Scholarly article on topic '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'

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 Academic research paper on "Media and communications"

Share paper
OECD Field of science

Academic research paper on topic "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"

H) Check for updates


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

Lynn P. Eldershaw CIHR Strategic Training Fellow University of Maine at Presque Isle U.S.A.

Maria Mayan University of Alberta Canada

Anne Winkler

CIHR Strategic Training Fellow University of Alberta Canada

© 2007 IIQM. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.


In a 2005 interview with doctoral and postdoctoral students Arthur W. Frank discusses a variety of topics related to qualitative research, such as methodology, narrative, power, rigor, and the peer review process. He reflects on his own work and the artists, philosophers and sociologists who have influenced him. He provides a selective history of research in the social sciences and discusses changes in health care and the practice of medicine.

Keywords: qualitative methodology, narrative, story, medicine, epistemology, structure


Arthur W. Frank is author of an illness memoir, At the Will of the Body (1991, new edition 2002); a study of illness narratives, The Wounded Storyteller: Body, Illness, and Ethics (1995); and an argument for the possibility of actual dialogue in health care, The Renewal of Generosity: Illness, Medicine, and How to Live (2004). He is currently writing a book about how stories

make experience possible and make life social. His longer term research involves changes in the narrative resources that shape illness experience. Dr. Frank is an elected fellow of The Hastings Center, a preeminent bioethics center, and an elected fellow of The Royal Society of Canada. He is professor of sociology at the University of Calgary and lectures internationally.

The following interview with Arthur Frank was conducted at the University of Alberta's International Institute for Qualitative Methodology in November 2005. The session was initiated by a transdisciplinary group of doctoral and postdoctoral participants of the EQUIPP: Enhancing Qualitative Understanding of Illness Processes and Prevention training program. Prior to the interview Arthur Frank was sent several questions related to method, research and his work, but as the reader will discover, the interview expands well beyond the bounds of these questions. In characteristic Frankian style, the conversation moves from René Magritte to Victor Shklovsky, drawing in Gustav Mahler, Martin Heidegger, Erving Goffman, and the ideas of Bruno Latour, Norbert Elias, Dorothy Smith, and many other influential thinkers and artists to inform his reflections on a wide range of topics, including habitus, dialogic method, fiction, and stories that intimately illuminates the social world we live in and how we come to represent it.

Following the interview is a select bibliography of Dr. Frank's publications.

Interview with Arthur Frank

We can start wherever you'd like.

I brought this picture. This is René Magritte's La Condition Humaine} I don't quite know how I'd do the graphics, but I'd love to have this on the cover of the book I'd like to write about narrative, because it exemplifies the methodological problem. There's actually a quotation from the great Russian formalist of the 1920s/1930s Victor Shklovsky, where he said that narrative is not like looking through an open window onto the world; it's more like a painted window. I don't know if Magritte was familiar with that quotation, with Shklovsky; he may very well have been.

That's really the methodological problem to me. The assumption in most methods is that you turn on the tape recorder and you open the shutters and suddenly it's like a film where somebody will start to tell a story and then the screen goes all wavy and then you're in the scene that they're telling. It doesn't work that way. It's like Magritte, and the longer you contemplate that picture, the more you realize that it's just layers of painting all the way back.

Now, that doesn't mean philosophical idealism, okay? Just because the world as I configure it is, as neurophysiologists tell us, my cerebral cortex mapping my body, it doesn't mean that if I get it wrong, I'm not going to fall into the hole or ride my bike into the cement pillar and that I won't really crack my head if I ride my bike into the cement pillar, whether I saw it or not. But it does mean that the world for me is the world that my cortex is mapping; it's this painting.

When people tell you stories, there are a number of viewpoints on the world. The world is full of multiple perspectives. We're each seeing the world from a physically separated perspective, an autobiographically separated perspective, a linguistically separated perspective, et cetera. The phrase I like, though, is back in the '60s, from old Herbert Blumer, the great symbolic interactionist: The world is obdurate. So while there will be a variety of stories, and each of those stories is, as Magritte shows us, a painting that we're seeing out there, the world will also resist some stories because the world is obdurate. Some stories are not going to fly over time. Of course, in health care we run into these all the time: either hypochondriac stories or what are pejoratively

labeled as denial stories. I mean, the body's going to do what it's going to do regardless of the story you tell about it. By the same token, all you can ever know of the body is the story you tell about it.

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 program2 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,

What I say ought to be taken as propositions, game openings, where those who may be interested are invited to join in. They are not meant as dogmatic assertions that have to be taken or left en bloc. (p. 246)

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 rigor3 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.

Can you further comment on the emphasis we put on "methods" and what Mills refuses to accept?

I think what Mills was objecting to was forgetting that in the first generation people like Paul Lazarsfeld, who was the dominant methodologist in the 1950s, went out and looked at a social field quite ethnographically and then said: So how can I gather some data that will allow me to confirm what I've observed ethnographically? Method was always ad hoc in that sense. Then, as Mills tells the story, people whom Mills calls acolytes came along after Lazarsfeld, and they'd take what was an ad hoc procedure and turn it into procedure with a capital P. Then they'd say, "This is how you go out and study various fields," which was to reverse the order of things.

Mills is quite clear that he finds both Parsons, in the grand theory chapter, and Lazarsfeld, in the abstracted empiricism chapter, both brilliant in their own ways. They really are. The problem, as he says, is with the acolytes. If there's a problem with the founders, it's that they were willing to let the acolytes come along and turn what they had done into a template which, you know, going back to the Foucault quotation, does ask you to accept and reject it en bloc, and says: If you want to be a sociologist, this is what you have to accept, period. It's what Zigmund Bauman calls "legislative," in the sense of: This is what you will believe.

The thing I love about Foucault is that he marks a refusal to be legislative. In that sense I find him genuinely dialogical in saying: Well, maybe I'll change my mind about this, maybe you ought to change your mind about this, but let me show you how I've been thinking about it. Unfortunately, most methods books certainly are highly legislative: This is what you do; this is the order in which you do it. If I may say so, even something like the referee template for QHR tends to affirm that in saying, "This is what articles have to have done. Have they done it? Have they done this, this, and this?" and so on.

Can you explain how sociological imagination connects with rigor?

Well, Sociological Imagination was the title of Mills' book in 1959. It connects with rigor because, I think, again, there's a continuity between Mills' critique and Foucault's critique. They're both saying that there are no freestanding criteria of rigor. What we have, a la Wittgenstein, is a language game, and rigor is a move in a language game. People use this word, this idea, in order to make certain claims, and one has to look at what those claims are and the people who want to make them for what.

A lot of what interests me about method, which really lies in the history of the social sciences, is How are you trying to make the world knowable in a particular way? and, as Foucault would say, What are the effects of power of making the world knowable in that particular way? What I find kind of scary about the proliferation of qualitative research articles on "patient experience," the "illness experience," is that they're claiming to make illness experience knowable as an object in reliable and verifiable ways. Who wants to use that information as an effect of power in what

institutional activities? That's where I find that the whole thing doesn't always interrogate its own possibilities or where it's leading to.

This is the perennial problem of the social sciences: What's the effect of making the world knowable in this way? When I teach social theory, I like to have students also read descriptive novels, ethnographic kind of novels, that are writing about the same period, you know, the World War I period in the classical sociological theory course. I like to pose for them the question, How does social science make the world knowable versus how does literary fiction make the world knowable, and what can you learn from both? Well, in the same way I'm very interested in how first-person illness narratives make illness knowable versus how qualitative studies of illness experience make the world knowable, and what are the effects of power, of taking seriously those different forms of making the world knowable.

So what are we to think about social science, about qualitative studies?

As I said, the Magritte painting is not an idealist view of the world and it's not an antirealist view of the world—I think it's quite a realist view. In the same way, what I'm saying is not an antiscience view. There are all sorts of things that science pretty well knows, and if you read Pierre Bourdieu's (2004) posthumous book, which was his last course of lectures at the College de France, it's a defense of science. But for Bourdieu science has to be reflexive, because science is itself a field, and as such, the way in which people play the game in the field is subject to a sociological analysis.

But Bourdieu also, I think, is one of the most articulate, most realist defenders of the genuine claim of science. The core of his argument is that in the culture of peer review, things are subjected to review by those who have the greatest interest in finding fault with the argument. And if peer review works as it should, science does have a distinctive claim, among other fields, of producing things that pretty well represent the obdurate world. So I'm not saying that we can't and we don't learn things from science. This is the case that Bruno Latour also makes, where people will talk to him about his social studies of lab science and say, "Why are you antiscience?" and he'll say, "Well, I'm not. You know, I just want to recognize exactly what it is."

In response to a question in which the student describes being criticized by nursing colleagues for her critical studies of nursing practice.

What you're saying is not in any sense anti-nursing, although some people, I think, would hear it that way because some people, as soon as they hear any kind of reflexive critique, make a jump that that's in opposition to whatever it's critiquing. Well, that's an immature, defensive attitude. And so, as I say, I'm no more anti-science than you are anti-nursing, but where that comes to a crunch in our shared field, I think, is probably health promotion, where we do know a good deal, like smoking and cancer, high-fat diets. We know a lot of things. We also understand that health promotion produces effects of power; it's highly class biased. It articulates with a reproduction of certain class privileges. And the problem is that there are certain immature minds that find those two recognitions incompatible. You know, I wish we published more on the problem of rising to a new level of understanding, where we can entertain both the critique and the validity of, you know, nursing knowledge or public health knowledge or whatever simultaneously and not have to go to one or the other, which very often is the position that people lapse into.

Can you talk some more about the culture of peer review?

Most insidious is the principle of homopoly, which means that I'm going to give a favorable review to work that looks like my work. Publishing this, in effect, reinforces the idea that what I'm doing is the appropriate way of proceeding. I also find it insidious moving toward tick box refereeing as opposed to what I would, for want of a better term, call big picture refereeing, where you ask this more fundamental question: What does this have to say, and what's the benchmark against which we measure what it has to say? For me, in looking at articles on so-called illness experience, the benchmark is, Do they tell me more than I can learn from reading personal narratives of that illness? Or if they tell me the same thing, do they at least have the decency to recognize that some of the people who've been ill, just writing about their own experience, have basically managed to say exactly the same things that the investigators said with a SSHRC grant?

I mean, it's curious what we count in terms of the so-called literature review and what we count in terms of the originality and the interest of the work that's claiming to be published. In the illness experience area, I find myself irked—my wife and I like that word whenever we see it— by the fact that people have gone out to study disease X and have not taken the trouble to realize that some people who have had disease X have written extremely thoughtful, reflective memoirs of what it was to live with that, and, as I say, what you did with your SSHRC grant has not particularly advanced my understanding beyond what I learned reading so-and-so's memoir.

Which gets us back, I guess, to this question of different claims of forms of knowledge and who makes what claim on what basis. Now, if they'd get to the end and say, "You know, it's interesting. All I've done is really confirm, and now this allows us to generalize so-and-so's memoir and realize that this wasn't just his or her experience. This is general," that would be fair enough. That would be interesting. It's the pretending that it's not there or being sincerely ignorant of it being there and feeling you have to reinvent the wheel: That's what I find cumbersome.

You commented that you were not in any sense a social constructionist. Can you elaborate on that?

I shouldn't say "in any sense," because there is a whole spectrum of social constructions, but I dislike that term. I don't find that a useful term. First of all, Bruno Latour's most recent book (2005) starts off asking this really interesting question of What do we actually mean when we append this word "social" as an all-purpose adjective to all kinds of things—social interaction, social construction, social mobility, whatever? What's going on with that adjective? Then, the whole notion of construction, I think, begs the question of exactly what is being built up, as opposed to what's already there, and it lends itself too easily to the idea—and there are some people who espouse this—that humans are building the whole thing.

We're meeting this year at the end of a season of extraordinary natural disasters, in which it becomes all too tragically obvious that the humans are not building up the whole thing. Now, the response to the disaster could in some ways be called a social construction, but I guess the whole the question is, Is that the most useful thing to call it? In what ways does that term lead us in how we think about these things?

I'd say the same sort of thing about this word "structure," which I think is an unfortunate word. I've never found the notion of social construction sufficiently precise to stave off its potential misunderstandings.

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.

In response to a question about narrative therapy

As I understand, what Michael White's5 trying to do, if I could phrase that in sociological terms, is make people aware of the habitus that's informing their story and the way in which that just really isn't working out for them and then finding ways of telling stories that are themselves the instruments of creating a habitus that's conducive to the life that they want to live.

Although we're always replanking the ship at sea, and it's not like we ever get out—first of all, there's no outside of habitus at all, and we never really get rid of the habitus that we came in with, these first crucial affiliations with our families of origin and our neighborhoods of origin and so on—you can speed up the process of their half-lives. That, to me, is a useful metaphor.

In response to a question about structure

I like structure better as an adjective or an adverb than I do as a noun. I worry when we start talking about "the structure," and that's generational for me. The thing is that you have this great advantage of being young enough so that you weren't trained in sociology at a time when Parsons was still hegemonic—there are ways in which I'll go to my grave still trying to get out from under the weight of the Parsonian hegemony—and by the time you came along, this was just another kind of classical theorist. It's sort of like the way in which parents who are toxic to their own children can be quite decent as grandparents because the grandchildren just regard this as some silly eccentricity of grandma and grandpa. It's not something they have to internalize and then work through in therapy and so on. It's just kind of: Oh, that's just grandpa; he's just, you know, what the heck. They can just enjoy; they can hold these people lightly. I hope we've gotten to a generation now, that you're speaking for, that can hold this idea of structure lightly, whereas my generation was called upon to take it very seriously.

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.

What is particular about health and body that's rich for narrative?

That was even tougher than How's medicine changed? But let me say three things. The first is that I think Brian Turner is right in his idea of somatic society. The moment we happen to live, what [Norbert] Elias would call the figuration, is a time when we play things out in terms of health and bodies as both metaphor and actual venue. This goes back to what I just said about how other kinds of questions of social equity will get played out in decisions over distribution of

access to health care resources. As I say, these are both metaphors and they're quite real. These have become the lingua franca, the most common denominator. Whatever field you're talking about—the political field, the economic field, whatever—it tends to reduce to bodies and health. It's why it makes perfect sense that of course half the provincial government budget would go for health care, because that's what it is to live in a somatic society; that's what we take seriously at this particular point in time.

The second point is, as I argued in The Wounded Storyteller (1995), that bodies are experiencing, but bodies are themselves mute; bodies need a voice. We're living in these bodies, we're highly conscious that things are happening, but it's difficult, it's really work, to give those bodies the voice that both brings these happenings into our own reflective awareness and expresses them to others.

The third point, then, is the need to articulate, in the sense of fit, my body into the needs and demands of other bodies. I think that's a lot of what we mean by this word "social." The work of the social, if we conceive of the social as an ongoing work, is to fit the needs, desires, demands of my body into the needs, desires, demands, structures of other bodies. That's what I mean by articulating: You have to make it fit with others. Stories, quite simply, are the media of this articulation. We do this articulation—we fit our bodies with other bodies—by telling stories about ourselves and hearing their stories about themselves. We negotiate through this telling of my story, your story, because experience is always a representation and what we call interaction is a representation, and that takes place through stories.

I think that stories crucially involve two kinds of bridging activities. Stories bridge the individual and the collective. They bridge the "me" with the "we," and stories also bridge the corporeal and the symbolic. Stories are ways in which we move from the tissue level to the symbolic level, and in the movement to the symbolic, again, we've moved from the me to the we. That's my basic point about why I think there seems to be this particular richness of health and bodies, a particular importance.


What about relations of power in storytelling?

That's my duality: that stories reflect all the dominance that makes certain topics "unstoryable," or the term I would use from Michael Berube, "non-narratable." Stories are also the best way of making something narratable. Berube made this argument very powerfully in his book Life as We Know It (1996), where he's writing about his son Jamie with Down syndrome. He uses this term "narratable," saying that the political problem of the disabled is How are their stories considered fully narratable? How are their stories considered just as important to tell and to hear as the stories of developmentally normal people, particularly Jamie's brother, Nick, a developmentally normal child? Berube says that the core problem in his family, which is also the core political problem of disability rights, is How are Jamie's stories just as narratable as Nick's stories, and, therefore, how does Jamie's life have the same entitlement claims that Nick's life is able to have? For Berube—and I think he's exactly right—political entitlement begins with a sense of the story being a perfectly tellable, important story, even though it doesn't fit the dominant paradigm of child development, which has very specific expectations of when your kid first rolls over, holds an object, crawls, walks, says its first words. These are all milestones, and they're milestones that people expect to have happen at particular times. It's that chronological scheme that makes the stories as narratable as they are. And that's a problem for people with disabilities: They don't hit

the milestone at the right moment, so how do we regard their stories at tellable? I think Berube's spot on about that.

That also gets back to the first question of why I think stories are, in fact, a privileged means for addressing the isms, and we could now include "ableism" among the other list that you sent me. I think some of the most important work of the disabilities movement has been telling about full lives lived with extraordinary disabilities. Gelya Frank's book—no relation—Venus on Wheels (2000) is an early example and a third-party telling of that story. Not everybody feels like telling their own story, especially in print, so I think there's plenty of room for the ethnographer or whoever. There are certain stories that are better told by a third person, in part because the first person just doesn't want to do the enormous physical work of producing this text, which—trust me—is a lot of work. They spend their lives elsewhere.

I have a question about the inherent dichotomy of storytelling: There is always a speaker and listener. Although this may be crude, an oversimplified way to think about stories like this, a lot ofpeople do tell stories to make truth claims and position themselves as moral authorities. They believe their story has value and, perhaps because they are in the position to articulate it, has more value than other stories.

Yeah, that goes back to my Foucault quotations, my epigrams at the beginning of this, that there are stories that are told in a legislative spirit and there are stories that are told as game openings. One of the things I'd hoped people would get out of reading The Renewal of Generosity (2004c) is that it goes into a lot more detail about what I said back in The Wounded Storyteller (1995), that stories are always co-creations. The whole notion of stories as dialogue suggests that the voice of the storyteller is always permeated with other voices, which is both constraining and liberatory. I mean, some of those other voices are going to be the voice of the dominant narrative that you want to get out of, and some of them are going to be voices of people who are enriching your possibilities. But the crucial thing is that we need to get away from this rather crude epistemology of one person having the story inside of him- or herself and then delivering the story like the goose laying the egg in the presence of the other person, who then goes: What an egg! In fact, it's a collaborative activity all the way through.

Anybody who lectures much comes to realize that the way you're lecturing builds on or fades away with the ongoing reaction of the audience even though they're silent in that particular frame, in Goffman's sense. Bakhtin's emphasis on the story as a constant process of anticipation and response is absolutely crucial here. Even if I'm the only one who's speaking as I tell the story, I'm telling it in constant anticipation of how you're going to respond, and my telling of the story is a response to what you've said before. This is the problem with taking stories out of their dialogical setting and treating them as freestanding texts, you know. They're not.

People beat up on other people with stories all the time. That's the damaged identities thing. It's what I've had to acknowledge, and, again, it's a really crucial part of this new book: that stories have their very dark side. Telling a story is not in itself any kind of high ground. Stories are used just as much to oppress and to justify violence as they are used to liberate. It's where we have to be willing to entertain that duality. I don't see any way around it.

You say that stories are co-creations. Could you expand on this idea?

Well, that's what I'm writing about in that recent QHR paper on dialogical research (2005). I think the important question is not Can you? The important question is, If you're committed to not producing something that's monologic, how do you find a style that does that? That gets back

to what I think we need to take a lot more seriously: that social sciences are not about producing certain knowledge about the world; they're about producing the terms in which people regard the world as being knowable.

Really, that's what the dialogical research piece is calling for. In describing the world this way, we're making normative claims about how the world is knowable. A lot of the ways we make those normative claims now reinforce the mythology of a knowing professional who is objectively separated from a known object of the professional gaze. Unfortunately, qualitative research has done very little to question or overcome that opposition. In many ways, in its calls for rigor and its attempts to justify itself as fully rigorous, qualitative research has simply recapitulated this knowing subject/known object problem of quantitative research. It's why I don't find the qualitative/quantitative distinction to be terribly useful any more in terms of what's been, really, the theme going back to my very first work in communications, when I first started social science, which is, How do we have what I would now call a dialogical social science, as opposed to a social science that posits this split of subject and object, the crudest form of naturalism, as it used to be called?

In response to questions concerning research involving ethnographers who study individuals or groups whose beliefs are different from their own.

Well, that's always the dilemma of ethnography. I don't think ethnography has any obligation to reaffirm people's own self-understanding, which would be a kind of way in which some of our ethics review is almost pushing us. By the same token, I also don't think that ethnography has an a priori obligation to repudiate people's self-understandings. I think Leslie Irvine (1999) is a very nice example of somebody who is able to say a lot about how it works that would not be immediately accessible to members while at the same time not finding it necessary to repudiate how it works. That's a balance that we need to have.

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 Mishler7 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.

1. The painting depicts an open window, looking out onto a meadow. An easel is in front of the window. On the easel is a painting of the meadow. The borders of the painting match perfectly with the (painted) meadow seen around the painting. Note: It is currently owned by the National Gallery of Art in Washington, DC, and can be viewed at the gallery's Web site (

2. The EQUIPP program is a certificate program in qualitative methodologies offered by the International Institute for Qualitative Methodology under the auspices of the University of Alberta's Faculty of Nursing ( Program participants submitted written questions to Dr. Frank before the interview in addition to asking questions during the session. The listed authors transcribed and reviewed the transcript with Dr. Frank, making necessary changes, and Moira Calder edited it.

3. This is in response to a written question.

4. See The Renewal of Generosity (2004) for further discussion of Bakhtin.

5. See, for example, White (2007).

6. Developed in Rankin and Campbell (2006).

7. For example, Mishler (1986).

8. For example, see Smith (2005).

Selected Bibliography of Books and Essays by Arthur W. Frank

Frank, A. W. (1993). The rhetoric of self-change: Illness experience as narrative. Sociological Quarterly, 34, 39-52.

Frank, A. W. (1995). The wounded storyteller: Body, illness, and ethics. Chicago: University of Chicago Press.

Frank, A. W. (1997a). Enacting illness stories: When, what, and why. In H. Nelson (Ed.), Stories and their limits: Narrative approaches to bioethics (pp. 31-49). New York: Routledge.

Frank, A. W. (1997b). Illness as moral occasion: Restoring agency to ill people. Health, 1, 131148.

Frank, A. W. (1998a). Just listening: narrative and deep illness. Families, Systems & Health, 16, 197-212.

Frank, A. W. (1998b). Stories of sickness as care of the self. Health, 2, 329-348.

Frank, A. W. (2002). At the will of the body: Reflections on illness (2nd ed.). Boston: Houghton Mifflin.

Frank, A. W. (2004a). After the methods, the story: From incongruity to truth in qualitative research. Qualitative Health Research, 14, 430-440.

Frank, A. W. (2004b). Asking the right question about pain: Narrative andphronesis. Literature and Medicine, 23, 209-225.

Frank, A. W. (2004c). The renewal of generosity: Illness, medicine, and how to live. Chicago: University of Chicago Press.

Frank, A. W. (2005). What is dialogical research, and why should we do it? Qualitative Health Research, 15, 964-974.

Frank, A. W. (2006). Health stories as connectors and subjectifiers. Health, 10, 421-440.

Frank, A. W. (in press). Five dramas of illness. Perspectives in Biology and Medicine, 50(3).


Austin, J. L. (1975). How to do things with words (2nd ed.). Cambridge, MA: Harvard University Press.

Bérubé, M. (1996). Life as we know it. Boston: Houghton Mifflin.

Bourdieu, P. (1984). Distinction: A social critique of the judgement of taste (R. Nice, Trans.). London: Routledge. (Original work published in French in 1979)

Bourdieu, P. (2004). Science of science and reflexivity. Chicago: University of Chicago Press.

Damasio, A. (2003). Looking for Spinoza: Joy, sorrow, and the feeling brain. New York: Harcourt.

Foucault, M. (1980). Power/knowledge: Selected interviews and other writings 1972-1977 (C. Gordon, Ed.). Brighton, UK: Harvester.

Foucault, M. (2003a). The risks of security. In P. Rabinow & N. Rose (Eds.), The essential Foucault (pp. 66-79). New York: New Press.

Foucault, M. (2003a). So is it important to think? In P. Rabinow & N. Rose (Eds.), The essential Foucault (pp. 170-173). New York: New Press.

Foucault, M. (2003b). Why the prison? In P. Rabinow & N. Rose (Eds.), The essential Foucault (pp. 246-258). New York: New Press.

Frank, G. (2000). Venus on wheels: Two decades of dialogue on disability, biography, and being female in America. Berkeley: University of California Press.

Heidegger, M. (1968). What is called thinking? (F. D. Wieck & J. G. Gray, Trans.). New York: Harper & Row.

Irvine, L. (1999). Codependent forevermore: The invention of self in a twelve step group. Chicago, University of Chicago Press.

Latour, B. (2005). Reassembling the social. Oxford, UK: Oxford University Press.

McIntyre, A. C. (1984). After virtue (2nd ed.). Notre Dame, IN: University of Notre Dame Press.

Mills, C. W. (1959). The sociological imagination. New York: Oxford University Press.

Mishler, E. (1986). Research interviewing: Context and narrative. Cambridge, MA: Harvard University Press.

Nelson, H. L. (2001). Damaged identities, narrative repair. Ithaca, NY: Cornell University Press.

Rankin, J., & Campbell, M. (2006).Managing to nurse: Inside Canada's health reform. Toronto, Canada: University of Toronto Press.

Sered, S., & Fernanopulle, R. (2005). Uninsured in America: Life and death in the land of opportunity. Berkeley: University of California Press.

Smith, D. (2005). Institutional ethnography: A sociology for people. Lanham, MD: Rowman and Littlefield.

Stein, J. (2001). The cult of efficiency. Toronto, Canada: House of Anansi.

White, M. (2007). Maps of narrative practice. New York: W. W. Norton.

Williams, W. C. (1951). The autobiography of William Carlos Williams. New York: Random House.