A Dialogue on the Ideas of “World” and “Field”

Alain Pessin: Howard Becker, the idea of “world,” which you have explored fully in Art Worlds (1982), has aroused great interest among sociologists of art, in France as elsewhere in the world. It appears in many works, but one nevertheless has the feeling that the uses it is put to are not always very clear and do not do it justice. It is often minimized, reduced in its range and significance to the single positive virtue of cooperation. It is sometimes purely and simply denied in its specificity when it is finally turned into a more optimistic variant of what Pierre Bourdieu has called “field.” Thus many authors, professionals as well as graduate students, think that the ideas of field and world are simply two interchangeable approaches which are equally useful in the same research project, one emphasizing conflict, the other the complementarity of actors and actions. In this view, sprinkling a little Becker on Bourdieu would produce good sociology, if only because it would make the world seem a little less desperate place. It seems to me that this would be too simple-minded, an insufficiently rigorous use of the idea of world. That’s why I think it is time to clarify this idea, and to see, with you, how it differs from and is opposed to the idea of field.

Let’s begin with this latter idea. What does the idea of field evoke for you?

Howard S. Becker: I’ve just finished reading Pierre Bourdieu’s autobiography (Bourdieu 2004), published after his death, and so I’ve had a chance to see how he uses the idea in practice. The book starts with a description of the champ universitaire as it existed when he entered it in the late 1950s. He describes it as dominated by Sartre and his followers. He says that philosophy was the important discipline, that sociology and social science were not taken seriously, except to be seen as dangerous tendencies to be suppressed. Sociology, in  particular, was seen by Sartre and his followers as too American, too positivist, too much opposed to the dominant myth of the solitary intellectual who achieved the great things he achieved by, as a friend of mine used to say, “thought and thought alone.”

He puts this description in the language of “field.” I’ll try to summarize the imagery he uses. First of all, the idea seems very metaphorical, the metaphor coming perhaps from physics. There is a defined and confined space, which is the field, in which there is a limited amount of room, so that whatever happens in this field is a zero-sum game. If I have something, you can’t have it. Naturally, then, people struggle and fight over the limited space. The people who control the limited space try to keep it all for themselves and their allies and prevent newcomers from getting any of it.

“Space” is a metaphor for anything that people want that is in limited supply. For Bourdieu, this is often esteem or recognition, but it can also be more material stuff like money or access to publication outlets, things like that, “real” things, you might say.

The field is organized as “forces” of various kinds, and one big force is power, which seems to be a question of the control of resources: in the case of the champ universitaire, these would be things like, as I said above, postes (permanent positions) in faculties and research centers, money to support research, access to publication outlets, and in a general way esteem, honor, recognition, and so on.

The people with power make judgments about newcomers, deciding whether they can be admitted to the circle of the powerful, perhaps in a subordinate role at first, or whether they must be rejected. He says that these determinations are made on the basis of the work people do but also on more personal criteria: their behavior, the way they dress, their accent, their political ideas, their friends, their lovers. (He doesn’t quite say that the latter are illegitimate criteria, although perhaps he does somewhere, but he certainly means that you should understand him this way.) Although the idea is meant to be completely general, the examples (naturally, since it is autobiographical) come from the French university system of the 1950s.

Alain Pessin: The idea of field should be generalizable to all areas of social life, including the one that interests us directly, artistic activity. Having proposed, with the idea of world, a very different approach, what point, would you say, separates you most clearly from Bourdieu’s approach?

Howard S. Becker: The idea of field seems to me much more a metaphor than a simple descriptive term. Bourdieu described the social arrangements in which art is made—what he calls a field—as if it were a field of forces in physics rather than a lot of people doing something together. The principal entities in a field are forces, spaces, relations, and actors (characterized by their relative power) who develop strategies using the variable amounts of power they have available.

The people who act in a field are not flesh and blood people, with all the complexity that implies, but rather caricatures, in the style of the homo economicus of the economists, endowed with the minimal capacities they have to have to behave as the theory suggests they will. Their relations seem to be exclusively relations of domination, based in competition and conflict. When I try to imagine such a field, I see a diagram: a square enclosing a space in which arrows connect units, creating invisible structures. Or, worse yet, I imagine a big plastic box with all kinds of rays shooting around inside it, like something you would see in a science fiction movie.

The repetition of the physical metaphor is very striking in The Rules of Art. For example, in the section at the beginning of the book where he treats “The question of inheritance,” he says:

In laying out the two poles of the field of power, a true milieu in the Newtonian sense, where social forces, attractions or repulsions, are exercised and find their phenomenal manifestation in the form of psychological motivations such as love or ambition. Flaubert institutes the conditions of a kind of sociological experimentation: five adolescents—including the hero, Frédéric—provisionally assembled by their situation as students, will be launched into this space, like particles into a force-field, and their trajectories will be determined by the relation between the forces of the field and their own inertia.. This inertia is inscribed on the one hand in the dispositions they owe to their origins and to their trajectories, and which imply a tendency to persevere in a manner of being, and thus a probable trajectory. And on the other in the capital they have inherited, and which contributes to defining the possibilities and the impossibilities which the field assigns them. (Bourdieu 1996, pp. 9-10)

Alain Pessin: What evokes such images is in some way the “compression” of the social. The virulence of the oppositions is inevitable because of the fundamental scarcity of the space and, as a result, the scarcity of positions anyone can occupy. The idea of world puts us in an extendable, open space, to which, moreover, it’s difficult to assign limits, insofar as the spatial metaphor is relevant to it at all.

Howard S. Becker: The idea of world, as I think of it, is very different. Of course, it is still a metaphor. But the metaphor of world—which does not seem to be at all true of the metaphor of field—contains people, all sorts of people, who are in the middle of doing something which requires them to pay attention to each other, to take account consciously of the existence of others and to shape what they do in the light of what others do. In such a world, people do not respond automatically to mysterious external forces surrounding them. Instead, they develop their lines of activity gradually, seeing how others respond to what they do and adjusting what they do next in a way that meshes with what others have done and will probably do next.

Above all, the metaphor is not spatial. The analysis centers on some kind of collective activity, something that people are doing together. Whoever contributes in any way to that activity and its results is part of that world. The line drawn to separate the world from whatever is not part of it is an analytic convenience, not something that exists in nature, not something which can be found by scientific investigation.

So the world is not a closed unit. Sometimes, of course, there really is a bounded area of activity, such as the university world, in which some set of organizations and people monopolizes the activity in question. Some forms of collective action have walls around them, not just the total institutions Goffman described but also all the companies where you have to have a badge to get beyond the reception area and, in the cases Bourdieu focuses on, those places where physical access isn’t limited but access to positions and activities is.

In these cases, you might say, the field, limited as it is by rules and practices which keep outsiders out, makes it impossible to be part of some collective activity unless you are chosen by the people who already are part of it. You can’t do sociology or intellectual work if you are denied access to the places where people are doing that sort of work together. So you can’t be a sociologist unless you can have a job in a sociology department or research center and can publish your work in the recognized places where sociology is published.

To say it that way raises obvious problems. Even in such cases, the monopoly is almost never complete and certainly is never permanent. So, as Bourdieu describes the world that was the setting for the beginning of his career, doing sociology was not confined to the places he seems to care about most. But it was not only at the Sorbonne or the College de France that sociological work got done. He never mentions, for example, Georges Friedmann, who was a friend of my mentor, Everett Hughes, and who studied factories, the industrial world.

I suppose a Bourdieusien might say that, well, of course, you could do something that would look like sociology and might even, from some point of view (maybe, as in the case of Friedmann, from the point of view of a visiting American industrial sociologist), be sociology, but, let’s face it, it wouldn’t really be sociology because the people who own the trademark wouldn’t recognize you as doing the real thing. “Congratulations, Friedmann, looks like interesting stuff, too bad no one knows or cares about you.” The equivocal term here is “no one,” because of course people knew about Friedmann but the people who counted, in Bourdieu’s view, didn’t accept him.

At this point it is, as we like to say, an empirical question: is it true that someone can control access to everything important in that way? Can your heterodox ideas be prevented from reaching some public if the “important people” ignore them? That depends. I think that probably it is not really very common, although it is common for people to feel that this is what’s happening to them and their ideas.

At this point I think it might be useful to consider the differences in institutionalized academic and intellectual life between the U.S. and France, and even to engage in some speculation about the sources of those differences. I have for years been telling people in France that to understand American sociology they must first understand that there are something like 20,000 sociologists in the U.S. and something like 2,000 departments of sociology (and many sociologists work in other fields—education, social work, nursing, etc.—thus making the number even larger). This is at least ten times the number of people and departments as exist in France, probably more like twenty times.

One consequence of this is that it is relatively easy to support a wide variety of sociological activities. No idea is too crazy or unacceptable to find a home somewhere. You name it and there is, somewhere, a department or a part of a department devoted to propagating that idea or point of view. You can always find some other people who think your idea, unacceptable as it is to “the leaders of the field,” whoever they are, is really good and are ready to march under your flag. If you can find two or three hundred of them (not so easy, but certainly not impossible when there are 20,000 among whom to recruit) you can organize a section of the ASA. If you can’t get that number you can start your own organization (e.g., the International Visual Sociology Association), publish your own journal, elect your own president, and give your own prizes.

It’s in that sort of setting that the idea of “world” seems like a “natural” way to think about organized activity.

Alain Pessin: One could summarize all this in one of your favorite ideas: “You could always do something else.” But this idea has to have a general application, it’s not only in the United States that you can do something else. Such a formula, when you apply it to any situation of social life, opens the way to a sociology of the possible, it stands in opposition to the idea of limited possibilities of action, and the blocked aspect of social systems. When you aren’t wanted in one place, you can always go someplace else, and do what you want to do there.

Howard S. Becker: Someone is monopolizing the field you want to work in? Move somewhere else and start your own field. You don’t even have to compete with the other people. You can criticize them to your followers, or ignore them, but they are not powerful enough and do not have enough of a monopoly to prevent you from doing anything.

Remember that even in totalitarian regimes there were almost always dissident intellectual movements doing things forbidden by the people who dominated the legitimate field for that kind of work. When the Brazilian military juntas forbade academic sociology, people organized research institutes—with outside help, of course—and began to practice “urban anthropology,” which was not forbidden. (Of course, there are extreme cases where it is impossible to escape the power of the leaders of a field, but I think that, empirically, that isn’t frequent, and certainly not at all in the case of artistic activities in most contemporary societies.)

So the idea of a world of people who collaborate to produce this or that result, a world in which people can find others to collaborate with even if the more powerful people in their discipline don’t approve or recognize what they do, a world in which the power to define what is important or acceptable does not rest stably with one set of actors—in that sort of situation, the idea of world makes sense and is analytically useful, because it takes account of what is there to be discovered, of what events there are to explain.

In contrast with the idea of “field,” the idea of “world” seems to me more empirically grounded. It talks about things that we can observe—people doing things rather than “forces,” “trajectories,” “inertia,’ which are not observable in social life, if you understand these terms in the technical sense given to them in physics. We cannot observe these things perfectly, of course, but well enough that we can argue about them, and the procedures of empirical science can give us provisional answers of the kind science gives.

Alain Pessin: A “world” is thus an ensemble of people who do something together. The action of each  is not determined by something like the “global structure” of the world in question but by the specific motivations of each of the participants, any of whom might “do something different,” create new responses to new situations. In these conditions, what they do together results from arrangements about which the least one can say is that they are never entirely predictable.

Howard S. Becker: A “world” as I understand it—and if my language elsewhere doesn’t convey this then I’ve failed to be clear—consists of real people who are trying to get things done, largely by getting other people to do things that will assist them in their project. Because everyone has a project, and the outcome of negotiations between them is whatever they finally all agree to, everyone involved in such an activity has to take into account how others will respond to their own actions. David Mamet, the playwright, said somewhere I can’t now find that, in a scene in a play, everyone in the scene has something they want. If they didn’t want something they wouldn’t be there, they’d be off someplace where they could pursue something they did want. The scene consists of each one trying to get what he or she wants, and the resulting collective activity is something that perhaps no one wanted, but is the best everyone could get out of this situation and therefore what they all, in effect, agreed to.

This means that while people are free to try to find other possibilities, those possibilities are limited by what they can force or persuade other people to do.

This approach perhaps makes social life seem more open to continuous change and spontaneous action than it really is. Social life exhibits, after all, substantial regularity, People do not do whatever comes into their heads at any moment. On the contrary, most of the time they do things as they have done them before. In a scheme that emphasizes openness and possibility, that regularity requires explanation.

I find that explanation mainly in the idea of “convention.” People often, but not always, know how things have been done in the past, how things are usually done, and they know that others know all these things too. So, if I do things as I know everyone knows they are usually done and is prepared to do them, I can feel confident that my actions will fit in with theirs and we will be able to accomplish what we are trying to do with a minimum of difficulty and misunderstanding. This is not to say that there is not, or never has been, conflict, but rather that in most cases the conflict has been settled, one way or another, and participants in the activity have agreed to do it this way rather than one of the other ways it might have been done.

That’s very abstract, so I’ll give an example, taken from my favorite domain of examples, music. Musicians and composers sometimes disagree on how many notes to include between the two notes of an octave. God did not decree that there should be the twelve notes of the Western chromatic scale. Musicians in other traditions have often made other choices and great musical traditions are founded on them. But Western musicians, over a very long time, did accept the twelve tone chromatic scale as the basis of their music. Now the instruments we play have that scale built into them, the notation we use to write music down for replaying, and everything else connected with Western music takes for granted, on the basis of shared conventional understandings, that everyone will be playing music written in that form on instruments built to play those notes. So it is always easier to play music based on that convention than music created in some other system. The cost in time and energy is much greater when you don’t accept these conventions. So—here, I’m afraid, is a physical metaphor!—a kind of inertia disposes people to do things as they have been done in the past, and that accounts for a great deal of the regularity of social life.

Among the conventional understandings that produce these regularities, we will of course, often find elements of coercion and force, open or disguised, which will produce inequalities and what we may feel are injustices. People often agree to things that are unfair, for lack of any better alternative.

Alain Pessin: The ideas of career and process, which are essential to understanding the functioning of a world, bring us back to the fact that personal trajectories, as they confront collective situations, go through stages and that, at each step, the actors have to make choices. Thus nothing is definitively promised to anyone. One can’t succeed in thinking in terms of process using the idea of field. Everything seems already settled in advance. The struggle is predefined as the normal framework of activity. And the weight of the habitus makes the behavior of those affected by it essentially predictable.

Howard S. Becker: Events and results are not determined that way. The history of attempts by social scientists to predict what will happen in this or that case should be sufficient to make us give up this dream. This is not just a problem of not having enough data or lacking sufficient computing power. It may be—but remember it is only an hypothesis of chaos theory, not something demonstrated—that a butterfly beating its wings in South America will produce a hurricane somewhere else in the world. But nothing like that has ever been demonstrated in social life and I don’t think it is a result we should aim for.

Imagine that we knew enough to predict some result, on the basis of habitus or something much clearer and more specific, a “variable” of the kind quantitative sociologists like to work with, that Mr. Jones will have an automobile accident tomorrow. He will be drunk, his brakes will be in bad shape, and it will be raining, all things which make an accident likely. But it will also be necessary for Mr. Smith (or Mr. Somebody) to “cooperate” to produce the accident. That is, Smith will have to be in the right place for the drunken Jones to hit him, and the possibility of predicting those two events is correspondingly less likely. When you multiply probabilities, they decrease. And the accident will involve not only Jones and Smith, but also hundreds of other people. So the practical possibility of predicting any event, considering the multiple specific events that are necessary and the diminishing multiplicative probabilities, approaches zero. That includes predictions about what people will do based on habitus and similar individual qualities. Such things aren’t meaningless, but they are just one among hundreds of things relevant to what people and organizations do.

You have pointed to something else important in your question. Things do not happen, events do not occur, people don’t choose, all at once. Rather, these things occur in steps, in stages, and that means that every step offers the possibility of going in more than one direction—there is more than one possibility at every juncture. That means that the possible outcomes are always numerous and varied, not easily captured in a formula.

Alain Pessin: It’s time now to put to rest once and for all the misunderstanding attached to the idea of cooperation. We sometimes hear it said that you are the sociologist who has forgotten conflict. But trying to do something together in no way implies an absolutely peaceful conception of social relations.

Howard S. Becker: I suppose that someone who wasn’t trying very hard to understand this point of view could characterize it as simply focusing on “cooperation.” But that wouldn’t be accurate. It could be true only if you understand “cooperation” in a very extended way, as encompassing anything that people do together in which they take into account, and respond to, what the others involved are doing. Collective action—two or more (usually a lot more) people doing something together—is not the same as cooperating in the more conventional, minimal understanding of that word, in which it has overtones of peacefulness, getting along with one another, and good will. On the contrary, the people engaged in collective action might be fighting or intriguing against one another or any of the other things that figure so prominently in Bourdieu’s descriptions of social fields.

But they might also be working together to do something (rehearsing for a concert they are going to give that night), or might be linked indirectly, one doing something necessary for what the other does, even though they might not know each other (as the instrument repair man fixes the broken saxophone necessary for the musician’s evening performance). They might have joined forces for this one occasion, as composers who otherwise compete with each other for scarce commissions and posts will cooperate to put on a concert of contemporary music (see Gilmore 1987). Or might routinely work together on the particular thing that brings them together, as the players in an orchestra with a long season do.

The nature of these relations between people is not given a priori, not something you can establish by definition. It’s something you discover by observing them in action, seeing what they do. If they are in conflict you’ll see that. If they are working together on a project you’ll see that. And if they do both—fight and work together on a project, you will see that too.

Alain Pessin: So one can thus easily integrate conflict into the idea of a world, as long as you integrate it as a situation and not as an priori overdetermination. From this perspective, situations are absolutely not reducible to some dynamic which overpowers them. The idea of field is characterized, on the other hand, not only by the omnipresence of conflict, but by the existence of the conflict of conflicts, the conflict of social classes, which overdetermines all other social relations. Conflict is, in this conception, a generating principle of social life. It seems that you don’t share this point of view, beginning with the very idea of a generating principle of social life.

Howard S. Becker: That’s right. I don’t think there is any one generating principle. It is more likely that many principles work together in one way or another to produce the messiness of ordinary life. But it’s not just a matter of my taste. It is also, I’m sure, true that this way of looking at things is a more fruitful guide to research, because it is more open to possibilities you hadn’t thought, which careful attention to the details of social life can suggest to you. It’s better not to decide before you begin what the “important things” are.

Alain Pessin: Readers of these two points of view are sometimes tempted to say that it is a photographic problem. Bourdieu uses a wide-angle lens while Becker focuses on micro-relations; one has an overarching global view, the other does case studies. And then people go on to say that, of course, case studies are inevitably partial, that they cannot get at what is really determining in social life. The answers you have already given show that it is the overarching view which is reductive, because it systematically ignores certain aspects and certain actors who are nevertheless essential and just as determining for the results of certain social arrangements.

Howard S. Becker: The language of a “world” points us toward an inclusive notion of which actors belong in an analysis of art works, makes us recognize that everyone who contributes anything to the work being what it eventually is participates in some way in its making. That’s tautological: everyone who participates in making a work participates in making it. The advantage of that tautology is that it shows us how to incorporate into our conception of art-making the people who are conventionally left out of such an analysis: the technicians, the money people, all the people I have called “support personnel.” Their participation in making the work shows itself through a little thought experiment. Remove any of them from the action (in your mind, no one would let you do it in real life) and see what happens. If the caterers don’t provide the meals for the people in the movie crew—well, they have to eat, don’t they? If they can’t eat right there, on the set or the location, they’ll go someplace else and take longer and the production’s costs will go up. That means that more money has to be raised or that something else won’t be paid for, either having serious consequences for the final form of the film.

The basic question of an analysis centered on “world” is: who is doing what with who that affects the resulting work of art? The basic question of an analysis centered on field seems to me to be: who dominates who, using what strategies and resources, with what results? Such questions can be, and often are (repeatedly in Art Worlds), raised in an analysis based on the idea of world, as a subset of the larger set of questions that might be asked. But that much larger set of questions cannot easily be raised by an analysis centered on Bourdieu’s notion of field. Most of them, it seems to me, are a priori set aside as trivial in comparison with the “big questions” of dominance and forces.

If this is all true, then the conventional notion that you can mix Bourdieu and Becker in whatever proportions you like, according to your taste for or tolerance of conflict, let’s say, is not accurate. In fact, they ask different kinds of questions and look for different kinds of answers and are not reducible one to the other.

Alain Pessin: They start out with two different intentions, which is easily seen in the fact that the one has to extract itself from common knowledge and must oppose itself to common sense to construct, in theory, the truth about the social, while yours must immerse itself in lived practices, observing and taking seriously the procedures by which social actors construct what you call “shared understandings,” which are the only truths that the social world can produce, those which create symbolic links between real people.

Howard S. Becker: This is an important difference. Many social theories start with the premise that reality is hidden from ordinary mortals and that it takes a special competence, perhaps even a magical gift, to be able to see through these obstacles and discover The Truth. I have never believed that. To quote my mentor Hughes again, he often said that sociologists did not know anything that nobody knew. Whatever sociologists knew about social life, they had learned from someone who was part of and fully engaged in that area of life. But since, as Simmel had made clear in his essay on secrecy (Simmel 1950), knowledge is not equally distributed, everyone doesn’t know everything., Not because people are blinded to reality by illusions, but because things have been kept from them by institutional arrangements (which may or may not have been put in place to achieve that end). Sociologists find out what this one knows and what that one knows so that, in the end, they can assemble the partial knowledge of participants into a more comprehensive understanding. The idea of “false consciousness” is a classic example of the theory of social knowledge opposed to my own practice.

Alain Pessin: A sociology of situations as opposed to a sociology of structures, process versus habitus, career versus disposition, openness versus closure, choice versus determination—the exercise of analysis we have gone through, it seems to me, shows very clearly that the idea of world is in no way a “soft version” of the theory of field. One could, moreover, add that it proceeds from observation, and is very suspicious of theory. These are not two differently nuanced versions of an approach which essentially refer to the same thing. They are two ways of thinking which are opposed in their intentions and, necessarily, in their results: the philosophico-sociological approach which searches for the essence of the social, which leads to the theory of fields, and the sociologico-ethnographic approach which tries to make explicit the circumstances in which social situations create links between actors, the idea of world.

Howard S. Becker: You have captured here all the essential differences between the approaches: the one open to multiple possibilities, discovered in the course of immersion in social life; the other focused on demonstrating, on the basis of a priori considerations, the truth of an already established abstract philosophical position. I have nothing to add.


Becker, Howard S. 1982. Art Worlds. Berkeley: University of California Press.

Bourdieu, Pierre. 2004. Esquisse pour une auto-analyse. Paris: Raisons d’agir editions.

Bourdieu, Pierre  and Loïc Wacquant. 1996. The Rules of Art: Genesis and Structure of the Literary Field. Stanford: Stanford University Press.

Gilmore, Samuel. 1987. ““Coordination and Convention: The Organization of the Concert World”.” Symbolic Interaction 10:209-28.

Simmel, Georg. 1950. “The Secret and the Secret Society.” Pp. 307-78 in The Sociology of Georg Simmel, edited by Kurt H. Wolff. New York: MacMillan