Ethnomethodological ethnography: Historical,conceptual,and methodological foundations

Abstract

This text discusses the relationship between ethnomethodology and ethnography and sketches what can be called an ethnomethodological ethnography. To do so, it shows that Garfinkel and his collaborators work ethnographically in order to adequately describe social phenomena and make their phenomenal field properties noticeable. To highlight the distinctive features of ethnomethodological ethnography, the text first discusses other ethnographic approaches. Differences between these approaches and the ethnomethodological ethnography become apparent through two argumentative steps: first, by discussing Garfinkel’s reflections on ethnomethodology and ethnography, and second, by discussing actual ethnographies by Garfinkel's collaborators. The text concludes with a general reflection on the methodological principles of ethnomethodological ethnography.

Introduction

The writings of Harold Garfinkel, the founder of ethnomethodology, deliberately focus on the methodological and practical foundations of sociology including ethnographic methodology. Within them, he formulates his own approach to ethnography, which we shall refer to as ethnomethodological ethnography, although Garfinkel himself refers to ethnomethodological studies. ¹

The starting point for Garfinkel’s reflections on the ethnographic method in the 1950s was an epistemological and methodological problem diagnosed by him within contemporary sociology. From our current perspective, this problem could be construed as a “crisis of representation” avant la lettre, considering the parallels between it and the subsequent anthropological “crisis of representation” in the 1980s (cf. Clifford and Marcus, 1986). ² The core issue with sociology is that the discipline itself forms part of the object of the discipline. This in turn implies that a method of study borrowed from natural science, which identifies regularities within the objects of study through external analysis—regularities that are not accessible to the objects studied—would be methodologically inadequate for sociology. Garfinkel draws on a principle endorsed by Alfred Schütz, 1953 according to which sociology generates second degree observations from first degree observations of the object of study, that is, the members of society. From this starting point, he then seeks to achieve a theory of science position that allows for congruence between theory and method on the one side and between method and object on the other side. This constitutes the basis for ethnomethodological ethnography: it must originate from the social phenomena that are to be inquired into.

With this text, we intend to position the ethnomethodological variant of ethnography within the spectrum of ethnographic methodologies. So far, this variant has hardly been noticed and has not been routinely presented in method books (cf. as exceptions Pollner and Emerson, 2001; Travers, 2001: 62–82; Button et al., 2015; Meier zu Verl et al., 2020). In our view, ethnomethodological ethnography is able to re-present and re-enact the orderliness of social reality. We therefore position it as an alternative to other variants of ethnography. As we will elaborate, these other variants methodologically frame the process of writing as a process of ordering in the first place.

In order to understand Garfinkel’s position, we will first discuss three other variants of ethnography. These variants are part of the mainstream methodological discourse and are also presented in current handbooks on ethnography. Subsequently, we present the theoretical foundations and historical development of ethnomethodological ethnography. We then go into the practical implementation of ethnomethodological ethnography on the basis of ethnographies made by Garfinkel’s collaborators. In a final step, we present the methodological principles of ethnomethodological ethnography.

Variants of ethnography

In order to contrast ethnomethodological ethnography, we distinguish three other variants of ethnography that we call here, roughly and in necessarily reduced way, naturalistic ethnography, textual ethnography, and bodily ethnography. This distinction does not claim to be exhaustive but serves as a heuristic contrast that allows us to better understand what we call ethnomethodological ethnography. ³

Naturalistic ethnography

The initial phase of ethnography was characterized by a naturalistic attitude towards the object of study: natural, social, and cultural worlds were seen as one and were to be described as they presented themselves to the external observer (cf. Stocking, 1992: 12–59). An attitude that established an epistemological difference between observer and observed largely persisted in both early founding figures of anthropological (Malinowski and Boas among others) and sociological ethnography (Jahoda’s and Lazarsfeld’s “sociography,” the Chicago School’s “field study”). They equally took a realist notion of description, albeit a cautious one, as the basis of their research (Malinowski, 1922: 18; Boas, 1920; Palmer, 1928: 7; Jahoda et al., 1971: 9). However, both the forms of documentation and the resources they each used to interpret observations varied. Boas preferred verbatim transcripts or technical documentation of expressive utterances. He advocated recording discourse data that was as uninfluenced as possible (narratives, chants, etc.) and working with key informants to describe history and cultural life in the researcher’s own words. He also chose public reenactment as a method of representing culture (Hinsley and Holm, 1976). Malinowski (1922: 18) recommended participation experiences during a long-term research stay unburdened by purposive motives (trade, mission, etc.), as well as language skills, so that the researched society could be described from “its own perspective” (ibid. 25) and in its own contexts of meaning (ibid. 19). The consequence of this, however, was that the ethnographer’s personality had a greater role to play than in other methods (ibid. 20), and thus Malinowski suspects in his diary, “it is I who will describe them or create them” (1967: 140). The Chicago School particularly saw familiarity with the field as an advantage of sociological ethnography (Palmer, 1928: 163), and the Austrian Marienthal study advocated “immersion into the situation” (Jahoda et al., 1971: 1) and assuming practical socio-political responsibility in the researched community (ibid. 5–6).

In these skeptical statements, methodological changes are announced, and subsequently a shift away from description and towards interpretation as the central ethnographic task took place.

Textual ethnography

Textual ethnography accomplishes this transformation, drafting culture as text. Geertz’s ethnographic project of “thick description” (Geertz, 1973: 3–30) consists in reading and writing down the eloquently pre-formulated text of culture (ibid. 452). The cultural text is pre-formulated because it is meant and pre-interpreted by the actors in the Weberian sense. Cultures or societies therefore consist of a web of interrelated intentional acts. As a student of Parsons, Geertz follows Parsons’ concept of action and focuses on the systemic connection of intentional but at the same time normatively oriented social webs of action (cf. Rawls and Turowetz, 2021). As an anthropologist (and student of Kluckhohn) he is mainly focused on culture as the system of meanings and symbols that normatively regulates social actions, thereby ultimately ensuring the latent maintenance and continuity of society (cf. esp. ibid. 17, 27–28, 44, 351).

“Thick description” means, first and foremost, inscribing the webs of intended meanings of observed social actions into the ethnographic descriptions produced. Unfortunately, Geertz gives us little concrete information on how this should be achieved. In no case should ethnographers include only “raw social discourse” in the “thick descriptions” because “we are not actors, we do not have direct access” (ibid. 20). To a small extent, however, ethnographers can gain access through their informants: they need to be questioned regarding the “intended meaning” behind the “raw social discourse.” As Geertz himself puts it: “The trick is to figure out what the devil they think they are up to” (1983: 58).

Behind the apparent disorganized plurality of social action, according to Geertz, the hidden intentional interconnectedness must be identified, which understands a social action or utterance as a response to or a commentary on a preceding action. In the course of their research, ethnographers slowly come to an understanding of the intentions and goals of those being researched (Geertz 1973: 27), which they then systematize (ibid. 15). Their aim is not, for instance, to discover an unfamiliar continent of meaning, but to conjecture meanings in order to formulate explanations (ibid. 20).

In many parts of his work, Geertz vehemently opposes the myth of the empathic ethnographer endowed with special sensitivities, this chameleon-like “walking miracle of empathy, tact, patience, and cosmopolitanism” (1983: 56), who understands the strangers by slipping into their skin and becoming like them (ibid. 9). What Geertz has in mind as ethnographic method is not psychic closeness and transcultural identification through participation, but understanding without empathy. An ethnographer must rather develop—and teaching example for this is Claude Lévi-Strauss—the ability “to penetrate the savage mind” by means of epistemological empathy. This means not to enter into an inner correspondence with one’s informants, but simply to find out their intentions (ibid. 58). Any understanding of what the researched are really like, what their inner life is like, actually comes from the ethnographer’s ability to interpret their expressions and symbols. Ethnography is more akin to grasping a proverb, catching a hint, getting a joke, or reading a poem than to achieving psychic communion (ibid. 70). Overall, then, Geertz believes that embodied participation is of little use in ethnographic research—it is more likely to create distrust (Geertz, 1973: 20)—and ethnographers should confine themselves to case-based “thick descriptions” of intentionally structured social webs of action, which they then analyze and interpret with the aid of theory. The knowledge necessary to identify the intentions and purposes of the researched is methodologically obtained through interviews and conversations with informants. Geertz’s “thick description” has been adopted by many anthropologists and sociologists as a methodological model, but not infrequently without adopting his background in Weber’s and Parsons’ theories of action and social systems.

Further variants of textual ethnography such as postmodern ethnography, which emphasizes either polyphonic and dialogical or literary and fictional aspects of ethnographic writing and is highly reflexive of the ethnographer’s self, are left out here (but see Clifford and Marcus, 1986).

Bodily ethnography

The approach of bodily ethnography differs from Geertz’ skeptical attitude towards the use of the body and its tacit field experiences. Drawing on the naturalistic tradition (Hammersley and Atkinson, 2007: 5–10), bodily ethnography explicitly calls for the use of the researcher’s body as main instrument of knowledge creation. Though, in line with Geertz’s ethnography, writing is characterized as the main ethnographic activity (cf. Lofland and Lofland, 1984; Hirschauer, 2006), what is assumedly written up is not the eloquent text, but, to the contrary, the originally silent social and cultural world of those researched. Only in the medium of writing can these private and tacit observations of social situations be made sociologically utilizable. ⁴

Howard S. Becker and Erving Goffman, among others, stands for this approach of bodily ethnography (“Second Chicago School”), in which ethnographers physically participate in an observed social situation. Becker, who was a student of Robert E. Park, instructed his students to conduct first-hand observation, for which it was important to not only get one’s hands dirty: “gentlemen, go get the seat of your pants dirty in real research” (cited in McKinney, 1966: 71). Goffman (1989: 125) emphasized in his methodological notes that in ethnography data are acquired “by subjecting yourself, your own body and your own personality, and your own social situation, to the set of contingencies that play upon a set of individuals, so that you can physically and ecologically penetrate their circle of response to their social situation, or their work situation, or their ethnic situation.” The participation or, as is sometimes said today, “immersion” that takes place in ethnographic research (Emerson et al., 1995: 2) means both entering with the researched into the many imponderable situations of everyday life and observing how they act, as well as experiencing psychologically and physically oneself their situations and circumstances. Ethnographers should engage themselves with their whole existence in the field. Even more, ethnographers must act as if they cannot return to the comfortable home of the academic at any moment, but as if they are as existentially affected by them as the researched. In addition to pleasant things, ethnographers must therefore especially endure unpleasant things. According to Goffman, this tunes the ethnographer’s body like a musical instrument. Only with a body tuned in this way, he argues, are ethnographers able to meaningfully observe the researched as they act and react in the face of life’s adversities. In particular, ethnographers acquire a capacity for empathy with their bodily participation, for “you’ve been taking the same crap they’ve been taking” (Goffman, 1989: 126). For Goffman, this is the core of ethnography, without which no serious ethnographic work is possible.

From this perspective, it would be a misunderstanding to assume, like naturalistic or textual ethnography, that the social and cultural world or even ethnographic experiential knowledge already pre-exists in data form—naturalistically observable and in direct correspondence describable in the case of naturalistic ethnography or meaningfully pre-structured and descriptively condensable in the case of textual ethnography (cf. Lofland and Lofland, 1984: 71–75). Observation from the bodily ethnography perspective is finding words and writing: a process of articulation, which represents a (re)construction of the object and “sociological artifact” (Hirschauer, 2006: 422). Its quality is determined by disciplinary conditions. Writing means transforming something into another form (ibid. 423, 427), which is primarily a matter of opening up phenomena that are not yet available in linguistic form in the first place, of putting them into words and writing them down.

In regard to the objects of study, this is directed first and foremost at those “events [that] occur in the life of a social group and the experience of an individual so regularly and uninterruptedly, or so quietly and unnoticed, that people are hardly aware of them […]; or they may never have become aware of them at all and be unable to answer even direct questions. Other events may be so unfamiliar that people find it difficult to put into words” (Becker and Geer, 1957: 30). Becker and Geer thus draw attention to the ethnographic task that Hirschauer later refers to as “verbalizing silent knowledge” (Hirschauer, 2006: 432).

In order for a verbalization of the social to succeed, ethnographers must use their own person and body with all its sensory apparatus as instruments of research and personal recording apparatuses. During fieldwork, opportunities inscribe themselves “into researchers and authors” (ibid. 437). Thus, a surplus of embodied knowledge emerges, of which only a part ultimately makes its way into verbal notes.

These three variants of ethnography discuss the role of the ethnographer and the process of ethnographic writing epistemologically in different ways. Nevertheless, they follow a correspondence-theoretical view that conceptualizes the relationship between ethnographic description and social reality as one of representation. The empathetic ethnographer, the “meanings-and-symbols” ethnographer, or the bodily sympathetic ethnographer is engaged in the process of writing to put the phenomena he or she observes into words. As a result, the orderliness of phenomena emerges only through the process of ethnographic writing. This correspondence-theoretical view is problematic for ethnomethodology and its concept of ethnography. In the following sections, we discuss ethnomethodological ethnography in its similarities and differences to the three variants of ethnography presented so far.

Garfinkel’s theory of science and epistemological position and the role of ethnography

In his writings, Garfinkel developed a methodological position from different theoretical sources including phenomenology (Husserl, Gurwitsch, Schütz, and Kaufmann), philosophy of language (Wittgenstein), and praxeology (Kotarbinski). He used this as a foundation for an ethnomethodological ethnography that in some respects departed from other ethnographic approaches in terms of goals, interests, and methodological approach. For example, phenomenology-based ethnography (vom Lehn and Hitzler, 2015) emphasizes the methodological relevance of subjective participant knowledge and draws on Schütz. Garfinkel’s philosophical starting point, however, is less Schütz’s egological, but Gurwitsch’s non-egological theory of object constitution (Garfinkel, 2021; cf. Meyer, 2022). And unlike other phenomenological approaches in sociology, Garfinkel furthermore draws on Merleau-Ponty’s chiasmic philosophy of the invisible (Garfinkel 1967: 182; Wiley, 2019) and on Wittgenstein’s theory of language games (Garfinkel, 2019).

Familiar with the Chicago School, Garfinkel started to engage with the ethnographic method at an early stage. Whilst still a student he published the article “Color Trouble” (1940), ⁵ in which he describes social practices of explaining and making understandable (what he later called “accounting practices”) within an everyday USA characterized by racial segregation. ⁶ Garfinkel describes an event he witnessed on a bus journey from Washington D.C. to Durham, North Carolina, where he was studying. At a bus stop in Peterburg, Virginia, two black passengers objected to discriminatory actions of the white driver and two white police officers: they were attempting to reseat the two passengers in the segregated area of the bus reserved for blacks. Within his ethnographic description, Garfinkel provides a highly detailed insight into the specific sequences of events comprising the interactions through which the co-participants generated and negotiated social phenomena such as racial segregation and racism.

In the Second World War, Garfinkel served in the US Air Force and was responsible for training recruits in Florida, whom he instructed to practice attacking tanks using photographs (rather than real tanks). During this period of service, he engaged in considerable depth with phenomenological literature and—drawing on the practical problems associated with training recruits—the epistemological and methodological problem of adequately representing social reality (cf. Rawls, 2002: 15). This “crisis of representation” experienced by Garfinkel not only in methodological terms as a sociologist but also in practical terms as a training instructor—how can the picture of a tank adequately represent a real tank?—was also a characteristic feature of the work that he continued in the Department of Social Relations at Harvard University from 1946 onwards. Garfinkel’s PhD thesis, “The Perception of the Other” (1952), was supervised by Talcott Parsons and also influenced by Alfred Schütz. It provides a theoretical and methodological contribution to the sociological theory of science and knowledge in relation to the question of adequate representation of findings (cf. vomLehn, 2014: 20).

Garfinkel criticized contemporary sociology for precisely not considering the fact that in sociology the observers are part of what they observe and the main tool for representing the object—language—is part of the object. As epistemological basis he used Felix Kaufmann’s (1944) critique of the correspondence theory of truth, which acts as if external observers represent facts in direct, realist equivalence between object and representation. Garfinkel directed his criticism above all at the contemporary canon of applicable methodological rules in sociology, which transformed specific everyday actors into abstract representatives of sociological regularities. This, he argued, assumes a correspondence between the observed actions of individuals, which were methodologically purged of all lifeworld features, and the principles of the social order as represented by the resulting abstract regularities.

In his PhD thesis (1952), Garfinkel laid bare the difficulty in transferring to science the criteria of meaning applicable within everyday practices and in representing them with scientific criteria of meaning without loss. Under the correspondence theory model of sociology, scientific assertions are generated by changing or removing the criteria of meaning of the social phenomena studied which are rooted in everyday practices. In order to purify itself of the lifeworld within which it was created—both as regards the empirical as well as the investigating theorists themselves—sociology must formulate abstract theories about both. However, these theories cannot exhaustively clarify, pre-empt, and determine empirical manifestations. They remain necessarily vague and indeterminate and inevitably refer to something that is absent. Garfinkel later brands this feature indexicality. They can thus no longer be converted back to their empirical context and are reliant on permanent life-worldly supplementing practices, which (Husserl, 1929; Schütz, 1945) termed “et cetera.” This results in an infinite regress fueled by the ascription of meaning through language that necessarily follows formalization. This in turn entails new lifeworld categories, which for their part need to be formalized again.

For Garfinkel, the contemporary concept of social structure offered a prime example of a line of research that had remained stuck in a rut with the correspondence theory assumption. Under this concept—and here he had, amongst others, Parsons’ concept in mind—the actor was, as Garfinkel later argued, construed as a genuine “cultural dope” (1967: 68). This is because, according to Garfinkel, on Parsons’ account only the sociologist can identify the ordering mechanisms and structures of society with reference to scientific methods, whilst these mechanisms and structures are not apparent to the actors themselves. Garfinkel branded the assumed basic population of society, comprised of rule-following actors unaware of this status of theirs, as “Parsons’ Plenum” (Garfinkel, 1988: 104). This notion can also be found in bodily and textual ethnography, which either puts the social into words and thus orders it, or orders an already textual social from a sociological perspective. In both cases, order first emerges in the process of ethnographic writing.

Garfinkel termed his alternative perspective, drawing on Schütz (1953), as congruence theory. It posits that the constructs and types of social science must be congruent with the constructs and typifications of everyday actors. A social theory must acknowledge that it is empirically rooted in second degree observations—observations made by actors who are themselves observing and interpreting the social world. The worldviews (“attitudes”) of observers are therefore no less in need of clarification than the worldviews of actors. This implies that sociological knowledge cannot be obtained from the outside, as it can within the natural sciences; on the contrary, it must be gleaned from within society. Garfinkel argued, drawing on Schütz, that social phenomena were already ordered in and by themselves. This is because participants themselves are confronted with the challenge of making their specific circumstances known and understandable to others. According to Garfinkel, it is fundamentally necessary to consider the perspective of members of a collectivity, their ethno-theories and ethno-scientific practices, which in the 1950s were an object of interest in particular for anthropologists. Harvard University, where anthropology and sociology were represented within the Department of Social Relations, hosted ground-breaking research into the knowledge stocks and internal perspectives of the objects of study. Ultimately, the prefix ethno within ethnomethodology was inspired by this context.

The scientific reality and the social reality that is to be described must therefore be considered to be deeply related, albeit with different interests in knowledge. Science and everyday practice interact with and influence each other. Therefore, it is necessary to enquire into scientific forms of representation where language is used both within everyday practice and science, remaining inextricably indexical and vague in both. Having rejected the correspondence theory approach, the aim can no longer be to depict the social reality as if through a mirror. On the contrary, the people studied must be taken seriously as first-order observers also within scientific representations.

This is where the ethnographic method becomes central, since it is the method of choice to comprehensively investigate the meaning-generating processes within everyday practices themselves. It is at this point that Garfinkel draws not on Schütz, but on Gurwitsch, Merleau-Ponty, and Wittgenstein. For him, meaningful social objects are not constituted by an empirical, biographical ego that ascribes meaning to objects of experience (as in Schütz). Rather, social objects self-organize through ongoing practices by socialized and encultured members (as in Gurwitsch and Merleau-Ponty; cf. Garfinkel, 2021; Meyer, 2022). Social phenomena, or objects—as “Durkheimian Things” (Garfinkel 2002: 124)—are thus primarily “organizational things” (Garfinkel and Wieder 1992: 192), produced as “gestalt contextures” (Garfinkel 2002: 281) through “details” that are interactionally provided by the actors. Being meaningful, social objects are intrinsically “observable-reportable” (Garfinkel 1967: 1) and explainable, that is, they are what Garfinkel calls “accountable” (ibid.). Ethnomethodological ethnography therefore requires to become familiar through participation with these practices of “producing” but also of “seeing” and “recognizing” organizational things within “phenomenal fields” (Garfinkel 2002: 73) in the social world.

Ethnomethodological ethnography is therefore in this sense “phenomenal fieldwork” through which the ethnographer becomes a competent member able to produce and recognize the details of this field. Part of this is to be able to appresent invisible and absent aspects (sequential or inferential) that are indexed by what is given in these fields. Here, Garfinkel refers to Merleau-Ponty’s philosophy of the invisible (Garfinkel 1967: 182). Since, moreover, a great portion of phenomenal fields consists in talk (and various forms of talkability) and much of the appresentation work occurs in the form of indexical expressions, a second part of member’s competences is the “mastery of natural language” (Garfinkel and Sacks 1970: 342). The ability to participate in natural language practices and in language games—here Garfinkel draws on Wittgenstein (ibid.: 348–350; Garfinkel, 2019)—has to be acquired by the ethnographer.

Partly, Garfinkel derived the second element of the term—methodology—also from the praxeology of Kotarbinski (cf. Garfinkel, 1956: 191–192; Hiz, 1954). This focuses on identifying and describing methods of action. Garfinkel later increasingly referred to practices rather than methods (and for a short period even considered re-baptizing ethnomethodology as “neopraxiology”). The key point is that, for him, practices always generate social order and structures in a methodological fashion—which means above all practically and recognizably. This implies that rules must in principle be accessible to and comprehensible for actors (or, as Garfinkel prefers to term them, members). Moreover, according to Kotarbinski rules do not in any sense emerge in an abstract manner, but rather through everyday forms of instructions, maxims and rules of thumb.

Meaning-generating criteria and practices must also be incorporated in the ethnographic study by establishing a congruent relationship between that work and the everyday criteria that generate meaning in relation to it. One solution for this proposed by Garfinkel is that scientific reports—to put it in today’s language—should become performative: ethnomethodological studies no longer represent social phenomena, but rather bring to life their specific form of reality. They reproduce them, thus making them comprehensible as a genuine aspect of the lifeworld. On a methodological level this again implies that, in conducting ethnomethodological research, the ethnographer must at the same time be a competent member of the collectivity whose ordered phenomena he is studying. This is because the ethnographer’s own understanding of the lifeworld studied must be based on the implicit practices of members of the collectivity; similarly, the methods of (scientific) representation must be consistent with the methods of (practical) accomplishment of meaning.

As Garfinkel (1967: 208–261) argues in an investigation into practices of coding within quantitative sociology, professional sociologists implicitly build on practices of lay sociologists (a term Garfinkel uses to refer to everyday actors). However, this occurs without reflection, not to speak of any explanation, within the sociologist’s methodological discussion. ⁷ Garfinkel concludes that the main task of a sociology informed by ethnomethodology must be to turn this everyday practice of knowing, explaining, and interpreting—within which also professional sociology implicitly operates—into the object itself (“topic”), rather than using it unreflectively as an analytical instrument (“resource”) (ibid. 31).

From the 1970s onwards, Garfinkel studied the congruence between methods for generating meaning and methods for representing in Hybrid Studies of Work (cf. Garfinkel, 2002: 100). The descriptions contained in these studies are characterized in strictly praxeological terms by their relevance for members of the collectivity who are being ethnographically studied as such in specific situations and who are not often particularly interested in ethnomethodological descriptions. For Garfinkel (ibid.: 100–103), the requirement of congruence is only met where the ethnomethodological studies of competent members can be read as practical instructions. The quality of an ethnomethodological description is thus established with reference to whether the members of a collectivity can use it to validly recognize ordered phenomena and subsequently create such phenomena themselves in practice (ibid. 101).

Thus, for Garfinkel, ethnography, as a method, provides solutions to the specific situation of sociology as a science that it is part of its own object of inquiry. In sociological research, the phenomena studied are always already methodically and practically organized and ordered in and by themselves by competent members of society. Sociologists, being among these members, therefore need to constantly reflect upon their resources of inquiry and interpretation, transforming them into a topic of research in its own right. The solutions ethnography provides especially concern its interpretive resources and modalities of scientific representation. For Garfinkel, social phenomena can only be described practically and reflectively using the ethnomethods specific to the phenomena. In this stance, ethnomethodological ethnography differs from all three ethnographic variants discussed in the previous section, as it systematically integrates the fact that (a) the assumption of a direct correspondence between object and representation is fictitious, (b) sociological descriptions must genuinely, and reflectively, relate to resources provided by their objects of study, and (c) the validity of these descriptions cannot be determined sociologically, but must be proven practically (as practical instructions).

Ethnographies of Garfinkel’s collaborators

Over the years in which he worked as a professor at UCLA, Garfinkel provided guidance to several ethnographies of his collaborators, who implemented in practical terms the socio-theoretical results and methodological implications of ethnomethodology within their own empirical studies. Garfinkel demanded three things: (1) they were required to acquire the competences of members, even so far as going native; (2) they should experimentally develop appropriate forms of representation for their field; (3) they should not fabricate an “ethnography of sentences” (Liberman, 2004: 40) whereby they only “stumbled around” in the field and collected in an isolated fashion whatever matched up with their research interests. The ethnomethodological ethnographies we present in this section are not comprehensive and do not claim to be. Rather, we want to make some of Garfinkel’s concepts visible through examples. How do his collaborators make their research intelligible as ethnomethodological ethnography? How do they deal with concepts such as congruence and from within? How do they learn practical skills in the course of their ethnographies? How do they make their descriptions performative and readable as instructions for practice?

For his study “The Social Organization of Juvenile Justice” (1968), Aaron Cicourel conducted research for more than 3 years as an ethnographer in two young offenders’ institutions, working as a probation assistant in one of the two institutions (cf. Cicourel, 1968: VIII). In order to give the reader an insight into the invisible background expectations of the members studied, which would otherwise remain “seen (but unnoticed)” (ibid. 15), he used recorded ethnographic data (written reports from the authorities and recorded conversations), which he then subjected to a process of “re-writing” (ibid. 18). His aim was “[to] enable the reader to understand how the participants and observer made sense of their environments as portrayed by the researcher” (ibid.). Cicourel’s ethnography methodologically puts into practice key elements of an ethnomethodological ethnography. It is an exemplar of a congruence-theoretical representation of the phenomena under study. In order to ensure adequate understanding by the reader, Cicourel takes into account their resources of understanding, which must also be made available through his ethnography. In this methodological sense, the translation process that Geertz’s textual ethnography frames as part of ethnographic writing is reversed. It is not the phenomena studied that have to be translated into sociological language, but the sociological reader must be enabled by the ethnography to practically understand the phenomena that are described.

In his thesis “The Convict Code” (1969), Larry Wieder studied the practices of staff and inmates at an open prison as well as, self-reflexively, his own practices when interacting with staff and inmates at the facility. In doing so he demonstrated how he was performatively induced by inmates, although also by staff, within discussions to interpret the actions of inmates as the expression of a “convict code”—a type of prisoner’s code including obligations of confidentiality (cf. Wieder, 1969: XIII). In his ethnography, Wieder makes the practices of narrating the “convict code” sociologically observable in making his own ethnographic enculturation the topic of his study. This way of ethnomethodological ethnography makes specific interactional resources observable that are relevant to accomplish social order within halfway houses.

David Sudnow’s study “Ways of the Hand” (1978) shows the difficulties associated with the ethnographic presentation of practical, embodied skills, focusing on jazz improvisation on the piano. His way to ensure an adequate description of this phenomenon was to build it up like a set of instructions—if they are followed in practice—for jazz improvisation: “My description is meant to be a guide to the looks of improvisatory hands” (Sudnow, 1978: 154). Sudnow therefore focuses on re-creating the orderly phenomenon of jazz improvisation and the practical resources it requires. His ethnography makes these implicit bodily skills explicit. As an ethnographer, he is no longer concerned with a bodily observation of others who play jazz, but with his own bodily participation in playing jazz. Sudnow’s ethnography is therefore fundamentally different from Becker’s ethnography, for which he observed and interviewed musicians without directly investigating the bodily practices of music-making (cf. Becker, 1951). In this way Sudnow ended up devising a highly successful and financially lucrative method for teaching how to play the piano (“The Sudnow Method”).

For his thesis “Art and Artifact in Laboratory Science” (1985), Michael Lynch participated in the activities of a neuro-biological laboratory as an ethnographer. As part of this process, Lynch attended informal laboratory experiments in order to become himself a competent member of the laboratory he was investigating. Although Lynch—by his own admission—never attained the skills of a competent biologist within the neuro-biological laboratory, during his field studies he acquired technical and analytical skills in dealing with the images produced by an electron microscope (cf. Lynch, 1985: 1–2). Lynch made these practical skills—such as being able to identify artifacts in these images—comprehensible for his readers by providing written transcripts of shop talk amongst biologists in the laboratory. As a follow-up to Lynch, Philippe Sormani’s PhD thesis “Respecifying Lab Ethnography” (2014) can be consulted. For this ethnography, Sormani spent over 3 years in an experimental physics laboratory. His “reflexive ethnography” makes observable the practices by which first degree constructs are transformed into second degree constructs. In doing so, reflexive ethnography serves to provide readers with a practical basis for the video analyses gathered in the book (cf. Sormani, 2014: 99). Thus, practical skills of experimental physics are taught in the form of a respecified lab ethnography.

In his thesis “The Ethnomethodological Foundations of Mathematics” (1986) Eric Livingston went even further and not only made the mathematical skills that he acquired through his own study of mathematics understandable but also offered a practical guide to readers for how to acquire these skills. Livingston made these skills understandable not by providing written transcripts of the technical practices of mathematicians, but rather with reference to in incremental genesis of mathematical equations as the most important element of mathematical practice (cf. Livingston, 1986: 7–20). Livingston extends his ethnography of mathematical reasoning into an “Ethnography of Reason” (2008). In doing so, he makes phenomena of a “reasoning in the wild” tangible not only as descriptions of playing checkers, tangram, doing puzzles, etc., but as instructions for doing them. The ethnographic, therefore, is not found in the book itself, but in the practical accomplishments of readers who close Livingston’s book and begin to do puzzles or play checkers. Nevertheless, this book describes the basics of practical “reasoning in the wild,” and as an instruction, reasoning becomes a hands-on experience for readers. “We look at skill and reasoning in activities such as playing checkers, working on jigsaw puzzles, making origami models, conducting experiments, and proving mathematical theorems. […] All these activities are ones in which skill and reasoning are witnessable features of people’s participation in them” (Livingston, 2008: 9). Livingston’s book encourages us to participate in order to witness phenomenal field properties of reasoning through our own skillfull bodies. In this sense, as readers, we literally become ethnographers.

To complete her thesis “Making Settlement Work” (2000), Stacy Burns first graduated in law (Yale University) in order to be able to conduct sociological study of legal practice thereafter as a competent member of the legal science community. Her thesis does not examine actual legal practices or present itself as a guide to such practices; on the contrary, she presents the implementation of legal practice in action both inside and outside the courts on the basis of recorded material and then makes legal skills and practices apparent to the reader with reference to these transcripts (cf. Burns, 2000: 8).

Kenneth Liberman has spent over 6 years learning the Tibetan language for his research on monks' reasoning and has conducted 3 years of fieldwork at a monastic university to understand the phenomenon by being able to perform it practically himself. This has enabled him to participate in debates between Tibetan monks as a competent member. In his book, “Dialectical Practice in Tibetan Philosophical Culture” (2004), he guides his readers to witness reasoning as a social phenomenon. To do this, Liberman not only presents transcripts of debates and guides readers to understand them, but he also makes these debates audio-visually observable in their elusiveness with an accompanying CD-ROM. The Tibetan culture of philosophy thus becomes comprehensible from the inside, as Liberman puts his readers in the practical position of being able to (partially) adopt even the ways of reasoning and to recognize, for example, inconsistencies in the reasoning of individual speeches. Liberman’s book is not only a description of the way Tibetan monks argue, but also pedagogy that encourages readers to cultivate his or her own skills of Tibetan reasoning while reading.

We have seen that the studies mentioned above of some of Garfinkel’s collaborators use ethnography in methodologically different ways in order to achieve adequate descriptions of social phenomena. The descriptions convey implicit knowledge concerning the skills and practices through which the ordered phenomena investigated are generated practically by actors themselves. Whilst early ethnomethodological ethnographies frequently engaged in non-participatory observation—albeit with reflexive references to their methodological shortcomings—subsequent ethnographies shifted this approach towards active, and in time competent, participation in the practice being investigated. As a result, the skills and practices acquired through participation shifted to the methodological center of ethnomethodological ethnography. Their descriptions are therefore formulated by researchers also in terms of instructions for putting into practice the ordered phenomena investigated by them. These instructions use the full repertoire of ethnographic methods (participant and non-participant observations, conversations and interviews, analysis of documents and artifacts, audio, and video recordings). As we have seen in relation to Sudnow (1978), Livingston (1986, 2008) and Burns (2000), they also frequently involve autoethnographic elements.

Methodological principles of ethnomethodological ethnography

We have discussed above the relationships between ethnography and ethnomethodology on the basis of Garfinkel’s studies and those of his collaborators. In this section we shall now discuss the methodological principles of ethnomethodological ethnography as developed by Garfinkel.

At the heart of ethnomethodological ethnography there are the conceptual views that: (a) the phenomena to be studied are always already organized in and by themselves; (b) this order is created methodically by competent members of a collectivity; and (c) these phenomena can only be described practically and reflectively using the (ethno-)methods specific to the phenomena. Ethnomethodological ethnography derives five core methodological principles from this, which will be explained below: “ethnomethodological indifference,” the “rendering theorem,” the “unique adequacy requirement of methods,” at least two forms of adequate description, and the “praxeological validity of instructed action.”

The principle of “ethnomethodological indifference” (Garfinkel and Sacks, 1970: 345–346) is directed against the practice of mainstream sociology of establishing phenomena a priori and thus generating scientific order that is not the order known to members. In terms of philosophy of science, then, it is not a matter of sociology of discovering reality as an ordered reality, but of re-creating an understanding of a reality that is always already ordered by members’ practices. Indifference means that no phenomena are excluded ex ante, no prior assumptions are fed into the material and no appraisals or even normative assessments are carried out. From the viewpoint of ethnomethodology, all social phenomena are ordered phenomena and are thus sociologically interesting.

The “rendering theorem” brings to light problems associated with the application of established methods of professional sociology. According to Garfinkel and Wieder (1992), these methods first isolate social phenomena and thereafter produce objects that are characterized by the methods. As a result, the authors developed two tutorial problems in order to enable students to appreciate in practical terms the loss of specific ordered phenomena (cf. ibid. 189–200; Garfinkel 2002: 264–269). By holding these practical tutorials, the crisis of representation is thereby practically generated and reflexively experienced by participants. On a methodological level, they imply that the ethnomethodological ethnographer must always find forms of ethnographic representation that preserve and present the social phenomena studied in their specific manifestation (such as Sudnow or Livingston, who instruct readers to practically produce the phenomena under study themselves).

Garfinkel and Wieder (1992: 182) refer to the fact that this applies not only for representation but also for the research methods used as the “unique adequacy requirement of methods.” This requirement can be construed as a praxeological attempt to establish symmetry between science and everyday practice, as formulated by Schütz (1953: 34) in his postulate of adequacy—albeit in relation to typifications and not practices. The basis for ethnomethodological reformulation is the insight that phenomena that are ordered for competent members of a collectivity are always only accessible in full as part of local production of meaning and within their “natural accountability” situated in this context (Garfinkel and Wieder, 1992: 184). The principle of the “unique adequacy requirement of methods” proposes ethnography as a solution to the crisis of representation that is based on the congruence of and reciprocal reflexivity between methods for generating meaning and representation. It follows from this that it is the specific attributes themselves of the social phenomena being studied that determine the methods suitable for studying it. The research goal of the ethnomethodological ethnographer is to recognize and describe order from within social phenomena (cf. also Garfinkel, 2002: 124). A distinction is drawn between two interpretations of the “unique adequacy requirement of methods”:

(a) the “weak” interpretation requires that researcher must have their own knowledge as members concerning the ordered phenomena to be investigated (becoming a member). This means that the researcher must be competent in dealing with the local production of meaning and “natural accountability” of the phenomena and also have the skills to (re-)produce them. The knowledge embodied in this manner enables the researcher to methodologically discover, identify, and describe implicit practices as ordered phenomena.

(b) The “strong” interpretation also requires symmetry between the methods applied in order to ensure adequate description. The researcher must not only be a competent member in order to identify and describe ordered phenomena. In addition, in his ethnomethodological presentations he must also explicitly use the practices that competent members also use in order to make their actions understandable to one another (Garfinkel and Wieder, 1992: 182).

Garfinkel (2002) claims that there are two forms of adequate description of ordered phenomena: (a) description as instruction and (b) description as an aid to memory.

(a) Description as instruction does not transform the phenomena studied into written text; on the contrary, it performatively directs exercises through which the reader can—if he carries them out—acquire practical skills in order to himself competently bring about the social phenomena studied, and in doing so, follow the ethnomethodological argumentation in the description. These texts quite certainly direct sociologists to acquire newly embodied social skills through practical exercise according to the instructions. The instructions should also be hybrid, being practically comprehensible both to competent members of a collectivity (experts) as well as for sociologists (novices) (cf. Sudnow, 1978; Livingston, 1986).

(b) Descriptions as aids to memory are preferably used in the form of audio and video recordings and related transcripts. These enable researchers first to remember in practical terms the embodied skills and implicit practices acquired in the ethnography in which they have participated. It is these skills and practices that generate the ordered phenomena investigated. Accordingly data within ethnomethodology serve as “aids to a sluggish imagination” (Garfinkel, 1967: 38) and “a source of insight into what we ‘already know’” (Lynch, 2002: 535). Second, they not only make ordered phenomena rememberable but also empirically referenceable while making the knowledge intersubjectively understandable. Third, their preparation in the form of transcripts gives rise to an alienating effect, which makes it easier for researchers to distance themselves from the ordered phenomena with which they are practically familiar (cf. Lynch, 1985; Liberman, 2004).

With the “praxeological validity of instructed action,” Garfinkel (2002: 105) formulates a criterion for ethnomethodological descriptions. On this view, a description is always adequate if it can be read not only as a description of a phenomenon but also as an instruction for how to practically reproduce the phenomenon studied. Praxeological validity is controversial within the field of practitioners. On the one hand, Sudnow was able to translate his ethnographic insights into a unique method of learning how to improvise jazz. On the other hand, Livingston’s insights into mathematical practice were not directly accessible to mathematicians and were therefore hardly noticed.

The principles of ethnomethodological ethnography set out here are the solution proposed by Garfinkel for the sociological crisis of representation. Ethnomethodological ethnographers put the accomplishment of implicit practices for generating ordered phenomena at the center of their descriptions. In some cases, the phenomena at which they are directed are also predicated for methodological reasons on a doing in order to ensure that the work on creating them is also kept conceptually visible.

Conclusion

Ethnomethodology grounds the principles of the ethnographic method reflexively and practically within the phenomena of order themselves that constitute its object. In this respect, ethnomethodological ethnography differs from all other variants of ethnography in fundamental ways. It is not conceptualized as an orderly process of writing but as one of re-presenting and re-enacting already orderly phenomena. For Garfinkel, due to its preference for acquiring member skills and as well as its versatility, ethnography is one way to meet the requirement of congruence between scientific and everyday methods within the self-reflexive discipline of sociology. In order to methodologically implement the congruence theory, research into how members produce descriptions and solve problems within everyday practice and science, along with their necessary limitations, must always be part of ethnomethodological ethnography. Ethnomethodological ethnography solves the methodological problem of adequate description thanks to the “unique adequacy requirement of methods.” This requires that descriptions be generated using the same ethnomethods as those that are used in practice by competent members to generate the phenomena of order studied. Thus, from the viewpoint of ethnomethodology, the ethnographer must therefore always be a competent member who can recognize and adequately describe the phenomenon of order. This enables readers to shift into the hybrid position desired, which is necessary for sociological research, in order to comprehend the phenomena of order described both practically as well as scientifically (“praxeological validity”). This requirement might appear at first sight to be highly demanding. However, it must be regarded above all as an instrument of necessary self-clarification in acknowledging that the epistemological resources and instruments of sociology are inseparably intertwined with the epistemological resources and instruments of the field of study.