What Theories of Deviance Need to Explain

A useful theory of deviance has to handle more than the bare fact that some people break rules. It has to explain a set of patterns that sit alongside the rule-breaking itself and that together constitute the actual subject matter of the sociology of deviance. Before comparing particular theories, it is worth being explicit about what the theories are trying to account for, because much of the apparent disagreement between them turns on which of these patterns each one foregrounds.

The first pattern is the distribution of rule-breaking across populations. Rates of offending vary sharply by age, sex, neighborhood, social class, historical period, and country. The age-crime curve — a sharp rise in offending from early adolescence, a peak in late adolescence or early adulthood, and a steep decline after the mid-twenties — is one of the most robust findings in criminology, documented across jurisdictions and centuries (Hirschi and Gottfredson 1983, pp. 552-558). The over-representation of young men in official crime statistics is similarly stable. Theories of deviance have to say something about why these distributions look the way they do, and not merely about why this or that individual offended.

The second pattern is the selectivity of labeling. Of the many rule-violations that occur, only a fraction come to the attention of authorities, and of those only a fraction result in official action. The filtering is not random: some acts are far more likely to be detected and punished (street-level property crime) than others (many forms of white-collar offense), and some actors are far more likely to be labeled even when behavior is held constant (Black 1976, pp. 21-36). A theory that only explains rule-breaking without explaining the differential application of labels will miss a large part of what sociologists of deviance actually study.

The third pattern is the career structure of deviant life-courses. Some people engage in a single episode of rule-breaking and then desist. Others move through identifiable stages — primary deviance, labeling, secondary deviance, commitment to a deviant identity, eventual desistance or persistence (Lemert 1951, pp. 75-77; Matza 1964, pp. 181-191). Life-course criminology has given this observation empirical shape, showing that most adolescent offending is time-limited and that a smaller group of life-course-persistent offenders accounts for a disproportionate share of serious and chronic crime (Moffitt 1993, pp. 674-678; Sampson and Laub 1993, pp. 18-29).

From these patterns fall out three analytically separable questions that the main families of deviance theory address in different proportions. (a) Why do some people break rules more than others? Strain theory, learning theory, and control theory are the principal answers to this question, each locating the explanation in a different mechanism — unequal opportunity, socialization, or the strength of social bonds. (b) Why do some acts get labeled deviant at all? Labeling theory and critical theories take this as the primary question, arguing that the definition of deviance is itself a social and often political process, not a self-evident reflection of harm. (c) Why do some people, given similar behavior, get labeled while others do not? This is the question of differential reaction, tied closely to conflict perspectives on law and to empirical work on discretionary enforcement.

A useful feature of this triad is that it makes clear why the major theories are not so much rivals as partial specialists. Strain theory has little to say about question (b) but a great deal to say about (a). Labeling theory has relatively little to say about why rule-violations occur in the first place, but a lot to say about (b) and (c). Critical and conflict theories speak principally to (b) and (c) by connecting the definition and enforcement of law to class, race, and state interests. Any comprehensive sociological account of deviance draws on several of these strands, because the object of study — rule-breaking, its labeling, and the resulting careers — has several components that are not reducible to one another. The rest of this topic walks through the main families of theory and returns at the end to the question of how they combine in contemporary research.