this is not a procedure for securing the understandings of the parties; those understandings are routinely in hand through the on-going work of next turns. Perhaps it is instead a procedure for detailing the analytic formations of the EP, an interpretative textual grammar, rather than one for the production of understanding in situ.
For CA, it is from the parties’ understandings that the analyst takes the measure of her own. Those understandings are not the only resource, but they are a central resource. And in the absence of the parties’ local, demonstrable orientations, the EP must then have some regular or stable procedure for disciplining the over-hearing analyst in how we shall hear the particulars of particles, their placements, indexing, markedness, upgrading, inappositeness, and the rest. The question becomes, what procedure is this? In a register both serious and playful, we want to sum our remarks on these exhibits and their Epistemic treatments through Sacks’ early work on membership categorization, and his formulation of a ‘Hearer’s Maxim’: ‘if two or more categories are used to categorize two or more members of some population, and those categories can be heard as categories from the same collection, then: hear them that way’ (Sacks, 1992: 221).
Borrowing on the expression, we want to suggest that when the EP turns to actual cases, we find an exercise that may be called an ‘Over-Hearer’s Maxim’. It runs roughly as follows: ‘if the analyst can hear an agonistic struggle, if she can hear a contest of informational or experiential possessions, or redundancy, or epistemic authority, or subordination, hear it that way’.
Both maxims underwrite interpretative degrees of freedom. For the ‘hearer’s maxim’, a vernacular practice of hearing categorical references, there are no particular rules to observe beyond the categorical boundaries in play. The membership categories must be heard and used sensibly, as any competent member would hear them. The rules lie in a grammar of apposite usage.
With respect to an over-hearer’s maxim, however, there do seem to be rule-resources attached, both formal and permissive, authorizing hearings that take the measure of things like a speaker’s independence, status, access, authority, subordination, and the like, and we have a particular formulation in mind that suggests the interpretative degrees of freedom involved. We find it in Heritage (2002), regarding Oh-prefaced second assessments: In sum, oh-prefacing in the context of agreements is a method persons use to index the independence of their access and/or judgment in relation to the state of affairs under evaluation … Thus, the basic claim here is that oh-prefacing, in and of itself, indexes epistemic independence: an independence that may or may not be elaborated by other elements of the turn that follows. The indexing is inexplicit, marked and optional. (p. 204)
There are at least two striking things about this proposal. The passage treats ‘Oh-prefacing in the context of agreements’ as an action that ‘in and of itself, indexes epistemic independence’. There are many things described and developed in sequential analysis. Many of them are natural language objects that any competent speaker might recognize, things like questions and answers, requests and refusals. There are also latches, hitches, positions, and technical noticings, and surmises made of still other production features. But ‘indexing epistemic independence’ seems an entirely different kind of noticing, and action. It seems to require an ‘elsewhere’ for the recognizability of the action, given that it ‘may or may not be elaborated by other elements of the turn that follows’, and especially so if those elements include the demonstrable orientations of the parties. ‘Indexing independence’ seems to stand wide of them. In this way it is a furtive thing, and as we may or may not find its evidence in the on-going turn and sequence production, it now seems that it is indexing that is ‘opaque’ and perhaps ‘occult’, removed from actual sequential organizations and available only sometimes, to certain eyes. ‘Indexing independence’ seems to formulate an action within an analytically constructed vocabulary of motives, and for every such conclusion one can fairly ask ‘has this been shown, and can it be shown or found as a move within the sequential grammar of natural conversation?’ 12
Second, there is a remarkable string of adjectives that concludes this passage: that the indexing work of an Oh-preface is ‘… inexplicit, marked, and optional’. There seem to be two readings that we can have for it: one is that these are adjectives for the ‘indexing’ work of ‘Oh’, and that that work – ‘Oh’s work’ – can be ‘… inexplicit, marked, and optional’. This seems to be the aim, although we then have the task of reconciling the ‘inexplicit’, the ‘marked’, and the ‘optional’. This is a diverse collection of categories (‘markedness’ is a production feature – what are ‘inexplicit’ and ‘optional’?), and even if we permit them, how, on any actual occasion, shall we know which it is, or was, this time? The question leads to a second reading.
The second reading is to see the string of adjectives as resources for developing a professional-analytic discourse, meaning that if the work of ‘indexing’ shows itself in these ways – again, as ‘… inexplicit, marked, and optional’ – then should the reader or the parties to the occasion fail to see or give evidence of it, it is because these are matters that are both ‘inexplicit’ and ‘optional’, and only sometimes ‘marked’; it might be done and not heard (inexplicit), or not done at all (optional). (Perhaps analysts and parties alike can see indexing that is ‘marked’.) This armament of adjectives permits the analyst to find indexing work where no one else can. And that is an extraordinary resource and privilege for the tasks of ‘animating transcript’, as it is an extraordinary retreat from the premise that the order of talk-in-interaction is throughout an orderliness available to the parties so engaged, as of their competence to its productions. The EP would seem to have it that the analyst can see what the parties cannot, and in this we can find a return to the familiar disposition of social science as a formal-analytic exercise (Garfinkel, 2002, passim).
Conclusion
Beyond the several remarks in Heritage (1984, 1998), the central role of information transfer as the traffic of interaction is not much developed for another 10 years. It is then formulated directly in Heritage (2012a, 2012b), although as argument without benefit of materials (see Lynch and Wong, this issue): [H]ow do utterances function as requests for information? How are requests for information as a specific form of social action built and made actionable as such? This is not an idle question. Requests for information are the ultimate paradigm of an adjacency-pair first action (Schegloff, 2007; Stivers and Rossano, 2010) that make response actionable and accountable without delay across many languages … (Heritage, 2012a: 3)
And, as Heritage (2012b) elaborates: The idea that ‘information’ is a key element in communication, motivating and warranting contributions to talk, is hardly a new one. It is a staple of communication theories from Shannon and Weaver (1949) onward, of a wide variety of functional linguistic theories focusing on the given-new distinction …, and many, comparatively diverse, pragmatic theories … dealing with sentence construction and interpretation … In general, however, conversation analysis (CA) stood aside from these trends, despite clear evidence that acknowledging new information as new and thereby enacting the updating of common ground is the first order of business transacted by many ‘sequence closing thirds’ (Schegloff, 2007) and related acknowledgments. (Heritage, 2012b: 31, selected citations omitted)
Perhaps. Yet somehow it seems completely unlikely that Sacks et al. were unaware of the flourishing trade in information studies reported above. It is entirely possible that they ‘stood aside’ for good reason, although what reasons they were are not taken up anywhere in the EP’s discussion.
Schegloff has written more frequently in recent years on the first tasks of making sense of single cases in their constitutive detail and organizations. This passage isn’t quite so recent: Among the most robust traditional anchors for the analysis of language beyond the level of syntax are orientations to information and truth. This position needs to be reconsidered … Especially (but not exclusively) in conversation, talk is constructed and is attended by its recipients for the action or actions it may be doing. Even if we consider only declarative-type utterances … the informativeness or truth of an utterance is, by itself, no warrant or grounds for having uttered it or for having uttered it at a particular juncture in an occasion. There is virtually always an issue (for participants and, accordingly, for professional analysts) of what is getting done by its production in some particular here-and-now. (Schegloff, 1995: 187)
A programmatic critique is quite clear. We want to extend it through a passage from Sacks. We offer it on behalf of the tasks and conceptual commitments that mark what Sacks (1984) referred to as ‘ethnomethodology-conversation analysis’. Those tasks lie very close to the achievements of common understanding and thus worlds-in-common, and the challenge of pursuing their production descriptions on any next occasion.
The passage is found in his lecture on ‘Conveying Information’ (Sacks, 1992), and needs to be read carefully to get his spoken phrasing just right: (P)eople suppose that what we’ve been talking about all along, you know in the way I told it to you, and I suppose that in producing any next thing I say. And without thinking about it, the work I do is to find for any item you say – no matter how grossly it misunderstands what I say – how well it understands what I say. (vol. 2, part III: lect. 2: 184, emphasis added)
His remarks are a talking formulation of long and thoughtful observations about ordinary things and occasions, and our questions are ‘Has the EP need or use for observations like these? Does it, can it, take interest in the praxeologies Sacks is referring to?’ Said differently, are the achievements that Sacks is speaking of, in their taken-for-granted vernacular grammars and organizations, expressions of ‘information possessions and transfers’? And if not, how does the EP stand to CA’s long-standing task and program of unpacking the achievements of common understanding on any actual occasion, while not taking leave of their occasion?
Drew (2012) observes more than once that the Epistemic proposal is a ‘radical’ one, and continues, ‘I find it difficult fully to conceptualize or express succinctly how it is that Heritage’s epistemic engine is so radical and profound a proposal’ (p. 63). We think he is quite right. The question then becomes ‘What kind of radical innovation is this?’. We want to respectfully suggest that it leverages a turn away from the corpus of Garfinkel, Sacks, Schegloff, Jefferson, their colleagues and students, a turn away from the analysis of occasioned productions and the play of temporality and sequence in the reflexive construction of social action.
Said differently, alongside CA’s regard for things like the parties’ ‘demonstrable orientations’, or ‘procedural consequentiality’, or ‘problems of relevance’, all as of ‘cases in their constitutive detail’ (see Schegloff, 1991), perhaps the emblematic heuristic of CA, a question intended to direct us back to the contingent productions we encounter in actual materials, has, by the lights of the Epistemic Turn, been answered in advance. ‘Why that now?’ is the question. The EP recommends epistemic status or stance or ‘balance’ as the answer in every case. But a heuristic so answered in advance no longer has heuristic powers.
The Epistemic turn may well have its rewards. The promise of a grand analytic amalgam is attractive, and the promise of an over-arching analytic consensus on our tasks, but first, on our questions, is something social science has pursued more than once. The path seems familiar in that way, whereby stipulations are required in advance in order to underwrite – in the particulars here – a prevailing structure of invidious, agonistic relations as the driver of conversation itself, its turns, sequences, and contingent productions. As Drew (2012) observes, we would then find an orderliness that is ‘constant, omnipresent, and omnirelevant’ (p. 64). Neither Sacks nor Schegloff, nor Garfinkel before them, seem to have had use for, or faith or interest in, such landscapes.
Epistemics thus writes in advance our analytic interests and our conclusions as well. A stable of concepts, such as ‘status’, indexing, authority and access, information and states, their changes and gradients, and other formal and extraconversational things are in principle available at any next glance. But what would then be required to claim them easily becomes a weakened version of ‘evidence’ (see Drew’s similar concerns). The Epistemic turn strikes us as a radical innovation indeed on the corpus works of Garfinkel, Sacks, Schegloff, Jefferson, their colleagues and students, and thus grounds for a very careful and cautionary reading. The questions we have raised are intended as useful measures for taking the measure of the continuities and departures of these Epistemic Program innovations from the extraordinary corpus of studies of natural language use leveraged by Conversation Analysis.
Footnotes
Declaration of conflicting interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.