Posts Tagged ‘ethnography’
What’s the difference between journalism and sociology? (I ask as someone who’s gotten article reviews that said my works reads too much like journalism).
I just finished, at Gabriel Rossman’s recommendation, Dreamland: The True Tale of America’s Opiate Epidemic, which is an incredibly ambitious (and beautifully executed) work of journalism by Sam Quinones. It won the National Book Critics Award for General Nonfiction and has gotten a lot of buzz from all sorts of people. It really is necessary reading: I feel like I finally have some sense of how and why the opiate epidemic is happening, as well as what we might be able to do about it.
And after I finished the book, I thought, how is what I read not simply a mixed-methods work of sociology, using interview and comparative-historical work? I’m not sure I have an easy answer—this is a loose set of categories—but if I had to make the distinction, I’d say that sociology suggests generalizable categories, mechanisms, or causal accounts. We’re looking for the logic beneath the story.
This gets to some of the grumbling I’ve heard about Matt Desmond’s book, or Alice Goffman’s, or many other works within the sub sub category of urban ethnography. These books and articles (but they’re usually books) are often criticized for simply being stories, or, in other words, not having clear theoretical payoffs. That’s certainly not always true of ethnography, or even urban ethnography. Iddo Tavory’s recent book, Summoned, is very much a theoretical contribution, and, I think, so is Alice Goffman’s, in that she subtly provides a theory of fugitive life that is, in fact, generalizable to other contexts. While Desmond can clearly do high level sociological work, his book appears to me a bit less transferrable, but, of course, I don’t think that’s the point. (And it’s clearly not what the National Book Critics Circle Award committee cared about: it gave Evicted the same award it gave Dreamland the year before).
Now not having some generalizable or transferrable theory might not be a problem of course—Andrew Abbot famously defends these forms of “lyrical” sociology, which he (somewhat confusingly) contrasts to narratives. (He doesn’t mean narrative in terms of the story of a particular community but narrative in terms of causal accounts.)
Could we generalize from Quinones’s book? Not really except inasmuch as we get access to a careful analysis of how various organizational structures happened, wholly unintentionally, to lead to one of the worst public health crises our country has ever seen. It’s a story of how drug marketing, changes in medical practice, adaptations of Mexican drug production and distribution, and the hollowing out of middle America all came together. Yet this isn’t all that different from a certain way of doing comparative historical work, which is to start with a neat empirical puzzle (why is the opiate epidemic so terrible?) and then provide a compelling empirical answer. The difference from sociology is that there’s usually some theorist we’re modifying as part of that answer, something like, well, Tilly would have said X and Sahlins would have said Y, but if you combine them with a little bit of our own magic, you get what explains this social outcome. That’s not what we get in Quinones: we just get the stories. This is another difference from urban ethnography: even if the authors aren’t as explicit about their sociological upbringing, even if they’re the most unrepentant grounded theorists who just figure it out as they go, you can always sense the Goffman (or any other theorist) just a little bit below the surface.
You don’t get that sense of a theorist lurking somewhere beneath the stories in Quinones. It makes the book easier to read, of course, and it makes it—sure—more lyrical too.
I sent the above to Gabriel Rossman, and he pointed out that there is underlying theory in Quinones about, among other things, unintended consequences and social capital. It’s a good point, and it makes me wonder about the difference between evidence of a theory (or even of a theory’s influence) and articulation and development of a theory. So I guess I have to restate what I write above: there is a sense of a theorist somewhere in Quinones—quite a few of them actually, as Gabe pointed out in his thoughts on the book at his blog, but it’s not an implicit theorist we see so much as data that, because it is so well documented, is easily theorized or related to theory.
Gabe argues that “every detail of the book illustrated and illuminated another aspect of sociology” and I think that’s right: it’s part of why I found the book so captivating as well. You really should read his post: it’s an excellent list of evidence of all sorts of sociological theories. Gabe describes themes in his post and I think he’s right to do so , so but themes aren’t the same as theoretical arguments, or, if they are, they’re implicitly there. Of course these distinctions don’t actually matter all that much: we could all agree it’s a great book that helps us understand the opiate epidemic, and just leave it at that. But figuring out the boundaries of a particular category is something both emic and etic within sociology, and, as scholars of boundaries point out, it’s a helpful way to determine (and maintain) who we are and what we do.
Because we start at the level of the social, sociologists tend to think questions of human universals are either irrelevant or wrong-headed. It’s empirically obvious that what appears to be universal usually is not and what might well be fundamental to all humans is generally pretty banal.
Often, but not always. And even if the first few steps in a proof are crushingly obvious, they’re still necessary for the later, more interesting stuff. So what do we need? And why does it matter? I’d suggest four starting points. First, to what degree can we understand humans as fundamentally self-interested? Second, to what degree can we understand them as tribal? Third, to what degree can we understand humans as fundamentally habituating? And beneath all of these, do we have a right to assume human life is fundamentally social?
I don’t have space here to get into all of these, but I hope it’s clear that these arguments have real stakes. For example, much of the hubbub over Jerolmack and Khan’s provocative article, “Talk is Cheap” came from their situationalist assumption about human nature (and, to be clear, even though I disagree with the article, I appreciate the conversations it encouraged, and I’m a big fan of both authors’ projects). The problem with situationalism is that it’s a nuclear bomb to sociology’s structuralist assumptions, including, ironically enough, Khan’s own argument in Privilege. If it’s true that human behaviors are basically situationally contingent (to which ethnographers, fairly enough, have the best access), then we have no idea what St. Paul’s is like the year after Khan left his fieldsite, nor do we have any reason to believe that the students he profiles will maintain the formation they have received. The Bourdieusian architecture his book depends upon would be blown to smithereens. Jerolmack and Khan might respond that their argument is not against habituation so much as that talk is poor evidence of habituation, and it’s a fair enough point that there’s a difference between behaviors and verbal self-descriptions. Yet that difference is not nearly as clean as it appears (what is a verbal self-description but a kind of behavior?) and much of their evidence for their argument is a series of situationalist critiques that are pretty devastating to any form of habituation, however it’s revealed (not to mention that much of the evidence in ethnography is, well, talk, albeit talk within situations in which the ethnographer has an interpretive understanding).
To be clear, social psychologists have been thinking about these questions for a long time, and the “Talk is Cheap” conversation originated in Steve Vaisey borrowing an argument about human universals from Jonathan Haidt. That’s a welcome development (even if I’m not at all convinced by those particular human universals), and it would be helpful to see more sociologists interested in larger (socially contingent) structures thinking about our social psychological assumptions of human action. You could easily think of similar assumptions about humanity that undergirds all sorts of sociological arguments, including boundary-work (tribalism), field position (self-interest, whatever that means), and sociology itself (sociality). Chris Smith has already started thinking about these things in Moral Believing Animals and the much longer What is a Person? (for my money the former is a sharper, cleaner argument). More importantly, the often criminally under-read subfield of social psychology has been asking these questions all the way back to Mead. So it’s not as though these conversations aren’t happening. But I think we would benefit from having more of them.
Alex Stewart and Howard Aldrich have published a thought-provoking piece about anthropologists and ethnography in management research. In “Collaboration Between Management and Anthropology Researchers: Obstacles and Opportunities” in Academy of Management Perspectives, the authors discuss several ethnographies and the institutional environment of the business school.
While anthropologists are employed at corporations, the authors claim that anthropologists are underrepresented among management researchers:
“To document the limited business school market, we examined the doctoral disciplines of faculty in “top” business schools. We found 751 tenure track faculty members in management in the 44 schools that are listed in the “top 25” by at least one of Business Week, The Economist, Financial Times, or U.S. News. Of these faculty members, about 60% obtained their doctorate in management; 16 % did so in psychology; 10 % in economics; and 7 % in sociology; but only 0.1% — one person — in anthropology.” (174)
The authors posit 8 barriers to the inclusion of anthropologists:
“To explore the possible reasons for anthropology’s surprisingly small impact, we draw on recent writings on applied anthropology and the emerging fields of business anthropology and practicing anthropology. Scholars in these fields work on the boundary between management and anthropology and experience the benefits and challenges of an anthropological approach. On the basis of these readings, we identify eight properties of anthropological scholarship that might limit anthropology’s integration into management scholarship. These are: (1) expertise about the remote and exotic, (2) sympathy for the remote and the less powerful, (3) ethnography as a primary data source, (4) challenges of fieldwork access, (5) lengthy fieldwork duration, (6) a tendency to solo authorship, (7) complex, contextualized findings, and (8) a higher value placed on monographs than on journal articles.” (175)
One on-going aspect of ethnographic work is the never-ending reflection and re-evaluation of conclusions made months, years, or decades prior. Retrospection invites extended analysis of findings that were otherwise cut short; it also facilitates shift from a worm’s eye to a bird’s eye contextualization of a case. Michael Burawoy’s “Ethnographic Fallacies: Reflections on Labour Studies in the Era of Market Fundamentalism” offers one such contemplation.*
In this research note, Burawoy re-examines several decades of his participant-observations in workplaces in various nations; he reveals the actual names of his most famous disguised field sites. Looking back, he summarizes six revelations while imparting a warning to those overly invested in the merits of particular methodologies:
From the ethnographer’s curse, therefore, I turn to the ethnographic fallacies that limited my vision of market fundamentalism. First, there are three traps that await the ethnographer who seeks to comprehend the world beyond the field site: the fallacies of ignoring, reifying and homogenizing that world. Second, there are three traps awaiting the ethnographer who fails to give the field site a dynamic of its own: the fallacies of viewing the field site as eternal or, when the past is examined, the danger of treating the present as a point of arrival rather than also as a point of departure; and finally the danger of wishful thinking, projecting one’s own hopes onto the actors we study.
I describe these six fallacies not to indict ethnography but to improve its practice, to help ethnographers grapple with the limitations of their method. No method is without fallacies, it is a matter of how honestly and openly we approach them. Being accountable to the people we study requires us to recognize our fallibility and, thus, to wrestle with that fallibility. The methodological dogmatists, who declare they have found the flawless method and spend their time condemning others for not following the golden trail, are the real menace to our profession.
While catching up on some reading during spring break, I ran across an Journal of Organizational Ethnography article by organizational ethnographer Gideon Kunda. In this article, Kunda’s reflections about his development as an organizational ethnographer seem pertinent to the on-going orgtheory discussion of ethnography. Kunda not only describes how he became drawn to organizational studies (hint: questioning a figure of authority about the differential treatment of patients based on class), but also how he arrived at his topic and research site, generating the now iconic study Engineering Culture.
During his training, Kunda worked on several projects using other data collection methods (i.e., surveys), during which Goffman’s work on Asylums was instructive:
Here once again was a science that starts with ready-made theories, selectively uses them in accordance with interests unrelated to (or even opposed to) the logic and spirit of scientific inquiry, collects data using a method that assumes it knows what and how to ask before encountering the world of its subjects, and disrespects or ignores their complex realities, or for that matter, their feelings about who is studying them and why. What factors effect quality is a legitimate question, if one takes the managerial perspective (although this is not the only perspective that could and should be taken). But in order to answer it, in fact in order to even know how to go about studying it, I began to realize, one has to find ways to collect valid data. And the data, if that was what the facts of life should be called, were found in the richness of the stories I heard and the complexity of the interactions I observed, in people’s sense of who they were and what they were up to, and in their willingness to convey it to an interested outsider. Whether or not all this could or should be ultimately reduced to numbers and statistically analyzed seemed much less important than finding ways to collect, understand and interpret evidence that was respectful of its complex nature. If this was the case, it seemed to me, then the scientific system I was enmeshed in, even by its own standards – the norms of science that demand respect for the empirical world – was woefully inadequate. And worse – its procedures and output were embarrassingly boring, to me at least, when compared to the richness of the world it set out to comprehend.
In conclusion, Kunda states:
Over the years I have continuously noted and wondered about the extent researchers in the early stages of their careers, and graduate students in particular, feel, or are made to feel, that while they are granted the methodological license, and sometimes looseness, of “qualitative methods” (a phrase that often replaces or refers to a watered down version of ethnography), the academic authority system (in terms of funding, supervision, publication requirements and career options) compels them to limit their questions, choice of theory and writing style to those that enhance the chances of approval, funding and quick publication. I encounter again and again the ways that this commitment comes at the expense of a willingness to let fly their own sociological imagination, to cultivate and trust their own interpretive resources and analytic instincts, to respect and develop their innate language and authorial voice, or, for that matter, to risk long-term ethnographic fieldwork.
The issue then is not, or not only, one of competing methods, and to overstate such distinctions is, I believe, to miss my point. Rather, I see my story as an invitation to acknowledge and explore the shared conditions of all scientific claims to knowing and depicting social reality, organizational and otherwise, under whatever theoretical and methodological guise, that together place limits on the depth, insightfulness and indeed the validity of interpretation: the endless complexity of data, the incurable subjectivity of the observer, the fundamental flimsiness of formal method and the prevalence of unsubtle yet often disguised institutional pressures to confirm to standards and ways of thinking outside and often against the pure logic of scientific inquiry.
If I am to formulate a conclusion, then, it is this: the continuing need to devise personal and collective ways – and I have suggested and illustrated some of mine – to release “discipline” from its misguided equation with an institutionally enforced a priori commitment to hegemonic theoretical discourse and methodological frameworks, and to apply it instead to its legitimate targets, the questions for which there can never be a final, authoritative answer, only continuing exploration and debate: What is data, what is a valid and worthwhile interpretation, how does it come about, what are and how to cultivate the personal sources of imagination that make it possible, how to report it and, not least, to what end.
Another major take-away for budding researchers is that peers can offer support. That is, scholarly development is not necessarily a hierarchical transmission of information from mentors to mentees, but the co-production of knowledge with peers.