I Outsiders

ALL social groups make rules and attempt, at some times and under some circumstances, to enforce them. Social rules define situations and the kinds of behavior appropriate to them, specifying some actions as "right" and forbidding others as "wrong." When a rule is enforced, the person who is supposed to have broken it may be seen as a special kind of person, one who cannot be trusted to live by the rules agreed on by the group. He is regarded as an outsider.

But the person who is thus labeled an outsider may have a different view of the matter. He may not accept the rule by which he is being judged and may not regard those who judge him as either competent or legitimately entitled to do so. Hence, a second meaning of the term emerges: the rule-breaker may feel his judges are outsiders.

In what follows, I will try to clarify the situation and process pointed to by this double-barrelled term: the situations of rule-breaking and rule-enforcement and the processes by which some people come to break rules and others to enforce them.

Some preliminary distinctions are in order. Rules may be of a great many kinds. They may be formally enacted into law, and in this case the police power of the state may be used in enforcing them. In other cases, they represent informal agreements, newly arrived at or encrusted with the sanction of age and tradition; rules of this kind are enforced by informal sanctions of various kinds.

Similarly, whether a rule has the force of law or tradition or is simply the result of consensus, it may be the task of some specialized body, such as the police or the committee on ethics of a professional association, to enforce it; enforcement, on the other hand, may be everyone's job or, at least, the job of everyone in the group to which the rule is meant to apply.

Many rules are not enforced and are not, in any except the most formal sense, the kind of rules with which I am concerned. Blue laws, which remain on the statute books though they have not been enforced for a hundred years, are examples. (It is important to remember, however, that an unenforced law may be reactivated for various reasons and regain all its original force, as recently occurred with respect to the laws governing the opening of commercial establishments on Sunday in Missouri.) Informal rules may similarly die from lack of enforcement. I shall mainly be concerned with what we can call the actual operating rules of groups, those kept alive through attempts at enforcement.

Finally, just how far "outside" one is, in either of the senses I have mentioned, varies from case to case. We think of the person who commits a traffic violation or gets a little too drunk at a party as being, after all, not very different from the rest of us and treat his infraction tolerantly. We regard the thief as less like us and punish him severely. Crimes such as murder, rape, or treason lead us to view the violator as a true outsider.
In the same way, some rule-breakers do not think they have been unjustly judged. The traffic violator usually subscribes to the very rules he has broken. Alcoholics are often ambivalent, sometimes feeling that those who judge them do not understand them and at other times agreeing that compulsive drinking is a bad thing. At the extreme, some deviants (homosexuals and drug addicts are good examples) develop full-blown ideologies explaining why they are right and why those who disapprove of and punish them are wrong.

Definitions of Deviance

The outsider—the deviant from group rules—has been the subject of much speculation, theorizing, and scientific study. What laymen want to know about deviants is: why do they do it? How can we account for their rule-breaking? What is there about them that leads them to do forbidden things? Scientific research has tried to find answers to these questions. In doing so it has accepted the common-sense premise that there is something inherently deviant (qualitatively distinct) about acts that break (or seem to break) social rules. It has also accepted the common-sense assumption that the deviant act occurs because some characteristic of the person who commits it makes it necessary or inevitable that he should. Scientists do not ordinarily question the label "deviant" when it is applied to particular acts or people but rather take it as given. In so doing, they accept the values of the group making the judgment.

It is easily observable that different groups judge different things to be deviant. This should alert us to the possibility that the person making the judgment of deviance, the process by which that judgment is arrived at, and the situation in which it is made may all be intimately involved in the phenomenon of deviance. To the degree that the common-sense view of deviance and the scientific theories that begin with its premises assume that acts that break rules are inherently deviant and thus take for granted the situations and processes of judgment, they may leave out an important variable. If scientists ignore the variable character of the process of judgment, they may by that omission limit the kinds of theories that can be developed and the kind of understanding that can be achieved.'

Our first problem, then, is to construct a definition of deviance. Before doing this, let us consider some of the definitions scientists now use, seeing what is left out if we take them as a point of departure for the study of outsiders.

The simplest view of deviance is essentially statistical, defining as deviant anything that varies too widely from the average. When a statistician analyzes the results of an agricultural experiment, he describes the stalk of corn that is exceptionally tall and the stalk that is exceptionally short as deviations from the mean or average. Similarly, one can describe anything that differs from what is most common as a deviation. In this view, to be left-handed or redheaded is deviant, because most people are right-handed and brunette.

So stated, the statistical view seems simple-minded, even trivial. Yet it simplifies the problem by doing away with many questions of value that ordinarily arise in discussions of the nature of deviance. In assessing any particular case, all one need do is calculate the distance of the behavior involved from the average. But it is too simple a solution. Hunting with such a definition, we return with a mixed bag—people who are excessively fat or thin, murderers, redheads, homosexuals, and traffic violators. The mixture contains some ordinarily thought of as deviants and others who have broken no rule at all. The statistical definition of deviance, in short, is too far removed from the concern with rule-breaking which prompts scientific study of outsiders.

A less simple but much more common view of deviance identifies it as something essentially pathological, revealing the presence of a "disease." This view rests, obviously, on a medical analogy. The human organism, when it is working efficiently and experiencing no discomfort, is said to be "healthy." When it does not work efficiently, a disease is present. The organ or function that has become deranged is said to be pathological. Of course, there is little disagreement about what constitutes a healthy state of the organism. But there is much less agreement when one uses the notion of pathology analogically, to describe kinds of behavior that are regarded as deviant. For people do not agree on what constitutes healthy behavior. It is difficult to find a definition that will satisfy even such a select and limited group as psychiatrists; it is impossible to find one that people generally accept as they accept criteria of health for the organism.2

Sometimes people mean the analogy more strictly, because they think of deviance as the product of mental disease. The behavior of a homosexual or drug addict is regarded as the symptom of a mental disease just as the diabetic's difficulty in getting bruises to heal is regarded as a symptom of his disease. But mental disease resembles physical disease only in metaphor:

Starting with such things as syphilis, tuberculosis, typhoid fever, and carcinomas and fractures, we have created the class "illness." At first, this class was composed of only a few items, all of which shared the common feature of reference to a state of disordered structure or function of the human body as a physiochemical machine. As time went on, additional items were added to this class. They were not added, however, because they were newly discovered bodily disorders. The physician's attention had been deflected from this criterion and had become focused instead on disability and suffering as new criteria for selection. Thus, at first slowly, such things as hysteria, hypochondriasis, obsessive-complusive neurosis, and depression were added to the category of illness. Then, with increasing zeal, physicians and especially psychiatrists began to call "illness" (that is, of course, "mental illness") anything and everything in which they could detect any sign of malfunctioning, based on no matter what norm. Hence, agoraphobia is illness because one should not be afraid of open spaces. Homosexuality is illness because heterosexuality is the social norm. Divorce is illness because it signals failure of marriage. Crime, art, undesired political leadership, participation in social affairs, or withdrawal from such participation—all these and many more have been said to be signs of mental illness.3

The medical metaphor limits what we can see much as the statistical view does. It accepts the lay judgment of something as deviant and, by use of analogy, locates its source within the individual, thus preventing us from seeing the judgment itself as a crucial part of the phenomenon.

Some sociologists also use a model of deviance based essentially on the medical notions of health and disease. They look at a society, or some part of a society, and ask whether there are any processes going on in it that tend to reduce its stability, thus lessening its chance of survival. They label such processes deviant or identify them as symptoms of social disorganization. They discriminate between those features of society which promote stability (and thus are "functional") and those which disrupt stability (and thus are "dysfunctional"). Such a view has the great virtue of pointing to areas of possible trouble in a society of which people may not be aware.4

But it is harder in practice than it appears to be in theory to specify what is functional and what dysfunctional for a society or social group. The question of what the purpose or goal (function) of a group is and, consequently, what things will help or hinder the achievement of that purpose, is very often a political question. Factions within the group disagree and maneuver to have their own definition of the group's function accepted. The function of the group or organization, then, is decided in political conflict, not given in the nature of the organization. If this is true, then it is likewise true that the questions of what rules are to be enforced, what behavior regarded as deviant, and which people labeled as outsiders must also be regarded as political.5 The functional view of deviance, by ignoring the political aspect of the phenomenon, limits our understanding.

Another sociological view is more relativistic. It identifies deviance as the failure to obey group rules. Once we have described the rules a group enforces on its members, we can say with some precision whether or not a person has violated them and is thus, on this view, deviant.

This view is closest to my own, but it fails to give sufficient weight to the ambiguities that arise in deciding which rules are to be taken as the yardstick against which behavior is measured and judged deviant. A society has many groups, each with its own set of rules, and people belong to many groups simultaneously. A person may break the rules of one group by the very act of abiding by the rules of another group. Is he, then, deviant? Proponents of this definition may object that while ambiguity may arise with respect to the rules peculiar to one or another group in society, there are some rules that are very generally agreed to by everyone, in which case the difficulty does not arise. This, of course, is a question of fact, to be settled by empirical research. I doubt there are many such areas of consensus and think it wiser to use a definition that allows us to deal with both ambiguous and unambiguous situations.

Deviance and the Responses of Others

The sociological view I have just discussed defines deviance as the infraction of some agreed-upon rule. It then goes on to ask who breaks rules, and to search for the factors in their personalities and life situations that might account for the infractions. This assumes that those who have broken a rule constitute a homogeneous category, because they have cornmitted the same deviant act.

Such an assumption seems to me to ignore the central fact about deviance: it is created by society. I do not mean this in the way it is ordinarily understood, in which the causes of deviance are located in the social situation of the deviant or in "social factors" which prompt his action. I mean, rather, that social groups create deviance by making the rules whose infraction constitutes deviance, and by applying those rules to particular people and labeling them as outsiders. From this point of view, deviance is not a quality of the act the person commits, but rather a consequence of the application by others of rules and sanctions to an "offender." The deviant is one to whom that label has successfully been applied; deviant behavior is behavior that people so label.6

Since deviance is, among other things, a consequence of the responses of others to a person's act, students of deviance cannot assume that they are dealing with a homogeneous category when they study people who have been labeled deviant. That is, they cannot assume that these people have actually committed a deviant act or broken some rule, because the process of labeling may not be infallible; some people may be labeled deviant who in fact have not broken a rule. Furthermore, they cannot assume that the category of those labeled deviant will contain all those who actually have broken a rule, for many offenders may escape apprehension and thus fail to be included in the population of "deviants" they study. Insofar as the category lacks homogeneity and fails to include all the cases that belong in it, one cannot reasonably expect to find common factors of personality or life situation that will account for the supposed deviance.

What, then, do people who have been labeled deviant have in common? At the least, they share the label and the experience of being labeled as outsiders. I will begin my analysis with this basic similarity and view deviance as the product of a transaction that takes place between some social group and one who is viewed by that group as a rule-breaker. I will be less concerned with the personal and social characteristics of deviants than with the process by which they come to be thought of as outsiders and their reactions to that judgment.

Malinowski discovered the usefulness of this view for understanding the nature of deviance many years ago, in his study of the Trobriand Islands:

One day an outbreak of wailing and a great commotion told me that a death had occurred somewhere in the neighborhood. I was informed that Kima'i, a young lad of my acquaintance, of sixteen or so, had fallen from a coco-nut palm and killed himself. . .

• I found that another youth had been severely wounded by some mysterious coincidence. And at the funeral there was obviously a general feeling of hostility between the village where the boy died and that into which his body was carried for burial.

Only much later was I able to discover the real meaning of these events. The boy had committed suicide. The truth was that he had broken the rules of exogamy, the partner in his crime being his maternal cousin, the daughter of his mother's sister. This had been known and generally disapproved of but nothing was done until the girl's discarded lover, who had wanted to marry her and who felt personally injured, took the initiative. This rival threatened first to use black magic against the guilty youth, but this had not much effect. Then one evening he insulted the culprit in public—accusing him in the hearing of the whole community of incest and hurling at him certain expressions intolerable to a native.

For this there was only one remedy; only one means of escape remained to the unfortunate youth. Next morning he put on festive attire and ornamentation, climbed a coco-nut palm and addressed the community, speaking from among the palm leaves and bidding them farewell. He explained the reasons for his desperate deed and also launched forth a veiled accusation against the man who had driven him to his death, upon which it became the duty of his clansmen to avenge him. Then he wailed aloud, as is the custom, jumped from a palm some sixty feet high and was killed on the spot. There followed a fight within the village in which the rival was wounded; and the quarrel was repeated during the funeral. . . .

If you were to inquire into the matter among the Trobrianders, you would find . . . that the natives show horror at the idea of violating the rules of exogamy and that they believe that sores, disease and even death might follow clan incest. This is the ideal of native law, and in moral matters it is easy and pleasant strictly to adhere to the ideal—when judging the conduct of others or expressing an opinion about conduct in general.

When it comes to the application of morality and ideals to real life, however, things take on a different complexion. In the case described it was obvious that the facts would not tally with the ideal of conduct. Public opinion was neither outraged by the knowledge of the crime to any extent, nor did it react directly—it had to be mobilized by a public statement of the crime and by insults being hurled at the culprit by an interested party. Even then he had to carry out the punishment himself. . . Probing further into the matter and collecting concrete information, I found that the breach of exogamy—as regards intercourse and not marriage—is by no means a rare occurrence, and public opinion is lenient, though decidedly hypocritical. If the affair is carried on sub rosa with a certain amount of decorum, and if no one in particular stirs up trouble—"public opinion" will gossip, but not demand any harsh punishment. If, on the contrary, scandal breaks out—everyone turns against the guilty pair and by ostracism and insults one or the other may be driven to suicide.7

Whether an act is deviant, then, depends on how other people react to it. You can commit clan incest and suffer from no more than gossip as long as no one makes a public accusation; but you will be driven to your death if the accusation is made. The point is that the response of other people has to be regarded as problematic. Just because one has committed an infraction of a rule does not mean that others will respond as though this had happened. (Conversely, just because one has not violated a rule does not mean that he may not be treated, in some circumstances, as though he had.)

The degree to which other people will respond to a given act as deviant varies greatly. Several kinds of variation seem worth noting. First of all, there is variation over time. A person believed to have committed a given "deviant" act may at one time be responded to much more leniently than he would be at some other time. The occurrence of "drives" against various kinds of deviance illustrates this clearly. At various times, enforcement officials may decide to make an all-out attack on some particular kind of deviance, such as gambling, drug addiction, or homosexuality. It is obviously much more dangerous to engage in one of these activities when a drive is on than at any other time. (In a very interesting study of crime news in Colorado newspapers, Davis found that the amount of crime reported in Colorado newspapers showed very little association with actual changes in the amount of crime taking place in Colorado. And, further, that peoples' estimate of how much increase there had been in crime in Colorado was associated with the increase in the amount of crime news but not with any increase in the amount of crime.) 8

The degree to which an act will be treated as deviant depends also on who commits the act and who feels he has been harmed by it. Rules tend to be applied more to some persons than others. Studies of juvenile delinquency make the point clearly. Boys from middle-class areas do not get as far in the legal process when they are apprehended as do boys from slum areas. The middle-class boy is less likely, when picked up by the police, to be taken to the station; less likely when taken to the station to be booked; and it is extremely unlikely that he will be convicted and sentenced.9 This variation occurs even though the original infraction of the rule is the same in the two cases. Similarly, the law is differentially applied to Negroes and whites. It is well known that a Negro believed to have attacked a white woman is much more likely to be punished than a white man who commits the same offense; it is only slightly less well known that a Negro who murders another Negro is much less likely to be punished than a white man who commits murder.1° This, of course, is one of the main points of Sutherland's analysis of white-collar crime: crimes committed by corporations are almost always prosecuted as civil cases, but the same crime committed by an individual is ordinarily treated as a criminal offense.n

Some rules are enforced only when they result in certain consequences. The unmarried mother furnishes a clear example. Vincent 12 points out that illicit sexual relations seldom result in severe punishment or social censure for the offenders. If, however, a girl becomes pregnant as a result of such activities the reaction of others is likely to be severe. (The illicit pregnancy is also an interesting example of the differential enforcement of rules on different categories of people. Vincent notes that unmarried fathers escape the severe censure visited on the mother.)

Why repeat these commonplace observations? Because, taken together, they support the proposition that deviance is not a simple quality, present in some kinds of behavior and absent in others. Rather, it is the product of a process which involves responses of other people to the behavior. The same behavior may be an infraction of the rules at one time and not at another; may be an infraction when committed by one person, but not when committed by another; some rules are broken with impunity, others are not. In short, whether a given act is deviant or not depends in part on the nature of the act (that is, whether or not it violates some rule) and in part on what other people do about it.

Some people may object that this is merely a terminological quibble, that one can, after all, define terms any way he wants to and that if some people want to speak of rule-breaking behavior as deviant without reference to the reactions of others they are free to do so. This, of course, is true. Yet it might be worthwhile to refer to such behavior as rule-breaking behavior and reserve the term deviant for those labeled as deviant by some segment of society. I do not insist that this usage be followed. But it should be clear that insofar as a scientist uses "deviant" to refer to any rule-breaking behavior and takes as his subject of study only those who have been labeled deviant, he will be hampered by the disparities between the two categories.

If we take as the object of our attention behavior which comes to be labeled as deviant, we must recognize that we cannot know whether a given act will be categorized as deviant until the response of others has occurred. Deviance is not a quality that lies in behavior itself, but in the interaction between the person who commits an act and those who respond to it.

Whose Rules?

I have been using the term "outsiders" to refer to those people who are judged by others to be deviant and thus to stand outside the circle of "normal" members of the group. But the term contains a second meaning, whose analysis leads to another important set of sociological problems: "outsiders," from the point of view of the person who is labeled deviant, may be the people who make the rules he had been found guilty of breaking.

Social rules are the creation of specific social groups. Modern societies are not simple organizations in which everyone agrees on what the rules are and how they are to be applied in specific situations. They are, instead, highly differentiated along social class lines, ethnic lines, occupational lines, and cultural lines. These groups need not and, in fact, often do not share the same rules. The problems they face in dealing with their environment, the history and traditions they carry with them, all lead to the evolution of different sets of rules. Insofar as the rules of various groups conflict and contradict one another, there will be disagreement about the kind of behavior that is proper in any given situation.

Italian immigrants who went on making wine for themselves and their friends during Prohibition were acting properly by Italian immigrant standards, but were breaking the law of their new country (as, of course, were many of their Old American neighbors). Medical patients who shop around for a doctor may, from the perspective of their own group, be doing what is necessary to protect their health by making sure they get what seems to them the best possible doctor; but, from the perspective of the physician, what they do is wrong because it breaks down the trust the patient ought to put in his physician. The lower-class delinquent who fights for his "turf" is only doing what he considers necessary and right, but teachers, social workers, and police see it differently.

While it may be argued that many or most rules are generally agreed to by all members of a society, empirical research on a given rule generally reveals variation in people's attitudes. Formal rules, enforced by some specially constituted group, may differ from those actually thought appropriate by most people.13 Factions in a group may disagree on what I have called actual operating rules. Most important for the study of behavior ordinarily labeled deviant, the perspectives of the people who engage in the behavior are likely to be quite different from those of the people who condemn it. In this latter situation, a person may feel that he is being judged according to rules he has had no hand in making and does not accept, rules forced on him by outsiders.

To what extent and under what circumstances do people attempt to force their rules on others who do not subscribe to them? Let us distinguish two cases. In the first, only those who are actually members of the group have any interest in making and enforcing certain rules. If an orthodox Jew disobeys the laws of kashruth only other orthodox Jews will regard this as a transgression; Christians or nonorthodox Jews will not consider this deviance and would have no interest in interfering. In the second case, members of a group consider it important to their welfare that members of certain other groups obey certain rules. Thus, people consider it extremely important that those who practice the healing arts abide by certain rules; this is the reason the state licenses physicians, nurses, and others, and forbids anyone who is not licensed to engage in healing activities.

To the extent that a group tries to impose its rules on other groups in the society, we are presented with a second question: Who can, in fact, force others to accept their rules and what are the causes of their success? This is, of course, a question of political and economic power. Later we will consider the political and economic process through which rules are created and enforced. Here it is enough to note that people are in fact always forcing their rules on others, applying them more or less against the will and without the consent of those others. By and large, for example, rules are made for young people by their elders. Though the youth of this country exert a powerful influence culturally—the mass media of communication are tailored to their interests, for instance —many important kinds of rules are made for our youth by adults. Rules regarding school attendance and sex behavior are not drawn up with regard to the problems of adolescence. Rather, adolescents find themselves surrounded by rules about these matters which have been made by older and more settled people. It is considered legitimate to do this, for youngsters are considered neither wise enough nor responsible enough to make proper rules for themselves.

In the same way, it is true in many respects that men make the rules for women in our society (though in America this is changing rapidly). Negroes find themselves subject to rules made for them by whites. The foreign-born and those otherwise ethnically peculiar often have their rules made for them by the Protestant Anglo-Saxon minority. The middle class makes rules the lower class must obey—in the schools, the courts, and elsewhere.

Differences in the ability to make rules and apply them to other people are essentially power differentials (either legal or extralegal). Those groups whose social position gives them weapons and power are best able to enforce their rules. Distinctions of age, sex, ethnicity, and class are all related to differences in power, which accounts for differences in the degree to which groups so distinguished can make rules for others.

In addition to recognizing that deviance is created by the responses of people to particular kinds of behavior, by the labeling of that behavior as deviant, we must also keep in mind that the rules created and maintained by such labeling are not universally agreed to. Instead, they are the object of conflict and disagreement, part of the political process of society.

1. Cf. Donald R. Cressey, "Criminological Research and the Definition of Crimes," American Journal of Sociology, LVI (May, 1951), 546-551.

2. See the discussion in C. Wright Mills, "The Professional Ideology of Social Pathologists," American Journal of Sociology, XLIX (September, 1942), 165-180.

3. Thomas Szasz, The Myth of Mental Illness (New York: Paul B. Hoeber, Inc., 1961), pp. 44-45; see also Erving Goffman, "The Medical Model and Mental Hospitalization," in Asylums: Essays on the Social Situation of Mental Patients and Other Inmates (Garden City: Anchor Books, 1961), pp. 321-386.

4. See Robert K. Merton, "Social Problems and Sociological Theory," in Robert K. Merton and Robert A. Nisbet, editors, Contemporary Social Problems (New York: Harcourt, Brace and World, Inc., 1961), pp. 697- 737; and Talcott Parsons, The Social System (New York: The Free Press of Glencoe, 1951), pp. 249-325.

5. Howard Brotz similarly identifies the question of what phenomena are "functional" or "dysfunctional" as a political one in "Functionalism and Dynamic Analysis." European Journal of Sociology, II (1961), 170-179.

6. The most important earlier statements of this view can be found in Frank Tannenbaum, Crime and the Community (New York: McGraw-Hill Book Co., Inc., 1951), and E. M. Lemert, Social Pathology (New York: McGraw-Hill Book Co., Inc., 1951). A recent article stating a position very similar to mine is John Kitsuse, "Societal Reaction to Deviance: Problems of Theory and Method," Social Problems, 9 (Winter, 1962), 247-256.

7. Bronislaw Malinowski, Crime and Custom in Savage Society (New York: Humanities Press, 1926), pp. 77-80. Reprinted by permission of Humanities Press and Routledge & Kegan Paul, Ltd.

8. F. James Davis, "Crime News in Colorado Newspapers," American Journal of Sociology, LVII (January, 1952), 325-330.

9. See Albert K. Cohen and James F. Short, Jr., "Juvenile Delinquency," in Merton and Nisbet, op. cit., p. 87.

10. See Harold Garfinkel, "Research Notes on Inter- and Intra-Racial Homicides," Social Forces, 27 (May, 1949), 369-381.

11. Edwin H. Sutherland, "White Collar Criminality," American Sociological Review, V (February, 1940), 1-12.

12. Clark Vincent, Unmarried Mothers (New York: The Free Press of Glencoe, 1961), pp. 3-5.

13. Arnold M. Rose and Arthur E. Pre11, "Does the Punishment Fit the Crime?—A Study in Social Valuation," American Journal of Sociology, LXI (November, 1955), 247-259.