Through a Painted Window: On Narrative,Medicine,and Method Interview with Arthur W. Frank Conducted by the International Institute for Qualitative Methodology EQUIPP Students,November 16,2005

Methods and Rigor

Let me say two other things that I wrote down. I was teaching Foucault in my theory class, and there are two quotations that I think are very useful for this program ² generally and in thinking about qualitative methods.

One is where Foucault is talking about his method, which has to be very much in quotation marks. He says, “On many points, I am still working, and I don't know whether I'm going to get anywhere” (2003c, p. 246). That's what I love about Foucault, this wonderfully ingenuous: don't know if this one's going to go somewhere. It's always a work of trial. Remember, he said an essay has to be understood in terms of the original meaning of essayer, to try something. You don't know. That's what makes all intellectual work exciting: it can fail, it can not go anywhere. Then he says,

That's very much how I hope you'll understand what I have to say this morning. It's the spirit in which I've offered the interventions I've published in QHR [Qualitative Health Research] as keynotes, or really everything I've published in QHR. I like Foucault's notion of game openings. Those who are interested are invited to join in, but you're also invited to take the parts that you find interesting; it's not a consistent position that you have to buy some kind of system.

The second thing is where Foucault talks about truth. We can get back to this when we get to the question you sent about rigor, but I'll just throw it out now: Truth isn't outside power or lacking in power; truth is a thing of this world. That's the crucial point: Truth is a thing of this world. It is produced only by virtue of multiple forms of constraint. The truth is not out there in itself; it is produced, and it's produced in multiple forms of constraint. And then Foucault (1980) says, “And it,” that is, truth, the idea of truth, the truth that we have produced, “induces effects of power. Each society has its regimes of truth” (pp. 72-73), and here we could add methods texts—they are, in a sense, regimes of truth—the techniques and procedures accorded value in the acquisition of truth, the status of those who are charged with saying what counts as true.

Now, if you go back and read old C. Wright Mills' 1959 The Sociological Imagination, I think there's a clear continuity between what Mills is saying about method in his chapter on abstracted empiricism, as he calls it, and what Foucault is saying. The issue of rigor ³ is very much one of what techniques and procedures are recognized by certain groups as affording the person who speaks the status of one who's charged with saying what counts as true. The fundamental question—the a priori question—about research and method is, Who are you talking to, and what kind of expertise are you claiming for yourself in that speech? and then, according to the procedures of the particular group that you want to talk to, What are they prepared to recognize? But you're always producing truth, and truth always induces effects of power, so after you've asked these questions, you get back to the question, Do I want to induce the effects of power that will accrue to this particular group if I play this kind of expert in their truth games?

That's where I refuse to accept rigor as some freestanding thing that can be guaranteed by a set of procedures. I insist that we have to place those procedures within the group that uses those procedures and that we have to interrogate the purposes of that group as they're involved in some sort of production of effects that Foucault calls power/knowledge. So in that sense I refuse to accept, as Mills refused to accept, method as a freestanding activity. Method is always itself subject to a sociological critique. That goes all the way back to when Foucault says: Truth is a thing of this world.

So I did end up giving my little speech on rigor. We started off on that. Maybe that's a good thing to start with because, in a sense, it's the most complex of these various issues. So having given my little diatribe, let me invite your reaction to that.

Stories and Structure

You were asking me: How does structure fit into narrative? Classism, sexism, racism: When you work with story, do you lose these? Well, if you did, it wouldn't be worth working with story. I mean, one has to say emphatically not; that if these get lost, then stories are not very useful. So the question is, How do you not lose them? Let me make a bit of an argument here.

The first point is that, again, I think this word structure was an unfortunate direction, and the theorists who I like—and I'm thinking particularly of Norbert Elias and Bourdieu—really don't use this word “structure.” You know, what do we actually mean when we talk about social structure? This goes back in some ways to Blumer's critique of Parsons: Like, where is it? You know, I used to joke in my theory classes about society as the “big railway station” view of things, where you: Oh, I'm in the structure now. What exactly is that word about? What does it make claims for? What it tends to make claims for is the reification of a theoretical scheme that describes the whole structure. That having been said, what I think there is out there—and I'm speaking here fairly carefully—is the predictable reproduction of access to forms of capital.

So when I think about your “isms,” I follow Bourdieu, in that society demonstrates predictable reproduction of access to forms of capital. Two of those that are salient here are education and health. We have very robust findings about predictions of educational attainment being based on parental educational attainment. Educational capital, we can show, is reproduced intergenerationally; this is, as I say, the most robust predictor of anybody's educational attainment. Students at institutions like the University of Calgary will all immediately say: Well, I'm the first one in my family. Okay, fine, you are, but there are two things that need to be accounted for. First of all is the population as a whole. This is where we do need to think demographically. Second of all, is your whole cohort advancing in its educational attainment, and will your comparative position within your cohort vary from that of your parents' comparative position in their cohort? Because it's obviously a comparative position that has to be reproduced. Health is where the social determinants stuff is really serious. We know that the biggest determinant of health is affluence and poverty. You want to be healthy, be rich. It's simple. Never mind the vitamins, never mind this other stuff; have money.

So, if we don't have structure, we do have this: these predictable regularities in reproduction of access to forms of capital. For these purposes I'm considering education and health as forms of capital. We also have what Hilde Lindemann Nelson calls damaged identities. When you get to your isms, you're talking about categories of people who are systematically deprecated in the public sphere, whose identities are systematically deprecated, by virtue of their membership in that group: ethnicity, gender, age, disability, other forms of isms. We can talk about those damaged identities and we can talk about those reproductions of access to capital without invoking a concept of structure. You just don't need that to be able to observe that those things happen.

So, How can you address what I'm now calling these predictable reproductions using narrative? is my rephrasing of the question How does narrative help us in addressing these things? I'll suggest three things. The first point, one that I hope will be a contribution of the book that I'd like to write on narrative, is to extend Bourdieu's notion of habitus and suggest that each of us has, as part of ourselves, a narrative habitus. Just as we have a habitus with respect to the kind of things that Bourdieu talks about—we have certain tastes in food, clothing, how we spend our leisure time, the kind of body that we're comfortable inhabiting; these are all matters of what Bourdieu calls habitus, and they're very much issues of taste and comfort, what we feel is right for us and what we then seek to reproduce as conditions of our being—we also have a narrative habitus.

By narrative habitus I mean two things specifically. One is that each of us has acquired through our lives a certain repertoire of stories we know—and there are also a whole lot of stories we don't know—and these stories direct us to have a certain sense of the world and our possibilities and constraints of action in the world. First of all, we have a repertoire of stories that tell us what the world is about and how we ought to act. For example, Marian Gray Secundy, who, unfortunately, died recently, a foremost African American bioethicist, gave very powerful lectures about belief systems—I would call them stories—in the African American community centering particularly on various understandings of the Tuskegee experiments and the pervasive ideas among African Americans that in hospitals they stood to be experimented upon, that they would receive systematically bad care, and they might be made into human guinea pigs, and how this played out in their experiences of health care. Now, that's having a repertoire of stories.

The other point, which is perhaps even more significant, is that our narrative habitus involves a sense of how to tell the story. We learn what's a proper story. We learn techniques of emplotment. We learn what counts as a tellable story. This also crucially affects our sense of the world through what I think social life is really all about, what I think we're really talking about when we invoke this adjective “social”: affiliation. I think that if you're looking for predictors of forms of capital, the generalized predictor is people's affiliations. And the social-determinants-of-health people seem to be substantiating this, the increasing importance of which neighborhood you live in in terms of a determinant of health.

Affiliations are, first of all, things that we're thrown into in Heidegger's sense, and then increasingly they're matters of selection and exclusion: who we choose and who chooses us and who we stay away from and who excludes us. So my second point, then—if the first point is that narrative habitus is the basis of affiliation and affiliation is the strong predictor of access to or exclusion from these forms of capital—is that stories are very much inclusion/exclusion devices. Just as people are included or excluded through the recognition of other forms of habitus—how they dress, their table manners, their speech, the references they throw in, the allusions that they use/recognize/fail to recognize—just as other forms of habitus, as Bourdieu shows exhaustively in Distinction (1984) and later work, are the basis of inclusion and exclusion, I would say that crucially the stories we tell, and the way we tell our stories, are absolutely essential for inclusion and exclusion. It's by the stories people tell and the way they tell them that people are known, recognized, and included or excluded as our kind of person.

My third point is that stories also, though, are reflexive; we can think about the stories we're telling and the effects that they have. Again I line up with Hilde Lindemann's notion of counter-stories (Nelson, 2001). Phrased in Bourdieu's terms, this means that we're able to tell stories that call for changes in habitus, and the story itself becomes a technique, a method, in which we change our own habitus. Here I'm thinking of work that's been done, for example, on the importance of autobiography in the African American tradition from slave narratives onwards, where these narratives were strong ways in which a change of habitus was not only called for in the content of the story, but the very fact of the story being told effected a change of habitus. The fact that Frederick Douglass could tell the story or Malcolm X or whoever: The reflexive act of telling the story itself effects a change in habitus.

Stories have this particular power not just to depict but to be performative acts in Austin's (1975) sense; stories actually do something in the act of telling them. When we get to the crucial Bourdieuian question of How does habitus change so that it isn't a form of determinism? well, a lot of it changes because people are able to tell different stories about their situations that imagine their situations differently, and in telling those stories, they create for themselves a new form of habitus. This is very much to me the therapeutic power of stories. It's another way of saying what I've said throughout my career about ill people telling the stories of their illness, because in that reflexive grasping of what's happening to them, people are able to get out of forms of habitus that are un-chosen and make habitus something more that they've chosen. So habitus, while it is durable, is also malleable, and that's what we really need to study: Exactly how are stories still permeated with earlier forms of habitus? and How are stories reflexive bids to create new forms of habitus?

As Bakhtin calls on us to study the other voices that we hear in the speaker's voice within the story, ⁴ so I think Bourdieu calls upon us to hear the forms of habitus that are still working as un-chosen choices in the story but also are emerging as chosen choices. That's where there's a research program of narrative analysis looking at this tension of habitus between the un-chosen choice and the chosen choice in the telling of the story. In Bourdieuian terms, How do I become aware of the ways in which society has imposed upon me a damaged identity, which is then reproduced through the un-chosen choices that attend the habitus of that damaged identity? versus How do I claim for myself an affirmative identity based on chosen choices, the primary one being to affirm that identity? These questions ground the stories by and about Frederick Douglass or Malcolm X or Simone de Beauvoir or Betty Friedan, and they themselves become icons of their stories. To me, stories are the privileged means of addressing things like classism, sexism, and racism, because these are all grounded in forms of habitus. Stories, which, on the one hand, do reproduce habitus, on the other hand are also the best means of altering habitus. So that's my response to your question about structure.

On Medicine

You asked me, and I feel somewhat pretentious responding to this because it's such a big question: How have you seen medicine evolve over the years? At first I thought I'd just wave that one off, but then I realized that, actually, I could say something about it, five things that sort of follow from each other.

The first is that there's been a redefinition of services, of things that physicians and nurses did, as products, and there's been an increase in the number of medical products that have a perceived benefit. What's more interesting than the increase in the number of these medical products is the fact that we've moved unquestioningly to this language of thinking of the things that people do as products. The products may be diagnostic scans or treatments, they may be pharmaceuticals, they may be the rise of cosmetic surgery, but all over there's a huge amount more that medicine has at least a perceived benefit by offering. That's the first thing. In a medical phrase, there's a lot more of it going around these days.

The second point is that there's been both a demographic and a democratic expansion of entitlement to these medical products. More people want more of what's available. Part of this, as I say, is democracy—there's this great line that managed care meant bringing lower class medicine to middle-class citizens—and part of it's demographic; that is, as you get more people with aging bodies, they become readier candidates for these products.

The third point, again, is on this theme of this language of product. Janice Gross Stein, in her Massey Lectures (2001), talked about what she called the cult of efficiency, or productivity, and health care was one of her main examples. I'm not sure that particular set of Massey Lectures has gotten the attention that I think it really deserves. But whether you call it the cult of productivity or whether you call it managed care or whether you call it bean counting, we've moved into an incrementalization model of health care. For all the talk about holism, in fact the structures of allocation, reimbursement, entitlement are based increasingly on incrementalization, so that you then can invoke various kinds of productivity measures. ⁶

My fourth point is the increasing tension between the demands of justice and the demands of the market. That's where I see the future, honestly. You know, I think the big issue coming up is going to be this tension of justice in the sense of What level of health care is a right that affords to human beings by virtue of their humanity? and What's the range of market allocation? This word “choice” as it's used in a neoliberal context to suggest that everybody is making choices as to what medical services they avail themselves of, or don't, is premised on an equality of resources that we know is just absolutely, patently false. Yet when I talk to a lot of health economists, this is the way their research is proceeding, so I think we have this increasing tension. This extraordinary book that came out last year called Uninsured in America (Sered & Fernanopulle, 2005) is a very interesting example of that.

The final point is the expansion of this category of health, which has now been differentiated from illness. Health used to just be a kind of less threatening proxy word for illness, but health has now come to mean something positive in itself, and health has been clearly expanded to lifestyle, so the two now travel together like salt and pepper. We have “healthy lifestyle,” and we say this without questioning what it means to have health expand into the sphere of lifestyle. All of this takes us back, again, to what I think is still the fundamental importance of Foucault, talking about what counts as care of the self and the ways in which effects of power operate by inducing people to believe in certain practices as appropriate care of the self and also what Foucault called the risks of security in this wonderful short interview he did with a French union representative (2003a), saying: Well yeah, of course we want the unemployment insurance, and we want the health insurance, and we want the disability, and we want all these benefits, but all of these come at the cost of turning oneself into a certain kind of subject who then accedes to various institutional requirements for the promised benefit of these forms of security.

Basically, Foucault's consistent message, that there's no free lunch out there, doesn't mean that you want to dismantle this system. Foucault doesn't at all. It does mean that you need to recognize the dangers of the system, and that's what's subtle about Foucault's politics. These things have very real benefits, and it's by no means clear how we do business otherwise, but we would do well to recognize what's dangerous about them. That's what I'd like to do with public health. It's not that I want to dismantle antismoking campaigns, but I do want to acknowledge that they've become highly class based and in that sense they're dangerous. We need to really keep that front and center in our thinking about these things. It doesn't mean that you don't do it, but it may, hopefully, mean you do it a little bit differently.

As I thought about this enormous question, those struck me as five fundamental points in how things have changed. There's one story that encapsulates that. I remember having dinner once at a conference with a group of young physicians, people who were still mostly in the postdoctoral, fellowship stage of their training. They were all talking about jobs, and I said to them: “You know, in my parents' generation a physician would have talked about starting a practice, or opening a practice. William Carlos Williams, in his [1951] autobiography, calls his chapter about his medical work ‘The Practice.’” I said, “I never hear this word practice any more. You guys all talk about jobs.” They thought for a minute, and they said: “Well yeah, that's exactly right. You know, you don't have a practice any more.” I don't know quite how you extend that over to nursing, but a beginning is this wonderful book by Marie Campbell and Janet Rankin at UBC (2006) that has a great deal to say about the ways in which nursing has changed.

Nobody knows where this is going to go. I mean, it's going to, I think, change very fast in both Canada and the U.S. in the very near future. And there are crucial stakes because what happens in health will then have big effects on the way we conceptualize justice and justice demands in other spheres, and that takes us back to classism, sexism, racism, ageism, all these other things. If we're willing to say that certain people, by virtue of their economic situation, will be asked to do without certain medical services, that's going to become a template in which we ask them to do without other things in their lives and we regard that as perfectly right and just and natural. So there are very heavy stakes involved in how this goes over the next while.

Nobody should underestimate the way in which medicine has become a venue in which all kinds of other social tensions and conflicts are acted out. It's kind of like the ways in which the automotive unions will target one of the big three for heavy labor negotiations, and the other two will fall in afterwards. The way we resolve this health care thing is going to have a crucial importance for all kinds of other standards of social equity, I believe.

That's my diatribe.

Storytelling

Reflections on Research

I'm conscious that it's already 11, and I just wanted to address this last question of yours—can I just take a minute and talk about that for a second?—which was the most embarrassing question because this was where you asked, “What has made you such a good listener/researcher?” I really don't think I am a very good listener. I've already interrupted most of you in the course of anything you tried to say this morning, but this is a seminar, not a research interview and not a clinical interview. I really don't interrupt when I'm doing either of those. And I don't know if I'm really such a good researcher, in that I perpetually get sidetracked in larger theoretical questions and I'm more interested, in many ways, in the possibility of research than I am in ever actually writing up my tapes, which I hope to get to when I retire. I have to become a researcher after I retire.

But with respect to this, I think what I maybe am actually, what I would claim a little bit, is that I'm not a bad reader. I was thinking about that. I think one thing I try to do is take authors seriously. I think too many people read just mining what they're reading for their own purposes instead of really trying to think through what the author is trying to do. For me, reading is very much a matter of getting to know someone so that they become a presence for me. I feel like they're walking around next to me. In the same way that you've gotten to know someone in your life, they become a generalized presence. You could imagine them in other situations and say, you know, How would my spouse feel if she were now at dinner with me? or Would she like this? You're able to project their presence into all sorts of other occasions, and they're with you in that sense. That's why I vastly prefer books over journal articles: because journal articles tend to wash out what I'm most interested in.

There are a few people who I've hung out with for my whole career, like Foucault, like Dorothy Smith. It's a wonderful privilege of an academic life to be lucky enough to find a few authors who, you know, maybe your life overlaps with theirs and for a while it'll be: What's so-and-so going to write next? Then your life will probably go on after theirs, and you continue to live with them and say, “Gosh, wouldn't Foucault and Bourdieu have just loved these riots in France this last week?” and “What would they have said about them?” and “How does it affirm particularly things Bourdieu was writing in The Weight of the World?”

You continue to live with these people, and I think that then does go over into how you analyze research materials. This is where I do have a difference of opinion with grounded theory. I think grounded theory moves so quickly to coding and chopping up that you don't spend time just really hanging out with the person who you've interviewed and dwelling on—brooding over—who they are. The way to do that is to keep asking yourself, What was going on in that story? Not What are the codable items? but Why did they tell me that particular story? Why did they tell me that particular story then? Why did they tell me that story with those emphases? And why were they telling it to me? How was the telling of the story in relation to me?

I think there's nothing wrong with a grounded theory analysis, but it tends to cut off this process of continuing conversation after the interview of What was it about?—this gets to some of your emotion questions—What was I really feeling? because even research interviews involve these qualities of transference and countertransference. Some of the most important things you can learn from the interview involve transference and countertransference: How were they making me feel, and how was I projecting my feelings back onto the way I was then shaping the rest of the session?

I mean, one of the really unfortunate aspects of grounded theory is that you just immediately leave out the questions. As Elliot Mishler ⁷ and others point out, responses are responses to particular questions: You invited people to talk according to how they understood that invitation. That takes us back to Magritte's painted window. Susan Bell, when she's talking about her research interviews, will talk about her question first of all, and she'll describe it as an invitation. Then she'll say: How was I inviting this person to talk, and how do I understand what they said as a response to their understanding of that invitation? That gets systematically lost in grounded theory. There's no possibility for that interpretive loop in the breaking down and coding of these things that people have said.

As I've thought about this word “hermeneutics,” what it really means to me is a continuing conversation in which I understand myself as a full participant: What are the people asking me to attend to? and What am I prone to attend to? and What am I missing in what they may want me to attend to but I'm not as likely to look for that? How can they teach me in a more significant sense than the content? How can they teach me to understand the direction of my own gaze, what it is that I tend to look for? What do I attend to and what do I leave, and how is that as much a reflection of me as of what's in the text itself?

And then their whole quality of voice: It's a matter of finally hearing this voice. When I'm listening to a composer, I like to try to get to the point where I can turn on the radio and come in the middle of something and say, “Oh, well that must be Mozart,” because I've just learned certain characteristics. Even if I can't identify the piece, there's just something about the compositional qualities that I've learned, you know, that's how Mahler does it. There's something in the rhythms; there's something in the juxtaposition of themes that's a Mahlerian voice. Well, that's an objective in learning to listen to people as you play back research tapes: How do you hear the characteristic ways in which they put it together, they express themselves? And beyond voice is this quality of thought, and Heidegger and then Foucault said some really important things. There's a little Foucault (2003b) interview called “So Is It Important to Think?” where he's talking about thought. I think that comes out of his reading of Heidegger and Heidegger's much more extended book What Is Called Thinking? (1968).

How does the person think? Well, that takes us all the way back around to the idea that most of us think in stories, and we produce stories out of the sense of story that we think in. Our thinking really is to me fundamentally narrative, in that, as Damasio (2003) says: The possibility of thought itself is this “movie-in-the-brain.” That's where I find the lay versions of neurophysiology really very exciting in suggesting the truth of what philosophers intuited, culminating in MacIntyre After Virtue (1984), that we really are narrative creatures. We're wired for narrative; our possibility of being as we are is a narrative possibility.

So as a researcher, because I don't think I immediately get caught up in somebody's method and feel a need to crunch the material some particular way, I stay longer with the question of What's the most important thing going on here? What is really important here? What matters? The problem I have with an awful lot of qualitative research that I read is that I don't think the researcher stayed very long with that fundamental question of What's the most important thing here? I think they were moving so quickly to a set of procedures—of What do I do with these materials? —that they didn't just live with the sense of the occasion.

The next question that I talked about in that little speech I published as “After the Methods, the Story” (2004a) is How do I tell the story? Given what's the most important thing going on, how do I tell the story of that important thing? One thing I sort of like about The Wounded Storyteller (1995) and The Renewal of Generosity (2004c) is that both involved seriously trying to think through how I told a story in which I'd gotten myself reasonably clear—and the titles express this—on what the most important thing was. But then I did a lot of work on How do you tell that story? How do you bring the pieces together to make that go?

I follow Mills, that there's no prescriptive formula for that. If you pick up somebody else's prescriptive formula, you've already given up a great deal of what, to me, is the fun of doing it, although as Foucault says so beautifully: I'm still working and don't know whether I'm going to get anywhere. When I start a book, for me the whole interest of taking on this enormous labor—with its consequential opportunity cost to my family, my friends, my possibilities of having a life—the whole reason for taking it on, following Foucault, is that I don't know. I don't know what it's going to end up looking like, and I don't know whether I'm going to be able to pull it off. In that sense, a certain analogy to having children is not entirely inappropriate, in that it really is a work of labor and it's a work of labor in which you don't know what it's going to look like, but you have a lot of things that you can do to affect that. That is the work of it and the fun of it, the excitement of it.

I mean, I hope I go out like Norbert Elias and Dorothy Smith and others. Well, Dorothy Smith hasn't gone out yet, but she's my model of what I want to look like at 79: still just completely excited about ideas. I hope you all heard Dorothy last February when she was here. ⁸ I've known her since 1975, and it's just so great to see somebody absolutely at the top of their game after the 30 years that I've known her, just as impassioned now as she was when we first met back in the mid ‘70s and having a lot more fun with it in many ways because she's just as impassioned. But the other thing is you get to a certain point where you don't have to be responsive to these things about, you know, What are the impact factors of your journal? and all this crap. I mean, there are some real advantages to getting old. You can be freed of so many expectations; not that you don't have others that take their place and can be very stressful, but there are some that you can just completely let go of.

Anyway, that was my answer to this question that just set me up for such incredible pretentiousness.