Differential Association Theory

Differential association theory, proposed by Edwin Sutherland, explains that criminal behavior is learned through interaction with others. It suggests that people become criminals not because of who they are, but because of the values, attitudes, and techniques they pick up from the people around them. In short, crime isn’t inherited — it’s taught through social relationships and everyday communication.

Sutherland’s theory of differential association

Differential association theory (DAT), developed by Edwin H. Sutherland, explains that criminal behaviour is learned through social interaction and communication within close, personal groups. The focus is on how people learn to commit crime – not on their background, income, or social class.

Challenging Class-Based Theories

Before Sutherland, many sociological theories linked crime mainly to poverty and social disadvantage.

For example:

• Functionalist strain theories suggested that people in lower social classes commit crime because they can’t achieve success through legitimate means.

• Social disorganization theory connected high crime rates to poverty, instability, and lack of community control.

Sutherland challenged this idea. He argued that focusing only on lower-class crime ignored the fact that people in all social classes commit offences.

According to DAT, crime is not caused by poverty or lack of opportunity – it’s caused by social learning.

Anyone, regardless of class, can learn to break the law if they’re surrounded by people who define such behaviour as acceptable.

---

The Social Context: Learning Through Interaction

Sutherland’s DAT is rooted in symbolic interactionism. This sociological perspective holds that people learn meanings and behaviours through social interaction. It emphasizes that how people define a situation shapes how they behave in it.

DAT treats crime as learned behaviour, not something innate. We learn how to behave, whether law-abiding or deviant, from the people closest to us, such as family members and friends.

Sutherland was influenced by the Chicago School of Sociology, which explored how social environments shape behaviour. His ideas drew from several traditions.

Cultural transmission theory explains how culture and values are passed on. Ecological theory explains how neighbourhoods shape behaviour, and culture conflict theory explains how competing group values can produce crime.

---

Learning Attitudes Toward the Law

Through social interaction, people learn:

When people spend time with those who break the rules, they begin to learn two key things:

• Techniques and motives: They learn how to commit crimes (like how to steal or cheat) and develop the motives and justifications that make those acts feel acceptable.
• Definitions favourable to law violation: People learn beliefs that support or excuse breaking the law. For example, thinking “Everyone cheats a little” or “It’s not really stealing if the company is rich”.

When someone has more definitions in favour of breaking the law than against it, they’re more likely to engage in criminal behaviour.

This balance – called an “excess of definitions favourable to violation of law” – is the heart of differential association theory.

---

Techniques of Neutralization

Sutherland’s fourth proposition says people learn more than criminal techniques. They also learn the motives, drives, and rationalisations that make an offence feel acceptable. Two criminologists, Gresham Sykes and David Matza, took this claim further in 1957.

Sykes and Matza noticed a puzzle. Most delinquents still say they believe in conventional morality, feel guilty about specific offences, and can be shamed back into line by people they respect. A person who had truly learned a whole set of anti-conventional values would not behave this way.

• Aim: Sykes and Matza (1957) set out to explain why delinquents keep expressing conventional values even while offending.
• Method: a theoretical analysis drawing on Sutherland’s own theory, research on delinquent subcultures, and earlier work on guilt and rationalisation.
• Results: delinquents learn five recurring techniques of neutralization that let them suspend their own morality just before an offence, rather than abandoning it altogether.
• Conclusion: delinquency is better understood as episodic “drift” between conformity and deviance, made possible by these learned rationalisations, than as a stable commitment to anti-conventional values.

Here are the five techniques.

• Denial of responsibility: “It wasn’t really my fault, I was pushed into it.”
• Denial of injury: “No one really got hurt.”
• Denial of the victim: “They had it coming.”
• Condemnation of the condemners: “The police are worse hypocrites than we are.”
• Appeal to higher loyalties: “I had to help my friend, loyalty comes before the law.”

The theory has stayed influential because it explains something DAT’s bare propositions leave open: occasional, situational offending by people who are not, in any simple sense, committed delinquents.

Its main limitation is that, like DAT itself, it rests on theoretical analysis rather than direct testing. It doesn’t directly test whether people invent these excuses before an offence, as the theory claims, or only afterwards.

---

How Peers and Groups Influence Delinquency

A key strength of DAT is how well it explains the role of peer groups in shaping behaviour.

• Gangs and Friendship Groups: Early studies found that delinquent behaviour often spreads through play groups and gangs. Young people in these groups not only learn how to commit crimes but also why they might see those acts as acceptable.

• Neighbourhood Culture: In some communities, new residents may learn delinquent behaviour from existing neighbours, passing on a local culture of crime.

• Peer Pressure and Drugs: Teenagers are especially influenced by friends. If a young person’s peer group uses drugs, shoplifts, or is sexually active, they are likely to learn that these behaviours are “normal” or acceptable.

• Rule-Breaking in Everyday Settings: The same process applies outside street crime. For instance, an athlete who uses steroids might persuade teammates to start “doping,” or a student who cheats might influence classmates to plagiarize. In both cases, deviance spreads through social learning.

---

Factors That Influence Learning

The strength of these influences depends on four key factors:

1. Frequency – How often you are exposed to criminal ideas or behaviour.

2. Duration – How long these influences last.

3. Priority – How early in life you are exposed to them (early experiences matter most).

4. Intensity – How emotionally close or important the people teaching you are.

For example, learning deviant behaviour from a close friend you admire has a stronger impact than hearing it from a casual acquaintance.

---

Factors That Strengthen the Learning Process

Not all social influences are equal. Sutherland identified several factors that affect how strongly someone learns criminal behaviour:

• Ratio of deviant to non-deviant people: The more a person interacts with people who commit crimes, the more likely they are to pick up deviant attitudes.

• Significant others: Influence is stronger when the deviant person is emotionally important — like a close friend, romantic partner, or family member.

• Age of exposure: Learning criminal behaviour early in life (during childhood or adolescence) tends to have a stronger and longer-lasting impact.

For example, a teenager who spends most of their time with friends who shoplift and look up to them is far more likely to copy that behaviour. Someone who only occasionally hears about it from a distant acquaintance is much less likely to.

---

Main Principles

Sutherland first presented DAT as seven main ideas, which he later expanded to nine propositions.

These outline how people learn criminal behaviour.

1. Criminal behaviour is learned: People aren’t born criminals. They learn to engage in criminal acts just as they learn any other kind of behaviour.

2. Criminal behaviour is learned through interaction with other people: Learning happens through communication and social interaction — by talking, observing, and participating with others.

3. The main part of this learning happens within close, personal groups: The most influential learning occurs in intimate relationships — family, close friends, or peers — rather than through media or casual contact.

4. When criminal behaviour is learned, it includes both techniques and motives: This learning involves: Techniques for committing the act (for example, how to shoplift or hack a system). Motives, drives, attitudes, and rationalizations that make the behaviour seem acceptable.

5. The direction of motives and drives is learned from definitions of the law as favourable or unfavourable: People learn how to view the law — either as something to respect or something to reject — through others’ opinions and attitudes.

6. A person becomes delinquent because of an excess of definitions favourable to law violation: Crime occurs when a person is exposed to more attitudes supporting crime than attitudes condemning it. This “balance” of learned values determines whether someone will conform or deviate.

7. Differential associations vary in frequency, duration, priority, and intensity: not all influences are equal, as explained in “Factors That Influence Learning” above. The earlier and closer the contact with deviant individuals, the stronger the learning effect.

8. The learning process for criminal behaviour is the same as for any other behaviour: Crime is learned through the same psychological processes as learning language, skills, or habits — such as imitation, reinforcement, and repetition.

9. Criminal behaviour is not explained by general needs and values: People commit crimes to satisfy needs like money, excitement, or social status — but these needs also motivate non-criminal behaviour. Therefore, the difference lies not in the needs themselves, but in how individuals learn to meet them.

---

Examples

• Peer groups and gangs: Young people who spend time with delinquent peers may learn both the methods and mindset for breaking the law — for instance, learning to shoplift or vandalize while seeing it as acceptable within their group.

• Drug use and peer influence: A teenager might start using drugs after watching friends do the same. Through these interactions, they learn the technique (how to use) and the rationale (“It helps me relax” or “Everyone else is doing it”).

• White-collar crime: crime committed by someone of high social status in the course of their job. Sutherland originally applied DAT to elite and corporate offending: businesspeople might learn unethical practices from colleagues, adopting justifications like “That’s how everyone gets ahead in this industry.”

---

Real-World Applications

Differential association theory isn’t only an explanation of why people offend. It also shapes how prevention programmes, workplaces, schools, and online platforms are designed.

Substance-Use Prevention

Differential association shows its strongest and most consistent effects for soft-drug use (Kruis, Seo, & Kim, 2019). That finding has a direct implication for prevention programme design.

Programmes built around reshaping peer norms and reducing exposure to substance-using peers work better. They are better targeted at alcohol, tobacco, and cannabis than at harder drugs.

Other risk factors likely carry more weight for harder drugs. Programme designers should calibrate their expectations rather than assume a uniform effect across every substance, and revisit that assumption as new evidence arrives.

This matters for how limited prevention budgets get spent. Money aimed at reshaping peer norms around alcohol and cannabis is well targeted. The same approach aimed at harder drugs, on its own, is likely to disappoint.

Group-Based Intervention Design

Perhaps the most consequential, and counter-intuitive, applied lesson from this research concerns interventions meant to help at-risk young people.

• Aim: Dishion, McCord, and Poulin (1999) reviewed a specific, unintended risk of a common intervention design: bringing groups of high-risk adolescents together for peer-based counselling or skills training.
• Method: a review of controlled evaluations of adolescent peer-group interventions, including long-term follow-up data and, where available, direct observation of what happened inside the sessions.
• Results: several programmes produced iatrogenic effects, meaning they increased problem behaviour compared with untreated groups. Deviant talk was often inadvertently reinforced by peer laughter and attention, a process called “deviancy training.”
• Conclusion: grouping high-risk peers together can accidentally manufacture the very conditions, frequent exposure to peers who model rule-breaking, that differential association predicts will increase delinquency.

This is one of the clearest real-world demonstrations that the theory’s learning mechanism can work in a direction a programme never intended. It is why many practitioners now favour individually delivered or family-based interventions over simply clustering already-delinquent youth together.

Corporate Ethics and Compliance

The white-collar crime section above shows that techniques of neutralization appear in professional as much as street-level offending. Many organisations now design compliance training to interrupt this learning process directly.

Training now names common excuses as they arise, such as “everyone does it” and “no one is really harmed.” Naming the excuse in the room takes away its power.

The reasoning follows straight from DAT. Misconduct spreads through informal workplace transmission of these rationalisations.

Interrupting that transmission is a more direct lever on misconduct than punishment alone. That is why modern compliance training is built around naming excuses, not just listing rules.

Rules alone don’t stop workplace learning. A colleague who models and excuses bad behaviour teaches it just as effectively as any formal policy discourages it.

Education and Digital Safety

Differential association identifies peer groups as the main channel through which norms spread among adolescents. This includes norms around cheating and bullying, not only crime in the legal sense.

Many school programmes deliberately work to reshape peer norms and peer leadership. They target the group, not just the individual student.

This follows DAT’s own logic. It is the balance of definitions circulating in a young person’s closest peer group, not the individual’s isolated attitude, that predicts behaviour.

The research on virtual peer effects has informed how schools and platforms think about online moderation too. Online contact tends to supplement, not replace, face-to-face influence.

Miller and Morris (2016) and McCuddy (2020), discussed above, both found that online peers mostly reinforce existing offline relationships. They rarely open some entirely separate channel of their own.

Interventions that address only online spaces, such as content moderation or monitoring apps, are therefore unlikely to work alone. Offline peer networks need addressing too.

---

Limitations

While DAT transformed how sociologists think about crime, it also faces criticism for being vague and difficult to test in practice.

---

1. Conceptual Vagueness and Problems with Testing

One of the most common criticisms is that DAT is too vague to test scientifically.

• “Excess of definitions” is unclear: It’s difficult to measure or quantify what counts as an “excess of definitions favourable to law violation.” How many pro-crime messages are “too many”?

• General vs. specific definitions: DAT doesn’t clearly specify whether definitions are general (e.g., “laws are unfair”) or specific (e.g., “it’s okay to steal from big corporations but not from individuals”).

• Changing attitudes: Critics also note that people’s definitions of right and wrong may change over time, making it hard to use them as stable predictors of criminal behaviour.

Because of these issues, researchers find it challenging to test DAT rigorously or to compare findings across studies.

---

2. Weak Empirical Support and Unclear Mechanisms

Although many studies support parts of the theory, evidence for its exact learning mechanism is weak.

• Attitudes vs. behavior: Research shows that the behaviour of delinquent peers often predicts crime better than their attitudes. In other words, what friends do may influence people more than what they say they believe.

• Missing explanation for how learning happens: Sutherland never clearly explained how people internalize deviant definitions. Later theorists, like Akers, filled this gap with Social Learning Theory, which added concepts from psychology such as reinforcement (reward and punishment).

• However, by doing so, SLT moved away from the purely sociological roots of symbolic interactionism, blending sociology with behavioural psychology.

A related, deeper problem is knowing which way the causal arrow runs. DAT assumes association causes learning, which causes delinquency. A rival explanation is just as plausible: already-delinquent teenagers might simply seek out, and be accepted by, other delinquent peers.

Matsueda and Anderson (1998) tested this directly, using statistical models built to disentangle the two directions of causality. They found the relationship genuinely reciprocal. But the effect of delinquency on later peer selection was larger than the effect of peer association on later delinquency.

DAT’s core causal claim survives this test. But the finding shows the theory has previously been given more credit than the evidence, on balance, supports.

---

3. Limited Scope and Incomplete Explanation

DAT also struggles to explain all types of deviance.

• Not all deviance is learned: It doesn’t account for behaviour that isn’t intentional or socially learned, such as that linked to mental illness or biological conditions.

• Uneven outcomes: Many people who are exposed to deviant peers don’t become deviant, while others with non-deviant backgrounds do. The theory doesn’t fully explain why.

• Missing the role of self and identity: Although DAT is part of the interactionist tradition, it pays little attention to the individual’s self-concept or how identity shapes behaviour. For example, it doesn’t explain how people interpret or resist deviant labels.

This limitation led to the development of Differential Social Control (DSC) Theory by Heimer and Matsueda.

DSC built on DAT by adding ideas about role-taking and reflected appraisals: how individuals internalize others’ views of them. It also linked social structure, identity, and delinquent behaviour.

Their research supported the DAT element (that anti-social attitudes predict delinquency) but provided a more complete model of how deviance develops.

---

Strengths

1. A Major Shift in Criminological Thinking

Sutherland’s theory marked a turning point in criminology.

Instead of focusing on individual defects or biological causes, DAT showed that crime arises from social processes.

It placed learning and interaction at the centre of understanding deviance.

2. Emphasis on Social Learning

DAT explains that people learn criminal behaviour the same way they learn conforming behaviour – through relationships with family, friends, and peers.

This helped normalize the study of crime, framing it as a result of everyday socialization rather than personal pathology.

3. Supported by Research on Peer Influence

Many studies since Sutherland’s time have shown that peer groups strongly influence delinquency, especially during adolescence. Young people often learn techniques and attitudes supporting deviance through close friendships, supporting the basic logic of DAT.

One study stands out. It remains the first large-scale empirical test of this idea, and one of the most influential in the field.

• Aim: Akers, Krohn, Lanza-Kaduce, and Radosevich (1979) set out to test whether the strength of a teenager’s association with substance-using peers predicted their own drinking, smoking, and drug use.
• Method: a self-report survey of several thousand adolescents across a number of Midwestern U.S. communities. It measured their peer associations, their attitudes to substance use, and how much reinforcement and modelling they received for it.
• Results: adolescents’ own drinking, smoking, and drug use closely tracked how much they were reinforced for it and associated with peers who modelled and approved of it.
• Conclusion: deviant behaviour is learned through essentially the same process as conforming behaviour, giving direct quantitative support to Sutherland’s claim that differential association drives delinquency.

The study’s main limitation is that it measured peer associations and personal use at a single point in time. This cross-sectional design cannot fully rule out that adolescents who already used drugs simply selected friends who did the same, rather than being led to start by their peers.

4. Foundation for Later Theories

Sutherland’s work inspired later developments. The most important is social learning theory (SLT), Ronald Akers’s reformulation of DAT that restates it around four interacting parts.

• Differential association: Sutherland’s original claim, largely unchanged: the groups a person associates with most closely, frequently, and earliest carry the most weight.
• Definitions: the specific attitudes a person attaches to an act as right or wrong.
• Differential reinforcement: people weigh the rewards and costs attached to a behaviour and are more likely to repeat what has been rewarded, or not punished, before.
• Imitation: drawn from Bandura’s psychological work on observational learning, this is the tendency to copy a behaviour one has seen an admired model perform.

SLT refined DAT by explaining how favourable definitions actually take hold, incorporating reinforcement and imitation from psychology. Not all four parts carry equal weight, though. A large meta-analysis has tested this directly.

• Aim: Pratt and colleagues (2010) set out to determine, across the full published literature, how strongly each of Akers’s four constructs actually predicts crime and deviance.
• Method: a meta-analysis pooling effect sizes from studies testing SLT’s constructs against measures of crime and deviance, coded separately by which construct each one measured.
• Results: differential association and definitions showed consistently strong relationships with crime, while differential reinforcement and imitation were, at best, modest.
• Conclusion: the theory’s sociological core, association and definitions, does more empirical work than the behavioural additions Akers introduced.

These ideas have even been applied beyond criminology, for example in health promotion and behaviour change campaigns.

Contemporary Research

Two threads run through the most important recent research on DAT and social learning theory. One asks how well the theory holds up for substance use specifically. The other asks whether it still works the same way when peer contact happens online.

Substance Use and Social Learning

The strongest evidence here is a meta-analysis, and it should carry more weight than any single study beneath it.

• Aim: Kruis, Seo, and Kim (2019) set out to pool the accumulated research testing SLT against substance use, a body of work that had never been quantitatively combined before.
• Method: a systematic review and meta-analysis of 83 studies published between 1974 and 2018, calculating weighted effect sizes and checking how they varied by construct, drug type, country, and age.
• Results: SLT showed a medium overall effect on substance use, with differential association the strongest predictor. Effects were larger for soft drugs such as alcohol and cannabis than for hard drugs.
• Conclusion: SLT, and differential association above all, may explain soft-drug use better than hard-drug use, a narrower claim than treating the theory as an all-purpose account of deviance.

Learning in the Digital Age

A newer strand of research asks whether close contact with delinquent peers still shapes behaviour when much of that contact happens through a screen rather than face to face.

Two studies test this directly. Miller and Morris (2016) found that virtual peer associations among college students predicted offending through much the same social-learning process as in-person ones.

McCuddy’s (2020) longitudinal study of over a thousand middle- and high-school students reached a similar conclusion. Online peers rarely exposed teenagers to genuinely new pro-delinquency messages. Instead, they mostly reinforced influences that traced back to existing offline friends.

Read together, these studies support the same conclusion this article has developed throughout. Differential association’s central mechanism, close and sustained exposure to peers who model and approve of rule-breaking, remains the theory’s best-supported claim, whichever channel that exposure travels through.

---

Cambridge Study in Delinquent Development 

The Cambridge Study in Delinquent Development (CSDD) is a famous long-term research project. It followed a group of boys from childhood into adulthood to understand how criminal and antisocial behaviour develops.

The claim that the CSDD supports Differential Association Theory (DAT), the idea that crime is learned through interaction, is sometimes made. This connection isn’t clearly supported by the research evidence.

In fact, the study’s findings are more commonly linked to Self-Control Theory (SCT) than to DAT.

---

What the CSDD Actually Examined

The CSDD, led by David P. Farrington, began in the early 1960s and tracked hundreds of boys from working-class areas of South London across several decades.

Its data has been widely used to test different criminological theories.

One specific study using this dataset – conducted by Polakowski (1994)—looked at Gottfredson and Hirschi’s Self-Control Theory (SCT) rather than DAT.

Here’s what that research involved:

• Data Source: Six waves of information from the CSDD.

• Behavioural Indicators: Reports from mothers, teachers, and peers about traits such as clumsiness, restlessness, and short attention span.

• Aim: To test whether these early behavioural traits could predict later misbehaviour and criminal convictions.

• Findings: The results supported Self-Control Theory, showing that low self-control in childhood was linked to higher levels of delinquency later in life.

It’s worth noting that Polakowski’s study didn’t include measures of social learning or peer influence (key components of DAT), so it couldn’t directly test Sutherland’s theory.

---

Why the CSDD Is Linked to Self-Control Theory (Not DAT)

The two theories approach crime from very different angles:

The CSDD’s findings—especially those showing that early traits like impulsivity and restlessness predict later offending – fit much more closely with Self-Control Theory.

They support the idea that crime stems from internal traits formed early in life, rather than being primarily learned from social groups, as DAT would predict.

---

Summary

While the Cambridge Study in Delinquent Development is one of the most important long-term studies in criminology, there’s no clear evidence that it directly supports Differential Association Theory.

Instead, most of its findings have been used to support Self-Control Theory, which emphasizes personal traits, early socialization, and self-regulation over the social learning processes proposed by Sutherland.

In short:

The CSDD helps us understand how personality and early upbringing shape criminal behaviour — but it doesn’t provide evidence that people learn crime primarily through association, as DAT suggests.

Comparison with Rival Theories

DAT doesn’t stand alone. Seeing how it agrees, disagrees, and combines with other major theories of crime shows what it explains well, and what it leaves for other theories to cover.

Social Disorganization Theory

Social disorganization theory and DAT are best read as complementary. They are not really rivals.

Social disorganisation explains where crime concentrates: neighbourhoods with weak community control and institutional instability. It says little about individual behaviour.

DAT explains the mechanism instead. It shows how living in such a neighbourhood actually produces criminal behaviour in a given person.

The mechanism is simple. Living there increases the odds that a person’s closest associations will model and approve of rule-breaking.

Sutherland explicitly drew on the Chicago School’s ecological tradition when he developed DAT. The two theories share intellectual roots, rather than competing for the same explanatory territory. Neither one really replaces the other.

A useful shorthand: social disorganization sets the stage, and DAT explains what plays out on it.

Self-Control Theory

As the Cambridge Study section above shows, self-control theory locates the cause of crime in a stable trait formed in early childhood, rather than in ongoing social learning.

The strongest longitudinal evidence available, Polakowski (1994), supports self-control theory’s predictions rather than DAT’s. That evidence never tested peer association at all.

The two theories make genuinely different claims about when the relevant cause takes hold. Self-control theory points to early childhood socialisation producing a lasting trait.

DAT points somewhere else entirely. It points to ongoing, potentially lifelong exposure to peers.

They are not mutually exclusive, though. A person with low self-control may also be more likely to seek out delinquent peers. Both mechanisms could then operate together on the same person.

Hirschi’s Control (Bonding) Theory

Hirschi’s control theory asks why people don’t offend. It answers that strong bonds to family, school, and other conventional institutions restrain most people from acting on impulses control theorists treat as more or less universal.

This is a genuinely different starting assumption from DAT, which treats the impulse itself as something that must be learned, rather than assumed to already be present in everyone.

The two theories have nonetheless been fruitfully combined. Weak bonds to parents make a young person more available for, and more likely to seek out, delinquent peers, at which point DAT’s learning mechanism takes over.

One prominent reviewer has called this integrated account “the single best empirically-substantiated theory of crime that can be offered by modern criminology” (Warr, 2001). That’s a striking endorsement given how many rival theories criminology has produced.

Rational Choice Theory

Rational choice theory is the shared parent hub for this whole cluster of crime theories. It assumes offenders weigh costs and benefits before acting.

It treats criminal preferences, and the techniques needed to act on them, as more or less already given. It doesn’t ask where they came from.

DAT supplies exactly what rational choice theory leaves unexplained. It explains where those preferences, techniques, and the perceived acceptability of offending actually come from in the first place.

Read together, the two theories describe different stages of the same causal chain. DAT explains how a person comes to see an act as an option worth considering.

Rational choice theory then explains the next step. That option gets weighed against its likely costs and benefits.

Radical Criminology and Labelling Theory

Radical criminology asks a different question entirely. It doesn’t ask how people learn to break rules already defined as criminal.

Instead, it asks who has the power to decide which acts get defined as “crime” in the first place. DAT takes that as already settled by the legal code.

Labelling theory raises a related point at the individual level. Once a person is publicly labelled deviant, that label itself can shape their future behaviour.

The label works through stigma and a changed self-concept, not through the original act itself.

DAT is focused on learning definitions from others before an offence, not on the social reaction after one. So it doesn’t directly incorporate this mechanism.

These perspectives are best read as addressing different questions. One asks how deviance is learned; the other asks how it is defined and reacted to. Neither one contradicts the other.

Biological Theories

Biological and constitutional theories locate the cause of crime in inherited or physiological traits. Sheldon’s constitutional theory is one well-known example.

Sutherland was a sharp, explicit critic of exactly this style of explanation. He rejected the idea that biology alone could explain crime.

He argued that biological accounts couldn’t explain why crime rates vary so dramatically. Rates vary across social groups, time periods, and neighbourhoods that share the same underlying biological makeup.

If biology were the real cause, that variation shouldn’t happen. People with similar biological traits should offend at similar rates.

DAT’s entire framework can be read as a direct sociological alternative. Sutherland regarded the constitutional theories of his day as scientifically discredited, and DAT offered a testable social mechanism in their place.

---

References

Akers, R. L. (1998). Social learning and social structure: A general theory of crime and deviance. Northeastern University Press.

Akers, R. L., & Jensen, G. F. (2017). Social learning theory and the explanation of crime: A guide for the new century. Routledge.

Akers, R. L., Krohn, M. D., Lanza-Kaduce, L., & Radosevich, M. (1979). Social learning and deviant behavior: A specific test of a general theory. American Sociological Review, 44(4), 636–655. 

Dishion, T. J., McCord, J., & Poulin, F. (1999). When interventions harm: Peer groups and problem behavior. American Psychologist, 54(9), 755–764.

Downes, D., Rock, P., & McLaughlin, E. (2016). Understanding deviance: A guide to the sociology of crime and rule-breaking (7th ed.). Oxford University Press.

Farrington, D. P. (2003). Key results from the first forty years of the Cambridge Study in Delinquent Development. In T. P. Thornberry & M. D. Krohn (Eds.), Taking stock of delinquency: An overview of findings from contemporary longitudinal studies (pp. 137–183). Kluwer Academic/Plenum.

Farrington, D. P., Coid, J. W., & Murray, J. (2009). Family factors in the Cambridge Study in Delinquent Development. Home Office Research Study.

Farrington, D. P., Ttofi, M. M., & Piquero, A. R. (2016). The Cambridge Study in Delinquent Development: A review of findings from 50 years of follow-up. Psychology, Crime & Law, 22(7), 609–624. 

Gottfredson, M. R., & Hirschi, T. (1990). A general theory of crime. Stanford University Press.

Heimer, K., & Matsueda, R. L. (1994). Role-taking, role commitment, and delinquency: A theory of differential social control. American Sociological Review, 59(3), 365–390. 

Kruis, N. E., Seo, C., & Kim, B. (2019). Revisiting the empirical status of social learning theory on substance use: A systematic review and meta-analysis. Substance Use & Misuse, 55(4), 666–683.

Matsueda, R. L., & Anderson, K. (1998). The dynamics of delinquent peers and delinquent behavior. Criminology, 36(2), 269–308.

McCuddy, T. (2020). Peer delinquency among digital natives: The cyber context as a source of peer influence. Journal of Research in Crime and Delinquency, 58(3), 306–342.

Miller, B., & Morris, R. G. (2016). Virtual peer effects in social learning theory. Crime & Delinquency, 62(12), 1543–1569.

Polakowski, M. (1994). Linking self- and social control with deviance: Illuminating the structure underlying a general theory of crime and its relation to deviant activity. Journal of Quantitative Criminology, 10(1), 41–78. 

Pratt, T. C., Cullen, F. T., Sellers, C. S., Winfree, L. T., Jr., Madensen, T. D., Daigle, L. E., Fearn, N. E., & Gau, J. M. (2010). The empirical status of social learning theory: A meta-analysis. Justice Quarterly, 27(6), 765–802.

Sutherland, E. H. (1939). Principles of criminology (3rd ed.). J. B. Lippincott.

Sutherland, E. H. (1947). Principles of criminology (4th ed., D. R. Cressey, Rev.). J. B. Lippincott.

Sutherland, E. H. (1949). White collar crime. Dryden Press.

Sutherland, E. H., & Cressey, D. R. (1978). Criminology (10th ed.). J. B. Lippincott.

Sykes, G. M., & Matza, D. (1957). Techniques of neutralization: A theory of delinquency. American Sociological Review, 22(6), 664–670.

Vold, G. B., Bernard, T. J., & Snipes, J. B. (2002). Theoretical criminology (5th ed.). Oxford University Press.

Walmsley, R., Howard, L., & White, S. (1992). The national prison survey 1991: main findings. HM Stationery Office.

Warr, M. (2001). The social origins of crime: Edwin Sutherland and the theory of differential association. In R. Paternoster & R. Bachman (Eds.), Explaining criminals and crime: Essays in contemporary criminological theory (pp. 182–191). Roxbury.

Further Readings

• The Current State of Differential Association Theory