Human Studies

, Volume 42, Issue 2, pp 221–237 | Cite as

From Documentary Meaning to Documentary Method: A Preliminary Comment on the Third Chapter of Harold Garfinkel’s Studies in Ethnomethodology

For Jörg Bergmann
  • Erhard SchüttpelzEmail author
Theoretical / Philosophical Paper


The text deals with Harold Garfinkels theorizing of what Karl Mannheim called ‘documentary meaning’, and established as a foundation of all historical disciplines, and what Garfinkel calls the ‘documentary method’ of lay and professional sociological reasoning. The commentary tries to establish the systematical position of the chapter in Garfinkel’s ‘Studies in Ethnomethodology’, and, indeed, in Garfinkel’s social theory at the time of publication. This position involves, and redefines, Weber’s definition of sociology, Schütz’s sociology of knowledge and especially, the very idea of a common stock of. typifications, and the discovery of ‘historical time’, or ‘Geschichtlichkeit’ in each micro-sequential interaction. Some consequences are drawn, or sketched, for ethnomethodology as a steady thorn in the side of sociology, and vice versa: for lay and professional sociology as an equally steady thorn in the midst of ethnomethodology.


Social theory Hermeneutics Ethnomethodology Mannheim Schütz Phenomenology 

The Systematic Significance of Garfinkel’s Experiment

At the end of the 1950s, Harold Garfinkel organized an experiment whose analysis forms the basis for the third chapter of Studies in Ethnomethodology (Garfinkel 1967: ch. 3). The experiment was derived from student exercises that Garfinkel used at the University of California at Los Angeles, in order to demonstrate to students the “method of understanding” that he found presented in Karl Mannheim’s essay on the interpretation of worldviews. With the help of his student assistant Peter McHugh, who would later publish his own analysis (McHugh 1968), Garfinkel converted his exercise into a laboratory experiment, which was adapted to the existing standards of social psychology of the time, in order to make possible a scholarly publication.

Garfinkel recruited ten students who were told that an alternative to psychotherapy was being tried out at the UCLA Psychiatry Department:

…to explore alternative means to psychotherapy ‘as a way of giving people advice about their personal problems’ (sic). (Garfinkel 1967: 79)

The experimenter, personified by Peter McHugh, was declared to be a student counselor in training. He was in an adjoining room and communicated with the subjects by way of an intercom system. The experiment was based on several deceptions, since neither the process of counseling nor the counselor himself was being tested, as the subjects must have assumed or, on the basis of the instructions, should have assumed. Thus, Garfinkel also does not write anything about the question of how the counseling itself should be classified or theorized, since in this case the counseling only served as a facade, in order to set in motion a discussion about social structures and to reveal the common culture and its form of common knowledge. This is at least what is promised by the reflections on the sociology of knowledge in the introduction to the chapter: “Sociologically speaking, ‘common culture’ refers to the socially sanctioned grounds of inference and action that people use in their everyday affairs and which they assume that others use in the same way…. Such socially sanctioned, facts of social life consist of descriptions from the point of view of the collectivity member” (Garfinkel 1967: 77).

The aim of the experiment was to investigate common everyday knowledge about social structures: the “common sense knowledge of social structures” mentioned in the title. The subjects were supposed, first of all, to present a personal problem and then pose the first question. Every question had to be posed in such a way that it could be answered by “yes” or “no”; there were ten questions in all. After each answer and at the end of the series of questions, the subject was supposed to think out loud about the answers and evaluate the counseling, while the channel to the experimenter was switched off. The entire course of the dialogues and monologues was tape-recorded. At the end of the series of questions, the experimenter came out of the adjoining room, in order to receive the subject’s judgment of how things went overall. The sequence of answers was always the same and had been determined at random in advance. This fact was explained to the subjects, and their reaction to the revelation was likewise recorded and analyzed.

There are several analyses by Garfinkel of this experiment. The most important is found in a long book manuscript entitled “Common Sense Knowledge of Social Structures,” which was distributed at the World Congress of Sociology in Stresa, Italy in 1959 (Garfinkel 1959) and then published in a radically abridged version in the proceedings of the congress (Garfinkel 1962). In this manuscript, Garfinkel also goes into the disappointment of the students, when it was revealed to them that the counseling was randomized or, if you like, never in fact took place. In this version, the experiment is even, by way of the analysis of the disappointment, part of the series of Garfinkel’s “breachings”: i.e., an example for the successful breaching of relationships that are presupposed as self-evident in everyday interaction. In the case of the “breachings,” the analysis focused on the phenomenological conditions of intersubjectivity and uncovered, with the help of the “breachings,” the morally sanctionable “fait social” of everyday interaction in them. This also applies to the end of the experiment about the “documentary method,” since the subjects feel morally disappointed and some of them remain morally outraged—and they do so indeed regardless of whether they, nonetheless, learned something that was very important for them or, in retrospect, approve of the experimental set-up (Garfinkel 1959).

In the third chapter of the Studies, the same two protocols are used and the same findings on the temporality of the process are gathered; but the analysis does without the presentation of the concluding confrontation with the experimental set-up and instead looks at the consequences of the experimentalization for the discussion of sociological method (Garfinkel 1967: 77–79, 94–103). And these passages are in fact the most detailed statements on professional sociology in the whole book, and they also contain enough explosive matter for all discussions of sociological method.

This variable ordering of topic, experiment and analysis is not unusual for Garfinkel’s experimental practice. What an experiment experimentalizes, what the findings are supposed to demonstrate, what concepts are supposed to be tested with its help, and what the results of this exploration are and how they can be arranged—all of this could be shifted around. Garfinkel’s experimental practice is reminiscent of a composer who goes into the studio to record a new song and tries out several musicians for each instrument—thus, several drummers, bassists, guitarists and brass soloists—in order finally also to change the composition little by little, until the ship has been completely remodeled on the open sea. The present case is not quite so extreme. But in the wake of Garfinkel’s essay, other interpreters have modified his experiment in this spirit, and they have done so indeed as a test case, as “experimentum crucis” for concepts like the social “situation” (McHugh 1969), as an experiment on giving advice (though not in the form of the conversational analysis that draws on Garfinkel); and in anthropology, Garfinkel’s essay is, in the meanwhile, known as the best theorization of random oracles and other processes of divination (Zeitlyn 1990).

And one should not forget that Garfinkel’s text was first published in a volume on “Theories of the Mind” and was not out of place there (Garfinkel 1962). Thus, at the beginning of the 1960s, Garfinkel’s text could also be read, following Fritz Heider and Wilfrid Sellars, in a social-psychological context and indeed as a contribution on the attribution of mental states: for instance, in terms of the question of whether the mental states of the subjects were only imputed to the randomized oracle or also to the person of the questioner him- or herself. It could be read as a “theory of mind” or also as a “theory theory”—TT, for short. Garfinkel himself, however, does not go into this discussion (Garfinkel 1962). If the text had been launched in Germany, one could also have read it in the 1960s as a contribution to the theory of everyday hermeneutics: as a text that came into being at the same time as Gadamer’s Truth and Method, but, unlike Gadamer’s book, really deals with truth and method. Perhaps this reception will still happen, when the Studies have been translated. The history of hermeneutics will never be the same again.

Moreover, there are in fact sociologists who have tried to turn the “documentary method” into a sociological method—which without a doubt was not Garfinkel’s intention (Bohnsack et al. 2001). These different strands of reception indicate that the experiment on the “documentary method,” i.e., on the use of evidence which is “typical,” “exemplary” or “a recognizable pattern,” itself remains “documentary”. But they thus contradict neither Garfinkel’s experimental set-up nor his analysis: The “documentary method” is and remains “typical” or “exemplary” for situations of counseling, for procedures of divination, for everyday hermeneutics, for the attribution of mental states of mutual understanding, for everyday sociological attributions, and also for professional sociology. And without a doubt it is also worthwhile to consider the history of “documentary film” in terms of how this genre is stamped by the “documentary method” and, at the same time, uses and exhibits it as a form of the everyday knowledge of the people shown. “Documentaries and the Documentary Method”—this could be the title of an essay or an essay film. In short, there are a multitude of possible research questions that connect up to Garfinkel’s experiment, but none of them contributes to a revision of his results.

Hence, I will now bracket all of these contexts that, sooner or later, have already been adduced and all still possible contexts, and I will concentrate on the text as part of the Studies. What is at issue in the third chapter, when one reads the text quite simply as a contribution to the book and its systematic objectives?

As is well known, in his preface, Garfinkel writes: “I regret a certain unity that was obtained in the collection by pondering and arranging texts” (Garfinkel 1967: ix). This was certainly not merely teasing the reader. Garfinkel had thought about the arrangement of the studies for many years and strived to achieve an integrated theoretical whole. The problem was rather that there were too many ways of organizing the material and that he always wanted to land several systematic hits at the same time: the typical problem (sic) of people who can only think in systematic terms and thus, in their creative work, pile up more and more systematic designs and alternative systematic variants. A basic outline of the study is described in the preface as follows:

The central topic of the studies is the “accountability” of social facts and their “reflexivity” as a feature “of practical actions, of practical circumstances, of common sense knowledge of social structures, and of practical sociological reasoning. (Garfinkel 1967: vii)

This basic arrangement is repeated again shortly thereafter—now indeed as the task of:
  1. (1)

    “learning how members’ actual, ordinary activities consist of methods to make practical actions, practical circumstances, common sense knowledge of social structures, and practical sociological reasoning analyzable”;

  2. (2)

    “and of discovering the formal properties of commonplace, practical common sense actions, ‘from within’ actual settings, as ongoing accomplishments of those settings” (Garfinkel 1967: viii)


There is thus a basic four-part arrangement and, moreover, a division between an investigation of the “members’… methods” or of the “ethno-methods,” on the one hand, and the “formal properties” of action sequences, on the other hand. This results in two major topics: How the everyday world of action is structured in such a way that it consists of “ethno-methods” and what formal means are used by this world to bring about its courses of action. Did the Studies fulfill this program?

If we consider the structure of the Studies, then there are some parallels to this arrangement, and the most obvious one concerns the chapter on the documentary method. For here the correspondence is already in the title: “Common Sense Knowledge of Social Structures”—thus one of the promised topics touched upon in the preface. Moreover, the “documentary method” is evidently an “ethno-method.” What else could it be? And indeed it is an ubiquitous method, which not even social scientists can shed.

If we look more closely, following study of the chapter titles and the main keywords, we can make the following assignments:
  • Practical activities, knowledge, and especially “practical sociological reasoning”—ch. 1

  • Practical actions or “everyday activities”—ch. 2

  • “Common sense knowledge of social structures”—ch. 3

  • Practical reasoning—”social inquiry”—ch. 4

  • Practical circumstances—ch. 5

  • Formal properties—ch. 8.

There remain chapters 6 and 7, which deal with “accountability” in an institution or, put differently, with “practical reasoning, knowledge and circumstances” all at once. This is to say that there is a certain cycle: first, three sketches on “practical activities,” “practical knowledge,” and “practical reasoning” in the first chapter.

Then four chapters on “practical action” (ch. 2), “practical knowledge” (ch. 3), “practical reasoning” (ch. 4), and “practical circumstances” (ch. 5).

And then a study in two parts about fully developed “practical activities, knowledge, reasoning and circumstances” (chs. 6/7); and an introduction to ethnomethodology in chapter 1, rounded out by a final part on scientific theorizing (Ch. 8).

If we compare this arrangement with the preface, one can assume in good faith that Garfinkel wanted to produce a unified book and also a sort of system, which, however, after the Studies—for whatever reason—he no longer insisted upon: in particular, the four-part schema.

Moreover, if we take the sociology of that time as starting point, there is a kind of ascent from his own basic concepts, by way of sociological basic categories, to more complex social phenomena, with a conclusion in the theory of science:
  • Ch. 1: presents the new basic concepts for defining ethnomethodology (accountability, reflexivity, indexicality);

  • Ch. 2: defines or redefines the “fait social”;

  • Ch. 3: presents the formal means of everyday sociology and those of a sociology of knowledge of everyday life;

  • Ch. 4: discusses what other sociologists call a “role,” replaces this concept with the question of how jurors become jurors;

  • Ch. 5: discusses what other sociologists call “identity,” replaces this concept by the question of “self-sameness”;

  • Chs. 6 and 7: studies behavior in “institutions” or in institutional “settings”;

  • Ch. 8; is devoted to the theory of science according to which there is a difference between scientific rationality and everyday rationality (whereas ch. 3 emphasizes that there is no such difference: the “documentary method” cannot justify such a distinction).

This is also a worthy arrangement for a book in sociological theory: or, in other words, for the claim first to present the basic concepts of “interaction” (ch. 1), the “fait social” (ch. 2) and of everyday sociological “understanding”; then what elsewhere in sociology are called “role” (ch. 4), “identity” (ch. 5) and “institution” (chs. 6/7); and, finally, the theory of science of sociology (ch. 8) and its distortion of the everyday rationality covered in the empirical chapters (chs. 1–7).

In this arrangement, the third chapter would represent the conclusion of the work on the founding categories: firstly, Garfinkel’s own new concepts are introduced in the first chapter, then Durkheim and Schütz are modified in the second chapter, and then Mannheim and Weber in the third, in the spirit of Max Weber’s well-known definition of sociology as “a science concerning itself with the interpretive understanding of social action and thereby with a causal explanation of its course and consequences”. Weber explains this definition in turn by definitions of social action: Social action is behavior to which a subjective meaning is attached and that is related to others to whom it is oriented. And in fact this definition is cited by Garfinkel as literally as a translation permits: “behaviors with a subjective meaning attached and governed thereby in their course”(Garfinkel 1967: 78). “In his concern for the sociologist’s problem of achieving an adequate description of cultural events, an important case of which would be Weber’s familiar ‘behaviors with a subjective meaning attached and governed thereby in their course,’ Karl Mannheim furnished an approximate description of one process” (Garfinkel 1967: 78).

The third chapter poses the Weberian question, but for everyday social action and knowledge: In everyday life, how do we succeed in interpretively understanding social action and in providing a causal explanation of its course and consequences? And the experiment radicalizes this question for the fundamental component of Weber’s definition: How do we manage at all to attach a subjective meaning to our action and to relate it to others to whom it is oriented? The test subjects attach a subjective meaning to their possible action: a meaning that is oriented to others, but with the peculiarity that, objectively considered (due to the randomization of the answers), the temporal sequence in which this happened was not oriented to others, but rather had the subjective meaning of being so—hence precisely served to isolate this “subjective meaning”. Garfinkel thus takes a step back before Weber’s definition of sociology and tries experimentally to reproduce it: What does it mean in everyday life, hence in an “ongoing accomplishment,” to want to bring about this accomplishment in an ongoing process: i.e., “interpretively to understand social action and thus to provide a causal explanation of its course and consequences”?

To answer this question, Garfinkel observes how people act socially in their interpretations: namely, by interpretively understanding the social action of others and providing other people a causal explanation of the course and consequences of this action as social action; and he observes how they attach a subjective meaning to their behavior that is directed toward objectively ascertainable facts: namely, toward the assumption of established social structures.

The Weberian definition of sociology is, in a sense, naturalized and is thus (like in Schütz) again available for reconstruction: Sociology already takes place in everyday life as an “everyday sociology” or “lay sociology,” and Weber’s definition is fitting for this sociology—also because this definition does not distinguish between everyday sociology and professional sociology. I think this is the main lesson of the chapter for sociologists: n.b., for sociologists, and indeed not only for sociological theory, but also for the assessment of sociological methods (as will be seen in greater detail below). Hence, if we take the first three chapters of the Studies as a whole, we arrive at the following systematic findings:
  1. (i)

    Durkheim’s definition of the “fait social” is taken over in modified form and, by way of the “breachings” of the second chapter, categorized with the technical phenomenological conditions of intersubjectivity. The first few sentences of the second chapter say: “For Kant the moral order ‘within’ was an awesome mystery; for sociologists the moral order ‘without’ is a technical mystery. From the point of sociological theory the moral order consists of the rule governed activities of everyday life” (Garfinkel 1967: 35).

  2. (ii)

    In the first chapter, Durkheim’s “fait social” is given greater precision by way of three new basic concepts, as the preface explains: “…the objective reality of social facts as an ongoing accomplishment of the concerted activities of daily life, with the ordinary, artful ways of that accomplishment being by members known, used, and taken for granted, is, for members doing sociology, a fundamental phenomenon. Because, and in the ways it is practical sociology’s fundamental phenomenon, it is the prevailing topic for ethnomethodological study”. (Garfinkel 1967: vii)


Even if it looks chaotic and overgrown at first glance, we can say that this sentence reproduces the arrangement of the first three chapters:

Chapter 1 deals with “the objective reality of social facts as an ongoing accomplishment of the concerted activities of daily life,” i.e., puts the accent on the “objectivity,” the “accountability,” as “ongoing accomplishment” in its “indexicality,” and on the “objective reality as an accomplishment of the concerted activities”—namely, the cooperative “reflexivity” of “accountability”;

Chapter 2 demonstrates “the ordinary, artful ways of that accomplishment being by members known, used, and taken for granted,” and, above all, the “taken for granted” is demonstrated here by way of the “breachings,” but also, of course, the knowledge and use of what “is presupposed as self-evident” and whose self-evidence can be shaken by “breachings”; and

Chapter 3 deals with “everyday sociology” or “practical sociology,” and Garfinkel here, as already cited, draws on Weber’s action-oriented definition of sociology. The everyday sociology of the world of action can be grasped using Weber. Conversely, this could mean that Weber’s definition is precisely not a definition of scientific sociology, but rather characterizes what everyday sociological knowledge and action and academic sociological knowledge and action have in common: what does not distinguish them from one another. As is presented in detail in the last part of the chapter, this is Garfinkel’s position: using a Weberian definition, and, in principle, also the phenomenological interpretation of the latter, we do not provide the bases for a scientific sociology, but rather we identify, above all, the everyday sociological faculty of knowledge about social structures, as well as the commonalities between the everyday and the scientific capacity for generalization that come into being thanks to this faculty.

Mannheim’s “Documentary Meaning” and Garfinkel’s “Documentary Method”

Could Garfinkel also have made the third chapter into the first? It appears that he did not regard this as didactically appropriate: he only deals with the commonalities between everyday sociology and professional sociology after the program of ethnomethodology has been laid out in the first two chapters. This is not the only precautionary measure taken by Garfinkel, in order to ensure that ethnomethodology is not simply confused with the documentary method.

In his reference to Weber, Garfinkel himself attenuates the connection, writing: “Karl Mannheim furnished an approximate description of one process” (see above). Thus, there is a double attenuation: an approximate description of one process of interpretive understanding (by way of “descriptions”). But shortly thereafter, Garfinkel notes in a footnote that the “documentary method” is the basis of everything that could be called “interpretive sociology”: “Attempts by sociologists to identify something called ‘interpretive sociology’ involve reference to the documentary method as the basis for encountering and warranting its findings” (Garfinkel 1967: 95, note 6). In other words, the documentary method is responsible for everything that could correspond to Weber’s definition of sociology—and for much more: in particular, for what Schütz characterized as ‘socially sanctioned’ typifications comprising a stock of knowledge. The beginning of the chapter refers explicitly to Schütz and his stock of typifications:

Socially-sanctioned-facts-of-life-in-society-that-any-bona-fide-member-of-the-society-knows… consist of descriptions from the point of view of the collectivity member’s interests in the management of his practical affairs. Basing our usage upon the work of Alfred Schutz, we shall call such knowledge of socially organized environments of concerted actions ‘common sense knowledge of socials structures. (Garfinkel 1967: 76)

Mannheim’s “documentary method” is experimentalized, but the experimental testing is oriented less to Mannheim than to Weber and Schütz: to what makes “social action” possible and to the “typifications” of everyday life. As is well known, Garfinkel took over the term “documentary method” with a certain displacement, which he did not himself discuss. The “Arbeitsgruppe Bielefelder Soziologen” [Working Group of Bielefeld Sociologists] has already provided detailed commentary on this displacement (Garfinkel 1973: 237ff.). Hence, I only want briefly to recapitulate it. Mannheim’s example was the following: “I am walking down the street with a friend; a beggar stands at a corner; my friend gives him an alms” (Mannheim 1951: 45). In this example, Mannheim distinguishes between the objective meaning, the “meaning…which belongs to the sociological field” (“assistance”), the subjective meaning, viz., “to convey a feeling of sympathy to me or to the beggar” (Mannheim 1951: 46), and the “documentary meaning”. The latter coalesces retrospectively, when the event appears as a proof for something else: for instance, inasmuch as “analysing all the implications of what I see, I may suddenly discover that the ‘act of charity’ was, in fact, one of hypocrisy” (Mannheim 1951: 47). Mannheim calls this retrospective interpretation “documentary meaning,” because it is not about what someone objectively did or subjectively expressed, but rather “what is documented about him, albeit unintentionally, by that act of his” (Mannheim 1951: 47).

Garfinkel does not make these distinctions, but rather concentrates only on the “documentary meaning”—i.e., on what is documented in an event for someone about others or for others about oneself—and on the fact that this meaning is continuously brought to bear in an interaction: not only in retrospect, but again and again and also during the ongoing interaction. He thereby also defines Weber’s task as the everyday sociological task that each of us has to tackle in his or her everyday life, and he does so by making the following modification: “interpretively understanding social action” means nothing other than extracting its “documentary meaning”.

To what extent is Mannheim’s text about a “documentary method”? Garfinkel’s reference to Mannheim is neither a misinterpretation nor a faithful reading, but rather an extrapolation that certainly has its basis in Mannheim. The word “method” does in fact play a central role in Mannheim’s text, since it is about the interpretation of a Weltanschauung by way of that which is typical and exemplary and hence about the methods of a typical historian of ideas or social historian in reconstructing the worldview of an era, a work of art or an artist. This is Mannheim’s question. It is by no means erroneous to call this method the “documentary method,” per a reading of Mannheim’s text. But Mannheim nowhere writes that the “documentary meaning” is what artists and people in their everyday lives have themselves to make into a “documentary method,” in order to be able to act in turn. For him, it is only about the viewpoint of historical reconstruction.

Garfinkel’s reading is thus, on the one hand, very precise: it is about the methods of that which generalizes a “documentary meaning”; and, at the same time, it goes against the grain of Mannheim’s question. But Mannheim’s essay also very explicitly deals with Max Weber’s sociology. What the “Bielefeld sociologists” overlooked in their commentary is the origins of Garfinkel’s concept in Mannheim’s discussion of Weber’s sociological writings. Mannheim writes, namely: “Weber postulates a mutual causal dependence among the various domains of culture and considers it necessary for purposes of a correct ‘causal account’ that the economico-material should at times be explained from the mental, and another time… the spiritual from the material…” (Mannheim 1951: 80). But the “documentary meaning” does not allow for such attributions of causality: “…there is no causal relation between one document and another; we cannot explain one as the causal product of the other but merely trace both back to the same global unity of Weltanschauung of which they are parts. Similarly, when we trace two actions of a person to the same personality trait, we cannot also treat one as being caused by the other, i.e., say that one kindness has caused another, instead of saying that two actions have been caused by the same kindness” (Mannheim 1951: 81). And here, in a footnote, the term “documentary method,” which was employed by Weber, in fact turns up:

“It should be stressed, however, that Max Weber’s actual historical analyses do not always correspond to his theoretical precepts. In his theoretical writings, he insists upon causal explanation; in his historical works, he very often proceeds according to the ‘documentary’ method” (Mannheim 1951: 81, note 1; corrected from the German original text). And this is the only passage in which Mannheim in fact speaks of a “documentary method”.

The relationship to Mannheim’s text is thus by no means as arbitrary or brutal as one might think. Garfinkel takes over from Mannheim the ‘acausal’ nature of documentary interpretation: that it does not make possible any causal explanations, but only concerns generalizations of the sort that something is “a document” of, “a symptom” of, or “typical” for a general pattern. And he also takes over the question of the temporal structure of the “documentary meaning,” turning it, however, from the macro-historical level of a history of reception to the micro-historical level of all interaction. Very much in the spirit of later aesthetics of reception, Mannheim writes:

“…documentary interpretation has the peculiarity that it must be performed anew in each period, and that any single interpretation is profoundly influenced by the location within the historical stream from the which the interpreter attempts to reconstruct the spirit of a past’s epoch” (Mannheim 1951: 61). Documentary interpretation is subject to historical irreversibility, of which Mannheim writes:

“…the temporal process of historical understanding, which does not add one item of knowledge to another but reorganizes the entire image around a new centre in each epoch, has positive cognitive value—this type of knowledge, in fact, being the only one a dynamically changing subject can have of a dynamically changing object” (Mannheim 1951: 62).

The documentary method of interpretation focuses on the interrelationship between patterns and instantiations or between typifications and evidence; the application of the method is not cumulative, but rather discovers new patterns in an irreversible order; and it is also prominently used in sociology. Against this background, it is somewhat easier to understand Garfinkel’s cryptic remark in the introduction to the experiment:

…a demonstration of the documentary method was designed to exaggerate the features of this method in use and to catch the work of ‘fact production’ in flight. (Garfinkel 1967: 79)

Discussed in relation to Mannheim, this means: The experiment was supposed to force the subjects to produce patterns and their instantiations; it was supposed to force them constantly to vary these patterns (and thus to accelerate and “exaggerate” micro-historical processes, i.e., “in flight” and reduced to the constant attribution of “typical features”); and it was supposed continuously to record the non-cumulative change in opinion and change in meaning.

By way of his randomized yes/no answers, Garfinkel does not produce just any sort of temporality, but rather a historical “flight,” in which what Mannheim says about the macro-historical interpretative relationship over decades takes place already within a few seconds. The temporality of an interaction is historical time. Already within just a few seconds, historical-hermeneutic events take place, whose peculiarities we can study in the history of reception of texts, testimonies and interpretations over decades or even centuries and about which Garfinkel generalizes as follows (Garfinkel 1967: 89–94):
  • New question are prospectively and retrospectively improvised on the basis of the course of the interaction and supplemented by interpretation of “what was meant” or “could have been meant”.

  • Questions posed and answers given are understood differently in retrospect; discursive segments are understood as answers to questions that were never posed or to several questions at the same time or they are made into a new question in turn. There is a constant retrospective-prospective reallocation of questions and answers.

  • Incomplete answers are kept ready for future completions; inappropriate answers raise speculations about the reasons for their inappropriateness; incoherence is registered or at least potentially conceived as learning process; contradictions are forgotten or swept under the rug, but sometimes also confronted.

  • The “documentary method” consists exclusively of typifications: hence of what is assessed as “typical of,” as “exemplary,” and as “pointing to”; but this method does not produce any stable typifications or any “stock of knowledge” that can be definitively established—at least not in this experimental test. The instability of what is passed down in the aesthetics of reception and its constant adaptation to the needs of the present also applies in the micro-historical time of interaction and does not allow for any final consolidation.

In The Social Construction of Reality, Berger and Luckmann, very much in the spirit of Schütz, note:

The social reality of everyday life is thus apprehended in a continuum of typifications, which are progressively anonymous as they are removed from the ‘here and now’ of the face-to-face situation. At one pole of the continuum are those others with whom I frequently and intensively interact in face-to-face situations—my ‘inner circle,’ as it were. At the other pole are highly anonymous abstractions, which by their very nature can never be available in face-to-face interaction. (Berger and Luckmann 1966: 33).

Garfinkel’s experimental testing undermines this view in all its aspects—precisely because what Garfinkel had in mind was nothing other than to recognize the peculiarity of Schütz’s socially sanctioned typifications: There is neither a fundamental difference between anonymous and intimate relationships in socially sanctioned typifications nor a “continuum of typifications”. When I—or anyone else—would present or present the relations of my “inner circle,” then this occurs using the means of typification that are also used for anonymous typifications; but these typifications can vary so greatly, depending on the addressee, that no thoroughgoing coherence is required anymore or becomes demonstrable. The stability of the “stock of knowledge” does not consist in the fact that in the course of our life we elaborate and consolidate a common, hierarchically ordered ensemble, but rather in the fact that we presuppose corresponding typifications in every conversation and elaborate them on an “ad hoc” basis, and we do so without mutually requiring any sort of binding form of consistency from ourselves.

Garfinkel’s foreshortening of typification and occasionality demonstrates that even an individual person in the course of just a few questions and answers does not draw on a coherent and stable stock of knowledge and that precisely the constant striving to elaborate a common pattern with others—i.e., the constant striving to achieve “documentary” coherence (as seen from a static perspective)—produces a single series of inconsistencies and, from a dynamic perspective, produces a series of continuously undertaken prospective-retrospective reinterpretations. This opportunism is the real “stock of knowledge,” since the whole cooperative art of typification would be neither feasible nor comprehensible without it. And if we sketch out very concretely—for instance, for Garfinkel’s two protocols—”wherein” the common “stock of knowledge” of typifications consists, we come to realize how little we contradict the possible calling-into-question, transformation or even the simple contesting of a single typification, as soon as it is put forward by others.

Garfinkel’s Conclusions and the Dilemma of Ethnomethodology

This result confronted Garfinkel’s reasoning with several bifurcations. In principle, I have already mentioned the first one: “Common sense knowledge of social structures” does not consist of a coherent “stock of knowledge”; it only consists of being able to come to agreement with others about social structures on an ad hoc basis and on the basis “of hearsay,” using the documentary method. A large part of the Schützian world thereby collapsed. And the question had to be posed of whether “social structures” could be generalized in a different way—at least scientifically. Garfinkel’s answer is found in the second part of the chapter and remains unanswered in its radicalism up to today: No, in this respect, there is no difference between everyday sociology and professional sociology. Both as concerns the prospective-retrospective reinterpretive temporal structure and the inconsistent form of generalization, the documentary method remains a documentary method: regardless of whether it is applied by lay persons or professionals.

The methodological steps of every social scientific investigation and the social scientific generalizations, which treat something as “typical of,” as “evidence” of a pattern, as “unmistakable symptom” of, remain “documentary”. It is not because they are more scientific than everyday generalizations that they are convincing, but rather because they are just as everyday as the sociological generalizations that are undertaken in everyday life. What, then, becomes of the concept of “social structures”? And, above all: What becomes of sociology?

At the end of the text, Garfinkel poses the question in radical form:

Is there, therefore, any necessary connection between the features of common sense situations of choice, the use of documentary method, and the corpus of sociological fact? Must the documentary method necessarily be used by the professional sociologist to decide sensibility, objectivity, and warrant? Is there a necessary connection between the theoretical subject matter of sociology, as this is constituted by the attitude and procedures for ‘seeing sociologically’ on the one hand, and the canons of adequate description, i.e., evidence, on the other? (Garfinkel 1967: 102)

Garfinkel says yes. As Garfinkel explains in detail, there is no professional sociology without the documentary method. It comes to bear wherever an incomplete description is supposed to stand for a larger context; where an exemplary case is generalized; where the application of methods requires improvisation, so that the proceedings, nonetheless, fit into a methodological schema and can be reported as part of a methodological approach; where a statistical measurement is analyzed as an indicator for a more wide-ranging generalization. In short, it comes to bear, in principle, everywhere: whether in qualitative and quantitative investigations, in case studies or in theoretical expositions. It determines the practice of research, the practice of generalizations and the practice of mutual criticism and correction.

And it does not distinguish everyday sociology from professional sociology. The applications of the documentary method in professional sociology perhaps appear to be more saturated in terms of their content and more mature in terms of their arguments than everyday statements about “social structure,” but as soon as sociologists subject these applications to testing among themselves, their everydayness becomes recognizable: they become assailable and revisable. And even this susceptibility to critique and revisability is nothing other than the everydayness of the documentary method: namely, its acausal and non-cumulative character. The success of a sociological essay or a sociological monograph is apparent in the fact that it is adopted by other sociologists with other typifications and other “documentary” generalizations and that the next analysis shifts the parameters of that “for which” the empirical or theoretical finding was established: that “for which” it was “evidence” and for which it was made. In other words, even in sociology itself, the foreshortening of typification and occasionality applies: both in the most successful and empirically saturated and in the most speculative and unsuccessful variants.

The “documentary method” is an unavoidable, ubiquitous, irreducible and incorrigible part of everyday life. It is as much an everyday occurrence in research as it is in everyday life, and it is never what distinguishes scientific generalizations from unscientific generalizations. Is it unscientific? This is also not so, since then there would not be any science. Should it to be scientized? No, its practice remains a matter of everyday knowledge. The emperor’s new clothes are expensive and have many fine-sounding names. But they are more than just transparent: they are so transparent that one does not see through anything else, as soon as one wants to look at them oneself.

This multiple generalization placed Garfinkel in front of a dilemma that influences the development of ethnomethodology up to today. The fable of the emperor’s new clothes is easy to tell, as long as one concentrates on the tailors, the king, the counselor and the people: the king loses his legitimacy, the tailors flee, the people begins to have doubts about the monarchy. This is also how it would be in an ideal world, if the social sciences had to acknowledge the share of the documentary method in what they do. It would be like in a fairy tale, as soon as reason has triumphed. But there still remains an unresolved problem: What in fact happens to the child who was the first to yell: “But the king isn’t wearing anything at all!” What does such a child do the next morning?

This is the situation that confronted Garfinkel after he had reworked his 1959 manuscript into the 1962 essay. Starting in 1963, he subsequently developed a new terminology: the terminology of “accounts” and of “accountability”. The “temporal structure” of the “findings” of the experiment becomes “indexicality,” but it was perhaps never again described in so multifaceted a way as in the experimental “findings” of the “documentary method”. And the micro-historical time of self-understanding that is derived from Mannheim’s macro-historical time becomes the “reflexivity” of interaction. But a problem still remained: Would ethnomethodology in its future scientific efforts likewise be subject to the “documentary method”? This question necessarily led to a dilemma. If the generalizations of ethnomethodology likewise take place in “documentary” fashion, it would not be better positioned than the rest of the sociology, which tries in vain to pull itself out of the swamp of everyday sociology by its own hair. If, however, the “documentary method” could be effectively switched off in an ethnomethodological presentation, or at least minimized to the greatest extent possible, would ethnomethodology still then be a sort of “sociology”?

What solution has ethnomethodology found for this dilemma? In fact, since the 1970s, ethnomethodology has been trying to undertake as few “documentary” generalizations as possible in the argumentation of its essays: i.e., to avoid statements like the described occurrence is “typical of,” “points to,” or is exemplary evidence for a larger context or even for the very “social pattern” of a given society, an institution, a distribution of roles, the class struggle or the generational conflict (etc.). In this respect, the greater part of ethnomethodology remains ascetic: The “documentary method” does not play any role in the published protocols, in the published analytical steps and in the arguments. And precisely where one could expect a “documentary” generalization in other sociological essays—namely, in the conclusion—one finds a conscious pushback in the other direction and a retreat to what can be indisputably established about vocabulary, method and subject. There can be something frustrating about this asceticism for other readers. But, as shown, there are good reasons for it: Every “documentary” generalization would be as everyday as the generalizations of other sociologies or other everyday sociological assertions. If what is at issue were such “documentary generalizations,” ethnomethodology would be a sociology like any other. This makes the cases in which an ethnomethodology in fact undertakes such “sweeping claims” (e.g., Bellman 1981) all the more surprising. But precisely in these cases, the following question arises for the recipient: Is this still “ethnomethodology” or just an ethnomethodologist that has exceptionally undertaken a sociological or anthropological generalization?

Does this mean that EM has managed to do without the documentary method? Of course not. This method possesses its everyday, unavoidable, ubiquitous, irreducible and incorrigible room for maneuver and place also and precisely in EM. In written publications, great effort is made to restrain and minimize this room for maneuver. But in the data session, the “documentary method” is given free rein among all those who take part in the interpretation: as a key means of exchanging ideas, in order to uncover the “reflexivity” of the events being studied.

In this way, as is anticipated in exemplary fashion in the fourth chapter of the Studies, ethnomethodology gets divided into the two typical aspects of every process of professionalization and every scientific study: in “work of mediation” and “work of purification” (Latour 1993). In the work of mediation, the documentary method remains the key means—the medium—of ethnomethodological research. In the classical ethnomethodological publication, it is the quantity to be purified: no “documentary” reasoning please—above all, not in the conclusion. In this respect, the Studies themselves are not yet purified, and this is perhaps the reason for their continuing fascination. They were situated at the threshold of the separation from the “documentary method,” but all of the studies can be read in a “documentary” fashion or, as others would formulate it, over and over as “case studies” for something new.

Ethnomethodology did not become a sociology like any other, but it paid a high price in order to avoid this: The more scientific it wanted to be, the more distant it became from the common interests of everyday sociology and professional sociology. Whereas the data session is situated in the middle of these common interests and methods—which is perhaps one reason why the data session can trigger such unexpected feelings of happiness, at least for me—the publications, and especially the conclusions, make great efforts to preserve an artificially thought-through estrangement. But do not the common interests of everyday sociology and professional sociology remain precisely what interests us in everyday life about sociology and in sociology about everyday life? And do they not remain in ethnomethodology itself, like in everyday life, the privilege of the most unprivileged, the shabbiest, the most ubiquitous of all methods?

I think that the most fruitful research and essays of ethnomethodology came into being by way of both possibilities: in the attempt to focus on the formal course of everyday interactions and to recognize its structuring; and in the insecure leap from a meticulous analysis of individual interactions to their documentary generalization—whether in the improvisation of a data session or in a published text. This would be reason enough to study anew the documentary method itself—and to do so indeed by way of ethnomethodological and conversation-analytical investigations—and for this purpose, to draw, for example, on those fields of application of Garfinkel’s experiment that we mentioned at the outset: counseling, divination, documentary film, psychotherapy, participant observation, and hermeneutics. But, this time, not in order to deceive the subjects about what is supposed to be studied, but rather to put the focus on the ethnomethodological question: How is a documentary generalization brought about in the everyday circumstances of counseling, divination, documentary-filmmaking, psychotherapy, and hermeneutics? Garfinkel’s presentation would no longer, then, be the underlying theory of these different accomplishments (like, for instance, in Zeitlyn 1990), but would rather be just in fact a first step or, to put it as Garfinkel does: “an approximate description of one process”. Or it would turn out that the “documentary method” is only a possible cipher for a universal linguistic faculty: whether that of “typification,” of “exemplification” or of the transitions between reference, predication and linguistic explanation. Everybody else’s remedy, and ethnomethodology’s “bête noire”:

The documentary method of interpretation is a convenient gloss for the work of local, retrospective-prospective, proactively evolving ordered phenomenal details of seriality, sequence, repetition, comparison, generality, and other structures. The gloss is convenient and somehow convincing. It is also powerful in its coverage; too powerful. It gets everything in the world for practitioner/analysts. Its shortcomings are notorious: In any actual case it is undiscriminating; and just in any actual case it is absurdly wrong. (Garfinkel 2002: 113).



  1. Bellman, B. L. (1981). The paradox of secrecy. Human Studies, 4(1), 1–24.CrossRefGoogle Scholar
  2. Berger, P. L., & Luckmann, T. (1966 [1967]). The social construction of reality. A treatise in the sociology of knowledge. New York: Doubleday. (Paperback: New York: Anchor Books).Google Scholar
  3. Bohnsack, R., Nentwig-Geesemann, I., & Nohl, A.-M. (2001). Die dokumentarische Methode und ihre Forschungspraxis. Grundlagen qualitativer Sozialforschung. Opladen: Springer VS.CrossRefGoogle Scholar
  4. Garfinkel, H. (1959). Common sense knowledge of social structures. In 4th world congress of sociology. Stresa (Italy), manuscript.Google Scholar
  5. Garfinkel, H. (1962). Common-sense knowledge of social structures: The documentary method of interpretation. In J. M. Scher (Ed.), Theories of the mind (pp. 689–712). New York: The Free Press.Google Scholar
  6. Garfinkel, H. (1967). Studies in ethnomethodology. Englewood Cliffs, NJ: Prentice Hall.Google Scholar
  7. Garfinkel, H. (1973). Das Alltagswissen über soziale und innerhalb sozialer Strukturen. In Arbeitsgruppe Bielefelder Soziologen (Ed.), Alltagswissen, Interaktion und gesellschaftliche Wirklichkeit (1), Symbolischer Interaktionismus und Ethnomethodologie (pp. 189–262). Reinbek: Rowohlt.Google Scholar
  8. Garfinkel, H. (2002). Ethnomethodology’s program. Working out Durkheim’s aphorism. Lanham: Rowman & Littlefield.Google Scholar
  9. Latour, B. (1993). Wir sind nie modern gewesen. Berlin: Akademie Verlag.Google Scholar
  10. Mannheim K. (1951[1921/22]). On the interpretation of ‘Weltanschauung’. In K. Mannheim (Ed.), Essays on the sociology of knowledge (pp. 33–82). London: Routledge & Kegan Paul.Google Scholar
  11. McHugh, P. (1968). Defining the situation. Indianapolis: Bobbs-Merrill.Google Scholar
  12. McHugh, P. (1969). Defining the situation: The organization of meaning in social interaction. Indianapolis: Bobbs-Merrill.Google Scholar
  13. Zeitlyn, D. (1990). Professor Garfinkel visits the soothsayers. Ethnomethodology and Mambila divination. Man (N.S.), 25, 654–666.CrossRefGoogle Scholar

Copyright information

© Springer Nature B.V. 2019

Authors and Affiliations

  1. 1.SiegenGermany

Personalised recommendations