Introduction

“What to do next?” Harold Garfinkel has taught us that this question is the practical question par excellence. After his death, such a question remains pertinent to social researchers trying to practice the “alternate, asymmetrical and incommensurable” approach to sociology he has initiated: what is the best way to make our heritage bear fruit? It is not easy to give a positive answer, because there is much disagreement today about EM’s program. Indeed, as Wes Sharrock (2004) wrote: “It is not easy to trace out the course or character of Garfinkel’s ideas since the 1960s, and there is, therefore, room for disagreement about it”. One current matter of disagreement is: should ethnomethodology continue to conceive itself as a contribution to the sociological tradition, even as an alternate approach?

Melvin Pollner’s Worries

In a posthumously published paper, recently edited by Robert M. Emerson and James A. Holstein, Melvin Pollner worries about the “dramatic changes” undergone by ethnomethodology after the “work studies” turn. He applies to Garfinkel’s writing Wittgenstein’s aphorism—“If a lion could talk, we could not understand him” (Wittgenstein 1958: 223)—: “Garfinkel may be something of a Wittgensteinian Lion whose form of life is so different (…) that when he speaks we cannot understand him. The sometimes convoluted, incoherent, and ambiguous writings from this point of view are not incidental or accidental; they reflect the effort to get to the something ‘more, other and different’ from within a disciplinary (and worldly) context which insists upon the very format which Garfinkel aims to problematize: rational accountability. If so, translation to understandable terms is to subvert Garfinkel; it is to translate him back to the very form of life—including its modes of representation (i.e., formal analysis)—which he attempts to avoid” (Pollner 2011: 2).

As a Wittgensteinian Lion, Garfinkel could have been a “dangerous man”. This was the qualification applied to him by Gerald Holton, in his comment on the Pulsar paper in the Toronto Conference on the philosophy of the social sciences (October, 1980): Garfinkel’s “implicit challenge seems to be this: our whole network of ideas on how science works is really useless if we don’t get the description of the basic work-done-on-a-given-night in the first place. We must be grateful that he is not disposed to tell us how to do our part of the work, either as philosophers or as scientists. But I predict his ideas will come back, will suddenly turn up again, in the minds of many of us as we study these encounters between the scientist and the event that has captured his attention. Professor Garfinkel will suddenly stand before us, and make us ask ourselves what it was that the scientist really saw and did” (Holton 1981: 159–161).

Pollner’s and Holton’s assessments are acute. Yet, we have few reasons to think that Garfinkel’s form of life was different from ours. But undoubtedly he was looking at things otherwise than we do when we reason sociologically, as laymen or as professionals; hence he could see things we don’t see. The idea of a Gestalt switch (another Wittgensteinian tool) may more appropriately define Garfinkel’s move. As Wes Sharrock explained many years ago, ethnomethodologists are not looking for different things than are “constructive sociologists”: “They are usually looking at exactly the same things, though in a different light, and from a different angle, and (…) a move somewhat comparable to a “gestalt switch” is required to get from one view point to the other. Like the duck and the rabbit, the faces and the vase, ethnomethodology’s and sociology’s topics are only clearly visible in alternation with each other” (Sharrock 1989: 668). Therefore, in a sense, as an alternate approach, ethnomethodology needs classical sociology’s “losing the phenomenon” to establish its own topics of inquiry, independently of those treated by extant sociologists: it seeks to grasp “the very things which make the phenomena what they are” (Sharrock 1989: 668).

If Pollner’s diagnosis is partly right, especially when he emphasizes Garfinkel’s concern for problematizing “rational accountability” and for avoiding the falsification of experience by treating it conceptually, or by producing “generically theorized accounts of theorized phenomena” (Garfinkel 2002: 166), it is also very harsh and unfair, especially when commenting on the evolution of his work: “Garfinkel’s work can be seen as a series of increasingly refined negations: of Parsons, of sociology, of formal analysis, and ultimately of ethnomethodology itself (as yet another instance of exogenous description). (…) Representation itself is increasingly seen as problematic (…). Relatedly, there are early intimations of avoiding description by, e.g., becoming the phenomenon or providing instructions to reproduce the phenomenon. Later: sociological description is hopelessly inadequate because it is done under the auspices of alien concerns—or because it is ‘representation’” (note 5).Footnote 1

What Pollner’s assessment leaves out is a net continuity in Garfinkel’s use of a phenomenological background, despite the fact that there has been an evolution, even a twist, in such a use—his references being more and more to existential phenomenologists (Heidegger, Gurwitsch or Merleau-Ponty) and less and less to Schütz. Though Garfinkel tried to misread those phenomenologists’ texts in order to reach the Gestalt phenomena in the “phenomenal field,”Footnote 2 those references show how deep was Garfinkel’s adherence (after the Studies) to the main catchwords of existential phenomenology, and to its use of Gestalt psychology: “To restore things their concrete physiognomy and to organisms their individual ways of dealing with the world,” “to frequent the phenomenal field,” “to rediscover the layer of living experience through which other people and things are first given to us, the system ‘Self-others-things’ as it comes into being” are such catchwords, as expressed by Merleau-Ponty. But Garfinkel tried to “praxeologize” them. He explained that phenomenologists’ texts were both relevant and vague, and that it was necessary to engage in inquiries about what they tell, using them as instructions, to be able to grasp what they are about—which is itself a very phenomenological recommendation (the world opens up to us only through our engaged perception and “coping”): “The investigations described in the text must be treated as tutorial problems. (…) The investigations are not optional. You needn’t feel that because I’m telling you about them you need not do them. To see what they are about you are obliged to do them” (Garfinkel 2002: 167f.).

Pollner perceives as so important the changes in ethnomethodological working that he distinguishes two versions of ethnomethodology: an earlier version, “concerned with a diverse array of “everyday” and institutional settings” (EM 1.0), and a more recent version (EM 2.0) which “seeks to make the strange familiar (by becoming an adept practitioner), focuses on the ‘foreground’ matters of interest to practitioners, values presence (haecceity), tries to avoid the distortion of member-analyst differentiation, urges ethnomethodological instructiveness involving hybrid relations with host disciplines, and turns away from radical reflexivity”. From one version to the other, he says, “the thrust of ethnomethodological methodology was virtually reversed”. For example, in “EM 2.0,” the “unique adequacy requirement” expresses the primacy of the competent practitioner’s experience and familiarity, and, from a methodological point of view, the abandonment of an exogenous perspective for an encouragement of immersion in work-site practices, while, according to EM 1.0, the inquirer had “to make the familiar strange”. So, from then on, “studies succeed to the extent they provide access to the rich immediacy of the lived order,” i.e., they “show,” “exhibit,” or “demonstrably reproduce the lived order”.Footnote 3 Such a heightened emphasis on the practitioner’s experience and perspective, and on the haecceity of the immediate circumstances as it is experienced by him, goes together, according to Pollner, with the “dumbing down” of the member, the deterioration of membership and the demotion of members’ identifying capacity, “mastery of the natural language”.

One major point of Pollner’s critique is about the change he perceives in E. M. 2.0’s relation to sociology. While “EM 1.0 engaged and situated itself within sociology, albeit uneasily and with grave misgivings,” and did its “explanatory work on behalf of sociology,” EM 2.0 has breached this allegiance and severed its ties to sociology; in it the social and the sociological grow progressively weaker: “The critique of formal analysis is extended to include any representation or analysis external to immediate haectic presence: Formal analysis (which by implication includes ethnomethodology so long as it is affiliated with the academy) irremediably distorts or misses the order of the phenomenal field. Thus, for example, Garfinkel cautions readers that a full understanding of what they are reading will assuredly be a misunderstanding. They must witness the phenomenon. In place of virtually any form of representation is the promotion of an immediate, unmediated presence in which the analyst seamlessly merges experientially with the phenomenal field. But furthermore, EM 2.0 explicitly disowns ethnomethodology as a sociologically affiliated stance or perspective. There are only hybrid ethnomethodologies, that is, ethnomethodologies entwined with and taking their issues from the host domain with which they are engaged and are instructive to them”. By following this path, ethnomethodology strives, according to Pollner, for its own disappearance “through immersion into the phenomenon”: “becoming the phenomenon” is its new catchword.

It is clear that Pollner’s two targets are Anne Rawls’ interpretation of Garfinkel’s work, and “post-analytical ethnomethodology” (Lynch’s expression). First, Pollner doesn’t believe at all in Rawls’ conviction that Durkheim is ethnomethodology’s ancestor: “The choice of Durkheim as the key foil and fountainhead for ethnomethodology seems somehow odd. It isn’t that connections can’t be made between Durkheim and Garfinkel. But aside from the most fleeting and promiscuous use of “social facts,” Garfinkel doesn’t make them (…). In 1967, Garfinkel acknowledged the authors who had provided him with ‘inexhaustible directives into the world of everyday activities’ (Garfinkel 1967: ix). The list includes Parsons, Schütz, Gurwitsch, and Husserl. Durkheim, however, is conspicuously absent, especially in light of the ways in which Ethnomethodology’s Program is now heavily cast as reflecting the ‘real’ Durkheimian heritage”. Moreover, for Pollner, “Garfinkel’s version of Durkheim seems contrived and eccentric. The reading of ‘concreteness’ and ‘thing’ and the implied version of Durkheim’s claims regarding the social foundations of logic, etc. are not compelling. True, Durkheim discussed the social foundations and origins of logic and the categories of understanding, but the analysis was different from what Garfinkel does or proposes. Husserl’s account of the Galilean turn (…) appears more cogent and relevant. Husserl talks about how these operations are grounded in the life-world. One could imagine that Durkheim might be used to ground Husserl, but the kinds of observations Garfinkel makes are not Durkheimian (…); they are Husserlian” (Pollner 2011). Pollner’s judgment may be quite right on that point.

Pollner’s description of “EM 2.0” is shaped by his grasp of Garfinkel’s and his students’ “studies of work” turn. His understanding is both distorted and perceptive. It is distorted, for it doesn’t appreciate the point of Garfinkel’s focus; it is perceptive where it sees very well the possible developments of certain forms of “post-analytical ethnomethodology”. It misses Garfinkel’s point, for if we want to account both for how the society works and how the social order is made to happen, we have to look not only for the ways people speak and account for what they do, but also look at the ways they do what they do, how they jointly organize their situated activities, or how they deal with their situations and produce their practical intelligibility. For doing that kind of exploration, we have to take care of the details of competent practices, of every situation’s local haecceities and of practitioners’ preoccupation with the appearances of things.Footnote 4 As Sharrock (2004) explained, “If one is interested in how social order is made to happen, how people put their affairs together so that the world of everyday life turns out to be, as it extensively does, much the same today as it was yesterday and will be tomorrow, then one cannot really be satisfied that somehow these things get done”: we have to make that “somehow” into an object of careful and insightful inquiry. The result could be that this kind of study will have no news to offer to disciplinary sociology, which seems to be totally indifferent to that kind of matter. Pollner’s judgment about the significance of the hybridation sought by Garfinkel is also distorted. This is so because he doesn’t seem to appreciate the problematic character of sociological descriptions which don’t “teach anything about themselves” to the practitioners whose practices they are about.

Nevertheless, Pollner’s evaluation is also perceptive. Undeniably, the focus of attention changes in what he calls “EM 2.0”: the object of inquiry is no longer “common sense understandings,” or “practical sociological reasoning” (be it lay reasoning or professional reasoning), or the uses of commonsense knowledge of social structures, but, rather, the situated production of social order through the mundane accomplishment of (concerted) daily activities and the forms of reasoning involved in it (a reasoning “in the wild,” as Livingston put it, which is no longer merely sociological). In the Studies, sociological reasoning was a matter of reasoning “about how the society works,” of establishing “matters of fact” in social settings, or of finding out how “social reality” is grasped “from within” and “from the point of view of an adult member of our society” (Garfinkel). As discursive reasoning, it was thought to pervade common sense understandings of social events and facts, and embedded in people’s conducts of everyday life.

“The title ‘ethnomethodology’ identified the project of capturing the practical ways in which society’s members implement their common sense understandings in courses of sociological reasoning that are embedded in the conduct of their everyday affairs. (…) The medium of Garfinkel’s innovative program was therefore to focus extensively on the ways in which people talk in the course of their activities, the examination of ‘indexical expressions’ in their home environments, with the aim of understanding how discourse conducted on the basis of common sense understanding (and by means of natural language) actually ‘works’; thus, the project could be summarized as studying the rational properties of indexical expressions” (Sharrock 2004).

One can argue that such a focus on sociological reasoning, grasping it through people’s talk in their activities, captures only a small part of that reasoning, or of the cognitive and material operations they do to organize their daily affairs. If such a judgment is right, then one can say that EM 2.0 breaks with a kind of “logocentrism,” or with a kind of “sociocentrism,” present in EM 1.0, and pays more attention to the diversity of practical operations implied in the accomplishment of activities, to other forms of reasoning than the discursive one, as well as to other places than talk and discourse to find out how society works: namely the many “ordinary doings,” with their “phenomenal field” properties.

However, Pollner saw very well some of the risks entailed in EM 2.0: the weakening of the social and the loss of the sociological; the breaking of E. M.’s ties to sociology; and a growing commitment to technological research—for example, in the field of Computer Supported Co-operative Work (CSCW) or in the design of large-scale computer systems. Recent publications from Eric Livingston, sharp as they are, indicate such an evolution.

Livingston’s Goodbye to “Ethnomethodology”

Indeed, Livingston speaks for “a simpler ethnomethodology, one whose origins no longer lie in, and which no longer remains in the orbit of classical sociology” (Livingston 2008b: 843). The task of such simpler ethnomethodology will be to “study, through descriptive analysis of situated, observable, material-specific detail, how people witnessably do things in ways that make what they are doing, for the practical purposes of doing them, what those things are” (Livingston 2008b: 843). Such an ethnomethodology will be a sociology of the witnessable order, which will have nothing to say to the disciplinary sociology, which is “a sociology of the hidden order”.

Here is how Livingston contrasts the two sociologies: “In sociologies of the witnessable order, the concrete and material are favored over the abstract; the idiosyncratic and the particular is honored above the universal and general; the goal (as elusive as it may be) is to obtain descriptive precision rather than produce master narratives and speak with the voice of authority. Studies of the witnessable order are fragmented, disjointed from each other, and eclectic; they don’t aim for the appearance of coherence, continuity, and systematic order made virtues in disciplinary work. As opposed to disciplinary research methods, we have practical techniques and a tinkerer’s craft. The centrality of a material culture and the ordinariness of the phenomenal domain take precedence over the importance of researchers expressing their opinions. Rather than the goals of independent truth and objectivity, there’s a celebration of the ambiguous and suggestive. The archaic and primitive are favored over contemporary intellectual fashion; a criterion of observability replaces a capability for documentation; physical presence to the phenomenal domain is given absolute priority over a current situation of disciplinary inquiry” (Livingston 2011). As it were, Livingston’s move is similar to that made in conversation analysis: both of them suggest that as of now each program can develop in an autonomous way, leave “the orbit of classical sociology” and say goodbye to “ethnomethodology”.

For Livingston, there is still too much theory in EM 2.0. He thinks that its inquiries have still been run under the aegis of ethnomethodological theory. Therefore EM has turned in on itself; it has started telling about itself, putting forward, in the sociological field, the worth of its own approach to social order, and assessing its inquiries in reference to “a totalizing conception of the sociological enterprise”: “In a curious way, ethnomethodology, in its critique of disciplinary sociology and its vision and reconstruction of Parsons’ problem of social order, might be seen as sociology’s last attempt at a totalizing conception of the sociological enterprise” (Livingston 2008b: 861). According to Livingston, the privilege granted to finding out an alternate conception of the sociological topic of social order has converted EM into a subfield of sociology of the hidden order. Hence, a major part of its original program, namely the detailed study of methods and procedures which impart to situated activities their identifying features, has been discarded; what he calls the “characterization problem”—for example: “How are queue members doing what they’re doing to make what they’re doing, then and there, what it is for the practical purposes of doing it” (Livingston 2011)—has been eluded. That is why he wants to say goodbye not only to classical sociology, but also to “ethnomethodology” (note the quotation marks).

Like Pollner, Livingston looks at current state of EM in a very critical way. He thinks that too many studies have strayed a long way from the core project: the analysis of what Garfinkel once called “radical phenomena”. This occurred notably in the ethnomethodological studies of science: failing to gain a technical mastery of the scientific practices he studies, the ethnomethodologist has not been able to reach the details of the work done in situ. Therefore, he has favoured the interviewing of the scientists, the collecting and analysis of accounts. But such accounts don’t render the concrete and detailed practices of the work. One consequence has been that “what a characterization of the identifying detail of a setting might be, how such a characterization might be given, how it might be assessed, and who might make such an assessment had all become problematic issues for ethnomethodological studies” (Livingston 2008b: 841). Another difficulty is that, when one tries to do a “sociology of the witnessable order,” one needs to acquire a peculiar skill: the ability to find “organizational thematics,” through which the lived work of the production of the social order, as a witnessable order, could be discovered. Here is, in Livingston’s terms, a list of such organizational thematics (another name for Garfinkel’s “perspicuous settings”): getting through doors, riding in elevators, waiting for elevators, ordering in a restaurant, shopping in a supermarket, picking and joining a checkout line, moving forward in a queue, arranging items on the checkout conveyer belt, etc.

In his recent book, Ethnographies of Reason, Livingston explains that he seeks « to find the work of a domain of practice that makes that work, for its practitioners, recognizably and identifiably the work of that domain » (Livingston 2008a: 240); or “to rediscover the witnessable world as it’s situated with a local course of action, how the witnessable world consists of social, embodied praxis, and how the properties of things are properties of the practices from within which they arise and to which they are inseparably wedded” (Livingston 2008a: 258). To succeed in this inquiry, the analyst must gain the technical mastery of the practices of the domain: “There’s no way to do this than by ‘going native’” (Livingston 2008a: 243). He must also acquire a practical understanding of those practices, i.e., understand them by doing them. So he has to get involved in their performance: “Rather than extracting ourselves from the technical doing of things, we want to find, by going more deeply into them, what is identifying of domain-specific practice for its practitioner” (Livingston 2008a: 258). “Going more deeply into them ‘means’ [to] find ourselves hopelessly embedded in and concerned with the detailed doing of things and the detail of things from within the doing of them,” and “in the midst of such doings, seeking therein the observability of those doings as the ordinary, practical things that they are for their practitioners” (Livingston 2008a: 258). Above all, he has to acquire the form of reasoning which is a constituent of the domain of mundane expertise he studies.

When Livingston says that the ethnomethodologist has to examine the “self-organizing, local work of engaging in a particular activity for the practical purposes of doing that activity” (Livingston 2008b: 844), he is just following one of Garfinkel’s recommendation: “Get respectful of just how the enterprises get done, just in any actual case” (Garfinkel 2002: 181). However, to take this line of action means, for Livingston, to develop an “ethnomethodology-in-the-small,” to make “small-scale studies” of activities after small-scale studies, in order to show how the social order is produced while avoiding a totalizing conception, the result being “a fractionalized collection of studies,” or a kind of “cabinet of curiosities” which he calls “a cabinet of reasoning”.Footnote 5

Through his ethnographies of reason, he shows very convincingly how different activities and practices involve their own form of reasoning, and their own skills, which unfold from within their accomplishment: “Rather than treating reasoning as an abstract, general, transcendent structure of cognition, we can begin to see reasoning as consisting of domain-specific phenomena distinctive to a domain of expertise” (Livingston 2008b: 859). So, “reasoning in checkers ‘belongs’ to the practices of playing checkers and consists of domain-specific phenomena” (Livingston 2008a: 28).

Livingston thinks that, by doing this kind of “ethnomethodology-in-the-small,” he can show more accurately how the social order is produced. But in what sense is the order locally and endogenously produced in the doing of activities a “social order”? How does that produced order render “the observable substance and reality of the witnessable society”?Footnote 6 Livingston takes up a serious challenge when he says that he wants not to judge beforehand what “the social” is and to discover it in the practitioners’ lived work: “We can no longer presume that we know what ‘the social’ is. We want to discover the social in and as the technical and, for practitioners, recognizable, identifying detail of domain specific skill and reasoning” (Livingston 2008a: 243). Or: “Getting through doors, riding in elevators, waiting for elevators, ordering in a restaurant, shopping in a supermarket, picking and joining a checkout line, moving forward in a queue, and arranging items on the checkout conveyer belt (…) indicate the ways that the social pervades our lives. Rather than treating these phenomena as facts of life, they’re better understood as things that we can discover as features of our lives together and rediscover and examine in increasing detail. For sociologies of the witnessable order, the social can’t be seen as the promised result of a programmatic course of research. Such studies are directed to finding the social, not as a conjectured underlying truth but, in its omnipresence, as the observable substance and reality of the witnessable society” (Livingston 2008a: 210).

Let’s consider one of Livingston’s instance: playing checkers. If we want to account for the social character of that playing, we can’t do that merely by reference rule following by players. According to Livingston’s approach, the players have “to see how moving one checker potentially changes the relationships between all the checkers, how these relationships change dynamically, and how all the checkers can work together to bring about a future state of play” (Livingston 2008a: 249). What is social in those moves? A classical response would say that to play checkers is an institutionalized practice, guided by impersonal ideas, rules and norms which are constituent elements of the game, and playing it requires some cultural know-how, skills and habits proper to that kind of practice. Livingston’s answer is more evasive: the logic of playing checkers is, he says, a natural logic, very different from the formal logic of mathematics; it is specific to crossboard checkers. It is social because it reigns in the social setting and the social circumstances where it arises and is cultivated: “It arises and is cultivated in and as crossboard play. It’s the natural, distinctive, indigenous logic emerging from and sustained by the crossboard checkers” (Livingston 2008a: 252). Livingston resorts then to the idea of a common culture, tied to the mastery of a specific way of reasoning (“Combinatorial reasoning,” which is a kind of visual reasoning): “‘Combinatorial reasoning’ isn’t a requirement of playing checkers; it’s a requirement of playing checkers seriously. It makes checker players recognizable to each other as members of the culture of serious checkers” (Livingston 2008a: 252).

Livingston says something similar when he describes the reasoning proper to a mathematical proof: its social character is embedded in the details of the lived work which is done. Proving something is a social activity, but to know what “social” means here, we have to discover it in the lived work of proving: “In our studies, we take the social, interactional settings where provers are at work proving theorems for other theorem provers and where provers are collaboratively engaged in mathematical discovery work as the primordial settings of mathematical activity. Returning to those settings, we can begin to see that provers are always looking into, inspecting, and seeking to find the pairings of descriptions and the work of proving that makes up a course of proving mathematical theorems. Those that engage in that work are recognized as mathematical theorem provers, and to be a theorem prover is to engage in that work. The work of pairing account and practice is the identifying work of proving theorems and, at the same time, that pairing constitutes the relevant detail of a proof account as an account of the theorem it claims to prove. In that such pairings are produced as accountable descriptions for other provers, we’ve begun to locate the primacy of the social in and as the details of provers’ work” (Livingston 2008a: 257f.).

In fact, the searched-for discovery falls short, for, in Livingston’s descriptions, “social” is always somehow implicitly predefined, and with good reason. Wittgenstein said that it is grammar which “tells what kind of object anything is”; and one doesn’t find by an empirical investigation the grammar of a concept one uses. What one can, possibly, find is the lived phenomenon in the phenomenal field rendered by the sign “social,” or how the social world opens up to the players, as constraints and resources they make exist by their doings, when they engage in play. But what Livingston rediscovers finally seems to be a relatively classical conception of the “social,” a mixture of Weber’s, Goffman’s, and Durkheim’s ones.

For Weber, an activity is social if it takes others’ presence into account, if it is oriented to others or if it is based on the recognition of others (notably as a member of a culture). Livingston says something similar: “[People] are always and already watching, monitoring and orienting to what they’re doing together. They’re constantly attending to and adjusting their distance and pace as they walk together; they’re continually monitoring visual attention, distance, interpersonal space, and bodily orientation when they have conversations” (Livingston 2008a: 212–213). For example, when two persons want to walk through a door, they have to cooperate to order their passing. That can be done in very different ways. But all these are a matter of collaborative, socially structured, work: “The joint production of getting through the door, once seen and entered into, seems inescapable. It’s an ephemeral social production: no sooner do the people pass through the door then the social structuring of their actions is forgotten. But in its production and accomplishment, in the details of its collaborative work, getting through a door together consists of intrinsically social phenomena” (Livingston 2008a: 298). Why?

This kind of explanation is near to Goffman’s conception of the coterminous character of social life and public life: a socially situated activity is one which occurs in a field of mutual perception. The “social” is, as it were, what shapes mutual relationships and reciprocal actions when people are co-present, i.e., in a perceptual range of one another (Goffman 1971). So, one doesn’t act in the same way when one is alone and when one is together with others, in a field of mutual perception. When they are alone within an elevator, people can do what they want and they can do very odd things. When they are with other persons, their conduct is constrained and they shape and control it so as display the appearances of an ordinary elevator, where many people cohabit temporarily, often anonymously: “They show each other that nothing that they’re doing—their proximity, their reason for being in the elevator—has anything to do with anybody else in that elevator. The anonymity of an elevator cohort is a produced anonymity, and it’s produced in an anonymous way as the ordinary looks of an elevator cohort. It’s a social phenomenon” (Livingston 2008a: 208). By underlining this produced character of the displayed anonymity, Livingston’s description becomes however slightly different from Weber’s and Goffman’s.

Livingston, however, following Garfinkel’s final respecifications, has better to offer: the discovery of “Durkheimian social objects” such as queues, conversations, the joint passage through a doorway, or through a four way stop intersection, etc. Such objects are objective and transcendent; they are naturally accountable; they are constraining and they are moral objects. Here we can recognize Durkheim’s criteria of the social. What is new in Garfinkel’s and Livingston’s approach of “social objects” is the idea that “at the same time [those objects] consist entirely of the actions and reasoning of their members: they are produced and maintained, supervised and exhibited as the witnessable work of their collaborative production” (Livingston 2011).

In his investigations of the forms of reasoning involved in the “detailed doing of things” by proficient practitioners, Livingston does not leave much room either for their sociological reasoning or for their common sense understanding of events, situations or mutual actions. It is as if practical intelligibility has been shrunk down to the “characterization problem”. So how people understand the circumstances and conditions of their concerted activities, how they use their commonsense knowledge of social structures, how they coordinate their perception and understanding of everyday life scenes, etc. are not analyst’s questions anymore. His justified insistence on the “centrality of a material culture” and on the necessary technical mastery of the practices of some domain may be one reason for this narrowing or withdrawalFootnote 7—unless he finds that such questions are still theoretical questions inherited from disciplinary sociology! One result is that there remains little room for most of the discoveries reported in Studies in Ethnomethodology, to which he has to say goodbye.

Another problem is that he follows Garfinkel’s late rejection of representation and late analysis of practices of “rendering”. Though Garfinkel’s critique of renderings and of generic theorizing is well grounded from a phenomenological viewpoint, it can lead to an exit from the language game which is ours when we try, in social studies, to say something about the dynamics of social life from an observer’s or an analyst’s perspective. We could use other media than words and concepts to show the phenomena of interest to us, for instance artistic media. But can we play our descriptive language game without using word, concepts and ideas? The best we can do is, as Bergson and James said, to use concepts and conceptions “for purposes of practice and not for purposes of insights,” and among these purposes of practice there will be those of observation and of use of ideas in a controlled inquiry. Such is precisely what the astrophysicists did in the pulsar inquiry: they were looking for something called a “pulsar”; their inquiry was moved by expectations and anticipations, shaped by their mastery of “pulsar” astronomical concept. They used this concept for orienting and organizing the lived-work of their discovery. So one can say that while individuals are involved in some lived-work, language gives also meaning to their practice.

Why to be so Suspicious of Language?

One can surmise that Livingston’s decision to say goodbye to “ethnomethodology,” or to so radically sever his ties with the sociological tradition, is motivated by something that appears to be a distrust of natural language or an undervaluation of the part played by language in practical understanding. To show this, we can start from his conception of the “characterization problem” and from his desire to “discover the social in and as the technical and, for practitioners, recognizable, identifying detail of domains specific skill and reasoning”. Livingston is certainly right when he says that an activity is what it is because of how it is done, “for the practical purpose of doing it,” or when he holds that “social” is only a linguistic sign rendering a phenomenon we have to discover in the life world. But for being able to understand which activity is done, what kind of relation is involved between people taking part in it, etc., one must use criteria for identifying actions, events, relations for what they are, and for producing their practical intelligibility. Now those criteria are to be found, firstly, in the system of practices, concepts and ideas of a form of life, in the ways of doing, thinking and talking embodied in the institutions which “govern the way the members of the societies (…) behave” (Winch 1958: 127), or in what Garfinkel called the “institutionalized features of the collectivity”. So practical understanding is not restricted to the practitioner, despite the fact that his relies on a very specific intelligibility: that which he has when he is acting—it is an understanding-for-doing, which arises from the midst of doing. It is locally and temporally produced. But his practical understanding is also larger, and, as we’ll see, language is central to it.

Undeniably, practical intelligibility is to a large extent scenic or visual. Lena Jayyusi explains that very well. One can see an action or an event in a scene for what it is, without interpretation; it is glance-available or glance-intelligible, because there is a “scenic transparence of the social world”: “The world is intelligibly available to our looking and seeing in specifiable ways, and its intelligibility is scenically organized. The scenic organisation and constitution is locatable in our category knowledge of persons, places, actions, and objects, and the ties between them. Indeed this category knowledge is itself a perceptual and visual knowledge” (Jayyusi 1993). That is why one doesn’t need to interpret the scenes in order to identify them; we just contextualize them, i.e., we grasp them in some naturally-occurring context: “As we encounter actual scenes in the real world, their contexts are available to us and are reproduced in the way that we treat scenes as intelligible”.

So, we can extend to many situations what Jayyusi says about photographs: “Where photographs are intelligible (and for the most part they are), it is routinely a contextualisation that is intended when we attend to the enframed scene, not an interpretation: we are looking to locate the ‘fragment,’ the ‘excerpt’ we see within a context that ‘belongs’ to it, or to which it ‘belonged’: the purposes, relevances and outcomes of the in-frame scene/action for example. Contextualisation is distinct from ‘interpretation’. What we do ‘interpret,’ more properly speaking, is the photographer’s, the exhibitor’s, the news editor’s purposes in using this photograph. In practice it is often the photographic object-in-use which is the locus of the query after ‘meaning’ or ‘signification’—yet that is often read back into the constitution of the image itself” (Jayyusi 1993).

From such a distinction between contextualization and interpretation, we are led to amend Garfinkel’s expression, when he says that members “use the institutionalized features of the collectivity as a scheme of interpretation” (Garfinkel 1967: 94). To speak of “interpretation” may be inappropriate, if we consider that to interpret is an activity: we interpret only when something is ambiguous, and when we interpret we form a hypothesis or a conjecture, which can be false. Most often perception and understanding are direct and immediate, and what takes place is more contextualization than interpretation.

The visual intelligibility of events, actions, persons and objects, is expected by members when they get involved in their activities and interactions. On the other hand, it is based on the perceptual availability of normality of courses of action. Such a perceptual availability is not banal. It has both a scenic and a conceptual component (see Coulter and Parsons 1990; Coulter and Sharrock 1998). Although Garfinkel has often underlined the scenic character of the social order, one can think that he was inclined, when influenced by Schütz, to intellectualize too much the cognitive dimension by using the idea of a scheme of interpretation—as if perceptions were mediated by intellectual operations of interpretation and subsumption.Footnote 8

Yet, his descriptions of how people use their common sense knowledge of social structures are still valuable ones. He explains that members not only treat social structures as “actually or potentially known in common”; they refer also to them, in an engaged mode, as “normatively valued social structures which the subject accepted as conditions that his decisions (…) had to satisfy. The social structures consisted of normative features of the social system seen from within which, for the subject, were definitive of his memberships in the various collectivities that were referred to” (Garfinkel 1967: 93).

If we follow Jayyusi’s perspective, such a reference produces the “scenic transparency of the social world”. And as the “natural facts of life” are “through and through moral facts of life” (Garfinkel), morality is part of the scenic transparency of the social world. For moral values or moral indices are witnessably embedded in most practices. Analyzing a photograph showing police beating a woman during a demonstration, Jayyusi explains that “there is a ‘moral order’ to vision, and that this ‘moral order of vision’ clearly partakes of, and is a constituent feature of the moral order at large. If you see a policeman beating a man vigorously, and you have other witnesses to the beating (in other words, if your veracity is not at issue) then that is perceivable and describable as just ‘what happened,’ and any judgment that does not incorporate this as fact may be accountably treatable as not ‘objective’. Of course, people may differ as to what the beating constituted—brutality or self-defence, etc., how it might be redescribed, what may or may not have justified it in terms of motivations, expectations, fears, threats, prior acts, possible consequences, etc. But all these will turn on the fact of the beating, and that fact is given in its witnessed and witnessable character or witnessed and witnessable indices. (…) Scenes, therefore, that constitute themselves as embodiments of certain kinds of morally implicative and morally articulable courses of action, are themselves morally implicative: they present themselves to our eyes in morally constituted ways: they are intelligible from within the moral order. But, as with all matters from within the cultural/moral order, there is no fixity, no logically determinable manner in which they may get taken up, judged or oriented to. This is where the notion of implicativeness is important. Vision then is ‘morally constituted’—it has a moral order” (Jayyusi 1993).

Like Livingston, Jayyusi views ethnomethodology as a “sociology of the witnessable order”. Her description of the perceptual availability of the social order echoes with what Livingston says about the visual character of “reasoning in the wild”. Yet there are significant differences between their approaches. On the one hand, for Jayyusi, the glance-intelligibility of the situations is rooted in the members’ “category knowledge of persons, places, actions, and objects and the ties between them,” this background knowledge been itself both “a perceptual and visual” one, and a cultural-linguistic one—hence it is not tightly tied to the technical mastery of a domain of practice. On the other hand, for sure, one can distinguish between scenes within which one is involved as a participant, and scenes which are given to one only as a witness; nevertheless each one of these is glance-available and glance-intelligible in its unfolding: “Clearly, even it is not given to you from within its endogenously developing order, as part of relevances of and activities within which you are ongoingly enmeshed as a participant, but rather given you as only a witness, or merely a passer-by, you may be able to hang around long enough (say at the scene of an accident, or a commotion with police cars arriving, etc.) to ‘see’ and ‘find out’ what happened, and to observe further happening and developments” (Jayyusi 1993).

So we can assume different perspectives in the environment of scenes we live in, those scenes being constituted by “a manifold of simultaneously ongoing unfolding activities”: some of them “are co-oriented to each other, and all of [them] may, in significant ways that are unspecifiable in advance, intersect, converge and co-constitute each other”. We can also take up different orientations—for example an orientation as to how some scenes might impinge on us, or as to how they might develop (out of curiosity, concern, anxiety, etc.). Particularly, we can take up different visual orientations, our looking and seeing being always organized by our relevances and our knowledge: “Clearly, there are many things that we may look at but not ‘see’; things that we ‘see’ but whose details we do not ‘notice,’ and things that we see or even take minute note of but do not engage” (Jayyusi 1993: 5). So the kind of visual availability or perceptual intelligibility of a scene depends also on the visual orientation taken up in the situation.

Is there then any serious reason to favour the practitioner’s perspective? It goes without saying that when one is involved in the accomplishment of an activity, such visual orientations are distributed otherwise than when one is merely witnessing various courses of embodied activities. But the practitioner’s perspective isn’t one. Think once more of playing checkers. The skills displayed have to be appropriate to the various aspects of the circumstances. So, nobody would understand that a proficient player behaves in the same way when he is playing in a tournament, in a training session with a friend of his club or in a light session with his young daughter in which he teaches her to play checkers. In each case his activity is formally the same; but it is not qualitatively the same, because the setting is different. So, in each case, he doesn’t exercise the same skills, look in the same way at the checkers on the board, work out his moves in the same way, see the same things, etc.

Livingston’s phenomenological argument seems to be that the social order from within which things and events are perceptually intelligible is only the order which governs the situated accomplishment of activities: the perceptual availability of the world is not so much a matter of “visual availability of categories within it” (Jayyusi), as of “embodied praxis,” so that “the properties of things are properties of the practices from within which they arise and to which they are inseparably wedded” (Livingston 2008a: 258).

The social order is thus tightly tied to the “reflexive settings” of the activities. The paradigmatic example of such settings is the “formatted queue,” which is both the product of a concerted organizational lived-work, and a transcendent object, which guides and constrains behavior: “There’s nothing there but the practical actions and reasoning of the queue members making the queue into an object seemingly external from them” (Livingston 2011: 4). Another example of reflexive setting is the turn-structure at a road intersection: “As drivers find and take their turns among the other turns crossing the intersection, they exhibit the structure of turns that other drivers, as they approach the intersection, look for and use to produce and exhibit the structure of their turns. Yet the structure they’re producing (…), while observably their own work and their achievement, is seen by drivers as an objective fact, witnessable by anyone who can drive and see what is there to be seen. It appears as the objective order of passage at that intersection. In summary, we might say that drivers drive in ways that produce and maintain the conditions that allow them to drive in those ways” (Livingston 2008a, b: 204).

Livingston’s descriptions are very persuasive. Nonetheless, one can say that glance-intelligibility for the situated doing is only one part of the overall perceptual intelligibility of actions, that practical understanding is larger than the intelligibility available to the practitioner when he acts, and that the social order doesn’t amount to the only order produced by practitioners in a “reflexive setting”. Why? One reason was given by Wittgenstein, in the way he answered the question: how could one describe the way men act? Here is his response: “Surely only by showing the actions of a variety of humans, as they are all mixed up to together. Not what one man is doing now, but the whole hurly-burly, is the background against which we see an action, and it determines our judgment, our concepts and our reactions” (Wittgenstein 1980a: § 629). Note that the question is asked from the point of view of a witness, but such a witness is not seeing or judging action in a disengaged mode; he is a committed witness, and he can rely on his practical understanding of the scene (One can add that such an understanding doesn’t carry with it the ability to execute specialized practices of a specific domain).

In the following aphorism, Wittgenstein ask a similar question: “How could you explain the meaning of ‘simulating a pain,’ ‘acting as if in pain’? (…) Should you act it out? And why could such an exhibition be so easily misunderstood? One is inclined to say: ‘Just live among us for a while and then you’ll come to understand’” (Wittgenstein 1980a: § 630). If such an understanding arises from a familiarity with the ways “we” behave (i.e., “we” act, talk and think), there is nothing to suggest that it requires mainly assuming the practitioner’s point of view. What is required is not a practical immersion in specific domains, but participation in “the whole hurly-burly” of a form of life.

In the Studies Garfinkel wrote that “a society’s members encounter and know the moral order as perceivedly normal courses of action—familiar scenes of everyday affairs, the world of daily life known in common with others and with others taken for granted” (Garfinkel 1967: 35). Farther, he explained that members’ “grasp of and subscription to the ‘natural facts of life in society’” comprise their commitment to motivated compliance with « a legitimate order of beliefs about life in society seen ‘from within’ the society” (Garfinkel 1967: 54). So one can say that the social order is partly available through “legitimate beliefs” about life in society, or about what is “perceivedly normal” [expected attitudes, appearances, affiliations, style of life, etc. with which persons “are held to compliance (…) regardless of their desires” (Garfinkel 1967: 125)]. But Garfinkel says neither that the social order is a system of beliefs, nor that the subscription to the “natural facts of life in society” occurs only when persons are doing some activity. For it would be surprising that such would be the case, since members can also think and talk about the social order, and, when they do so, they do it not only in an engaged mode, but also with linguistic means.

We can formulate the point in Charles Taylor’s words: “The understanding of what we’re doing right now (without which we couldn’t be doing this action)” implies a wider grasp of our whole situation, through the ways we imagine our social existence, our social surroundings and the moral order, through how we imagine belonging to different wholes, through the ways we imagine how we stand to each other, how we fit together with others, how things go on, and how they ought to go on, between us and our fellows, and through “the expectations that are normally met, and the deeper normative notions and images that underlie these expectations” (Taylor 2004: 22). All those things constitute the “legitimate order of beliefs about life in society seen ‘from within’ the society” (Garfinkel).

Taylor adds three points. The first is a reciprocal relation between practice and this background understanding: “If the understanding makes the practice possible, it is also true that it is the practice that largely carries the understanding” (Taylor 2004: 25). The second is that to make the sense they do to the participants, the practices must have certain ideas “internal to them,” ideas which are nobody’s (Winch said something similar). And the third is that those ideas are formed in and through a language—hence language contributes to practical understanding. In a recent (not wholly convincing) paper, in which he distinguishes an “interactional expertise” from a “contributory expertise,” Harry Collins wrote something similar: the meaning of an activity, “and therefore the way we conceive of it, and practice it, is formed by language”. So, practice alone isn’t “a sufficient explanation for understanding practice and for the acquisition of practice skills. (…) A practice can never be learned from someone else in the absence of shared language” (Collins 2011: 279).

Garfinkel wrote that members identify their practical circumstances by using “background expectancies” and “institutionalized features of the collectivity” as a “scheme of interpretation”. That means that the social order, as it is known in a commonsense manner, either through “subscription to the natural facts of life in society,” or through “knowledge of socially organized environments of concerted actions,” operates within perception and cognition to produce an organization, and to make things appear with meaning: “With their use [the use of ‘background expectancies’ and ‘institutionalized features of the collectivity’], actual appearances are for [the member of society] recognizable and intelligible as the appearances of familiar events” (Garfinkel 1967: 36). Garfinkel said there something similar to what Peter Winch explained in The Idea of a Social Science: ‘When one is dealing with intellectual (or, indeed, any kind of social) ‘things’ (…) their being intellectual or social, as opposed to physical, in character depends entirely on their belonging in a certain way to a system of ideas or mode of living. It is only by reference to the criteria governing that system of ideas or mode of life that they have any existence as intellectual or social events. It follows that if the sociological investigator wants to regard them as social events (as, ex hypothesis, he must), he has to take seriously the criteria which are applied for distinguishing ‘different’ kinds of actions and identifying the ‘same’ kinds of actions within the way of life he is studying. It is not open to him arbitrarily to impose his own standards from without. In so far as he does so, the events he is studying lose altogether their character as social events’ (Winch 1958: 108)”. Studies in Ethnomethodology is a very perceptive analysis, with a Durkheimian flavor, of such ‘a system of ideas or mode of living’ (“natural facts of life as a morality” and mores), grasped “from the point of view of the collectivity member’s interests in the management of his practical affairs” (Garfinkel 1967: 76).

There are Many “Action Regimes”

Finally, there can be some more problems in the twist Garfinkel gave to his investigations after the Studies. One of them can be seen in Livingston’s analysis, which followed Garfinkel’s lead. In his Ethnographies of Reason, Livingston identifies more or less with existential phenomenologists. Like them, he emphasizes embodied coping, non conceptual perception, visual reasoning, direct responses to affordances, motor intentionality, and he conceives perceptual receptivity as a skilled accomplishment. Like them, he thinks that for understanding the practices of a specific domain, we must get a practical immersion in it. But, at the same time, he leaves aside one aspect accounted for by those phenomenologists: the capacity to transform our unthinking and transparent “coping with affordances” into an “explicit coping with objects” (Dreyfus 2006: 48). Hubert Dreyfus, who speaks in behalf of those phenomenologists, refers to Heidegger’s remark that “we have skills that enable us, step by step, to transform our perception of affordances into the perception of context-free objects, and the content of our skilled responses to perceived whole patterns into articulable conceptual content” (which implies linguistic skills). And he recalls Heidegger’s comment “that when there is a problem with an affordance we can change our relation to it. For example, when hammering is going well, the hammer is not what I focus on. The hammer simply affords hammering; the less I perceive it the better. If, however, the hammering is unusually difficult, I may experience the hammer as having the situational aspect of being too heavy under these conditions. And should things go even more badly so that I have to abandon my activity, the hammer may appear as an object that has the context-free property of weighing five pounds” (Dreyfus 2006: 48). So, Dreyfus concludes, “analytic attention brings about a radical transformation of the affordances given to absorbed coping. Only then can we have an experience of objects with properties, about which we can form beliefs, make judgments and justify inferences. At the same time, however, this transformation covers up the non conceptual perception and coping that made our openness to the world possible in the first place”.

Garfinkel and Livingston emphasize the covering up done by such a radical transformation, and describe it, perspicuously, with their “rendering theorems”: when the lived, order-productive, work is rendered through discursive accounts (collections of signs), it is lost, just as “social objects and their Gestalt contextures” are then left out; and it is impossible to recover any of them by improving the practices of linguistic rendering. They are certainly right. But we have here a kind of dramatization of existential phenomenologists’ perspicuous analysis. And, if we follow the argument, we are led to think that the reflexive, linguistically mediated, inquiries into which beliefs are formed, practical judgments made and inferences justified, are outside practice. Indeed such inquiries transform the action mode or “regime”. However we can’t say that when practices are guided or shaped by them, they aren’t genuine actions. What is true is that, when we act, notably when we act reflexively, we can never do without some unthinking and transparent bodily coping with the situation.

We should be wary not to dissolve the internal tension we meet in members’ relations to their own activities and to their social environment, a tension which appears in the fact that, when they account for what they do, they have no alternative than to cover up their “concerted organisational lived-work”. In his chapter on jurors’ activities, Garfinkel noted that one of the main preoccupation of persons, when the outcome of what they did was in hand, was to give their decisions and courses of action some order which justified them, or to give them their “officialness,” to assign “outcomes their legitimate history”. When they did so, they couldn’t but hide (unwittingly) the lived-world of the production of such outcomes. Something similar occurred in the pulsar case, where the official account reported an inquiry supposedly controlled from the beginning by the certainty that such an object, conceptually defined, existed somewhere and had discoverable properties, while the lived-work done by the astrophysicists during their night’s work was fraught with contingencies and uncertainties, indexicalities and incongruities, which didn’t appear in the account. Such a tension can be seen as one occurring between the contradictory requirements of the two registers in which the social order is produced and rendered manifest, which are also the two registers of accountability.

There could be another problem in carrying on Garfinkel’s concern for Gestalt phenomena. It needs also a Wittgensteinian twist for avoiding undesirable consequences. Most certainly a central aspect of the structuration of situations and activities is the organization of their details into Gelstalt-contextures (Gurwitsch 1964). When we perceive a Gestalt, we grasp more than mere outlines and shapes; we perceive an ordered and differentiated whole, or an arrangement of parts. As a form, a Gestalt is a unity of order: it holds together and orders the differentiated parts of a whole. In a Gestalt, as Gurwitsch (1964: 134–135) said, there is an interdependence and an interdetermination of the parts, due to the functional significance of each of them: “the constituents may be said to exist through each other”. But a Gestalt is also, and mainly, a unity of meaning, so that what we perceive are meaningful objects or events, scenes or situations. So, when one defines a Gestalt mainly as “a system of functional significances which all complement, and fit with, one another” (Gurwitsch), one can be inclined to insert the grasping of an organized whole as an intermediary operation in the perception of a meaningful thing or scene.

Wittgenstein vigorously criticized such an interpolation, as we can see in his Remarks on the Philosophy of Psychology, when he questioned this aspect of Gestalt psychology: “It is, contrary to Köhler,—precisely a meaning (eine Bedeutung) that I see” (Wittgenstein 1980b: § 869). Farther, he continues: “Isn’t what Köhler says roughly: ‘One couldn’t take something for this or that if one couldn’t see it as this or that’? Does a child start by seeing something this way or that, before it learns to take it for this or that? Does it first learn to answer the question, ‘How do you see that?’ and only later ‘What is that’?—Can we say it must be capable of grasping the chair visually as a whole, in order to be able to recognize it as a thing?—Do I grasp that chair visually as a thing, and which of my reactions shews this? Which of a man’s reactions shew that he recognizes something as a thing, and which, that he sees something as a whole, thingishly (als ein Ganzes, dinglich)?” (Wittgenstein 1980b: §§ 977–978).

So, the ability to identify or to recognize what is occurring for what it is requires the ability to see in the material details of concrete achievements “more than what one sees,” namely, a meaningful event or action, meaning entering directly into the particular event or action to qualify it. Perception doesn’t comprise any intermediary operation: we don’t first grasp visually something as an ordered whole, and, second, recognize it as such and such meaningful object.

Conclusion

Undeniably, Garfinkel and “post-analytical” ethnomethodology have opened new relevant perspectives for the study of social order and social action. In particular they have shown, in a new and rich way, how social settings are “reflexive settings” in which people produce locally the normative milieu in which their practical actions and reasoning take place, this milieu being made objective and transcendent, while it consists of nothing other than those situated practical actions and reasoning. They have also shown how talk, lay or professional, about social order is structured by “renderins theorems,” which cover up the radical phenomena which develop in the “phenomenal field,” phenomena which need to be observed and described for themselves. But one can also think that such an innovation might make irrelevant part of the horizons opened up for inquiry in the Studies. Why? Mainly for two reasons: the first is that practical intelligibility is now grasped from the only viewpoint of the practitioner, as he or she is involved in the doing of his activities, and the second is that there is little room, in the new concerns, for the part played by language and “legitimate beliefs” in the practical understanding which makes possible the accomplishment of practices. So, in this respect, Pollner’s diagnosis seems right: there is, after the “work studies” turn, a kind of demotion of members’ identifying capacity, “mastery of the natural language”. The challenge for us in the future is to hold together Garfinkel’s former and later “policies,” and to pluralise the approach so as not to eliminate the intractable tension noted above between different “action regimes” and between different registers of accountability.Footnote 9