I THE CONCEPT OF NORMSRobin M. Williams, Jr.
II THE STUDY OF NORMSJack P. Gibbs
A norm is a rule, standard, or pattern for action (from the Latin norma, a carpenter’s square or rule). Social norms are rules for conduct. The norms are the standards by reference to which behavior is judged and approved or disapproved. A norm in this sense is not a statistical average of actual behavior but rather a cultural (shared) definition of desirable behavior.
To the extent that a particular social norm actually is effective, one will be able, of course, to observe a marked actual regularity of social acts in recurrent situations of a particular kind. Thus there will be more or less standardized ways in which people are seen to behave when conducting trade, engaging in religious worship, or playing games. A sheer uniformity in behavior, however, does not necessarily mean that a norm is involved. The uniformity may simply represent such separate individual reactions to a common stimulus as fleeing from fire. Nevertheless, the great majority of important social interactions are guided in part by norms.
Related concepts. The history of the concept of norm goes back as far as men have recorded their ideas of proper conduct. Although the term has been fairly common in philosophical analyses of ethics, in social science use of the specific word “norm” has become extensive only in recent times (the Encyclopaedia of the Social Sciences, published in the 1930s, did not even include the term). Closely related concepts, however, have long been current in the discourse of the social sciences: for example, convention, etiquette, law, custom, folkways, and mores.
Custom. The very broad term “custom” frequently has been used to refer to the whole body of those shared behavior patterns in a particular group or society that are regarded as the traditional and established ways of the people. To speak of a “customary” way of acting is to suggest that the conduct in question has come down from the past and is not regarded solely as a matter of immediate expediency. Customs are more than mere aggregates of individual habits; a custom is always tacitly supported by social approval. Negative social sanctions, ranging from ridicule to extreme punishment, may be applied when there is deviation from custom. Of course, sanctions may not occur when the deviation is as slight as failure to take a siesta in the tropics; but they will certainly occur when it is as great as marrying a forbidden clan-sister. Thus, the actual enduring regularities in social behavior, accepted by and approved by most members of a group or society, may be called customs. The standards used for judging conformity are the norms for the customary behavior.
Folkways. The concept of folkways was developed by the American sociologist William Graham Sumner and by his followers. Sumner considered folkways to be group habits or customs which originated in the repetition of adjustive acts by individuals responding to a similar social or individual need. The folkways, thus, are concrete patterns of behavior; they have usually been interpreted as the outcome of largely unreflective trial-and-error adjustments.
Mores. Customs that are regarded by general agreement as highly important and obligatory, as evidenced by strong sentiments against deviation and by severe punishments for violation, are usually called mores. When certain folkways become well-established and are supported by the belief that they are proper, right, and indispensable, they become mores. They are ordinarily thought to be supported by diffuse common agreement. Examples of mores include incest taboos and rules against in-group murder, rape, cannibalism, and other practices generally regarded as especially heinous. Some definitions of mores include as a defining criterion the presence of some overt philosophical justification of these customs as indispensable to social welfare.
Characteristics of norms . Social pressures arising from group acceptance of norms have definite consequences for behavior. When, for example, at a given time an individual is confronted with incompatible norms held by different individuals or groups that are important to him, his behavior is likely to show signs of psychological conflict or tension. Unanimity of support for norms within a group or society powerfully checks tendencies of individuals toward deviance (Asch 1952, chapter 16).
Norms are learned by individuals in social intercourse with others—that is, in the process of socialization. By definition, then, norms are shared by two or more individuals (Sherif 1936). Some norms are particular to quite small groups, such as a husband-wife pair or a clique of friends, while other norms may be held by a large collectivity that is one of several existing in one nation (for example, Muslims and Christians in Lebanon). Still other norms may be shared by most adult members of a nation or of an entire civilization (for example, those pertaining to monogamous marriage).
Norms are always to some extent both generalized and generalizable. They may refer to all human beings at all times and in all places, or they may refer only to a specific category of person in a specific type of situation. A norm calls for “right action” and implies a generalizable reason for the rightness of the indicated conduct. Ultimately this propriety or Tightness traces back to some standard of value that is taken without further justification as valid by the individual or group in question.
Norms thus are more than an idiosyncratic expression of the wants or desires of a particular person. Even quite specific norms imply a basis for assent by someone other than the norm sender himself (Rommetveit 1953, pp. 11-42). At the same time, norms actually can guide conduct only if they prescribe or proscribe identifiable courses of action; therefore norms are more specific and socially imperative than values or ideals. For example, “honesty” is a generalized value (a “conception of desirability”), but it is still found socially necessary to have specific rules for concrete situations such as students’ behavior on examinations or the financial responsibilities of banking officials.
Norms may arise in relation to any aspect of human activity and experience that comes to be regarded as of any importance or consequence. There are norms for perceiving, feeling, thinking, judging, evaluating, and acting. Thus, there are cognitive and logical norms for carrying on scientific investigations, aesthetic norms for judging music or sculpture, and norms of grammar and style for writing in a particular language. A religious believer relates himself to supernatural or transcendent entities in accordance with certain norms. Convention, etiquette, fads, and fashions are all regulated by norms. Finally, there are the norms of moral conduct which guide direct social interaction. The term “social norm” may refer to any of these, but most commonly is used to designate rules for social interaction.
Norms are exceedingly diverse not only in their objects but also in respect to their important properties in different societies and in different historical periods. Some norms are widely known, accepted, and followed, whereas others are characterized by low consensus and only partial conformity. Some norms are learned early in life, through identification with parents or other primary agents of socialization; others are acquired in later life through secondary social relations. To understand any particular norm it is important to know whether it has been “internalized” so as to become part of the conscience or self-ideal of the individuals in question; if so, there will be much conformity even if there is no external surveillance of conduct or punishment for deviant behavior. Norms not thus internalized can be enforced only through external rewards and penalties. Those norms that are primarily enforced through punishment and threat contrast with standards that are maintained by a flow of positive social rewards such as wealth, prestige, or social approval (see Rommetveit 1953; Parsons 1951; Thibaut & Kelley 1959). Great variations exist in consistency of enforcement, source of authority, degree of allowable variation in conformity, extent of deviance, and type of enforcing agency (Morris 1956).
Norms and social needs. What is it that makes a norm normative? How does a norm acquire the obligatory quality which distinguishes it from a simple habit or preference? How does it occur that a man accepts the legitimacy of a criterion for conduct that results in his disadvantage in the immediate situation of action? (cf. Sorokin 1947, chapter 4, on law norms). The answer seems to be that norms arise to meet recurrent problems. They tend to be initiated or proposed by someone who finds an immediate agreement to be in some way advantageous. They persist when, on the basis of the basic values and beliefs accepted in a society, acceptance of norms “cuts costs” (Thibaut & Kelley 1959; Homans 1961). Of course, the types of advantages so obtained exhibit great qualitative diversity.
Types of norms . There are several intrinsically different kinds of norms that vary according to the locus of human activity and experience to which they are applied.
Technical norms have to do with effective means for the attainment of specific goals (Pepper 1960). They define effective action in dealing with physical and biological as well as social elements in immediate situations. When the satisfaction of individual desire for pleasure and affection is the goal we may call such norms hedonic. When the individual must choose between different personal purposes and satisfactions, his choices may be guided by personality-integrative norms. For example, “eat, drink and be merry” may be countered by “moderation in all things.” Likewise, long-term goals may generate considerations affecting immediate gratifications.
Hedonic or technical norms and the norms of personality integration, however, are insufficient for guiding social interaction, which involves the interests and claims of many persons and groups. Hence, distinctively social rules emerge, specifying rights, duties, disabilities, and privileges. Furthermore, most if not all cultures contain norms which regulate what are considered to be man’s relations with the superhuman and supernatural.
The most definite, strongly held, and socially obligatory norms regulating conduct may be called institutional norms. In any group or society there are certain norms that are widely known and accepted, supported by legitimate authority, and incorporated into the individual’s social conscience early in life. However, there is no sharp line between institutional and noninstitutional norms. The degree of institutionalization can be represented statistically as a profile exhibiting more or less of each one of the characteristics just reviewed.
To the extent that there is consensus concerning the norms regulating conduct, a pragmatic basis is available for integration among the persons and other units making up social systems (Parsons 1951). Choice among various norms regulating interaction may in turn be decided with reference to generalized norms that transcend the particular context and shade over by degrees into cultural values. It is generally thought that a cultural system generates complex pressures toward consistency among its beliefs, values, and norms. A central empirical problem in the sociological study of norms is to ascertain whether norms combine into systems, and, if so, what conditions and reasons govern these combinations.
Thus, there are six main foci of normative regulation of human conduct: the internal functioning of the individual personality; relations with the world of physical and biological phenomena external to the actor; interpersonal and intergroup relations within a particular society; intersocietal relations; intracultural integration; and man’s relations with supernatural entities and events.
Conformity. The greater the agreement upon a norm among the members of a social system, the greater in general will be the pressures for conformity to that norm. Some norms acquire great force by sheer unanimity; for example, in a society in which nearly everyone speaks only Chinese, one must speak Chinese to be understood. Conformity is rendered more likely if the would-be deviant has no alternate group to support his deviance or to which he could escape to avoid penalties (Thibaut & Kelley 1959, chapter 10). If compliance with norms is induced primarily by punishment, much surveillance is required; conformity induced by rewards, by intrinsic satisfactions, or by consensus does not need to involve detailed monitoring and supervision. When persons (or collectivities) are highly interdependent within a closed social boundary, the implicit threat of withdrawal of reciprocity is an effective sanction against nonconformity.
In ordinary interpersonal relationships, conformity to every detail of every specific norm is not rigidly required; rather, sanctions tend to focus upon the total role relationship (Goode 1960, p. 252). When consensus is low, personal mobility high, and individuals carry on many of their activities in nonoverlapping social networks, both the demand for conformity and the actual degree of conformity tend to be low.
Conformity in inner belief or in affective intensity and depth of commitment is more difficult to observe than overt behavior. Therefore, any type of “thought control” is difficult to maintain. On the other hand, any expression of disbelief in a norm (or normative system) and any overt evidence of lack of emotional commitment is likely to be especially threatening to those who are committed to enforcing the norm. The sinner can always repent; the heretic challenges the very existence of the norm (cf. Goode 1960, p. 257).
Norms and freedom. Consideration of institutionalized norms raises questions about the meaning of “freedom” in human affairs; a necessary first step in examining this concept is to draw a distinction between psychological and structural freedom in social systems. Psychologically, a man is “free” if he can do whatever he wishes to do; if his desires are more limited than the possibilities for action that are open to him, he will believe and feel himself to be free. It is possible for him to have this subjective freedom even though a detached observer with a more extended knowledge of the circumstances might regard him as closely constrained by institutionalized norms and sanctions. This kind of freedom is not, in short, descriptive of any clearly identifiable feature of social structure.
Different social structures vary widely in the extent to which they enforce a monolithic system of norms. Pluralistic societies contain a variety of relatively autonomous norm-creating and normenforcing groups, associations, and governmental units. Maximum structural freedom exists when the individual has many alternative choices between groups with varying norms and varying types and degrees of demanded conformity.
Origins of norms . Much research will be required to establish a clear understanding of the conditions and processes associated with the emergence and acceptance of norms. Some sources of norms are known. Many rules for behavior seem to arise as a direct consequence of recurring problems that are found to be socially costly. Examples of such conditions of high “cost” (in terms of frustration, deprivation, and conflict) include: persistent failures of coordination in important group tasks, mutual misunderstandings, chronic interferences in achieving consensus, low predictability, recurring interpersonal conflicts, and incompatible claims to legitimate authority. Persons in weak positions, fearing exploitation by the stronger, often favor predictable norms and sanctions. Persons in positions of authority and power often desire reliable conformity that is not dependent upon continual surveillance and coercion; they may therefore favor establishment of norms even if their own caprices are thereby curbed.
A demand for norms is likely to arise from persons who find their interactions confusing or vaguely defined; for this reason, unstructured situations often create a pressure for the development of new norms. Enduring social conflicts, when not of too great an intensity, also generate new norms, developed out of negotiation, compromise, mediation, and related social processes. Another major source of new norms lies in collective reactions to shared “strain” experienced in relation to old norms.
Every assertion of a demand or the communication of an expectation from one person to another contains some, however slight, normative element, even if the norm is not stated. To communicate an expectation to another person about his own behavior is to implicitly claim the right to influence his conduct. Acquiescence in the expectation by the receiver establishes a shared commitment. Such shared expectations form the initial basis for development of a behavior standard for later recurrent situations of the same kind.
Needed research . Contrary to a priori theories which would deny the presence of normative elements in man’s behavior, modern social science finds a central place for the concept of norms in its analysis of social systems (Parsons 1951, pp. 3-45). But we need to know more about the consequences of partial consensus and the relation of consensus to social cohesion. Under what circumstances do old norms cease to be effective because of gradual alienation, loss of support and concern? When does such “withering away” lead to anomie (normlessness), and when and how does it prepare the way for new norms? In addition, we need rigorous study of the sources of rebellion, deviance, and creative reformulation of norms. Such research will add to knowledge of those conditions under which self-realization is possible in an increasingly complex and interdependent world.
Robin M. Williams, JR.
Angell, Robert C. 1958 Free Society and Moral Crisis. Ann Arbor: Univ. of Michigan Press. → See especially Chapters 2 and 5.
Asch, Solomon E. (1952) 1959 Social Psychology. Englewood Cliffs, N.J.: Prentice-Hall. → See especially Chapters 12 and 19.
Ginsberg, Morris (1932-1955) 1956 Essays in Sociology and Social Philosophy. Volume 1: On the Diversity of Morals. London: Heinemann.
Goode, William J. 1960 Norm Commitment and Conformity to Role-Status Obligations. American Journal of Sociology 66:246-258.
Homans, George C. 1961 Social Behavior: Its Elementary Forms. New York: Harcourt.
Morris, Richard T. 1956 A Typology of Norms. American Sociological Review 21:610-613.
Parsons, Talcott 1951 The Social System. Glencoe, III.: Free Press. → See especially Chapters 1 and 2.
Pepper, Stephen C. 1960 Ethics. New York: Appleton. → See especially Chapters 1, 2, and 13.
Rommetveit, Ragnar (1953) 1955 Social Norms and Roles: Explorations in the Psychology of Enduring Social Pressures With Empirical Contributions From Inquiries Into Religious Attitudes and Sex Roles of Adolescents From Some Districts in Western Norway. Oslo: Akademisk Forlag; Minneapolis: Univ. of Minnesota Press. → See especially pages 11-42.
Sherif, Muzafer (1936) 1965 The Psychology of Social Norms. New York: Octagon.
Sorokin, Pitirim A. (1947) 1962 Society, Culture, and Personality: Their Structure and Dynamics; A System of General Sociology. New York: Cooper. → See especially Chapter 4.
Thibaut, John W.; and Kelley, Harold H. 1959 The Social Psychology of Groups. New York: Wiley. → See especially Chapters 8 and 13.
Williams, Robin M. JR. (1951) 1960 American Society: A Sociological Interpretation. 2d ed., rev. New York: Knopf. → See especially Chapter 3.
No concept is invoked more often by social scientists in explanations of human behavior than “norm.” However, despite the plethora of “normative” explanations, the conceptual treatment of norms remains unsatisfactory. No particular generic definition of norms is widely accepted in the social sciences, and consensus is lacking as to the differentiation of types of norms. Moreover, for the most part normative explanations are ad hoc—that is, they do not stem from systematic comparative research. In fact, given the state of the concept and the paucity of systematic studies of norms in general (as opposed to particular norms, such as the incest taboo), normative explanations of behavior are often dubious. The alleged prescription or proscription may be merely assumed to exist, and the conceptual and methodological issues inherent in any such explanation are rarely discussed.
Issues and questions
Any normative explanation presupposes the existence of norms, and the adequacy of the explanation therefore depends partially on the validity of this presupposition. At first glance the only problem might appear to be that of perfecting techniques for identifying norms, but techniques alone cannot resolve the conceptual issue of how to define norms in general. Most definitions emphasize that a norm necessarily represents a high degree of consensus in a social unit as to what the conduct of members ought to be. But, if so, what is the “normative” status of laws that govern a country under military occupation or, more broadly, the status of all unpopular laws? Are we to conclude that such laws are not, after all, norms? It is far better to recognize that, by definition, some but not all norms represent collective evaluations (that is, consensus as to what conduct ought to be). Other norms represent nothing more or less than collective expectations, meaning consensus as to what the behavior of members of a social unit will be. Further, an evaluation or expectation (collective or noncollective) of conduct may or may not be associated with attempts to prevent, revenge, or rectify contrary behavior; and such attempts may or may not be made only by persons in a particular status through means that may or may not include the use of force.
The point is that a truly generic definition of norms must encompass a wide range of definitional attributes. If this is borne in mind, it enables one to distinguish types of norms, such as customs, mores, rules, laws (Gibbs 1965). Equally important, if it is recognized that what is a definitional attribute for one type of norm is a contingent attribute for another type, then what was once a sterile conceptual issue gives rise to some significant empirical questions on norms in general. For example, why are some laws more consistent with collective evaluations than others? How much do societies differ with regard to the consistency between legal and extralegal norms? Why do they differ? Why do some collective evaluations and collective expectations diverge, as when members of a social unit expect other members to drink alcoholic beverages but have no definite collective evaluation of the act? Why are some but not all collective evaluations backed by force? How much and why do social units differ as to the use of coercion to revenge or prevent violations of norms?
Social scientists are not completely ignorant in the face of such questions as these. Some notable studies have been conducted in jurisprudence, social anthropology, and sociology (Diamond 1935; Durkheim 1893; Hoebel 1954; Sumner 1906; Weber 1922); but demonstrably valid answers are wanting. Indeed, for the most part, truly systematic research on normative phenomena is now largely limited to the study of small groups (see Rommetveit  1955 for references); and comparative studies at the macroscopic level, such as that conducted by Richard D. Schwartz and James C. Miller (1964), are now rarer than in the heyday of evolutionary theory in the social sciences.
Regardless of how the conceptual issue is resolved, the empirical identification of norms presents some complex technical problems. Consider laws as a case in point. Analytical definitions (e.g., Weber  1954, p. 5) stress the point that a legal norm is characterized by a “high probability” that an attempt will be made, by means which may include the use of force, to prevent, revenge, or rectify contrary behavior. However, these definitions do not specify a numerical criterion for a “high probability,” and any criterion (for example, 50 per cent) would be arbitrary. Moreover, it is difficult to identify coercion as an attribute of laws as such. The frequency with which force is employed may not be relevant because violence in the enforcement of laws may be rare when there is a high degree of consistency between legal and extralegal norms. Additional complications stem from the fact that force does not play the same role in all types of laws (for example, in criminal as opposed to tort law). Finally, one could argue cogently that a law is identified not only by a high probability of enforcement but also by a low probability of retaliation as a reaction to enforcement; yet, currently there is no defensible sociological rationale for stating exactly how probable retaliation may be.
The pressing problem in the identification of legal norms is that of moving from general definitions to specific empirical criteria that are applicable in all societies. It is most unlikely that this problem can be solved without extensive research of a special kind on laws in widely diverse societies. While it is possible to identify some norms in a given society as “legal” and others as “extralegal” by virtually any definition, we know very little about the related frequency of enforcement, use of force, and retaliation for enforcement. Such knowledge could provide a basis for achieving the ultimate goal—specification of empirically applicable criteria for identifying legal norms.
The technical problems posed in identifying extralegal norms are different from the case of laws but no less difficult. On the whole, the problems fall into three classes—the semantical, the dimensional, and the statistical. Semantical problems stem from the fact that most definitions treat norms as prevalent beliefs as to what conduct ought to be. Accordingly, any study of such beliefs (e.g., Stouffer 1955) is an investigation of norms. The crucial technical problem in conducting such a study is that “normative” opinions can be solicited in a variety of ways. For example, consider the alleged proscription in the United States of sexual relations outside of wedlock. How can the existence of this norm be demonstrated? One approach would be to ask a representative sample of the American public a question worded along the following lines: Should a man and a woman who are not husband and wife have sexual relations with each other? Observe, however, that the question could be phrased in several other ways. For that matter, an entirely different kind of question could be posed, such as whether or not the person being questioned would commit the act, or what the person thinks should happen when the act is committed (i.e., the “reactive norm” as described in Clark & Gibbs 1965). One could argue, of course, that the results would be substantially the same regardless of the kinds of questions; but research to date does not provide a basis for such a generalization, let alone a conclusion as to the best way to solicit opinions about norms.
Apart from the choice of words in formulating normative questions, there is the problem of structuring the solicitation of opinions in such a way as to avoid ambiguity. To illustrate, consider the following question as it might be posed in a survey of norms in the United States: Should Negroes and whites be permitted to have sexual relations with each other? Some respondents probably would be confused and/or give an answer that is misleading, because nothing is said in the question about whether or not the couples are married. Thus, a respondent of liberal convictions may assume that the question refers to racial intermarriage and answer Yes, while another may assume that extramarital relations are the point of the question and answer in the negative. Indeed, two respondents could agree in their answers even though their interpretations of the question differ. The general point is that most prescriptions or proscriptions of conduct are relative to statuses or situations, and related opinions can be solicited in a meaningful way only by possessing prior knowledge of the norms involved. The solution may require interviews in depth, as in John Ladd’s study of Navajo ethics (1957), before normative questions can be formulated in such a way as to elicit answers that can be easily classified and analyzed in quantitative terms. Without systematic classification and quantitative analysis, we have, as in the case of most anthropological studies, only the investigator’s intuitive assessment of the norms in a social unit.
One dimensional problem in the identification of norms is that statements as to what conduct ought to be do not necessarily reveal the intensity of norms. By way of illustration, consider opinions in the United States concerning two acts—(1) a bride not assuming the surname of her husband and (2) a man engaging in sexual relations with his mother. Now in all probability, the overwhelming majority of Americans would state that both acts “ought not be” but it is equally probable that the sense of “oughtness” is much stronger for the latter act, and statements merely as to what ought to be would not reveal this differential intensity. However, social scientists have not perfected a standard technique for assessing not only opinions as to what conduct should be but also the intensity of such opinions.
The viability of norms is much more difficult to treat than the dimension of intensity. It can be argued that norms are something more than mere statements of opinion. The point is that a person may state emphatically what conduct ought to be and yet be ostensibly indifferent to contrary behavior. Stated otherwise, a person may manifest no signs of true commitment to his normative opinions; and, so the argument goes, if a large proportion of the members of a social unit do not exhibit such a commitment, the related opinion is not a norm, regardless of its prevalence. There is merit in the argument, but it raises some very difficult conceptual and technical problems.
One can take the position that a norm is viable if and only if a certain degree of conformity to it is realized. But this position is tantamount to asserting that norms are to be inferred from actual behavior. They can be so defined, but the result is a tautology when one then employs the concept to explain behavior. The alternative is to focus on the characteristics of reaction to contrary behavior as a test of the norm’s viability. Put simply, if the reaction is typically negative, then and only then is the related opinion a norm. The technical problem in this instance is that it is difficult to specify what constitutes a negative reaction (for example, must it be overt?), and little is gained by declaring that the behavior must be subject to a sanction, because the term is no more precise than the notion of a negative reaction. The conceptual problem is that certain types of norms (customs in particular) are often defined without reference to negative reactions to contrary behavior.
The statistical problem
Whatever the solution to the semantical and dimension problems, the identification of norms requires the formulation of statistical criteria. Despite the superorganic quality ascribed to norms, opposing normative opinions may prevail among members of a social unit, and negative reactions to contrary behavior are seldom either certain or uniform. Thus, most Americans probably believe that first cousins should not marry, but it is improbable that all instances of such marriage provoke a negative reaction, certainly not the same type of negative reaction. Now the absence of normative consensus and/or uniformity in reaction to contrary behavior does not imply that there are no norms, but it does pose a problem in the identification of norms. Some statistical criterion must be formulated, and the formulation is most difficult. To be sure, one can assert that at least 50 per cent of the members must subscribe to a particular opinion as to what conduct ought to be before that opinion is a norm, and the same proportion can be used as a standard to assess the viability of a norm in terms of reaction to contrary behavior; but the number obviously is arbitrary in both cases. Accordingly, in the absence of a rationale for statistical standards, there is a real need for studies of normative phenomena from a purely methodological perspective in order to explore alternative ways of identifying norms.
Even though most sociological investigations deal with norms directly or indirectly, systematic attention to the problems of conducting research on normative phenomena is rare (Zelditch 1962 is an isolated exception). The few existing techniques for the analysis of norms are most applicable to small groups (e.g., Jackson 1965) or to the study of normative phenomena in organizations (e.g.,Organizational Stress 1964). Nevertheless, there are no corresponding techniques at the macroscopic or societal level. Conceivably, the methods developed by F. H. Allport (see Katz & Schanck 1938) could be applied at any level, but they presuppose the inference of norms from actual behavior—which is contrary to typical sociological usage—and do not provide a basis for identifying all of the commonly recognized types of norms.
Conformity and deviance
Even if the demonstrability of norms is granted, all normative explanations are debatable insofar as they ignore deviant behavior. Thus, to illustrate, while reference to the prescription of monogamy provides a facile explanation as to why an adult American male may confine his sexual relations to one woman, bigamous and adulterous husbands cast doubt on the explanation, whether applied to a particular case or to a statistical construct labeled “the average American.” The point is that a normative explanation of behavior presupposes not only the existence of a norm but also conformity to it. Accordingly, the explanation is never complete or adequate without reference to a theory of conformity.
It follows that theories and research on deviant behavior are directly relevant to the study of norms. Indeed, speculation as to the causes of deviance poses still another central normative question and one that often underlies conceptual issues. For example, analytical definitions of a law stress the role of coercion as a legal element, and they have been criticized on the ground that the use of force does not explain either conformity or the subjective aspects of legal obligation (see especially Hart 1961). Such criticism may be erroneous, because it calls for an answer to an empirical question (that is, why persons do or do not conform to laws) from what is nothing more than a definition.
While investigations of deviance are relevant to the study of norms, it does not follow that such fields as criminology have or will provide answers to all normative questions concerning conformity. Theories of deviant behavior typically take norms as given and regard certain characteristics of individuals, such as their personality traits, or certain social conditions, such as the rate of unemployment, as possible etiological factors in deviation; but a purely normative perspective focuses on the characteristics of the norms (see, for example, Mizruchi & Perrucci 1962). In brief, the central question from a purely normative point of view is not why some individuals violate norms or why the rate of deviant behavior varies from place to place and time to time but, rather, why the amount of conformity varies from one norm to the next in the same social unit.
The maintenance of norms
The prevalence of normative explanations in sociology and anthropology reflects the fact that there are only two fundamental questions for these fields: Why do persons conform to norms? And, for any given social unit, why are the norms what they are?
No general theory provides answers for the above questions, but there are at least two schools of thought. Functionalists from Durkheim to Parsons have viewed norms as reflecting consensus and a common value system. Correlatively, norms form or tend to form an integrated system, that is, inter alia, a minimization of conflict among and over norms. In opposition, the “conflict” school of sociology (the Marxian variety in particular) views norms as primarily reflecting the historical or present power of one division of a society over the others. As such, the normative system is not generated by consensus and common values; indeed, at least one type of norm—law—is interpreted as reflecting dissensus and divergent values. Given the irreconcilable views of the two schools, it is not surprising that conflict sociologists and functionalists have opposing ideas on the fundamental questions in the study of norms.
The question of conformity is the central issue in attempting to explain the maintenance of norms. Observe, however, that norms remain viable not because persons conform to them, but rather persons conform to them because they are viable. In any event, the difference between the conflict and the functionalist schools on the subject of conformity (or more generally the maintenance of norms) is striking. Coercion and sanctions as bases of social order are emphasized much more by the conflict than the functionalist school, as E. A. Hoebel (1954), for instance, has pointed out in his critique of Malinowski’s conception of law. For the functionalist school the crucial factors in maintaining social order are the effectiveness of socialization, and the rewards obtained through reciprocity as it operates in conformity to norms (Gouldner 1960).
As far as the maintenance of norms is concerned, the truth seems to lie somewhere between the extreme positions of the conflict and the functionalist schools. But the debate between the two schools has not generated the kind of research necessary to resolve the issue. In all societies at least some norms are supported by coercion; therefore, the question turns largely on differences among societies and other social units as to the proportion and kinds of norms that are subject to organized sanctions. General surveys (Hoebel 1954; Diamond 1935) have indicated that such differences are considerable, and additional research could confirm the suggestion that maintenance mechanisms vary from norm to norm and from one social unit to the next. The goal of theory construction would then become that of specifying the conditions under which formal sanctions, internalization of norms, reciprocity, etc., will be most prevalent.
The origin of norms
In the nineteenth and early twentieth centuries, the sociologists and anthropologists who tried to discover the origins of norms were concerned mainly with widely diffused beliefs and doctrines of long standing, such as monotheism. Since the relevant historical evidence was usually lacking, most of the proffered “explanations” were necessarily speculative or, at the least, one-sided; and with the growth of field research, together with increasing theoretical sophistication, this type of inquiry fell into disrepute. Contemporary social scientists are aware that the origins of norms need to be explained, but they are not usually content to focus on a single institution to the exclusion of all others, nor do they see any need to delve into the remote past in order to explain what is happening now. Instead, they tend to concentrate first on identifying the norms of a given social unit and only then to ask, in terms precluding a purely historical explanation, why, for this type of social unit, the norms are what they are.
The functionalist and conflict schools differ in their response to this kind of question, but the difference cannot be summarized neatly. For one thing, the functionalist school is largely indifferent to the question. Functional analysis is prone to take a norm as “given” and then show how it contributes to the functioning of the larger social system; thus it explains the proscription of incest by asserting that the incest taboo “functions” to prevent the socially disruptive sexual exploitation of daughters by fathers. In a sense this perspective does explain the origin of norms, but only insofar as one assumes that a social system has certain “needs” and that norms inevitably develop to meet these needs. However plausible such an explanation may appear, it does not account for normative differences among societies. These differences cannot be explained in strictly functional terms without further assuming that all social systems do not have the same needs. But this assumption runs contrary to the emphasis in functional theory on universal needs. In any event, the functionalists have not identified the distinctive needs of various types of social systems, much less the norms which supposedly meet these needs. Finally, if different norms can meet the same need, then the functionalist perspective cannot generate predictions as to what norms or types of norms will prevail in a given type of social unit.
The position of the conflict school on the origin of norms is also subject to criticism on both substantive and methodological grounds. Insofar as this school can be said to accept any general principle, it tends to argue that norms represent means by which one social class or caste achieves the domination and/or exploitation of others. But, considered in substantive terms, it is most difficult to see how some norms (for example, the mother-inlaw taboo or certain dietary prescriptions) have any connection whatever with domination or exploitation.
The methodological problem in dealing with the conflict interpretation is that it does not generate any specific propositions about norms and normative differences among social units. Given the undisputed fact that norms vary from one social unit to the next, it follows, if we use the terminology of conflict theory, either that domination and exploitation necessarily assume various forms or that domination and exploitation are essentially the same in all societies but can be maintained by different norms. But in either event, the members of the conflict school have not specified in a systematic manner how domination and exploitation create differences in the norms of societies. The same criticism also applies to the Marxian variety of conflict sociology, which goes beyond the facts of domination and exploitation in asserting that the material forces of production are the ultimate causes of economic relations, which in turn determine all other social and cultural characteristics of a society, including its normative system. However, the Marxian view of norms is significant in one particular respect. Whereas functionalists tend to take norms and values as given and treat them as the basic determinants of behavior, the Marxists view them as, at most, only links between the material forces of production and the postulated effects of the forces. Since the Marxist view goes beyond norms and eschews normative determinism (see Blake & Davis 1964 on this particular issue), it is a much more general theory than that achieved by the functionalists. Nonetheless, neither Marx nor his followers have cast the theory in terms which enable one to predict systematically the attributes of a normative system from knowledge of the material forces of production.
Jack P. Gibbs
[Directly related are the entries Conflict, article onSocial Aspects; Deviant Behavior; FunctionalAnalysis; Social Control; Values. Other relevant material may be found in Interaction, article on Social Exchange; Legal Systems, article On Comparative Law and Legal Systems; Marxist Sociology. ]
Blake, Judith; and Davis, Kingsley 1964 Norms, Values, and Sanctions. Pages 456-484 in Robert E. L. Faris (editor), Handbook of Modern Sociology. Chicago: Rand-McNally.
Diamond, A. S. (1935) 1950 Primitive Law. 2d ed. London: Watts.
Durkheim, Emile (1893) 1960 The Division of Labor in Society. Glencoe, 111.: Free Press. → First published as De la division du travail social.
Gibbs, Jack P. 1965 Norms: The Problem of Definition and Classification. American Journal of Sociology 70:586-594.
Gouldner, Alvin W. 1960 The Norm of Reciprocity: A Preliminary Statement. American Sociological Review 25:161-178.
Hart, Herbert L. A. 1961 The Concept of Law. Oxford: Clarendon.
Hoebel, E. Adam SON 1954 The Law of Primitive Man: A Study in Comparative Legal Dynamics. Cambridge, Mass.: Harvard Univ. Press.
Jackson, Jay 1965 Structural Characteristics of Norms. Pages 301-309 in Ivan D. Steiner and Martin Fishbein (editors), Current Studies in Social Psychology. New York: Holt.
Katz, Daniel; and Schanck, Richard L. 1938 Social Psychology. New York: Wiley. → See especially Chapter 3.
Ladd, John 1957 The Structure of a Moral Code: A Philosophical Analysis of Ethical Discourse Applied to the Ethics of the Navaho Indians. Cambridge, Mass.: Harvard Univ. Press.
Mizruchi, Ephraim H.; and Perrucci, Robert 1962 Norm Qualities and Differential Effects of Deviant Behavior: An Exploratory Analysis. American Sociological Review 27:391-399.
Organizational Stress: Studies in Role Conflict and Ambiguity, by Robert L. Kahn et al. 1964 New York: Wiley.
Rommetveit, Ragnar (1953) 1955 Social Norms and Roles: Explorations in the Psychology of Enduring Social Pressures With Empirical Contributions From Inquiries Into Religious Attitudes and Sex Roles of Adolescents From Some Districts in Western Norway. Oslo: Akademisk Forlag; Minneapolis: Univ. of Minnesota Press.
Schwartz, Richard D.; and Miller, James C. 1964 Legal Evolution and Societal Complexity. American Journal of Sociology 70:159-169.
Stouffer, Samuel A. (1955) 1963 Communism, Conformity and Civil Liberties: A Cross-section of the Nation Speaks Its Mind. Gloucester, Mass.: Smith.
Sumner, William G. (1906) 1959 Folkways: A Study of the Sociological Importance of Usages, Manners, Customs, Mores, and Morals. New York: Dover. → paperback edition was published in 1960 by New American Library.
Weber, Max (1922)1954 Max Weber on Law in Economy and Society. Edited, with an introduction and annotations by Max Rheinstein. Cambridge, Mass.: Harvard Univ. Press. → First published as Chapter 7 of Max Weber’s Wirtschaft und Gesellschaft.
Zelditch, Morris JR. 1962 Some Methodological Problems of Field Studies. American Journal of Sociology 67:566-576.
Social scientists invoke the concept of norms to explain a broad range of human behaviors. No universally agreed-upon definition of norms exists, but many definitions share three components (Horne 2001, pp. 3–5). First, norms are rules that prescribe or proscribe a behavior or set of behaviors. Second, norms are enforced by external sanctions (rewards and punishments furnished by a source other than their target). These sanctions can be material (e.g., financial bonuses or fines) or symbolic (e.g., expressions of approval or disapproval). Third, norms are consensual, group-level phenomena. Group members recognize the existence of norms and feel entitled to enforce them. Some debate remains over whether norms, once established, are unconditional, clear, and generally followed (as in many rational choice theories), or conditional, unclear, and constantly negotiated (as in many symbolic interactionist theories) (Hechter and Opp 2001, pp. 394–396).
This definition distinguishes between norms and similar concepts, such as values and attitudes. Norms differ from values in that they are enforced by external sanctions, whereas values are enforced only by internal sanctions, such as feelings of pride or shame (Hechter and Horne 2003). Norms differ from attitudes in that norms are consensually held and legitimated by the group, whereas attitudes are a property of individuals. Norms also differ from laws in that a central authority (such as the state) formally creates and enforces laws, while people informally create and enforce norms. Some scholars distinguish between norms and conventions, the primary difference being that the direction of conventions is arbitrary (Coleman 1990). For example, it does not matter whether people drive on the left or the right side of the road, so long as everyone follows the convention of driving on the same side.
Norms shape social behavior by constraining action, and may be beneficial or harmful. Beneficial norms make society possible by protecting people from exploitation. Humans existed for thousands of years without the protections of a formal legal system (de Quervain et al. 2004), and the actual exercise of law remains limited due to resource constraints (Ellickson 1991). In the absence of law, norms regulate behavior and prevent people from routinely using force and fraud for private gain (Ellis 1971). Such norms include those that ensure that people look after their neighbors’ children or livestock, punish laziness at work, respect cease-fire and arms-control agreements, reciprocate favors, help strangers, and otherwise contribute to a stable society.
Harmful norms may constrain individual achievements. Such norms include leveling norms that prevent advancement by members of disadvantaged ethnic groups (Portes 1998, pp. 15–18), or norms mandating that women confine their activities to domestic pursuits and forego the labor market. Norms can also sustain harmful practices that a majority of the group opposes, including informing on one’s neighbors in repressive regimes and binge drinking on college campuses (Centola et al. 2005). Some norms also encourage dangerous practices such as dueling (e.g., Axelrod 1986, p. 1095).
Norms constrain the behavior of corporate actors as well as individuals. Firms sometimes adopt rules or structural changes that conform to norms in order to enhance their legitimacy, even when these practices run counter to market pressures (Meyer and Rowan 1977). Territories gain recognition as states in part by conforming to international norms defining the requirements for statehood. Failure to do so can result in loss of statehood and the attendant privileges (Meyer et al. 1997).
Given the widespread use of norms as an explanatory device, an important task for the social sciences is to explain the conditions under which norms emerge, change, and persist. Early functionalist explanations of social norms argued that norms arise because they benefit society, but this view has been discredited and generally abandoned. Because actors enforce norms through sanctioning, a number of theories seek to understand why actors sanction particular behaviors.
Structural features of situations can influence norm emergence. Conformity tends to increase as the size of the majority in favor of the norm increases (Asch 1951). Such factors as low levels of trust and a high risk of exploitation motivate the creation of norms that reduce the risk of exploitation (Yamagishi 1988). Network density may help coordinate sanctioning (Coleman 1990), and the threat of collective punishment can produce norms that mitigate that risk (Heckathorn 1988). In addition, actors with greater structural power possess greater ability to create, enforce, or undermine norms that serve their own interests at the expense of vulnerable actors (Coleman 1990).
People also follow and enforce norms in order to gain approval and signal that they are trustworthy interaction partners, thus encouraging others to profitably exchange with them (Homans  1992; Horne 2004). People increasingly enforce metanorms (by rewarding those who sanction deviants) as the direct and indirect benefits of exchanging with others increases (Horne 2004). Enforcing norms to gain approval can backfire, as when people enforce norms that the majority privately disdains because they falsely believe the norm to be popular (Centola et al. 2005).
Norms may also develop through a process of cultural evolution in which a norm, once established, provides greater average benefits to those who follow the norm than those who do not (Axelrod 1986; Bendor and Swistak 2001). Similarly, people may also learn to support norms via trial and error (Macy 1993).
Affective processes motivate norm enforcement. Emotional responses lead people to punish theft, walk away from profitable but unethical business deals, and help strangers in need, even when these actions contradict their short-term material interest (Frank 1988). Neurologically, people experience greater levels of activation in a reward center in the brain when they punish people who have behaved in an untrustworthy manner (de Quervain et al. 2004).
A number of important questions regarding norms remain. Broadly, it is important to continue to develop explanatory theories and empirical tests of norm emergence. Other pressing questions include how norms acquire content (Hechter and Opp 2001), how harmful norms emerge (e.g., Centola et al. 2005), and how norms and laws interrelate. Answers to these and related questions will deepen our knowledge of this fundamental but often opaque concept.
SEE ALSO Culture; Lay Theories; Social System; Values
Asch, Solomon E. 1951. Effects of Group Pressure upon the Modification and Distortion of Judgments. In Groups, Leadership, and Men: Research in Human Relations, ed. Harold Guetkow, 177–190. Pittsburgh, PA: Carnegie Press.
Axelrod, Robert. 1986. An Evolutionary Approach to Norms. American Political Science Review 80: 1095–1111.
Bendor, Jonathan, and Piotr Swistak. 2001. The Evolution of Norms. American Journal of Sociology 106: 1493–1545.
Centola, Damon, Robb Willer, and Michael W. Macy. 2005. The Emperor’s Dilemma: A Computational Model of Self-enforcing Norms. American Journal of Sociology 110: 1009–1040.
Coleman, James S. 1990. Foundations of Social Theory. Cambridge, MA: Belknap.
de Quervain, Dominique J.-F., et al. 2004. The Neural Basis of Altruistic Punishment. Science 305: 1254–1258.
Ellickson, Robert C. 1991. Order without Law: How Neighbors Settle Disputes. Cambridge, MA: Harvard University Press.
Ellis, Desmond P. 1971. The Hobbesian Problem of Order: A Critical Appraisal of the Normative Solution. American Sociological Review 36: 692–703.
Frank, Robert H. 1988. Passions within Reason: The Strategic Role of the Emotions. New York: Norton.
Hechter, Michael, and Christine Horne. 2003. Values and Norms. In Theories of Social Order: A Reader, eds. Michael Hechter and Christine Horne, 91–100. Stanford, CA: Stanford University Press.
Hechter, Michael, and Karl-Dieter Opp. 2001. What Have We Learned about the Emergence of Social Norms? In Social Norms, eds. Michael Hechter and Karl-Dieter Opp, 394–415. New York: Russell Sage Foundation.
Heckathorn, Douglas D. 1988. Collective Sanctions and the Creation of Prisoner’s Dilemma Norms. American Journal of Sociology 94: 535–562.
Homans, George C.  1992. The Human Group. New Brunswick, NJ: Transaction.
Horne, Christine. 2001. Sociological Perspectives on the Emergence of Social Norms. In Social Norms, eds. Michael Hechter and Karl-Dieter Opp, 3–44. New York: Russell Sage Foundation.
Horne, Christine. 2004. Collective Benefits, Exchange Interests, and Norm Enforcement. Social Forces 82: 1037–1062.
Macy, Michael W. 1993. Backward Looking Social Control. American Sociological Review 58: 819–836.
Meyer, John W., John Boli, George M. Thomas, and Francisco O. Ramirez. 1997. World Society and the Nation-State. American Journal of Sociology 103: 144–181.
Meyer, John W., and Brian Rowan. 1977. Institutionalized Organizations: Formal Structure as Myth and Ceremony. American Journal of Sociology 83: 340–363.
Portes, Alejandro. 1998. Social Capital: Its Origins and Applications in Modern Sociology. Annual Review of Sociology 24: 1–24.
Yamagishi, Toshio. 1988. Seriousness of Social Dilemmas and the Provision of a Sanctioning System. Social Psychology Quarterly 51: 32–42.