Edwin Sutherland's differential association (1947) holds that criminal behaviour is learned in interaction with people close to us, within intimate groups, and that the mass media play a secondary part. Daniel Glaser's differential identification (1956) corrects that point: what is decisive is not who you deal with, but who you identify with. A person pursues criminal behaviour to the extent that they identify with real or imaginary persons from whose perspective that behaviour seems acceptable. The model may be a friend, but also a character, a public figure or someone never met.
| Differential association (Sutherland) | Differential identification (Glaser) | |
|---|---|---|
| Author and year | Edwin H. Sutherland, 1939–1947 | Daniel Glaser, 1956 |
| What transmits crime | Interaction with others in intimate groups | Identification with a model, near or distant |
| Is contact needed? | Yes: face to face | No: the reference group is enough |
| Role of the media | Minor | Significant: they supply models to identify with |
| Borrowed concept | Learning in interaction (interactionism) | Reference group and role-taking |
| What it explains best | Crime learned in gangs, families and firms | The lone offender and the copycat |
// What they shareThe common ground
Glaser was not trying to refute Sutherland, but to mend a crack in him. Both are learning theories and share everything essential: crime is neither inherited nor invented, it is acquired like any other behaviour, and what is acquired is not only techniques but a way of seeing things that makes it acceptable. Both explain why, in the same neighbourhood and the same poverty, some offend and others do not: it depends on who they take as their reference.
// Where they partThe differences that matter
The crack Glaser fills. It had been objected to Sutherland that there are people surrounded by criminals who do not offend —prison officers, lawyers, police— and others who offend without ever dealing with any. Glaser solves it by changing the variable: what counts is not exposure, but the choice of model. You can live alongside someone without wanting to be like them.
Models without contact. By allowing identification with "real or imaginary" persons, Glaser opens the door to something Sutherland left out: that cinema, the press or fame can teach. It is the forerunner of what Akers will call imitation and of what is studied today as the copycat effect.
The individual's part. In Sutherland the person is almost a vessel: they receive definitions and the heavier side wins. In Glaser there is an active part, because to identify is to choose. That makes it more flexible and also harder to measure: how do you establish who someone identifies with before they act?
// One villain, two readingsRupert Pupkin
Rupert Pupkin is the offender Sutherland cannot explain and Glaser can. With differential association there is no way in: he belongs to no criminal group, nobody has taught him to kidnap, and he lives with his mother rehearsing monologues in a basement. He has not received a single definition favourable to crime from anyone close, because he has no one close. With differential identification everything fits. His reference group is television, and his model, a host he has spoken to exactly once. Rupert holds entire conversations with cardboard cut-outs, and from the perspective of that imagined world, kidnapping Jerry Langford to get ten minutes on air is not a crime: it is the price of admission. He learned from no one. He identified with someone who did not know he existed.
The mistake everyone makes
Confusing differential identification with differential association because the names are alike, or crediting Sutherland with the influence of the media. Sutherland played down the role of the mass media; it is Glaser who brings it in. And it should not be mixed up with a third: differential anticipation is Glaser's own later reformulation (1978), which adds expectations about the consequences of the act.
Which villains each one explains
Case files the archive analyses with each theory. The number is the total; the list shows the first ones.
Sources · further reading
- Sutherland, E. H. (1947). Principles of Criminology (4th ed.). J. B. Lippincott.
- Glaser, D. (1956). "Criminality Theories and Behavioral Images". American Journal of Sociology, 61(5).
- Glaser, D. (1978). Crime in Our Changing Society. Holt, Rinehart and Winston.
Theory questionsFrequently asked questions
What is the difference between differential association and differential identification?
Sutherland's differential association holds that crime is learned in direct interaction with people close to us. Glaser's differential identification holds that what is decisive is who the person identifies with, whether or not there is contact: the model may be real or imaginary, close or seen only in the media.
What problem in Sutherland's theory does Glaser solve?
He explains why people in constant contact with criminals do not offend, and why others offend without having had that contact: what matters is not exposure, but the choice of reference group or person.
What is differential anticipation?
It is Glaser's 1978 reformulation of his own theory: a person tends to act according to the consequences they expect from the behaviour, and those expectations depend on their bonds, on what they have learned and on the opportunities they perceive.