Maynard: "An Intellectual Remembrance of Harold Garfinkel"

An Intellectual Remembrance of Harold Garfinkel: Imagining the Unimaginable, and the Concept of the “Surveyable Society”
Douglas W. Maynard
Human Studies
A Journal for Philosophy and the Social Sciences
ISSN 0163-8548 Volume 35 Number 2
Hum Stud (2012) 35:209-221 DOI 10.1007/s10746-012-9226-0
Your article is protected by copyright and
all rights are held exclusively by Springer Science+Business Media B.V.. This e-offprint is for personal use only and shall not be self- archived in electronic repositories. If you
wish to self-archive your work, please use the accepted author’s version for posting to your own website or your institution’s repository. You may further deposit the accepted author’s version on a funder’s repository at a funder’s request, provided it is not made publicly available until 12 months after publication.

Hum Stud (2012) 35:209–221 DOI 10.1007/s10746-012-9226-0
An Intellectual Remembrance of Harold Garfinkel: 
Imagining the Unimaginable, and the Concept
of the ‘‘Surveyable Society’’

Douglas W. Maynard

Published online: 8 May 2012
Ó Springer Science+Business Media B.V. 2012

In October 2011, a commemoration of Harold Garfinkel’s life and achievements was held at the University of California, Los Angeles, where Garfinkel had been a professor for the duration of his career. I presented a version of this intellectual remembrance at that event. My purpose was and is to recount aspects of a collaborative research experience with Garfinkel when he visited my home department (Sociology) at the University of Wisconsin during the spring semester of 1990, and to convey something of his remarkable presence and intellectual personage. Assuming Garfinkel is known mostly through his nonpareil writings and contributions to the social sciences, I mean to provide a different view, something a little closer and more personal but still about his ultimate and deeply sociological concerns.

The visit in 1990 was the longest of three separate occasions over the years after his 1987 retirement from UCLA when he was free to spend time in our department. During this particular visit, he taught a graduate seminar in ethnomethodology and also made himself available to students and faculty for consultations. Garfinkel in many ways could be a contradictory person. In an obituary for Garfinkel published in London’s The Guardian newspaper, Mike Lynch (2011a) writes,

Garfinkel was a remarkable and volatile character who kept his interlocutors off balance with startlingly original arguments, unique examples and puzzling turns of phrase. During seminars and tutorials, he would ponder questions visibly, almost theatrically, pausing for an inordinate amount of time while the novices waited on his words. Often, he would break the silence with enigmatic pronouncements and anecdotes that left his students with problems to work out.
D. W. Maynard (&)
University of Wisconsin, Madison, WI, USA e-mail:
D. W. Maynard

Indeed, in our seminar, Garfinkel could sometimes talk almost in a monologue for two hours, where it was difficult to get in a word edgewise, even to ask a question. His presence was a reminder of how the philosopher Malcolm (1962: 27) depicts the lectures of Ludwig Wittgenstein: ‘‘One knew that one was in the presence of extreme seriousness, absorption, and force of intellect’’. And, we learned a great deal in that seminar. In the UW Department of Sociology, we have tapes and transcripts of those seminars and our individual meetings with Garfinkel as part of a growing archive in what we have aptly named the Harold Garfinkel Laboratory for Ethnomethodological Research. Anne Rawls is compiling a much larger archive in Boston and our hope is eventually to make the unpublished work and conversations publically available.

Along with the seminar and individual meetings, something else endures strongly from that semester-long visit—a collaborative spirit, reflecting Lynch’s (2011b: 939) observation that Garfinkel ‘‘often was very generous with time and credit and... repeatedly emphasized that the legacy he wanted was not a monument to him as a great theorist, but a continuation of the unfinished business of ‘the company’’’. This essay recounts Garfinkel’s intellectual generosity, and his concern specifically to address the unfinished business of understanding the practical basis of a major form of social scientific inquiry, survey research. After all, when writing about the documentary method of interpretation, Garfinkel (1967: 96–100) had already considered the ‘‘practical exigencies’’ to which interviewers and respondents accommodate in the actual course of a survey. Interviewers, through ‘‘managing the stepwise course of the conversation,’’ adjust their ‘‘questions in profitable sequence while retaining some control over the unknown and undesirable directions in which affairs, as a function of the course of the actual exchange, may actually move’’. Through their methods of handling exigencies, investigators make possible the ‘‘reasonableness’’ of the findings that survey research generates.

Fortuitously, when Garfinkel was visiting Madison I was just embarking on what has turned out to be a longtime collaboration studying the exigencies of interaction in the survey interview with my colleague, the sociologist and survey methodologist Nora Cate Schaeffer. At the time, Schaeffer (1991) was just writing about the relevance of conversation analytic studies for survey research. Garfinkel was intrigued by our burgeoning efforts and enthusiastically joined a number of our conversations over the weeks of the semester as we talked productively about the survey enterprise and how to study it ethnomethodologically and with the tools of conversation analysis. He was an incredibly good listener, attending what we had to say about the survey with great intensity and finding in our remarks the possibilities for deeper understanding of the practices that make the interview happen. He often would elaborate matters with great insight and with inimitable clarity and articulation of thought. Eventually we landed on a pivotal concept, what Garfinkel proposed as the ‘‘surveyable society’’. In this essay, I will map how we got to this concept, what it meant and could mean, and how it leaves us with some ‘‘unfinished business’’. In addition, I will draw on his seminar talks, and an unpublished paper by Garfinkel that provides a particularly good reminder about the nature of his intellectual and sociological legacy.


An Intellectual Remembrance of Harold Garfinkel 211
Studying the Survey Interview: Beginnings

At the time of Garfinkel’s visit, we were operating in a scholarly context where investigators such as Briggs (1986), Cicourel (1974), Mishler (1986) and others— most recently Suchman and Jordan (1990)—had leveled critiques of the survey because, for example, it poses standardized questions, questions that have their provenance in the survey researcher’s own theories, concerns with social policies, and ideas about what is important to know. Therefore, it does not allow for respondents to express matters in their native, or folk, or commonsense ways of understanding the world. These investigators also saw the survey as imposing a structure on the interaction that is somewhat alien to ordinary social life and distorts what can be learned from those whose experiences are the focus of inquiry. We (Maynard and Schaeffer 2000) have our own respectful take on what this critique misses, based in part in our discussions with Garfinkel and comments he made in the seminar. The issue, in his view, was that too many had approached the survey with so-called ‘‘remedial’’ interests in hand, wanting to fix it:
Now if you’re going to get into the remedial enterprise then you can be... disappointed with the workings of things.... Imagine that you’re in fact anthropologists, and you’re looking then as anthropologists to come upon this of the parties in whose midst you’re now working out your inquiries. Okay, they may know what they’re doing, but not to suit you. And after all you know better than they,... because if they’re doing something crazy like adminis- tering poison to a chicken in order to make decidable whether they’re in fact going to recover from this illness, you can already see that it’ll get them no where. What they need to do is get to the nearest airport, fly to New York and see a specialist (Seminar, 3/7/90).

The reference to the famous study by Evans-Pritchard (1937) is evocative because, instead of engaging in a ‘‘remedial enterprise’’ and telling the Azande to go New York with their medical concerns, Evans-Pritchard immersed himself in the culture in Ethiopia where the group lived and came to understand how the members oriented to contradictions (according to Western standards) inherent in using a poison oracle for deciding both mundane and momentous issues in their daily lives, such as determining the causes of illness. Analogously, what Garfinkel asked (4/9/ 90, 39:001) was not how to remediate survey research, but rather, ‘‘How does the [survey] industry get the job done?’’ He was avid to get immersed in the work that goes on in the industry and, from the practitioners’ point of view as embodied in their practical actions, to learn what the work is. Elsewhere of course, Garfinkel (2002: 175–176) called this the ‘‘unique adequacy’’ requirement of ethnomethod- ological methods.

So then, Garfinkel said, ‘‘What we need to be doing is putting Nora on the dock—tell us how this thing works’’. Informed by her extensive training and
1 Unless otherwise noted, dates refer to occasions of recorded discussions among Garfinkel, Schaeffer, and Maynard. Time notations are the approximate point in the discussion where a particular quotation is situated.
D. W. Maynard

experience in the survey enterprise, Schaeffer spoke at length about the problems of interviewing—measurement and bias, attaining understanding between interviewers and respondents, time to complete the survey, associated costs, boredom, interactional problems, and more. Survey methodologists have been working on these problems for years. Garfinkel was enamored of what she had to say, and remarked, ‘‘I feel as though we’re running through a beautiful field! Let’s look at this, what’s that, well there’s something over there!’’ He was interested in, his imagination was activated by, as he put it, ‘‘what the industry has to offer as intractable problems.... They seem to be massive and unresponsive to classical ways of dealing with them, and it could be that that’s where our studies could be attractive. Something left over to look at’’ (4/9/90, 57:00).

Imagining the Unimaginable

What Garfinkel really wanted to do with us was discover what the survey interview is as a thing in the world that professionals and laypersons, including researchers, designers, interviewers, and respondents, assemble collaboratively and accountably in their practices for the varied purposes the survey industry may have. But of course, he was constantly admonishing that practices cannot be stipulated, cannot be hypothesized, and cannot be imagined (Garfinkel 2002: 111). One of his ways of putting it was that imagining the features of practical actions ‘‘only gets you imagined practices’’. In other words, practices are only ‘‘discoverable’’ and ‘‘empirically researchable’’ (Garfinkel 2002: 96, 111). To come upon practices requires being ‘‘in situ’’ and getting access to the ‘‘in vivo’’ or lived experience (seminar 2/14/90), although he also recoiled from using these slogans too wantonly.

Paradoxically, however, despite Garfinkel’s frequent admonishments against trying to imagine practices, he in fact did just that with us. In Studies in Ethnomethodology, Garfinkel (1967: 38) famously quoted Herbert Spiegelberg to the effect that we need aids to the so-called sluggish imagination: demonstrations that ‘‘produce reflections through which the strangeness of an obstinately familiar world can be detected’’. In our conversations, besides talking about the survey industry and focusing empirically on interview data that Nora Cate and I had collected, he suggested many possibilities for us, activating our sluggish imagina- tions in ways that would make our inquiries both possible and richer.

One day (4/2/90, 48:00), as we were talking about the interviewer-respondent relationship, Garfinkel brought up his experience with scientist Gerald Holton and learning from Holton that methods don’t travel between labs, unless the receiving lab adjusts the methods to make them operate in the local lab. He elaborated Holton’s point, ‘‘just to have a slogan to work with,’’ with the notion of ‘‘custom fitting’’ methods to their local circumstances. And after telling about Holton, Garfinkel went on (my emphasis):
Well let me give you another example. Years ago I was visiting John O’Neill [now Professor Emeritus, York University Toronto]. We went to a department store one day. He loves department stores, because the goods there are,

An Intellectual Remembrance of Harold Garfinkel 213
because they are industrially manufactured, are all alike. Nevertheless what happens is, that the customers in the flow go through to find ‘‘what they’re looking for,’’ unquote. That is to say, they’re going to have to go to the counter, and there, in the array of sweaters, all of them are green, all of them are red, all of them in the small size are all in that small size. So what he proposes, just imagine there at the counter, they’re finding just what they’re looking for. And to be finding just what they are looking for means they are, with just this sweater, if it’s somebody they’re buying it for, then they’re buying just the thing that is appropriate for that particular somebody off the array that look like that. So here’s more of that custom fitting. (4/2/90, 50:00)
There it is: ‘‘Just imagine there at the counter....’’ An upshot of the story was to use the imagination to consider how, just as there is custom fitting of the particularities of a person to the generic sweater, the survey and the work of doing the survey on a case by case basis obtains for the survey industry, as Garfinkel went on to say, the ‘‘great generically recurrent features of whole populations’’.
These three-way discussions would eventually lead to a resonant phrase that captured our developing sense of the phenomena we were studying. Having posed the possibility of getting access to what he also termed the ‘‘physiology’’ of whole populations in terms of the practices that provide for the findable and documentable features of those populations, Garfinkel suggested the concept of the surveyable society.
The Surveyable Society
The concept came to be articulated in various ways. For example, Garfinkel offered this explication:
Here... is the industry, its requirements—speak of the work, their work, of it being everywhere in the society thought of now as the innumerable many, endlessly many agencies that must deal administratively with gathering huge amounts of material in short order at pre-decidable costs in such a way that they can point to and justify their procedures as adequate, effective procedures... and that means that the uses that they’re being gathered for are very very critical. And that the society could be such that persons collaborate to lend to their affairs their reproducible character (4/25/90A; 24:00).
There is a lot packed into this statement, including what we (Maynard and Schaeffer 2000: 323) have glossed as the ‘‘ubiquity’’ of the survey, or the enormity of the survey industry when we consider its deployment and the expense of its deployment around the world, from country to country to country, east, west, north south, on a day-by-day basis. In his statement, Garfinkel also strikes a note about collaboration in the survey process whereby persons—interviewers and respondents as well as other parties to the survey industry—act in concert with one another, just here, just now, to make available what emerges through the repetitive enactment of
D. W. Maynard
Author's personal copy
question-asking and -answering in detail: the countable, code-able, quantifiable and associational characteristics of the very society in which they are members.
At one of our meetings (4/16/90B, 15:45), we had been listening to a tape of an interview and we heard a very curious interchange in which a respondent first gives one answer and then, a few seconds later, a different answer to the question she is asked. The question is at lines 1–3 below, and immediately after the interviewer finishes, the respondent offers an answer (line 4). This, however, turns out to be a violation of the interview protocol, which requires that respondents not answer an inquiry until the interviewer gives the answer categories. Thus, at line 5, the interviewer does what she is supposed to do in circumstances when a respondent answers prematurely, and that is to go ahead and read the full set of answer categories (lines 5–6; FI = female interviewer, FR = female respondent):
Roadblock Survey
But then, after the interviewer did read the categories, a rather long repair sequence ensues, starting at line 7 and continuing through lines 13. Subsequently, FR gives the answer (line 14, ‘‘somewhat’’) that is different from what she had first stated (‘‘a little’’). As we listened to this several times, we were trying to figure out how or why she could switch categories so readily. Garfinkel remarked, ‘‘Well she was offering her opinion. She was being asked and she was offering her opinion’’. He went on with another story:
The craziest one, we have- Arlene2 and I have this thing that happened at the time of Kennedy’s assassination. There was this guy- a young guy was being
2 This is Arlene Garfinkel, Harold’s wife.
Author's personal copy
An Intellectual Remembrance of Harold Garfinkel 215
interviewed. You know on TV. He’s asked, ‘‘Well what do you think of this assassination’’. And he’s going through something like this, ‘‘Well uh uh, I don’t know, uh well I don’t know. But that’s MY opinion’’. (4/16/90B, 21:30)
After we laughed heartily at the story, Garfinkel put its upshot in relation to our concerns in this way: ‘‘Well when it’s ‘will you do this for me,’ it’s entirely appropriate to do that shtick’’. He was suggesting, in other words, that the surveyability of the society exists in things like respondents’ sometimes non- substantive compliance with the form of a question–answer sequence on behalf of the survey’s collaborative completion.
What Actually Happens: ‘‘Would-Say’’ as a Phenomenon
Now let us turn away from the mode of imagined practices and anecdotal evocations to examine what happens in the real survey interview—the ‘‘surveyable society’’ in terms of actual practices. If interviewers custom fit their instruments to respondents, and if respondents as well are knowingly complicit in bringing a survey to fruition, then how do these participants complete their tasks in interaction with one another, especially when we know that survey questions often miss their mark in terms of tapping respondents’ native, folk, or commonsense understandings?
The phenomenon, investigated in a master’s thesis by Horgan (2004), is what we are calling ‘‘would say’’. When respondents are given a question, and they struggle to answer it using the survey categories, ‘‘would say’’ is a device that interviewer or respondent or both may invoke. Respondents often offer ‘‘would say’’ as a way of approximating an answer—distancing it a little from the instrument’s categories. Or interviewers, in the face of a struggling respondent, will solicit such approximations by probing with the device. ‘‘Would say’’ does not get used when the respondent can and does straightforwardly answer the question.
The illustration below comes from a single question on a rather long interview that the University of Wisconsin Survey Center formerly conducted called The Wisconsin Continuous National Survey. It covered such topics as politics, economic expectations, fairness of taxes, performance of governmental bodies, mental health, and others.
The illustrative question is from the section of the interview that implements a depression scale with several interrelated questions. This is how the question would look on the interviewer’s Computer Aided Telephone Interview (CATI) screen:
D. W. Maynard
Author's personal copy
Here are two relatively unproblematic iterations of the question (MI = Male Interviewer, MR = Male Respondent, etc.):
The respondents produce relatively simple one-word answers. However, when a respondent expands her answering beyond a single word or number, a regular practice is to use a ‘‘would say’’ phrase in the process.
FR first acknowledges the question about ‘‘how many days’’ with the ‘‘oh,’’ and then pauses before suggesting ‘‘five’’ with a ‘‘would say’’ preface. The answer is followed by an account showing that her answer is less precise than the question may have presumed. Whereas the question asked ‘‘how many days’’ her sleep was ‘‘restless,’’ she does not ‘‘sleep too good some’’. Although ‘‘would say five’’ suggests an approximate rather than a precise answer, the interviewer accepts it and goes onto the next question. ‘‘Would say’’ answers regularly have such approx- imating accounts built into them, and interviewers most always register them.
In the next example, it is the interviewer who, in probing the respondent’s initial answer (line 4), embeds the ‘‘would say’’ device in a longer phrase, soliciting a choice between the ‘‘one or two’’ that the respondent initially stated. Recalling Garfinkel’s story about buying sweaters, it’s a little like having to choose just one from a select few pre-fabricated colors. There are no blends.
Author's personal copy
An Intellectual Remembrance of Harold Garfinkel 217
After the interviewer’s follow-up question (‘‘which would you think you’d say ...,’’ lines 5–6), the respondent tells the interviewer to ‘‘make it one’’ (line 8), proposing that the interviewer go ahead and register that answer, which he does. This ‘‘make it one’’ deals with the ‘‘would say’’ invitation, and like Garfinkel’s metaphor about buying industrially manufactured goods, she chooses an answer from those on offer. It has a manufacturer’s ring to it. However, her answer is not constructed independently, but rather in relation to the ‘‘would say’’ question. Interviewer and respondent co-produce a codable answer. Of consequence for the survey research enterprise, moreover, it may be that such interactions pose one of what Garfinkel, in our earlier discussion, termed the ‘‘intractable’’ problems of doing surveys. Invocation of ‘‘would say’’ and related devices may increase measurement error, decreasing reliability and accuracy in answering and heightening interviewer variability (Schaeffer and Dykema 2011).
‘‘Would say’’ formulations and versions thereof have many orderly properties that we are presently investigating. One more brief example is that ‘‘would say’’ enables respondents to formulate answers in terms of contingency. Answering a question about whether federal income taxes are excessively high, about right, or low, a respondent says, ‘‘I guess if those are the three choices, I would say high’’. Overall, would-say devices enable interviewers to record answers on the computer officially or according to the protocols of the survey instrument, despite facets of interactional co-construction that the devices index. Thereby, the answers in their digital form become the respondent’s individual or unilateral answers for the survey. This is not an ironic comment. In terms of what Garfinkel called custom fitting, ‘‘would say’’ devices facilitate the production of answers that are precise enough for the interview while accountably in the interaction showing the relation between that answer-for-the-survey and the answer at which a respondent, given the question, potentially might arrive if not constrained by the instrument’s own categories.
And that is the point. Respondents are constrained, because of the needs of the survey and standardization of measurement, which is something interviewers teach and respondents come to know in situ as the survey is being conducted. This raises
D. W. Maynard
Author's personal copy
another Garfinkelian point, one that we also discussed during our meetings, having to do with ‘‘instructed’’ and ‘‘instructable’’ actions’’.3 The survey instrument by itself is a set of ‘‘docile’’ instructions—questions and answer categories to be read verbatim during the interview with the explicit and implicit directive to respondents to answer only in the way that the survey permits (Fowler Jr. and Mangione 1989). But of course, when the survey becomes incarnate and embodied it means that interviewers and respondents have to find their way through the question and category sets in such a way as it could be said that the docile survey instrument, the skeleton that they start with, in addition to being the generator of questions and answers, is also the product or outcome of participants’ incarnate practical actions in questioning and answering in detail. Through what interviewer and respondent do in concert with one another, using would-say and other practices, the skeleton acquires its flesh.
Drawing on Horgan’s (2004) thesis, we are doing further investigations of the ‘‘would say’’ and related devices as practices that Garfinkel anticipated when discussing the surveyable society. Independently, Halkowski (2012) has been investigating similar phenomena in medical settings, having to do with the metrics that doctors and patients use for questions about smoking and drinking, and what he calls ‘‘approximation elicitors,’’ would-say being one of the elicitor devices. So the practices of would-say may be robust across settings in a context sensitive kind of way such that, especially in an ‘‘interview society’’ (Gubrium and Holstein 2002) and in an age of online administration, the concept of the surveyable society has wider tentacles than we might have imagined. The surveyable society is not just about the survey industry, in other words, but also about how we collectively measure individual attributes in medicine, in many other institutional settings, and by many different media.
The Ethnomethodological Legacy
A ‘‘premier aim’’ of ethnomethodology, in Garfinkel’s (1996: 7, 2002) terminology, has been to ‘‘respecify’’ the work of the social as well as the natural sciences—to capture the concrete details of the practices and methods existing in the circumstances of ordinary activities that undergird objectivized, procedural inquiry. Garfinkel speaks of these practices and methods as the ethnomethodological alternates, referring to the structures of social action that make the doing of science possible in the first place and that have untold effects on what the science is as well as what its findings are. Analytically to capture these methods and practices in studies of interviewing, one has to follow interviewers and respondents as they alternate away from the docile format of the survey instrument and engage the embodied work of interactional conduct—incarnate instrument use through which facts are produced—in real time (Maynard and Schaeffer 2000).
With the ‘‘would say’’ phenomenon, we are analyzing the most ordinary if not minute bit of talk in a telephone survey interview—those moments when a
3 See also Garfinkel (2002: 105–106). 123
Author's personal copy
An Intellectual Remembrance of Harold Garfinkel 219
respondent is struggling with the impositions of precision in a question, and the interviewer and/or the respondent invoke a would-say phrase as a tactic for generating an answer that is good enough for this survey’s measurement task while also preserving for the here and now the accountability of that answer. Such local accountability is stripped away as the interviewer enters a category on the computer. This official response subsequently takes on a life of its own that is exogenously accountable in that it otherwise has been produced by way of the instrument that the investigators designed for their attaining their research goals. Necessarily left behind are ethnomethodological struggles with questioning and answering as sites of social action and as also as sites of order and organization, at infinite depths of detail.
A matter that social science has yet to fully recognize is that there simply is no time out from the ethnomethodological alternates and commonsense reasoning at any level of a practical endeavor, even when the endeavor (e.g., survey research) has become extremely sophisticated by way of its developed and developing technol- ogies. Consider the trope in the subtitle of a recent book that Duncan Watts (2011) has authored, the title of which is Everything is Obvious Once you Know the Answer. It is actually an extremely insightful book that every sociologist should read, but in my view it has a misleading subtitle: How Commonsense Fails Us. Ethnomethod- ologically, it is not the case that commonsense fails us. Commonsense cannot fail us because if it did we would not have society. Ethnomethodologically, the task is not to document the failures of commonsense. Rather the job is to understand the achievements of commonsense competence at every point at which it is manifest: how through commonsense actions we get our society done and how, more specifically, we produce our scientific and other types of professional work. Given this fundamental focus on how social life and social inquiry come to be, it would seem that sociology could benefit more widely from ethnomethodological inquiry.
That professional social science might be more interested in ethnomethodology than it has been is something that Garfinkel felt and conveyed in different kinds of ways. In an unpublished manuscript, he (n.d.) suggests that from the standpoint of the profession, the organization of ordinary activities—the ethnomethods or practices—are trivial phenomena. When compared to the enormous significance that we accord to gangs in society, role relationships, family structures, bureau- cracies, economic conditions, and the like, it would seem that the organization in everyday activities is of miniscule importance. But Garfinkel (n.d.:98) wanted to reverse that attitude—to claim magnitude for EM studies—and he did so by proposing what he called a ‘‘figure of speech,’’ a kind of metaphor, where ‘‘the missing what’’ dwarfs the typical concerns most central to sociology:
With respect to the circumference of the earth, Mt. Everest has the prominence, importance and relevance of a pockmark on a billiard ball. The missing what that ethnomethodologists have been at the work of discovering is to available lay and professional versions of organization as the earth stands to the presence of Mt. Everest. Orderings of ordinary activities are unimaginably extensive phenomena; they are essentially other than we do imagine or could ever imagine them to be, and they await discovery.
D. W. Maynard
Author's personal copy
While this metaphor returns to the eschewal of imagination and the need for discovery, it is worth considering that ethnomethodological practices, analytic alternates, commonsense actions, or ‘‘formal structures of practical actions’’ (Garfinkel and Sacks 1970) in relation to the problems of gangs, the economy, stratification, and so on, are as the earth is to the presence of Mt. Everest. Society exists concretely, pervasively, and globally in ways that transcend eruptions in its manifold spheres. So when stating, for example, that the would-say phenomenon is ordinary and minute, it may be a misleading statement, for it and its allied forms of orderliness are massive in terms of their presence for the produced structure of the survey interview, the survey industry, and other industries and professions where social measurement is an issue. Moreover, they affect what such enterprises generate as sociological data, administrative data, economic data, and other information. This massive presence of orderly, commonsense actions also means that there is an infinite variety of practices whose presence for our everyday, every night, ordinary, occupational, workaday, playful, and other social worlds have yet to be investigated and understood. There is a huge legacy of ‘‘unfinished business’’ that Garfinkel has bequeathed to all of us.
At various points in our discussions with Garfinkel, we were amused as he made fun of or mocked the familiar criticisms of ethnomethodology of which he was only too keenly aware.4 On one occasion, he quoted his critics: ‘‘They say, ‘You’re just turning away from the achievement of the science’. Of course that’s not true. Look, what we’re doing is brainstorming’’. (This brainstorming is what the title of this essay means to convey: ‘‘imagining the unimaginable’’.) Garfinkel continued, ‘‘Okay so we’re looking to see well what the hell could be going on that would be interesting to do AND would be of use to the profession’’. Despite the controversy ethnomethodology raised, and the rejection Garfinkel felt, his aims always were to contribute to the enterprise. Elsewhere, it can be noted, Garfinkel (2002: 114, original emphasis) states, ‘‘Ethnomethodology is applied ethnomethodology’’. This is not the same thing as saying that ethnomethodology has a remedial stance, for which he critiqued the received qualitative assessments of survey research. Whereas a remedial investigation suggests the reparability if not replacement of formal inquiries and standardization, applied studies in ethnomethodology mean acquiring expertise in such a way as to explicate usefully the taken-for-granted underpinnings of such inquiries. Garfinkel (4/2/1990, 53:30) was vitally interested in how, with such acquired expertise, ethnomethodology could make a relevant difference for the endeavors it studies: ‘‘What we’re doing is not the cultist ethno that’s been charged with only writing for each other with the ideas. You know, after all, it’s sociology that we’re doing’’.
4 For a compendium of such criticisms, see Rawls (2002: pp. 50–51). 123
Author's personal copy
An Intellectual Remembrance of Harold Garfinkel 221
This quotation is something of a rare understatement by Garfinkel: It’s sociology that we’re doing. There is no doubt that Garfinkel’s sociology is a wonderful and awesome intellectual bequest to the social science disciplines, and one that will unfold in manifold ways for years to come.
Acknowledgments My thanks go to Steve Clayman, Virginia Gill, Tim Halkowski, and Nora Cate Schaeffer for extremely helpful comments on an earlier version of this essay.
Briggs, C. L. (1986). Learning how to ask: A sociolinguistic appraisal of the role of the interview in social science research. Cambridge: Cambridge University Press.
Cicourel, A. V. (1974). Theory and method in a study of argentine fertility. New York: Wiley. Evans-Pritchard, E. E. (1937). Witchcraft, oracles, and magic among the Azande. Oxford, UK: Clarendon
Fowler, F. J. Jr., & Mangione, T. W. (1989).
Standardized survey interviewing: Minimizing interviewer-
related error. Newbury Park, CA: Sage.
Garfinkel, H. (1967).
Studies in ethnomethodology. Englewood Cliffs, NJ: Prentice-Hall.
Garfinkel, H. (1996). Ethnomethodology’s program.
Social Psychology Quarterly, 59, 5–21.
Garfinkel, H. (2002).
Ethnomethodology’s program. Lanham, MD: Rowman & Littlefield.
Garfinkel, H. (n.d.). Sources of issues and ways of working: An introduction to the study of naturally

organized ordinary activities. Unpublished manuscript.
Garfinkel, H., & Sacks, H. (1970). On formal structures of practical actions. In J. D. McKinney & E.

A. Tiryakian (Eds.), Theoretical sociology (pp. 337–366). New York: Appleton-Century Crofts. Gubrium, J. F., & Holstein, J. A. (2002). From the individual interview to the interview society. In J. F. Gubrium & J. A. Holstein (Eds.), Handbook of interview research: Context & method (pp. 3–32).
Thousand Oaks, CA: Sage Publications.
Halkowski, T. (2012).
Approximation elicitors and accountability in the pursuit of amounts. Unpublished
manuscript, University of Wisconsin, Stevens Point.
Horgan, M. (2004).
‘What Would You Say?’ Some practices for answering questions in telephone survey
interviews. Unpublished master’s thesis, Department of Sociology, University of Wisconsin,
Madison, WI.
Lynch, M. (2011a, July 13). Harold Garfinkel obituary: Sociologist who delved into the minutiae of daily

life. The Guardian, London.
Lynch, M. (2011b). Harold Garfinkel (29 October 1917-21 April 2011): A remembrance and reminder.

Social Studies of Science, 41, 927–942.
Malcolm, N. (1962).
Ludwig Wittgenstein: A memoir. New York: Oxford University Press.
Maynard, D. W., & Schaeffer, N. C. (2000). Toward a sociology of social scientific knowledge: Survey

research and ethnomethodology’s asymmetric alternates. Social Studies of Science, 30, 264–312. Mishler, E. G. (1986). Research interviewing: Context and narrative. Cambridge: Harvard University
Rawls, A. W. (2002). Editor’s introduction. In H. Garfinkel (Ed.),
Ethnomethodology’s program (pp.
1–76). Lanham, MD: Rowman & Littlefield.
Schaeffer, N. C. (1991). Conversation with a purpose—Or conversation? Interaction in the Standardized

Survey. In P. P. Biemer, R. M. Groves, L. E. Lyberg, N. A. Mathiowetz, & S. Sudman (Eds.),
Measurement errors in surveys (pp. 367–391). New York: Wiley.
Schaeffer, N. C., & Dykema, J. (2011). Response 1 to Fowler’s chapter: Coding the behavior of

interviewers and respondents to evaluate survey questions. In J. Madans, K. Miller, A. Maitland, & G. Willis (Eds.), Question evaluation methods: Contributing to the science of data quality (pp. 23–39). Hobken, NJ: Wiley.
Suchman, L., & Jordan, B. (1990). Interactional troubles in face-to-face survey interviews. Journal of the American Statistical Association, 85, 232–241.
Watts, D. J. (2011). Everything is obvious once you know the answer: How common sense fails us. New York: Crown Business.