American sociologist Howard Becker stands out for his rejection of theory and his attachment to ethnographical observation of worlds in which he is himself an actor. One of his key lessons to sociologists is that they should ask “how” rather than “why”.

Long professor at Northwestern University and at the University of Washington in Seattle, he is one of the key figures of interactionism. Drawing on art, philosophy and music, Becker demonstrates, in this interview, his eclecticism and his unique way of analyzing complex social facts in a simple and pragmatic way, using concrete examples to suggest abstract ideas.

Shared Understandings

Books and ideas: The question of social order, norms and the emergence of norms is central to all your work. How did you become interested in this question?

Howard Becker: I don’t use the notion of norm, I use the notion of « shared understanding ». A norm is like a rule, whereas shared understanding is what you and I find that we can agree on in order to do whatever we want to do. It’s not a rule we have accepted. It’s what we accept for now. We want to have an interview, so, OK, this is what we will do, this is what we can do. The notion of “shared understanding” does not suppose there is an agreement, maybe there is, but just on the object – yes, we will have an interview – but otherwise, we will find out what we agree on.

Books and ideas: Where does your interest in the question of agreement and shared understanding come from?

Howard Becker: For me, it is the basic idea. It explains how a society can exist. People understand. I know that you know. I have an idea that if I do this, I can expect you more or less to do that. Maybe you don’t. Usually you don’t do exactly what I expect. But you do something I can respond to. There is a wonderful expression I found in Wittgenstein “if a lion could speak, we could not understand him,” [1] by which he means that’s because we don’t have a way of life in common, but I would like to change this sentence this way: “if a lion could speak, we would not understand him at first.” But if we’re talking, then we are beginning to have a life together… This is an image for what I am talking about. I don’t know exactly where I found this way of thinking but it must have been when I was a student in Chicago.

Books and ideas: This question is present in all your work. It runs as a red thread in all your research.

Howard Becker: Yes and no. I did not start with this idea. The idea gets richer for me all the time, it gets more connotations, more overtones (in the musical sense of that word). I understand more and more what it could mean. I don’t like the idea of treating my ideas, my thinking and my work comme un tout. It is more like a journey. Things change. Twenty-five years ago, I spent six weeks at the University of Manchester. After I finished teaching there, I made a tour of the British islands. I gave six talks at different universities. I had the feeling that everywhere, whatever I talked about, when I was through, some young person would stand up and say: “Professor Becker, in 1965 you wrote this, but in 1972 you wrote this. The two conflict. How do you explain this?” To me, that’s not a problem. I found out something new, something I hadn’t known, so I changed my ideas. I’m always writing something new. I’m always changing my ideas. And I think this is a more realistic picture of scientific work. Because I do think that what we do is some kind of science. It is not business. The subject matters. We are trying to find out how the world works, not to create a perfect structure of ideas. That’s my idea of scientific work. So in this respect I think the way I think is very similar to the way other people in the sociology of science write, people like Latour. It is a very active field. When they describe science, it sounds to me more like what I do than building this beautiful structure of perfectly harmonious ideas. I didn’t start with an idea and follow it. I started with an idea and changed it, and changed it again and again.

The last book I published in English, with Robert Faulkner, is called Do You Know… ? [2]. To me it’s a big step forward in thinking about how people operate and share understandings. The book starts with a question. There are four guys who show up in a bar or some place, they don’t know each other, they have no written music, it is almost time to start. What are they going to do? They manage to play successfully. Everybody in the bar thinks they must have played together for years. How do they do that? That was our research question. The first answer we had, from our own experience, was: they all know the same songs, so all they have to do is say the name of the song and then they can play it. When Robert Faulkner started his fieldwork, he wrote down what happened. But that’s not what happens. What happens is, somebody says “Let’s play ‘Let’s Fall In Love’”. Actually, he would say, “Do you know ‘Let’s Fall In Love’?” And someone else might say “Sure, what key? B flat. OK, let’s play”. But sometimes, someone says “No, I don’t know that”. So now, what do we do? What you see then is a process of working together to find what they can do. It’s a perfect metaphor of shared understanding: we know some things together, some things we don’t, but there are some ways of working with what we know so that we can play together. Another answer could be: “No, I don’t know that song, but if you play the first chorus, I can play the second one”. They have a body of skills in common that is enough for them to play together. That is something I learned. I’ve had this idea of shared understanding for a long time. But it’s only now that I really understand what it means. In that way, you can think of all social life as improvisation. People improvise, and when they find something that works, that’s it.

Ordinary Kinds of Work

Books and ideas: You have worked on the professions. In France, the interest in the professions is currently expanding under the influence of Anglo-American theories, but also under the influence of the sociology of professional groups; it proposes variations to American functionalism and interactionism. How would you position your work on professions (for example your work on teachers) vis-à-vis this new French sociology of professions? [3]

Howard Becker: When I was a student, my mentor, Everett Hughes, used to teach a class called sociology of professions. But he said it was a mistake: people were interested because they had the idea that these professions (law, medicine, religion…) were different from other kinds of work. They thought these kinds of work were better, more noble. Hughes thought this was a very partial view of the world of work. When I came to him and told him I would like to do my research for my master’s thesis on musicians in bars, well, he was looking for people like me who would study things that did not have so much prestige or honour. Ordinary kinds of work. He very soon changed the name of his class, first to “Sociology of occupations and professions”, and finally “Sociology of occupations”. And finally, it became Sociologie du travail.

Professions in American sociology had this peculiar relation with functionalism. Talcott Parsons had the idea that there were these basic problems, these basic profound values, etc., and the professions embodied, incarnated these highest ideals, the values of the society. They were about important things, such as medicine, the body, health. That was not a finding of research. That was a conviction. But if you know doctors and if you know lawyers, they are just like everybody else, they are not very noble.

For me and for my colleagues, like Eliot Freidson, professions were something that people who were organized to do a certain kind of work wanted to have as a label, because it allowed them to have more autonomy, to do as they wished. Whereas if you are a musician playing in a bar, you do what the owner wants you to do. You can’t say no, you can’t say “I represent the musical ideals of our society”, you play what they want. So for me, a profession is just a small portion of a larger idea of the sociology of work. For Hughes, the sociology of work was everything. One of the things he liked to say was that everything that happens is somebody’s work

“Whatever Allows Us to Understand More”

Books and ideas: In your book Art Worlds [4], you use the concept of “world” to talk about the art sphere, but also about other professional worlds. The concept of “world” allows you to emphasize the notion of “shared understanding” we have just talked about. Doing so, you distinguish yourself from another tradition that uses another concept to explain the rationality of the art world, Bourdieu’s concept of “champ”. You never mention it in your book.

Howard Becker: I had not read Bourdieu when I first wrote about art worlds, because he had not written about these things yet. When I first proposed the idea of art world, in a 1974 article, it was not about rupture. At the time, I did not read French. That came later. I have said this several times, but it’s true: The first book I read “en français” is Raymonde Moulin’s book on the art market [5]. For me that was the French sociology of art. Bourdieu’s work was not useful for me. The concept of “world” deals with a completely different reality than the idea of “field.” The concept of “champ” is filled with things you can’t observe, invisible forces. The idea of world refers to visible, common things you can see easily. Everywhere I looked, I found the workings of the connections that made up art worlds. There is this wonderful story I found in a biography of Picasso, about the man who made the ceramics that Picasso decorated. This craftsman said that the things Picasso wanted done were impossible, but Picasso manipulated him so that he finally did what Picasso wanted, the things he had said he couldn’t do. When the ceramist said, “This is impossible, Monsieur,” Picasso replied, “Oh, I’m sorry, I thought you were really good, I see you aren’t as good as I thought you were.” This was part of his skill. As an artist, he had to know what he could say to this man to get him to do what he could not do. There’s no place for that in the idea of “champ”. I don’t think there is even a conflict between me and Bourdieu, I just think it is another way of thinking. For me, the idea of champ isn’t useful [6]. I don’t have this conception of social science as a battlefield. It happens of course, often enough, but the heart of the enterprise is to find things out. How does this work? How does this bar here... how do these people here work together? That requires explanation. That’s what I would like to explain. From this other perspective, the problem is, “When I look here, surely there is some form of domination. Where is it?” But if you actually look at what’s happening there, can you see domination? That’s a problem. Surely, there probably is, but that’s not everything. This kind of question isn’t interesting to me. I am just interested in pursuing these ideas and seeing what we can do with them. I learned this from Hughes. It’s very pragmatic. Whatever works. Whatever allows us to understand more.

Deviance is a Judgement

Books and ideas: In the definition of deviance you propose in Outsiders [7], there is an important part of arbitrary. You give the example of the Federal Bureau of Narcotics that becomes, after the Prohibition, a moral entrepreneur. You seem to draw the same conclusions as Foucault, though through a very different itinerary. Has Foucault’s work contributed to your reflection on deviance?

Howard Becker: No. I found out many years later, through a young French scholar, that what I wrote about the development of these laws was very ignorant. François-Xavier Dudouet wrote this wonderful thèse about the international control of illicit drugs. [8] In the United States, the laws forbidding the use of narcotics were all justified by one idea, the reason given for them was: we have an international treaty with other countries requiring such laws, so the United States had an obligation to have these laws because of the international agreements it signed, first in the League of Nations, then in the UN. But that wasn’t the origin of these prohibitions. No one who was involved in making those agreements had any interest in the problem of addiction or in the illicit use of drugs. They were interested in something quite different: maintaining the monopoly that France, Great Britain and the United States had on the legal market in drugs, opiate drugs used in medical practice and in hospitals. That’s where the money is. These countries were interested in maintaining that monopoly which is far more profitable. It’s not junkies in the streets. It’s the hospital and the doctor’s office where, every day, there is so much cocaine, so much morphine used. That’s where the money is. The representatives of these countries wanted to make sure that nobody brought illicit drugs into that market and threatened their monopoly. So everything that had been written about this by everyone was wrong. I suppose that Foucault didn’t know this very important part of the story. So now, I have a completely different understanding of what happened. It is really a matter of political and economic actors protecting their interests.

Books and ideas: Your definition of the norm is relative – relative to a historical moment, relative to a particular group. People one would spontaneously label as “deviant” are also following a set of norms, that are specific to the group. But are you not afraid that the concept of “deviance” gets diluted?

Howard Becker: I never use the word “norm”. Instead, I use the idea of “shared conventions”. This isn’t just a question of usage, it’s the idea behind the expression that I don’t accept. People don’t have norms, they have ideas. If you do it this way, you probably have a reason. If you don’t do it this way, this is a waste of your time, so I am going to show you how it works so you can do it. It is not a norm, it is practice. Norms always have this overtone of “this is the right thing to do.” But nobody will tell you that you have to smoke marijuana because this is the right thing to do, or tell you that you have to smoke it this way otherwise you’re doing something wrong. But they will say, “If you want to achieve this result—getting high—it would be better if you smoked this way.”

Deviance is the result of a work of definition. Nothing is inherently deviant. You know, it is like the duck-rabbit optical illusion. You look at it one way, it looks like a duck. You look at it another way, it looks like a rabbit. There is an excellent definition of this idea in French. What is a mauvaise herbe? (Literally, “a bad plant” and in ordinary English a “weed”) It’s not bad in itself, it’s just not where someone thinks it should be. For me, for instance, a blackberry plant isn’t a weed, it’s a plant that produces wonderful berries. But for someone who has a garden it’s a terrible weed. This is a good model to use in thinking about things like this. In fact, one of the early sociologists of deviance in America, David Matza [9], wrote a paper which he never published on “weeds”. It is a perfect metaphor.

Books and ideas: So what you mean is that one is always someone else’s deviant? But does deviance really exist if it depends on one’s standpoint?

Howard Becker: Deviance is not a fact, it’s really a judgment. It’s like the optical illusion. I know that people think this way, they can’t see how they think, I know they do, I accept that some people think this. So to me, it is obvious, just as it is to them. If I say to them, show me the deviance, they say “don’t be silly. You know what I mean”. No. I am like, you know, Latour’s first chapter of La science en action [10]: the man who doubts, “I don’t understand this, show me the machine, how do you know this machine works?” Finally the scientist says “See, we put the morphine in a solution and the muscle reacts like this”. And the man asks “But how do you know this is morphine?” The scientist says “Look at the label, it says morphine, it was made by such and such laboratory in Chicago”, “But how do you know they are not lying?”, “Well, go away”. At a certain point, no one argues except someone who is crazy. Of course, it sometimes happens that morphine is not morphine. Some companies cheat. But some things we accept. We are not going to argue, we accept that if it says morphine, it is morphine. It’s the same thing with art. How do we know this is art as opposed to everything else that isn’t art? When I was a student for a short time in an art school in San Francisco, when I learned photography, there was a French visiting artist who captured and took away the tops of restaurant tables covered with the dishes and uneaten food from real meals. Is that art? This is why Marcel Duchamp was such a key figure. The question is not: is what Duchamp did art? It is art because it’s in the Museum. That’s the working definition of art: it’s in the museum, collectors will pay for it. Philosophically, it’s a real difficulty. The world of art has taken a position on Duchamp, they all agree this is art. So, practically, it is not a problem, but philosophically it is.

Books and ideas: In Outsiders, you show that there is a career of deviance. According to you, the relevant question is not ”why” but “how”. However, the “why” seems to resist, as shown by the multiplication of epidemiological studies, political speeches…

Howard Becker: The reason why I say don’t ask “why”, ask “how”, is very practical. If I say to you “why did you become a sociologist”, you have to give me a good answer. Most of us didn’t know what we were doing when we chose this profession. I never had the intention to be a sociologist, I wanted to be a terrific piano player. Sociology was a sort of hobby. But after some years, I realized I could be a good piano player, but never a terrific piano player. I had to acknowledge the truth: I was a sociologist, not a professional musician. I still could not tell you why I became a sociologist, but I can tell you how. If I ask why, I get a defensive response, the person I question wants to give me a “good” reason, a defensible reason. But if I ask: How did it happen? You can give me practical answers. This way of thinking has a precise origin in the United States, C. Wright Mills’ 1940 article, “Situated Actions and the Vocabulary of Motives.” [11]

Interview with Howard Becker conducted by Olivier Pilmis and Nadège Vezinat, 12 October 2010 in Paris.

To quote this article :

Pauline Peretz & Olivier Pilmis & Nadège Vezinat, « Social Life as Improvisation. Interview with Howard Becker », Books and Ideas , 31 January 2011. ISSN : 2105-3030. URL :

Nota Bene:

If you want to discuss this essay further, you can send a proposal to the editorial team. We will get back to you as soon as possible.

by Pauline Peretz & Olivier Pilmis & Nadège Vezinat , 31 January 2011

Related articles

Giving Gifts

Debate between Anne Bory, Nicolas Guilhot, Sabine Rozier and Olivier (...)
by Pauline Peretz

The Empirical Sociology of Critique

An Interview with Luc Boltanski
by Nicolas Duvoux


[1L. Wittgenstein, Philosophical Investigations, 1953, II, xii.

[2Do You Know . . .? The Jazz Repertoire in Action (Chicago: University of Chicago Press, 2009), with Robert R. Faulkner

[3D. Demazière, C. Gadéa, Sociologie des groupes professionnels. Acquis récents et nouveaux défis, (Paris : La Découverte, 2009).

[4H.S. Becker, Art Worlds (Berkeley: University of California Press, 1982)

[5R. Moulin, Le marché de la peinture en France. Paris, Minuit, 1967 (Abridged English translation: The French Art Market. A Sociological View. New Brunswick, Rutgers, 1987).

[6H.S. Becker, “A Dialogue on the Ideas of ‘World’ and ‘Field’ with Alain Pessin,” Sociological Forum, 21 (2006), 275–86.

[7H.S. Becker, Outsiders: Studies in the Sociology of Deviance. (New York: The Free Press, 1963)

[8F.-X. Dudouet, Le grand deal de l’opium: Histoire du marché légal des stupéfiants. Paris, Syllepse, 2009.

[9See. D. Matza, Becoming Deviant. Englewood Cliffs (NJ), Prentice Hall, 1969.

[10B. Latour, Science in Action. Cambridge (Mass.), Harvard University Press, 1987.

[11C. W. Mills, “Situated Actions and Vocabularies of Motive”, American Sociological Review 5 (6), 1940: 904-913.

© - Any replication forbidden without the explicit consent of the editors. - Mentions légales - webdesign : Abel Poucet