An Introduction to the Danish and Brazilian editions of Outsiders

Outsiders did not invent the field of what is now called “deviance.” Other scholars had published similar ideas earlier (especially Edwin Lemert (1951) and Frank Tannenbaum (1938), both mentioned in the book). But Outsiders was different from earlier approaches in several ways. One was that it was written rather more clearly than the usual scholarly text. I take no credit for this. I had good teachers and my mentor, Everett Hughes, who supervised my dissertation and with whom I worked closely on several research projects later, was fanatical about writing clearly. He thought it was quite unnecessary to use empty, abstract terms when plain words were available that would say the same thing. And he reminded me of this frequently, so that my reflex was always to search for the plain word, the short sentence, the declarative mode.

In addition to thus being more understandable than much sociological writing, half of Outsiders consisted of empirical studies, reported in detail, of topics that were “interesting “ to the generation of students then entering American universities in a way that more abstract theorizing was not. I wrote about musicians, who worked in bars and other lowly places, playing music that had a sort of romantic aura, and I wrote about the marijuana some of them smoked, the same marijuana that many of those students were experimenting with and whose effects they were learning to enjoy (just as the analysis in the book suggested they might). These topics, intersecting more or less with their own lives, made the book one that teachers, many of whom shared the student interest in drugs and music, liked to assign to students to read. And so the book became a sort of standard text in classes for younger students.

Something else was happening at the time. Sociology was going through one of its periodic “revolutions,“ in which older theoretical frameworks were reassessed and criticized. At the time, in the early 1960s, sociologists typically studied crime and other forms of bad behavior by asking what led people to act that way, violating commonly accepted norms and not leading the “normal” lives which, all our theories told us, they had been socialized to accept as how they too should live. The theories of the time varied in what they took to be the main causes of such anti-social behavior as excessive drinking, crime, drug use, sexual misconduct, and a long list of other misdemeanors. Some blamed the psyches of the people who misbehaved—their personalities had flaws that made them do it (whatever “it” was). Others, more sociological, blamed the situations people found themselves in, which created disparities between what they had been taught to strive for and the possibility of actually getting the prizes they sought. Working class youth who had been taught to believe in the “American dream” of unlimited social mobility, and then found themselves held back by socially structured impediments (like the lack of access to the education that might make that mobility possible), might then “turn to” such deviant methods of mobility as crime.

But these theories did not ring true to a new generation of sociologists, who were less conformist and more critical of the social institutions of the day, less willing to believe that the criminal justice system never made mistakes, that all criminals were bad people who had done the bad things they were accused of, and so on. They turned for theoretical back-up to a variety of sources. Many found explanations in Marxist approaches to the analysis of the pathological effects of capitalism. Some—I was one—found a firm foundation in old-fashioned sociological theories, which had somehow been forgotten when researchers approached the field of crime and what was then called “social disorganization.”

Briefly, research on these areas of social life had been captured by the people whose profession and daily work was to solve “social problems,” activities that made trouble for someone in a position to do something about it. So crime sometimes became a problem for someone to solve. (Not always, because much crime was, as it had always been, tolerated because it was too much trouble to stop or because a lot of people profited from it.) That “someone” was usually an organization whose members took care of that problem full time. So what came to be called the criminal justice system—the police, the courts, the prisons—was conventionally given the job of getting rid of, or at least containing, crime. They put together the apparatus of crime fighting and containment.

Like all professional groups, the people in these criminal justice organizations had their own interests and perspectives to protect. They took it as a given that the responsibility for crime lay with the criminals, and they had no doubt about who the criminals were: the people their organizations had caught and jailed. And they knew that the important research problem was: “Why do the people we have identified as criminals do the things we have identified as crimes?” This approach led them, and the many sociologists who accepted that as the important research question, to rely heavily, for their understanding of crime, on the statistics these organizations generated: the crime rate was calculated on the basis of crimes reported to the police, not necessarily an accurate measure: people often did not report crimes and the police very often “adjusted” the figures to show the public, insurance companies, and politicians, that they were doing a good job.

An alternative approach was available in the sociological tradition that traced its roots to W.I. Thomas’ famous dictum: “If men define situations as real, they are real in their consequences.” (Thomas and Thomas 1928, p. 572) That is, people act on the basis of their understanding of the world and what’s in it. Stating the problems of social science this way makes the question of how things are defined problematic, points research toward finding out who is defining what kinds of activities in what way. In this case, who is defining what kinds of activities as criminal and with what consequences? Researchers working in this tradition did not accept that whatever the police and courts said was crime was “really” crime. They thought, and their research confirmed, that being called a criminal and treated as one had no necessary connection with anything the person might actually have done. There might be a connection, but it wasn’t automatic or guaranteed. Which meant that research that used the official statistics was filled with errors, and fixing those errors might lead to quite different conclusions.

Another aspect of this tradition insisted that everyone involved in a situation contributed to what happened there. Everyone’s activity had to be part of the sociological investigation. So the activities of the people whose business was to define crime and deal with it were part of the “crime problem” and a researcher could not simply take what they said at face value, or use it as the basis for further work. This ran counter to common sense, but it produced interesting and novel results.

Outsiders followed this path. I never thought that it was a novel approach. Rather, it was what a good sociologist, following the traditions of the craft, would do. It’s common these days to say that every new approach has produced what the historian of science Thomas Kuhn called a “scientific revolution” (Kuhn 1970). But I would say that this approach to deviance was not a revolution at all. If anything, you might well say that it was a counter-revolution, which got sociological research in this area back on the right track. (See the interesting discussion in Cole, 1975.)

I began by talking about crime. But now, in the last paragraph, I have referred to this area of work as focusing on “deviance.”  That is a meaningful change. It redirects attention to a more general problem than the question of who commits crime. It directs us to look instead at all kinds of activities, noting that everywhere people engaged in collective action define some things as “wrong,” what should not be done, and usually take steps to stop people from doing what has been so defined. By no means all of these activities will be, in any sense of the word, criminal. Some rules are restricted to a specific group: Jews who observe the rules of their religion should not eat food which isn’t kosher, but others are free to. Some rules govern a restricted area of activity. The rules of sports and games are like this: it doesn’t matter how you move a chess piece unless you are playing chess with someone who takes the rules seriously, and any sanction for violating the rules is in force only in the chess community. But, within those communities, the same sorts of processes of making rules and finding people who violate them operate.

In another direction, some behavior will be widely regarded as not right, but no laws apply and no organized system of finding people who break the informal rule Some of them, apparently trivial, might be seen as failures to follow the rules of etiquette (belching where you shouldn’t, for instance). Talking to yourself in the street (unless you are holding a portable telephone) will be seen as unusual and lead people to think you are a little odd, but most of the time nothing will be done about it. Occasionally, these out-of-the-ordinary actions do provoke others to decide that you might be “mentally ill” instead of just “rude” or “odd,” and then sanctions may come into play and off you go to the hospital. Erving Goffman, a colleague of mine in graduate school, explored these possibilities thoroughly, especially in his study of mental hospitals (Goffman 1961) .

The term “deviance” was meant by Goffman, by me, and by many others to encompass all of these possibilities, using a comparative method to find a basic process which took many forms in many situations, only one of them being criminal. The various formulations we proposed attracted a lot of attention and a lot of criticism, some of which I responded to in the last chapter of the revised version of the book. But over the years a large literature has grown up around the problems of “labeling” and “deviance” and I have not revised the book to take account of them.

If I did revise it, I would give a lot of weight to an idea that Gilberto Velho, the distinguished Brazilian urban anthropologist, added to the mixture (Velho 1976; Velho 1978), which I think clarifies some ambiguities which have given some readers difficulty. His suggestion was to re-orient the approach slightly by making it a study of the process of accusation, raising these questions: Who accuses who? What do they accuse them of doing? Under what circumstances are these accusations successful, in the sense of being accepted by others (at least by some others)?

I have not continued to work in the area of deviance. But I have found an even more general version of the same sort of thinking of use in the work I have been doing for many years in the sociology of art. Similar problems arise there, because it is never clear what is or isn’t “art,” and the same sorts of arguments and processes can be observed. In the case of art, of course, no one minds what they do being called art, so it is the same process seen in the mirror. The label does not harm the person or work it is applied to, as is usually the case with deviance labels. Instead, it adds value.

This is just to say that the terrain that I and others mapped out in the field of deviance still has life and is capable of generating interesting researchable ideas.

Howard S, Becker
San Francisco
February, 2005


Cole, Stephen. 1975. “The Growth of Scientific Knowledge: Theories of Deviance as a Case Study.” Pp. 175-220 in The Idea of Social Structure: Papers in Honor of Robert K. Merton, edited by Lewis Coser. New York: Harcourt Brace Jovanovich.

Goffman, Erving. 1961. Asylums. Garden City: Doubleday.

Kuhn, Thomas. 1970. The Structure of Scientific Revolutions. Chicago: University of Chicago Press.

Lemert, Edwin McCarthy. 1951. Social pathology; a systematic approach to the theory of sociopathic behavior. New York,: McGraw-Hill.

Tannenbaum, Frank. 1938. Crime and the community. Boston, New York,: Ginn.

Thomas, W.I., and Dorothy Swaine Thomas. 1928. The Child in America: Behavior Problems and Programs. New York: Knopf.

Velho, Gilberto. 1976. “Accusations, Family Mobility and Deviant Behavior.” Social Problems 23:268-75.

________. 1978. “Stigmatization and Deviance in Copacabana.” Social Problems 25:526-530.