Theories of crime


Notes for PowerPoint


1.   Classical and positivist theories

2.   Radical criminology

3.   Labelling theory

4.   Constitutional theories

5.   Genetic Explanations

6.   Environmental explanations

Classical and positivist theories

The classical theorists believe in the concept of free will when explaining crime.  If the rewards for being a criminal are greater than the retribution it would bring then criminal behaviour seems more likely.  This theory would predict that extreme punishments such as flogging or death would deter people from all crimes.

Cesare Beccaria in the 18th century successfully argued that the punishment should fit the crime.  Excessive punishments could be counter productive as a criminal would reason that if they were to be hung for stealing a sheep they might as well commit more serious offences such as murder and robbery.  Jeremy Bentham, the British philosopher of the early eighteenth century, justified punishment in terms of its being able to prevent further crime rather than encouraging it.  Why would Bentham, if he were alive today, argue that the Bulger killers should not go to an adult prison?

Classical theory was prominent in the late 18th and 19th Centuries but positivist theories then became popular instead.  Since the 1970s there has been a resurgence of interest in the classical theory of crime, with many people calling for harsher punishments.

Positivist theories have been criticised for failing to discover the causes of crime and to develop effective strategies for controlling crime.  Positivist theories discount the role of free will; instead it takes into account factors such as genetic transmission, personality, learning and moral development.  The sociological perspective is also taken into account. Emphasis is placed on anomie (a lack of moral standards in society) and strain resulting from poverty imposed by a rigid class structure.  Strategies to reduce crime would involve treatment at an individual level or intervention at a social level.

Radical Criminology

Radical criminology was first proposed by Taylor et al. (1973) who based his views on the Marxist position.  No act is naturally immoral or criminal; definitions of crime are socially determined, reflecting current social values.  Crime is therefore seen as socially determined.  Criminal law is designed to suit the purposes of the wealthy and powerful.  Those without money result to crime in order to enjoy the luxuries of the wealthy.  The wealthy also commit crime in order to gain further wealth.  However because the legal system flavours the wealthy they are less likely to be arrested and punished.

The solution to crime would occur at economic, political and social levels.  If wealth was redistributed then any crime that occurred would be the result of individual psychopathology.

Labelling theory

Labelling theory accords well with radical criminology as it sees criminal behaviour as being defined by society.  People without power are labelled by those with power (for example judges, parents, police and, teachers, etc.).   Behaviour is not seen as right or wrong rather as a deviant behaviour.  This argument applies not only to criminal behaviour but also to other groups in society that have been labelled, such as alcoholics, the mentally ill, the deviant, etc.

There are two consequences of labelling - the creation of stigma and the modification of self image (Gove 1975).

Stigma refers to the public attitude of condemnation and the subsequent exclusion of the criminal.  The criminal is seen as a person to be avoided and treated with suspicion.  The criminal is barred from certain types of employment, the family may make them unwelcome, the police may give them an undue amount of attention.

The modification of self-image comes about because of the stigma the criminal experiences.  We have a self-fulfilling prophecy whereby the individual becomes the person so described by the label.  The label becomes a role and the individual changes that life to suit that role.  More crimes are committed and the individual forms an identity, that of the criminal along with all its associated values, Attitudes, and beliefs.  Lemart (1951) calls this process deviance amplification.  Any intervention such as punishment or treatment simply reinforces the individual's perception, that of being a criminal.

Arguments against labelling theory would be its failure to specify how some individuals started on a path of crime and that some crimes, such as rape and murder, are universally judged as wrong.

Becker (1963) says that deviance is a normal part of adolescent life.  Such behaviour is labelled deviant and the adolescent then sees himself as a ‘deviant’, which encourages him to continue his deviant behaviour into adulthood.  The initial deviance might be some minor theft (primary deviance) and this through labelling turns into more serious crime (secondary deviance).  Secondary deviance is when the adolescent accepts the label and follows the deviant lifestyle (Lemert, 1972).  Labelling theory cannot explain all criminal behaviour but might help to explain crime amongst the lower economic social groups.

Chen,-Xiaoming (2002) discusses the underlying philosophy and functions of social control in Chinese society, and applications of reintegrative shaming theory. Social order in China is based on moral socialization and not on the deterrence of law. The Chinese system of social control reflects the 2 basic Western approaches to crime control, informal social control and opportunity reduction. Labeling theory holds that formal and informal societal reactions to delinquency can influence delinquents' subsequent attitudes and behaviours. Reintegrative shaming theory and practice aim at maximizing shame while maintaining community links and providing opportunities for the offender to make amends to those injured. The evidence presented supports reintegrative shaming theory as a valid model of Chinese social practice, rather than labeling theory.

There are suggestions recently that schools should provide the police with information about pupils who are ‘potential criminals’ even though they have not committed a crime.  There are concerns that this will label youngsters.


Constitutional theories

Lombroso in 1876 argued that the criminal is a separates species, a species that is between modern and primitive humans.  He argued that the physical shape of the head and face determined the "born criminal".  These people were primitive and were unable to adapt to modern morality.  His view was based on genetics.  The atavist (primitive genetic form) and had large jaws, high cheek bones, large ears, extra nipples, toes or fingers, and were insensitive to pain.  Lombroso went further and suggested that from the surveys he had carried out in prison, he could detect physiological differences between different types of criminal. Thus, murderers were said to have:

Cold, glassy, blood-shot eyes, curly, abundant hair, strong jaws, long ears and thin lips

Whilst sex offenders have:

Glinting eyes, strong jaws, thick lips, lots of hair and projecting ears.'

Lombroso's views received some indirect support from Goring, who in 1913 studied 3000 English convicts and found that although anatomical differences were not as extreme as Lombroso had suggested, a common factor in his subjects was low intelligence. At this time intelligence was regarded as genetically determined and so criminal behaviour too was seen to be linked to genetic inheritance. Within such a reductionist biological determinism, characteristics such as intelligence and criminality are seen as fixed and immutable. Any possibility of social factors influencing outcome is minimised. From this perspective it is a short step to compulsory sterilisation of those who are deemed immoral' or mentally deficient, a policy which not only existed in Nazi Germany, but in Sweden, Britain and the USA (see Rose et al., 1984).

It is easy now to criticise Lombroso's theories; for instance, his research methodology was not particularly rigorous (he did not use a proper control group, often relying on large groups of soldiers, and his criminal samples contained large numbers of the mentally disturbed) and his conclusions sound bizarre. Perhaps the most important criticism, however, is Lombroso’s failure to recognise that correlation does not imply causality. Simply became his criminal subjects shared a significant number of physical anomalies does not mean that this made them criminal. It could be that poverty and deprivation produce physical defects, rather than the defects being the result of genetic transmission. There are a large number of people with unpleasant looks in the prison population.  However this may be caused by the stereotype of an unpleasant looking person being a criminal.  If from an earlier age less attractive individuals have been rejected by others then they might turn to criminality.  A study that involved giving prisoners plastic surgery in America, found that they were less likely to commit crimes upon release (Kurtzberg et al, 1978).

But in spite of the obvious criticism and lack of political correctness, could Lombroso have had something? Maybe we can rule out the possibility of genetic transmission, for as Rowe (1990) says:

No responsible geneticist would argue that a specific gene exists for crime, as specific genes may be identified for Huntington's disease or eye colour.

William Sheldon believed that people could be classified into three body shapes, which correspond with three different personality types.

  1. endomorphic (fat and soft) tend to be sociable and relaxed.
  2. ectomorphic (thin and fragile) are introverted and restrained
  3. mesomorphic (muscular and hard) tend to be aggressive and adventurous.

Sheldon, using a correlational study, found that many convicts were mesomorphic, and they were least likely to be ectomorphic (Sheldon et al 1949).

Although Sheldon's work was criticised on methodological and subjective grounds (because he rated his subjects' body types himself) his theories were supported in the area of criminal behaviour by Glueck and Glueck (1956) who found that in a sample of delinquents 60% were mesomorphs, whilst in a non-delinquent sample only 31% were, and by Cortés and Gatti (1972) who found in a sample of 100 delinquents that 57% were mesomorphic compared with 19% of controls.

Wadsworth (1979), in contrast to Sheldon's finding that large and muscular convicts predominate, found that British delinquents who committed serious crimes tended to be smaller than average and were late in reaching puberty!

West and Farrington (1973) in their longitudinal study of London working-class boys, found no association between delinquency and body shape or size.

Feldman (1977) points out that these relationships between shape and criminality can be explained by expectations.  Such features might catch the attention of police or influence juries or sentencing.  The mesomorph is more likely to use aggression because size is on their side.

In 1939 Thornton showed 20 photographs of criminals to people who were asked to choose one of four crimes each of the criminals might have committed. He found that people could match faces to crimes more reliably than would have been predicted by chance. Bull (1982) achieved similar results, this time using photographs of non-criminal individuals but still producing an association between some faces and some crimes, which was stronger than could have been predicted by chance.  Masters and Greaves (1969) surveyed the incidence of facial deformities in 11,000 prisoners and concluded that 60% of them had facial deformities by comparison with 20% in a non-criminal population.

Genetic explanations

1.   Twin studies

2.   Adoption studies

3.   Family studies

4.   Chromosomes

5.   Neurochemical explanation

6.   Neurophysiological

Twin studies

An 'MZ apart' study is when two monozygotic children have been brought up apart.  If both turn out to be criminals then this would be support for the genetic explanation.  The degree of similarity between two twins is known as the concordance rate.  This rate can then be compared with dizygotic twins who are brought up together ('DZ together').

Looking at a number of studies the average concordance rate is 55% for MZ twins and 17% for DZ twins (Bartol, 1999).  


  1. Different studies define criminality in different ways (e.g. traffic violations, military offences, treason during World war 2).
  2. Quasi-experimental designs are not so controlled as experimental designs.
  3. Age of separation of MZ twins.
  4. Misclassification of twins as MZ or DZ.
  5. · MZ twins look alike and may therefore generate more similar social responses than DZ twins. This means that in addition to sharing the same genes, they may also share an almost identical social environment.
  6. · MZ twins often have a very close: relationship and may therefore develop similar interests, which might include criminal behaviour.
  7. · Very small sample size in some studies, because of the inherent difficulties in obtaining access to criminal twins.
  8. · Variable definitions of criminal behaviour.

Recent studies looking at measures of personality and intelligence in relation to MZ twins reared apart have found some striking similarities (Bouchard et al., 1990).

Adoption Studies

If a child who has criminal parents is adopted and later the child becomes a criminal then this will lend support to the genetic explanation.

A study by Schulsinger (1972) found that only 3.9% of such children developed criminal tendencies.  This was compared with control children (adopted children from non-criminal biological parents), where 1.4% became criminal.  The small difference was not significant.  The sample was large at 57, so there was a good prospect of detecting significance.  The psychopathy was loosely defined as impulse-ridden behaviour.

Crowe (1974) found that in a sample of 52 adopted children of imprisoned women; seven of them had at least one criminal conviction, by comparison with only one in a matched control group.

A retrospective study by Mednick et al (1987) looked at court convictions in a small European country and found 14,000 adoptees amongst them.  The criminal records of their biological and adoptive parents were then investigated.  Many of the adoptees had criminal biological parents (particularly strong relationship for sons and fathers).  There was no relationship in the types of crime committed.  Where there was an improvement in social conditions there was a reduction in crime (going against the genetic explanation).

Biological parents have criminal record

Adoptive parents have criminal record

% of sons with criminal record (Mednick 1987)

% of sons with criminal record (Bohman, 1995)


















  1. Age of adoption
  2. Amount of contact with biological parents (contamination effect)
  3. Adoptive family selected to be similar to biological family.
  4. Small sample sizes.

Family Studies

40% of sons of criminal fathers became criminal compared with13% of sons of non-criminal fathers (Osborn and West, 1979).

A case study of a Dutch family found a link between a genetic mutation on the X chromosome, passed from mother to son, and uncontrollable rage.

The Cambridge Study: problem families produce problem children

Farrington, D.P., Barnes, G.C. and Lambert, S. (1996)  See study 3

Farrington, D.P. (1996) The Development of Offending and Antisocial Behaviour from Childhood to Adulthood. In Cordella, P & Siegel, L. (eds) Readings in Contemporary Criminological Theory. Boston: Northeast University Press.   Aim:    To describe the development of delinquent and criminal behaviour in inner city males, to investigate how tar it could be predicted in advance, and to explain why juvenile delinquency began. The original study was begun in 1961—62: this is a follow-up study of 411 London boys, born mostly in 1953.   Method: Design:       This is a longitudinal study based on interviews and tests conducted at various ages over, to date, 24 years.   Participants:          The great majority of the sample was chosen by taking all the boys who were then 8—9 years old and on the registers of six state primary schools in one location in London. The boys were almost all white and predominantly from working-class families.   Measures used: 1          Tests and interviews at school at ages 8, 10 and 14 years. Interviews in the research office at about 16, 18 and 21 and in homes at about 25 and 31. The tests in schools measured intelligence, attainment, personality and psychomotor skills. The interviews collected infor­mation concerning employment history, relationships with females, leisure activities such as drinking and fighting, and offending behaviour.   2            interviews with the parents about once a year from when the study began (when the boys were about 8) to when they were 14 or 15 years old, It was mainly the mothers who were inter­viewed. The parents provided information on family size, employment history, childrearing practices, degree of supervision and whether there had been any temporary separations.   3            Questionnaires completed by the boys’ teachers when the boys were 8, 10, 12 and 14. These concerned troublesome and aggressive behaviour, attainment, and truancy. Peers provided information on popularity, daring, dishonesty and troublesomeness.   4            Records From the Criminal Record Office to gain information on convictions of the boys, their parents, their siblings and, later, their wives or cohabitees. Minor offences such as common assault, traffic offences and drunkenness were not included in these statistics.   Results: Statistics on crimes committed:   1   By the age of 32, 37 per cent of the males had committed criminal offences. The peak age was 17. Nearly three quarters of those convicted as juveniles were reconvicted between the ages of 17 and 24, and nearly half of the juvenile offenders were reconvicted between the ages of 25 and 32.   2            Offending was very much concentrated in families. Just 4 per cent of the 400 families accounted for 50 per cent of all convictions of all family members.   3     The worst offenders tended to be from large-sized, multi-problem families.   4                        Most juvenile and young adult offences occurred with other people, but this co-offending declined with age. Co-offending with brothers was not uncommon when the siblings were close in age but co-offending with fathers (or mothers) was very rare.   5     The most common crimes in late teens were burglary, shoplifting, theft of and from vehicles, and vandalism. All of these declined in the twenties, but theft from work increased,   6          Self reports showed that 96 per cent of the males had committed at least one crime that might have led to conviction, so criminal behaviour was not deviant.   Predictors of crime at age 8—10: 1     Antisocial child behaviour including troublesomeness, dishonesty and aggression.   2          Hyperactivity-impulsivity-attention deficit.   3   Low intelligence and poor school attainment. 4 Family criminality.   5        Family poverty, including low family income, large family size and poor housing.   6           Poor parental child-rearing techniques, poor supervision, parental conflict and separation from parents.   Conclusion:    Any one of these factors independently predicts offending and the following suggestions are made as to the reasons for this. Children from poorer families are more likely to offend because, due to poor school attainment and an inability to manipulate abstract concepts, they are less able to achieve their goals legally. Impulsive children cannot see the consequences of their actions and desire immediate gratification. Children who are exposed to poor child­rearing practices, conflict or separation do not build up inhibitions against antisocial behaviour. Lastly, children from criminal families and those with delinquent friends develop anti-estab­lishment attitudes and the belief that it is justifiable to offend. This research demonstrates that problem children grow into problem adults who in turn produce problem children. Sooner or later serious measures must be taken to break this cycle.

Farrrington's findings suggest an interaction between environment and genetics.  Mednick et al's (1983) study also supports this view as where biological and adoptive parents were criminal 25% of the children became criminals, compared with 20% for children with non-criminal adoptive parents.

Walters (1992) concluded from a meta-analysis of 38 studies of twins, families and adoptions, that genetics played a small but significant part in determining criminality.  Many of the studies suffered from methodological flaws.

Established risk factors for antisocial behaviour (which may or may not result in criminal convictions) include parental criminality, ineffective and inconsistent discipline, family discord and a deviant peer group (Rutter, 1971a: Patterson. 1982) but the mix is so volatile that it is almost impossible to differentiate between the contribution of genetics and the environment. Nevertheless Farrington (1995) is quite clear about the areas that should be targeted to prevent the development of antisocial behaviour:

1.      improving school achievement (to avoid low intelligence and school failure);

2.      improving child rearing methods (to reduce erratic discipline and promote positive parent-child interaction);

3.      reducing impulsivity (teaching cognitive and interpersonal skills to combat egocentricity) and

reducing poverty (providing increased economic resources for deprived families).


Size of family

Larger families - less attention.  Also younger siblings model their behaviour on that of older siblings - the 'contagion effect'.

Interactions within the family

Correlation between crime and poor interaction within the family.  Parental conflict, poor use of language and communication, mistrust of other family members.  Patterson (1982) observed interactions between anti-social children and other family members.  Amongst 14 factors that contributed to coercive exchanges (giving orders) was 'nattering', which is a constant scolding of the children with no particular focus or specific threats for non-compliance.

Disruption of family home

John Bowlby (1946) maintained that it was coming from a broken home that led to delinquency (see Hodges and Tizard).  However, Rutter (1971) said that it was the amount of discord and distress rather than the 'broken home'.  Loss of a parent through death does not necessarily lead to delinquency.

Child rearing practices

Hoffman (1984) found that 'power assertion' style of child rearing was associated strongly with delinquency.  Power assertion involves physical punishment and criticism with little rewarding or praise.  Other strategies involve 'love withdrawal' and 'induction' (explaining and reasoning with the child).

Physical punishment encourages the child to consider aggression as acceptable as an authority figure uses it.  Inconsistent punishment has a profound effect on criminal behaviour.  Coopersmith (1968) found that inconsistent punishment was linked to very low self-esteem and delinquent behaviour.  Lack of parental supervision is associated with inconsistent punishment.  Harsh punishment is effective in the short term, but in the long run can lead to violent delinquency (Straus, 1991).

Loeber and Dishion (1983) considered the following factors to be associated with male criminal behaviour: family size, quality of parental supervision, parental alcoholism, employment history.

School and peers

Low academic achievement is associated with criminal behaviour.  Failure at school may be because of low intelligence but also could be due to undiagnosed learning difficulties such as dyslexia.  The failing (or 'failed') child then joins an anti-school sub-culture of truancy and delinquency.  The child will seek approval that is lacking from school.  The child might be good at lighting fires and get approval from his peers.  This boosts his self-esteem and leads to more fire lighting behaviour.  It would be a mistake, however, to assume that peer pressure alone can explain all juvenile crime. Some offences are committed alone and when Agnew (1990) asked 1400 adolescents what had led them to engage in specific offences, he found that pressure from peers was only one factor, the others being a rational choice to obtain money or kicks or a result of anger or provocation.

Schools with high staff turnover, low staff commitment, streaming and social disadvantage tend to have the highest (Hargreaves 1980).

It is clear from preschool enrichment programmes such as Operation Headstart in America that potential social and educational disadvantage can be tackled at an early age, but a recent study indicates that the particular type of preschool programme provided can reduce the chances of criminality developing (Schweinhart and Weikart, 1997). This study traced the progress of 68 children from poor backgrounds who were randomly placed in different preschool programmes at the age of four years. The conclusions of this study were that children from economically deprived backgrounds who are educated using traditional teaching methods are four times more likely to have a criminal record at age 23 than those who are given greater autonomy and are allowed to participate in the planning and review of their own learning activities.  Interestingly, the children who were taught in a more structured learning setting initially demonstrated higher IQ scores than the children in the other programmes, but were not encouraged to develop problem-solving, decision-making or interpersonal skills. They subsequently developed more severe behavioural problems than their peers.

A major predictor of delinquency among adolescents is the delinquency of close friends; few youngsters engage in solitary offending, most work in groups of two or three (Aultman, 1980). Hargreaves (1980) argues that status deprivation’ of those who fail in the school system leads to both negative attitudes towards education and the motivation to join delin­quent groups. Gold (1978), however, argues that different processes are acting in the two situations. School failure results in lowered self-esteem whilst antisocial behaviour amongst peers helps to present an attitude of defiance, which is rewarded by peer admiration. When boys are unloved and maltreated at home and have been made to feel a failure at school, it is not surprising that they turn to a group who provides friendship, admiration and a sense of belonging, as well as excitement and some material rewards.

For violent/property crime, cannabis, and alcohol abuse, Fergusson et al. (2002) found there was significant evidence of age-related variation in the strength of association with deviant peer affiliations, with affiliations having greater influence on younger than older participants.


There is disagreement amongst theorists about the extent to which certain adolescents are passively recruited into delinquent groups and how much they seek out those who they perceive as similar to themselves. It would appear that unsocialised youths tend to seek out similar partners for friendship (Bandura & Walters 1963; Dishion et al, 1991) and that once these alliances are formed, the associates tend to become more alike over time, suggesting that they influence each other. This is consistent with research on interpersonal attraction showing that people form relationships with those who share their attitudes and that stable friend­ships are maintained by mutual influence (Dwyer, 2000).


Others argue that the peer group is a relatively unimportant influence in delinquency and occurs only after early life experiences in the family and school have led the adolescent towards a life of delinquency. Finding friends with whom to commit crimes is incidental to this process. However Hirschi (1969), who originally espoused this view, later acknowledged that association with delinquents could influence even those boys who had strong family ties. The strength of peer group influence is as yet still not established. The evidence does suggest that whilst the peer group may facilitate criminal behaviour, a propensity for delinquency is already present in individuals who join such groups (Gottfredson & Hirschi, 1990).

Economic factors

There is a link between unemployment, poverty and crime.  The most persistent offenders had not had a stable employment record; changing jobs often (Farrington and West 1990).  An analysis of data from 42 police forces in the UK demonstrates an association between crime and male unemployment, the amount of easily stolen and resalable property and high wage disparities (Witt, Clarke and Fielding, 1999).

Note that it would not be true to say that only poor people commit crimes.

May not be just lack of money, unemployment brings loss of status, loss of self-esteem, boredom and alienation.

Sociologists draw our attention to the importance of economic inequality rather than poverty as an important influence on much social behaviour, including crime. Economic inequality entails a comparison of the levels of material wealth between different groups within society. Goodman et al. (1997), in an analysis for the Institute for Fiscal Studies, reported that the level of economic inequality has risen massively in the last 20 years: the combined income of the top 10 per cent of earners was equal to that of the bottom 50 per cent of all earners. It has been suggested by several theorists (e.g. Stack, 1984) that there is a relationship between crime and inequality. From an analysis of 16 studies of crime and economic inequality between 1974 and 1985, Box (1987) found that eleven showed a positive correlation between economic inequality and crime and he hypothesised that this inequality was responsible for the rise in crime during periods of recession, especially amongst the poorest — the young, women, and ethnic minorities. From an analysis of the situation in the US, Hagan (1994) suggested that the increasing inability of the poorest citizens to attain any upward social mobility is likely to lead to increasing competition and violence in the underworld of vice and racketeering. Ironically, this is liable to give the wealthier more opportunity to exploit the poor, both by legal and illegal means.

Horton (2002) examined decreases in violent crime rates in the US. Crime and economic data of 77 community areas of Chicago, Illinois, were analysed. Results show that an increase in concentrated poverty was directly related to violent crime, evidenced by homicide rates. Concentrated poverty led to violent criminal behaviour, regardless of racial factors. It is concluded that some decrease in violent crime has been the result of endogenous variables including a booming economy and a decline in the crack-cocaine drug wars. However, Chicago has experienced an increase in its homicide rate despite a downward national trend in the US, the result of concentrated poverty.

Lone-parent families

Being a lone-parent in itself is not likely to increase the child's chances of being a criminal.  However, being a lone parent does bring the greater likelihood of having a number of problems that are associated with juvenile delinquency, such as the trauma of a divorce, poverty and poor mental health. [Youth who have been exposed to community violence or who had a mother with mental illness were 4 times as likely to commit serious criminal behaviours (Preski and Shelton 2001)].  If these factors are matched then there is no difference between lone-parents and married parents in the level of delinquency of their children.

Is there a link between being an unwanted child and becoming a criminal?  If so where abortion is freely available there should be seen a reduction in crime.  Researchers have found an association between increased abortion and a decrease in crime years later (when the child becomes a young adult).  The data was taken from individual states in America (each with a different abortion rate) for after 1973, and compared with the level of crime in those states in the 1990's.  Can you think of other explanations?  Being unwanted could not explain all crime; white middle class children from privileged backgrounds have become mass-murderers.

Social Roles

Young men who are macho, believing in a traditional male role ideology, are more likely to become criminals.  Hollywood's depiction of maleness is one where violence is integral to maleness.  The power of social roles is demonstrated in Zimbardo's prison experiment.

Hirschi’s control theory

In his control theory, Travis Hirschi (1969, 1995) claims that the essence of social control lies in people’s anti­cipation of the consequences of their behaviour. Hirschi assumes that everyone finds at least some deviance tempting. Imagining condemnation from family or friends is sufficient to deter most people from temptation; concerns about how transgressions will affect their careers will give others pause. By con­trast, individuals who have little to lose from deviance are most likely to become rule-breakers.

Hirschi asserts that conformity arises from four types of social controls.

1.         Attachment. Strong social attachments encourage conformity; weak relationships in the family, peer group and school leave people freer to engage in deviance.

2.         Opportunity. The more one perceives legitimate opportunity, the greater the advantages of confor­mity. A young person bound for university, one with good career prospects, has a high stake in con­formity. By contrast, someone with little confi­dence in future success drifts more toward deviance.

3.         Involvement. Extensive involvement in legitimate activities — such as holding a job, going to school and completing homework, or pursuing hobbies —inhibits deviance. People with few such activities —who simply ‘hang out’ waiting for something to happen — have time and energy for deviant activity.

4.         Belief Strong beliefs in conventional morality and respect for authority figures restrain tendencies toward deviance. By contrast, people with a weak conscience are more vulnerable to temptation.

Not having a stake in society could lead to criminal behaviour.  Research has shown that juveniles often stop offending once they settle down and establish social bonds through work or a close personal relationship (e.g. marriage).


Jacobs et al (1965) suggested that men with the XYY syndrome were more aggressive than normal 'XY' men.  XYY men are over-represented in the prison population.  There are 15 sufferers per 1,000 in prisons and 1 per 1000 in the general population.

XYY men have lower intelligence and it could be this that leads to aggression.

A Danish study, which screened 4591 men for the presence of XYY, found only 12 cases (Witkin et al., 1976). Whilst these individuals were indeed more likely to be involved in crime than chance would have predicted (41.7% of them, by comparison with 9.3% of the XY individuals), it was not involvement in violent crime. Their conclusion was that the over-representation of XYY males in prisons and special hospitals was more likely to be the result of other characteristics - low intelligence and above-average height - and the social reaction that these characteristics may have produced.

Some men thought to have the XYY syndrome have later been found to be XY, causing problems with studies.  Key symptoms of XYY are being unusually tall and suffering from acne and scars; some men have been incorrectly classified as XYY on this evidence alone.

Another chromosome theory is the 'long Y' theory.  Individuals with an elongated Y chromosome were higher on a delinquency scale.

Neurochemical explanation

Though there may not be a criminal gene it is possible that certain genes may influence the brains chemistry (neurochemicals), and this may account for the criminal behaviour. The chemistry of the brain can be affected by environmental as well as genetic factors.  Lower levels of serotonin have recently been blamed for anti-social behaviour.  Low levels of serotonin have also been blamed for depression and eating disorders.  Serotonin is linked to control or inhibition of behaviour.

The brains chemistry can be influenced by diet, for example, food additives, pollution or hypoglycaemia (low blood sugar levels associated with forms of diabetes).  Dawn Stanton attacked her husband with a knife when hypoglycaemic.  But not all diabetics without insulin act criminally.

Individuals with low serotonin levels, who also do not eat (and therefore their blood glucose levels drop) are more susceptible to impulsive behaviour at that time.  One study has found a link between low blood glucose levels and arson.

David Garabedian worked with strong pesticides and attacked and killed a woman where he was working.  He was a quiet man who changed after the influence of the chemicals.  The reason for the change was an increase in acytecholine in the hypothalamus.  Brain chemistry in itself is not a sufficient explanation to account for all types of criminal behaviour.

The link between diabetes and crime could be seen as racist as the main groups in America who suffer from diabetes are African Americans, Hispanics and Native Americans.  Biological explanations to crime allow politicians to avoid spending money to combat environmental causes such as poverty and inner city problems.

Women who smoke more than 10 cigarettes a day are more likely to have sons with behaviour problems.  177 boys, who were referred to clinics for conduct disorders, had their records analysed by researchers.  80% of mothers who smoked half a packet a day during pregnancy had sons with conduct disorders.  Conduct disorders in children can become antisocial personality disorders in adults in one third of cases.  Such adults commit 50% of all crimes.  However there may not be a causal relationship as the mother's smoking might indicate other problems in the family.


Individuals who take large amounts of steroids can become extremely violent (known as "roid rage").  Steroids, usually taken to increase muscle growth, also increase testosterone levels. Horace Williams, a body builder, beat a man to death after taking two thousand times the recommended dosage of steroids.

Correlational studies looking at the relationship between aggression and levels of testosterone have tended to give conflicting findings; some show positive correlations, others negative correlations or no correlation.


When Charles Whitman killed 21 people during one day in 1966 shooting 16 of them from a Texas university tower an autopsy subsequently revealed that he had a large brain tumour, which might have been affecting the area of the brain responsible for controlling aggressive, urges, the amygdala. Such conditions or serious brain injuries may cause dramatic personality and behavioural changes which can result in the acting out of violent impulses. In the 1970s, it was suggested that surgical intervention in such cases might he beneficial (Delgado, 1969) though ideological objections to such drastic and often irreversible action were legion. However, this did not prevent Mark and Ervin (1970) suggesting that brain dysfunction related to focal lesion was the cause of violence and should be treated by psychosurgery.

Another suggestion is that certain individuals, as a result of genetic predisposition or brain damage at birth, suffer from a cluster of symptoms which render them incapable of moral control and because of cortical underarousal, they are constantly seeking stimulation. The symptoms appear in early childhood and are subsumed in the term attention deficit hyperactivity disorder (ADHD). They include inappropriate degrees of inattention, impulsiveness, challenging behaviour and hyperactivity. The symptom cluster is also known as MBD or minimal brain dysfunction.

Moir and Jessel (1995) have suggested that this brain dysfunction can account for impulsive and seemingly irrational crimes, some of which involve violence. Moreover, they suggest that brain scans could identify the disorder in young children who are already showing behavioural problems and that treatment (Ritalin (a stimulant), biofeedback and parental training) could possibly prevent these children growing up to be seriously antisocial.

Another neurological disorder, developmental co-ordination disorder or dyspraxia, has also been seen as a possible contributor to antisocial behaviour. Sufferers will typically tend to be clumsy and accident prone, slow to reach milestones, uncoordinated and unpopular because they do not fit in. If these symptoms are not recognised and addressed then the social consequences of ostracism may lead to seriously disruptive behaviour. When Madeline Portwood, an educational psychologist, assessed a group of young offenders she found that 61% of them were dyspraxic and yet none of them had been diagnosed as such (Portwood, 1996).

There are serious problems ahead if young children are to be labelled as potential criminals, and this kind of biological determinism often fails to take into account the complexity of criminal behaviour and how it is socially constructed.

2000 persistent offenders in Canada were studied.  90% were found to have minor damage in the frontal or temporal regions of the brain (Yeudall, 1982).  Raine 1994 used PET scans to study the living brains of impulsive killers.  Damage was found in the pre-frontal cortex, which controls impulsive behaviour.  The task used involved sustained attention.  It involved watching a screen for 32 minutes and responding every time a 0 appeared.  Impulsive individuals also missed many of the 0s.  Oddly enough, pre-frontal under arousal has also been found in politicians!

It could be that criminals have a left hemisphere that is poorer at processing Verbal information.  Therefore they might make poor use of verbal code (or self talk) and have less development of social rules (conscience).

Hare (1970) has found that 14% of aggressive psychopaths showed slow wave activity in the temporal lobe of the brain compared to 2% of the general population.  We cannot be sure whether the slow wave activity causes a person to become psychopathic or whether being psychopathic produces slow wave activity. 

EEG readings taken of 600 Swedish children with no history of delinquency.  12 years later those with slow brain wave patterns were more likely to have had a criminal record (Mednick et al 1981).

Raine A. and Lui (1998)  See study 1

Low intelligence

There is no clear link between low intelligence and criminal behaviour, or at least no direct link.  Low intelligence could lead to failure at school and the consequent drifting away from what is for many a route to a successful career.  Failure at school could lead to low self-esteem.  This leads to emotional disturbance, conduct disorders and criminal behaviour.  The average IQ of criminals is 92, which is just below average.  Remember IQ probably reflects schooling rather than innate 'intelligence'.

Conclusions on biological explanations.

The evidence does not prove that criminals are born that way.  The evidence can still be explained in terms of a poor environment.  It may suit politicians to believe in biological explanations, as the cure could be simpler and cheaper than accepting that poverty and a poor environment are major determinants of criminal behaviour.  Setting to rights the ills of society could be politically and financially difficult.  Drugs and surgery have dealt with genetic problems in the past.  In the earlier part of the 20th century, the eugenics movement led to 27 US states having compulsory sterilization of the feeble minded.  The Nazis killed those whom they deemed to possess an inferior genetic makeup.

Environmental explanations

Some modern research points to a variety of biochemical factors which may be involved in criminal behaviour, such as allergies, environmental conditions (e.g. the presence of lead or radiation from artificial lighting) and diet (the use of food additives or deficiencies in vitamins or essential fatty acids) - Hypoglycaemia (a state of low blood sugar) has also been suggested as a factor which may affect the functioning of the brain and contribute to antisocial behaviour (Virkkunen, 1986), as has low brain serotonin (Virkkunen et al., 1996). These attempts to explain crime are very difficult to put to the test because of methodological difficulties and are thus not without criticism (e.g. Gray, 1986). Moreover, it is difficult to see how they could possibly explain the diversity of criminal behaviour, nor its onset and cessation.

More crimes are committed in the hot months of the year.  This may be due to the heat making people more aggressive and short tempered or it could be simply to do with there being more people out of doors.  Anderson (2001) considers that hot temperatures increase aggression by directly increasing feelings of hostility and indirectly increasing aggressive thoughts. Results show that global warming trends may well increase violent-crime rates. Better climate controls in many institutional settings (e.g., prisons, schools, the workplace) may reduce aggression-related problems in those settings.

Learning Theories

Learning theory would explain criminal behaviour in terms of rewards (reinforcements) and punishment.  Such an explanation does not take account of the individual’s cognitive reasoning.  Trasler (1978) suggests that ineffective parental strategies or, in some individual cases, an inability to acquire conditioned fear responses may produce inadequately socialised children who then go on to offend.

The view that criminal behaviour, in common with all other behaviour, is learned was most clearly expressed by Sutherland (1939) in his theory of ‘differential association’, which states that:


·        criminal behaviour is learned;

·        the learning is through association with other people;

·        the main part of the learning takes place within close personal groups;

·        the learning includes techniques to carry out certain crimes and also specific attitudes and motives conducive towards committing crime;

·        the learning experiences — differential associations — will vary in frequency and importance for each individual;

·        the process of learning criminal behaviour is no different from the learning of any other behaviour.


Sutherland also attempted an analysis of the gender differential in crime, arguing that boys are more likely to become delinquent than girls because they are less strictly controlled and are taught to be aggressive and active risk seekers, all characteristics likely to bring success in the criminal world and, indeed, in the world generally.


Subsequent developments of Sutherland’s theory emphasise the role of rewards and punishments, whilst the acceptance of social learning theory allows for the consideration of learning by imitation or observation of others’ behaviour and the importance of the social environment (Bandura and Walters, 1963). Peers who are held in high regard are very influential role models (Bandura, 1977), as are media stars, whilst rewards that might maintain certain behaviours take the form of status, self-esteem, financial incentives or just plain excitement. Akers (1990) goes even further, whilst sticking with Sutherland’s original theory, and looks to the meanings of crim­inal behaviour for offenders, thereby adding a useful cognitive dimension. Thus offenders may take on beliefs about offences, which allow them to maintain their behaviour. These beliefs may be positive, so car thieves might argue that joy riding is good fun and is the only way for them to access cars, whilst other beliefs allow the negative consequences of some offences to be ‘neutralised’. Other offences are described as not having ‘real’ victims, e.g. shops, or as so trivial they don’t really count.

Jeffery (1965) stated that such individuals would need to feel reinforced and not punished for associating with certain others. This could explain why some people from a particular background become criminal and others do not.

Bandura's Social learning theory, instead of looking at how the past affects behaviour says instead that violent behaviour is modelled from others.  This occurs mainly through media such as films and television.  Social learning theory is an extension of operant principles.  While operant theory maintains that behaviour is acquired through reinforcement and punishment from the environment, social learning theory holds that behaviour can also be learned at the cognitive level through observing the actions of other people.  Once learned the behaviour may be reinforced or punished by its consequences like any operant behaviour.  Bandura (1977) suggests that there are three aspects to motivation:

1.      External reinforcement (as in operant theory)

2.      Vicarious reinforcement – the observation of other people being rewarded or punished for their behaviour

3.      Self-reinforcement – gaining internal satisfaction from an activity, which therefore motivates the individual to behave in a similar way in the future.

Models who are regarded as successful or of a high status are more effective in motivating an individual to copy their behaviour.  Observational learning is thought to take place primarily in three contexts:

1.      In the family

2.      in the prevalent sub culture

3.      Through cultural symbols such as television and books (Bandura 1976).


Learning theory and social learning theories of crime evaluation

For ·        The theories are based on carefully conducted empirical research

·        They can help to explain why criminality does, to an extent run in families.

·        Males and females are socialised very differently, with females encouraged not be aggressive and to conform, males are encouraged to "stick up for themselves" and be independent.


·        The studies on which these theories are based are conducted in laboratories or other artificial environments in carefully controlled conditions.  They therefore lack validity in that they may not be sampling behaviour as it occurs in real life situations.

·        The fact that criminal behaviour tends to run in families is not necessarily due to imitation but may be due to circumstances such as social deprivation, or to genetic propensity to behave in an anti-social way.

·        There are biological differences between men and women, especially hormonal ones, and this, rather than socialisation, may account for differential rates of crime between men and women.

Sapsford (1996) Multi-causal model

An example for burglary

Intra-personal domain

Neurochemistry predisposes individual to take risks

Interpersonal domain

People known to the individual (friends and family) who commit crimes

Social domain

An unequal society.  Many people in society lacking the means to obtain highly desirable possessions.  Advertising can cause people to feel inadequate if they do not have such possessions.

Other factors such as alcohol would be considered when using a multi-causal approach.  Drunk males gave more electric shocks when under social pressure to do so compared with sober males (Placebo group) (Taylor and Sears, 1988).  Drunk participants gave severe electric shocks in retaliation for shocks regardless of how mild the shocks were, whereas sober participants took into account the level of shock received and adjusted their retaliatory shock accordingly (Leonard, 1989).

Eysenck's Personality theory and crime 

Eysenck, S.B.G, Rust, J. and Eysenck, H.J. (1977) See study 2

Extraverts need excitement so they are more likely to seek crime in order to gain excitement.

Eysenck proposed that extraverts do not condition easily.  Whereas others would learn that crime does not pay through classical and operant conditioning, the extravert would not learn these associations.

A third dimension was added; psychoticism.  Those high in psychoticism are 'egocentric, aggressive, impulsive, impersonal, cold, lacking in empathy and concern for others, and generally unconcerned about the rights and welfare of other people' (Eysenck 1982).  Eysenck predicted that criminals would score highly on extraversion, psychoticism and unstableness (Eysenck 1977).  Much research has been generated in attempts to verify Eysenck’s prediction that criminals should achieve high scores in E (extraversion), N (neuroticism) and P (psychoticisin). Whilst there has been some support for an association between P scores and criminal behaviour, there has been little support for the configuration of E, N and P scores. Moreover, there has been serious criticism of the authority with which this particular theory has been presented when there are misgivings on any evidence for its theoretical foundation (Trasler. 1987).  Also, there is little evidence to support unstableness or emotionality and crime.  However, Eysenck has provided a testable theory that has stimulated international research.

Gray (1981) has revised Eysenck's theory.

Moffit (1993) identified four types of young male offender:

Stable early-starters (ES)

Start offending from an early age and commit many crimes.  Childhood problems and low IQ

Adolescent-limited (AL)

less violent than ES; have social skills to succeed in later life

Adult-starters (AS)

Unusual.  Non-violent offences.  No childhood problems.  More likely to be females (Kratzer and Hodgins, 1999)

Discontinuous offenders (DO)

Crimes appear for a limited period only


The central concept of psychoanalysis, and the one that Freud first put forth, is the unconscious.  The concept had been around before Freud, but he was the one that made the most out of it, arguing that traumatic experiences in early childhood left their mark on the individual despite the fact that the individual was not aware of these experiences.  The idea of unconscious determination of behavior flew headfirst against the idea of free will, and was quickly jumped on by positivistic criminology. 

The next most important idea is conflict, and Freud postulated the existence of a three-part personality (an idea going back to Plato) consisting of id, ego, and superego which operated in constant conflict with one another (primarily between the id and superego) producing the basic problem of guilt which required the use of one or more defense mechanisms.  The idea of personality conflict as a cause of crime became quite popular among both scientists and the general public.

The id is a part of the unconscious that contains all the urges and impulses, including what is called the libido, a kind of generalized sexual energy that is used for everything from survival instincts to appreciation of art.  The id is also kind of stubborn, for it responds only to what Freud called the pleasure principle (if it feels good, do it), and nothing else. 

The ego is the only part of the conscious personality.  It's what the person is aware of when they think about themselves, and is what they usually try to project toward others.  The ego is dominated by what Freud called the reality principle (an orientation to the real world in which the person lives).  It is continually trying to mediate the demands of the id and prohibitions of the superego.

The superego is a part of the unconscious that is the voice of conscience (doing what is right) and the source of self-criticism.  It reflects society's moral values to some degree, and a person is sometimes aware of their own morality and ethics, but the superego contains a vast number of codes, or prohibitions, that are issued mostly unconsciously in the form of commands or "don't" statements.  The superego is also somewhat tricky, in that it will try to portray what it wants the person to do in grandiose, glowing terms, what Freud called the ego-ideal, which arises out of the person's first great love attachment (usually a parent).  The assumption is that children raised by parents experience love conditionally (when they do something right), and the child internalizes these experiences as a series of real or imagined judgmental statements.

Guilt is a very common problem because of all the urges and drives coming from the id and all the prohibitions and codes in the superego.  There are a variety of ways an individual handles guilt, and these are called defense mechanisms (see table for complete list).


Desires of the id are diverted to healthy outlets approved by the superego


Desires of the id are stuffed back into subconscious and the person denies they exist or engages in Freudian slips


Desires of the id are followed impulsively to escape from hearing the superego (reality)


Anxiety about following desires of the id goes unacknowledged or treated unemotionally


Prohibitions of the superego are applied as standard for judging others and not oneself


Prohibitions of the superego are so strong that the person develops fears/phobias


Superego is so strong that the person continually makes amends or apologies for what they do

Reaction formation

Both id and superego are so strong that person does the opposite of both, sometimes identifying with aggressors


Both id and superego are so strong and ego is so weak that person settles for second best or any available substitute (something better than nothing)

Of the defense mechanisms, psychoanalysts have put forward displacement as their number one choice for explaining crime.  A few criminologists have explored the others, most notably, reaction formation, but the list remains largely unexhausted because, essentially, the ideas are untestable.

Freud also provided a theory of human development.  These ideas revolved around his terms for the three stages of early childhood.  The oral stage (age 0-2) is when a person develops their sense of satisfaction or satiation, of figuring out how much is enough (e.g. of food and drink) which has implications for much of an eater, drinker, or smoker a person is in adulthood.  The anal stage (age 2-3) is when a person develops their sense of orderliness, of figuring out their preferred levels of cleanliness (e.g. as in toilet training).  The genital stage (age 3-4) is when a person develops their sexuality, of figuring out their sexual preferences and whether they are capable of real love or not.  Male children go thru what is called an Oedipus Complex (with a comparable Electra Complex in females).  Both complexes involve sexual feelings for the parent of the opposite sex (and are called "inverted" complexes in cases of homosexuality), and how these feelings are resolved determines how capable the person is of real love later in life.

The primary technique of psychoanalysis is transference, the process of the patient, through free association and "talk therapy", re-enacting or reliving their early childhood experiences with the therapist.  The therapist, in short, becomes a love object, for the patient, but without the therapist going too far or becoming too involved, which is called counter-transference.  Without therapy, a person will engage in amateur transference with other people, a process whereby they play out a "script" or replay the same failed relationships over and over again.

Freud never really had much to say about crime, other than it was most likely motivated by guilt, committed by people with overdeveloped superegos, and characterized by unconscious errors (Freudian slips) which appeared to represent a desire to get caught and be punished.  The inconsistencies in this (is it Repression, Fixation, or what) is why it’s often said that there is no purely Freudian theory of crime (other than the idea the criminals want to get caught).  It was up to the followers of Freud who revised his theories (the Neo-Freudians) to shed light on the psychoanalytic explanation of crime.

One of the first neo-Freudians to do so was August Aichorn, author of Wayward Youth, who took the position that it was not overdeveloped superegos but an underdeveloped superego that primarily caused crime.  He believed that some criminals, raised as children without loving parents or parents at all, developed unregulated ids.  Others were overindulged at the oral stage and required different treatments.  In any event, Aichorn's ideas popularized the notion that delinquents needed unconditional love rather than a punitive, institutionalized setting.  The ideas of maternal deprivation or love deprivation as a cause of crime are still popular.

Redl & Wineman were another group of neo-Freudians who studied Children Who Hate, and took on the Freudian notion of Oedipus Complex.  According to orthodox Freudian theory, criminals should hate their fathers more than their mothers, but Redl & Wineman found that criminals hate both their parents, both father and mother.  In fact, they hadn't gone thru a genital stage at all.  Their egos were therefore undeveloped, and with nothing to mediate between the id and superego, their personalities were nothing but an endless series of raging conflicts, and this is what they called the "delinquent ego".

Healy provided what is perhaps the greatest contribution to Neo-Freudian theory.  He clarified that Displacement (compensating, substituting, ditching) was the most common defense mechanism used by delinquents.  He discovered this by using the "life history" method at his psychiatric clinic in Chicago where he pioneered the process of a nurse taking vitals, a physician examination, a social worker taking a social history, a clinical psychologist testing, and a psychiatrist treating.  He estimated that 91% of delinquents were emotionally disturbed, 50% because of a broken home, and the rest because of too much or too little parental discipline.

Alexander and Healy (1935) suggested that children need to progress from the pleasure principal (being id dominated and therefore needing instant gratification) to the reality principle (where the ego is dominant).  Criminals are those children who do not make this transition.  According to Freud the child needs a stable home environment in order to successfully make this transition.  Research has supported the fact that most criminals come from unstable homes.

The reality principle was central in the explanation of crime offered by F. Alexander and Staub (1931) and F. Alexander and Healy (1935). The criminal from this perspective is someone who is unable to postpone immediate gratification in order to achieve greater long-term gains. In other words, the criminal is one who has failed to progress from the pleasure principle to the reality principle. The antisocial, criminal behaviour in adulthood is seen as a display of characteristics formed during childhood. Healy and Bronner (1936) applied another psychoanalytic concept, sublimation, to the explanation of crime. Sublimation is the process by which instinctual impulses are channelled into other thoughts, emotions, and behaviours. Thus the criminal act, it is argued, results from inner unsatisfied desires and dissatisfactions; these unsatisfied wishes, in turn, stem from a failure to experience strong emotional ties with another person, usually a parent. Thus the delinquency is an 'acting out' or sublimation of inner processes. Healy and Bronner provided evidence to support their theory with two groups of children from a child guidance clinic. Compared to a non-delinquent group, the children who had committed offences had less stable families and also showed greater signs of emotional disturbance. Other psychoanalytic explanations for crime have focused on the inability to control impulsive, pleasure-seeking drives (Abramson 1944); unconscious parental permissiveness which gives approval to delinquent behaviour, hence leading to a poorly developed superego and thereby a lack of control over antisocial impulses (Johnson and Szurek 1952); and acting out of feelings of oppression and helplessness (Halleck 1971). Redl and Toch (1979) and R.J. Marshall (1983) provide reviews of the psychoanalytic perspective on crime.

John Bowlby (1946) (see Hodges and Tizard) studied 44 juvenile delinquents and compared them with non-criminal disturbed juveniles.  39% of the delinquents had experienced complete separation from their mothers for six-months or more during the first five years of their lives compared with 5% of the control group.  Problems with Bowlby's research:  unrepresentative samples, poor matching for control group, low reliability in the interviews with participants (Feldmann 1977).  Also whether or not the effects are reversible are contested (see Hodges and Tizard).  Koluchova (1976) studied a pair of Czech twins and reported that although they were severely neglected for the first 7 years of their lives they were deemed normal by the time they were 14 after being cared for.  Clarke and Clarke (1976) studied children from deprived backgrounds using a longitudinal study and found there were many factors that contributed to the child becoming a criminal, not just whether or not they were maternally deprived.

Psychoanalytic explanations are difficult to test.  However, the effects of emotional or sexual abuse can well be believed when we find that 'serial killers' such as Frederick and Rosemary West suffered terribly as children (Wansell 1996).  Out of 36 sex murderers interviewed in the USA 42% were found to have been sexually abused as youngsters (Ressler et al 1988).  Dietz and Warren (1995) found that 76% of the 41 serial rapists that they interviewed were abused when young.  Having said this though only about 10% of abused children go on to commit crimes.

Psychoanalytic theory would predict that 'over-controlled' individuals could save up all their emotional turmoil and then explode into a torrent of violence!  However, one of the essential assumptions of psychoanalytic theory is that females will have a less developed superego/conscience and will therefore commit more crime and this is simply not borne out by criminal statistics.

Other theories


Plomin (1990, 1994)


Plomin (1990,1994) has convincingly argued that genetics alone cannot explain adult behaviour and focuses much more on the way genes interact with the environment. He further suggests that the environment cannot be taken as a constant and that the ‘non-shared environment' is a crucial factor in determining adult outcomes for children in the same family i.e. what most affects children are the things they do not share with their siblings, their own unique individual experiences. Parents tend to think they raise their children in the same way but families alter over time, sometimes quite significantly, e.g. chronic illness, unemployment, high mobility, divorce, etc., and children’s experiences vary too, e.g. having an influential teacher, developing a new interest, losing a friend, etc. Plomin suggests that it is these differences and the way they interact with genetic make-up, which produce personality differences.

Jefferson (1996)

Jefferson (1996) analyses the life story of Mike Tyson, who came from a traditionally deprived background, became phenomenally successful as a boxer, was convicted as a rapist, served a sentence of imprisonment and managed to return to boxing success only to ruin his reputation by an horrific assault on his opponent in the ring, during which he almost severed the man's ear. Jefferson suggests that Tyson's need to adopt a tough guy discourse to match his appearance in order to succeed involved learning to enjoy being ‘bad' and that the subsequent psychological confusion led to inevitable problems which were impossible to contain. By extension, he argues that if boys learn how ‘good' it can be to be bad they not only reduce their fears of inadequacy but also gain themselves status; thus male involvement in crime begins to become more understandable.


Are violent criminals more angry and paranoid than non-violent criminals?   Katz, R.C. & Marquette, J. (1999) Psychosocial characteristics of young violent offenders: a comparative study. Criminal Behaviour and Mental Health, 6, 339—348 Aim: To compare a group of violent offenders (murderers), non-violent offenders and non-­offenders on measures of anger, hostility, paranoid ideation and global psychopathology. The hypothesis was that those guilty of violent crimes would, compared to the other two groups, have more anger, rage or paranoid symptoms.   This hypothesis was based on several theories of crime. The psychodynamic approach sees homicidal behaviour as stemming from unconscious conflict, an ego weakness, displaced anger or repressive coping mechanisms. From a biological interactionist viewpoint, Lewis & Pincus (1989) suggest that many extremely aggressive young men have suffered from abuse and family violence resulting in damage to their central nervous system. Psychosocial variables such as poverty, Family turmoil, gang involvement, excessive alcohol intake, physical abuse, educational difficulties and exposure to violent role models either within the Family or through the media have also been found to be related with adolescent aggression.   Method: Design:  The method used was psychometric testing. This was an independent groups design comparing three groups of young men:      convicted murderers        non-violent offenders   non-offending high school juniors and seniors.   Participants:         There was a total of 77 participants, all males, from Stockton in California, ranging in age From 16 to 23 with a mean age of 19. All were volunteers who were told that the purpose of the study was to assess various attitudes and beliefs in young people; they all received $5 for participating. There were 29 convicted murderers; 15 non-violent offenders imprisoned for repeated robbery and similar acquisitive crimes or for violation of probation; 33 junior and senior school students from a state school, all studying psychology.   All three groups were ethnically diverse and consisted of roughly equal numbers of Caucasians, Blacks, Hispanics and Asians. The two offender groups had a somewhat larger number of Blacks and Hispanics.   The psychometric tests: Three tests were used:               MMPI-A: a revision of the MMPI for use with adolescents. This is a personality test, which measures adolescent anger, family problems, alienation and conduct problems.        SCL-90-P: used in this study to measure hostility, paranoid ideation, interpersonal sensitivity and level of distress.   State—Trait Anger Scale: a self-report measure that assesses anger in two ways: as a relatively stable personality trait (trait anger) and as a current emotional state (state anger) reflected in feelings of tension, annoyance, irritation or rage.   The participants were also asked to complete a questionnaire asking the number of times they had been arrested, the number of times they had been arrested for a violent crime, whether they had ever been a gang member and whether anyone in their Family had ever been imprisoned. In order to encourage honest responses, they were asked not to put their names on the tests.   Results: The prediction that young murderers would show increased levels of anger, hostility, paranoid ideation, feelings of alienation and global psychopathology was not supported. There was no significant difference here between any of the three groups.   However, there were some differences between the groups. Compared with the school students, a significantly higher proportion of the offenders had grown up in families where at least one other family member had spent time in jail or prison.   Murderers were more likely to be gang members and more inclined to express feelings of self-consciousness, inferiority and inadequacy.   Compared with school students, young offenders expressed less global distress and were less troubled by feelings of anger and hostility; they were more likely to have been gang members or to have come from a family where antisocial behaviour was a problem.   Conclusion: There was no evidence for differences in anger, hostility and psychopathology between violent offenders, non-violent offenders and non-offenders. A history of gang membership was more likely to distinguish between the three groups. This supports a report From the US Department of justice (Blumstein, 1995; Snyder & Sickman, 1995) that sociocul­tural variables are the driving force behind the considerable and steady increase in the amount of juvenile crime and violence in the US.


Three studies in detail with evaluation


Explanations of Criminal Behaviour

Study One

Raine A. and Lui (1998) Biological Predispositions to Violence and their

implications for Biosocial Treatment and Prevention. Psychology, Crime and Law,

Vol. 4, pp107-125

Aim: This review article describes four recent studies conducted by the authors that

identify possible biological risk factors for violence and crime. One of these studies is

the ‘murderers’ brains’ study that you may have looked at in the OCR AS course and

two of the other three studies are summarised below.

Study 1: Are low levels of physiological arousal a predictor of offending


Sample 101 15 year old boys

Method: This is a correlational study which looked for a relationship between a

number of physiological measures (skin conductance, EEG and heart rate) taken at

age 15 and the numbers of offences that they had committed by the age of 24.

Results: The authors report a strong correlation between the two measures. Those

committing crimes had significantly lower heart rates, reduced skin conductance and

more slow wave EEG theta activity than non-criminals. The authors claim that these

measures correctly classified 74.7% of all participants as criminal or non criminal.

Study 2: Are birth complications combined with early maternal rejection a

predictor of offending behaviour?

Sample: 4,269 consecutive live male births in Copenhagen, Denmark

Method: the researchers collected data on the following:

birth complications (including breech, forceps and anoxia)

maternal rejection (mother not wanting pregnancy, attempted abortion or

institutionalisation by 4 months)

Violent Crime (data collected when the sample were aged 18). These crimes included

murder, attempted murder, assault, rape, armed robbery, illegal possession of a

firearm and threats of violence

Results: Those boys who had experienced both birth complications and early

childhood rejection were most likely to become violent offenders, although there was

no effect for non-violent offending. There was also no interaction between poverty

and birth complications. This is thought to be the first study to provide evidence from

a large birth cohort to show that birth complications in association with rejection are

associated with violent crime. The authors conclude by stating that although they have

demonstrated a link between physiological functioning and criminal behaviour, it is

the interaction between biological and social predisposition that is crucial to

understanding violent crime.

Study Two

Eysenck, S.B.G, Rust, J. and Eysenck, H.J. (1977) Personality and the

Classification of Adult Offenders. British Journal of Criminology. Vol. 17 No 2.

Aim: an attempt to classify criminal behaviour in relation to personality variables.

The personality variables measured in this study are Extraversion, Neuroticism and

Psychoticism. The authors argue that these variables show strong evidence of genetic

determination as well as being linked with criminal behaviour.

Sample: 156 prisoners aged 18-38 divided into 5 groups on the basis of their crimes:

1. Violent crimes: Prisoners who had committed two or more violent crimes (but no

sexual crimes)

2. Property crimes: Prisoners with three or more convictions for breaking and

entering and other convictions for theft only.

3. Confidence Crimes (fraud). Prisoners with three or more convictions for fraud.

No convictions for violent or sexual crimes.

4. Inadequates: Prisoners with ten or more convictions in three years and serving an

average sentence of less than 18 months. No convictions for robbery and not more

than one conviction for a violent or sexual offence.

5. Residual: Prisoners who did not fall into any of the above categories, i.e. who

committed a variety of crimes in combination.

They were all tested on the Eysenck Personality Questionnaire (EPQ) and also on a

variety of physiological measures including skin conductance, EEG and eye blink


Results: The authors claim that it is possible to classify offenders by personality

types. The first major distinction was seen with the Psychoticism scores, with the

conmen having very low scores. Neuroticism scores separate the violent and property

offenders (low scores) from the inadequates and the residuals (high scores). Finally

the Extraversion scores distinguish between the violent and residual offenders (high

scores) and the inadequates and property offenders (low scores) although this

difference did not reach statistical significance.

Some physiological differences were also found between the property and inadequate

offenders and the other three groups.

Study Three

Farrington, D.P., Barnes, G.C. and Lambert, S. (1996) The concentration of

offending in families. Legal and Criminological Psychology, 1, 47-63

Aim: To test the hypothesis that problem families produce problem children. The

research here is one part of a much larger longitudinal study, namely the Cambridge

Study in Delinquent Development which has followed one group of males from the

age of eight to thirty-two in interviews and from the age of ten to forty via their

criminal records. This paper attempts to relate convictions of these males to the

convictions of their biological parents and full biological siblings.

Sample: 411 boys from inner city areas of London, mostly born in 1953. Participants

were selected by taking all 8 and 9 year old boys from the registers of 6 state primary

schools in one location in London. The boys were predominantly white and from

working class families. As there were 14 pairs of brothers in the original sample

(including five pairs of twins) there are 397 different families involved in the

research. To avoid counting the same family more than once, one of each pair (the

younger brother or a randomly selected twin) was excluded from this analysis.

Method: The Cambridge Study involved interviews with the children, their parents

and questionnaires completed by their teachers. These results have been reported in

numerous papers and books. In this study the focus is on criminal records. Searches

were carried out in the central Criminal Record Office in London to locate evidence

of convictions of the males, their biological fathers and mothers and their full brothers

and sisters (and also their wives / partners but these findings are not discussed here).

The Criminal Record Office contains records of all relatively serious offences

committed in Great Britain or Ireland and also holds records of minor juvenile

offences committed in London. The records would not include details of common

assault, traffic violations and drunkenness. Most commonly the offences were theft,

burglary and unauthorised taking of motor vehicles, although they also included

violence, vandalism, fraud and drug abuse.

Results: Earlier reports from the Cambridge Study have confirmed the hypothesis that

criminals are likely to have criminal relatives. When the original cohort were aged 20,

48% of those with convicted fathers also had convictions compared to 19% of those

without convicted fathers. This was not dependent on when the father had committed

the offence (before or after the son was born) suggesting that there was no direct

behavioural influence. 54% of those with convicted mothers also had convictions

compared to 23% of those with unconvicted mothers. This link remained even when

males with both mother and father with convictions were removed from the analysis.

These results are confirmed in this analysis which took place when the study males

were aged 40. 64% of the families contained one convicted person or more, just 6%

(23 families) of the families accounted for over half of all the convictions.

Convictions of one family member is strongly linked to conviction of another family

member. About 75% of convicted parents had a convicted child and having a sibling

who had been convicted (especially an older sibling) was a strong predictor of

conviction. These results are not the result of co-offending with family members.

There were 26 study males who had co-offended with father, mother or sibling and

when this data is excluded from the analysis the effect remains strong.

Overall the results provide strong support for the notion that offending is concentrated

in families and tends to be transmitted from one generation to the next. However, as

the authors point out, these results do not establish whether this is due to the influence

of nature or nurture.


1. The Nature-nurture debate.

Considering the relative influence of nature and nurture in the determination of

criminal behaviour is obviously crucial. We have looked at some research that

considers the importance of genetic factors (Raine & Lui, Eysenck) and some that

considers the importance of environment and upbringing (Farrington). Farrington’s

research clearly suggests that environment plays a major part in criminal behaviour

and results from the Cambridge Study as a whole identify social / environmental

factors such as poverty, low family income, large family size and poor child-rearing

techniques as major influences. In the study reported here, it is clear that offending is

concentrated in families and passes from generation to generation. However, this

could be due to the kind of social / environmental factors mentioned above or it could

be due to genetic factors. The research conducted does not allow us to discriminate

between these two explanations. Raine and Lui offer strong support for role of

genetic factors in determining criminal behaviour, although they stress that it is the

interaction between these factors and social environment that is crucial. Interestingly,

they suggest that poverty does not contribute significantly to criminal behaviours. It

is unlikely that research will ever resolve this debate. Firstly it is likely that criminal

behaviour (as all other behaviours) is the result of a complex interaction between both

nature and nurture and the vast numbers of different crimes make it unlikely that one

explanation fits all crimes. Secondly it is virtually impossible to think of a piece of

research that could be conducted that would answer this question. The current

viewpoint would appear to be that psychology has identified a number of genetic

factors (physiological, neuro-physiological, personality etc) which appear to

contribute to the likelihood of someone committing a criminal offence, but that these

factors are mediated by a vast array of social and environmental factors.

2. Correlation not causation.

This is a very important evaluation issue to consider in many areas of criminal

psychology. Most of the research in this area looks for correlations between measures

of different aspects of behaviour (for example, between personality and offending

behaviour, or between physiological measures and offending behaviour). Such

research can only ever show a relationship between two variables and do not show

cause and effect. The relationship may be spurious or there may be a third factor

involved that has not been considered. For example, although Raine and Lui have

shown a correlation between physiological measures and offending, this does not

necessarily mean that physiological factors determine offending. There may be some

environmental factor which has affected the physiological functioning of the

individual which also contributes to an increased likelihood of offending. In a similar

way, Farrington et al have shown a relationship between parental offending and

offspring offending but they have not been able to explain the causal relationship.

The only way to determine a cause and effect relationship between two variables is to

conduct an experiment where you control one variable and measure its effect on

another variable and for many practical and ethical reasons it is not possible to

conduct this kind of research here.

3. Types of measurement:

The three studies reported here have all collected their data in very different ways.

Raine and Lui have used a variety of physiological measures which are reliable

measures and unlikely to be strongly affected by demand characteristics, although it is

possible that having heart rate monitored and so on would make people anxious. They

also collected medical details about births which are likely to be highly accurate. They

correlated this data with official records of crimes which are less valid as a measure of

offending behaviour. The data is actually a record of convictions, not criminal activity

and it possible that people may have committed crimes that they have not been

convicted of or been convicted of a crime that they did not commit. This last comment

would also apply to the research conducted by Farrington et al as their data too came

from the Criminal Record Office and may be inaccurate. However as part of the full

Cambridge Study, the researchers did ask participants for their self reports of criminal

activity and they state that these were remarkably similar to the official statistics.

Finally, Eysenck uses the Eysenck Personality Inventory, a psychometric test that has

been developed over a number of years. This test has been well validated and appears

to be reliable. However there are problems with demand characteristics and social

desirability bias in any psychometric test (although the EPI does contain a lie scale

designed to pick up contradictory responses which might indicate that someone is



4. Usefulness / Implications.

Clearly, determining the causes of criminal behaviour is an extremely useful thing to

do. In simple terms this could lead to the development of prevention programmes for

children deemed to be at risk or to treatment programmes for offenders. You will see

in the final section of this chapter that the reason many treatment programmes are

unsuccessful is that they do not address the causes of the problem. However, there are

several problems with this. Firstly, it is highly unlikely that there are single factors

which can be identified as the causes of offending behaviours. It is far more likely that

offending behaviour is the result of complex interactions between factors. This might

make some of the research in this area reductionist and over simplistic. Secondly,

there are ethical implications to consider. The type of research conducted by Raine

could be applied to the development of ‘screening’ programmes. It would be very

easy for children to be labelled as ‘potential criminals’ just because they have the type

of physiological functioning identified by Raine and Lui, or the personality type of a

conman as identified by Eysenck, or for children born to convicted parents to be

treated very differently by teachers and social workers. This could lead to the selffulfilling

prophecy where children develop the behaviours attributed to them by

others. Raine suggests more worrying possibilities. He suggests that the technology

exists to change the brain functioning of individuals. This raises serious issues of

social control and it is important that the results of research like that of Raine and Lui

is treated with caution.



Crime linked to absent fathers - Guardian 5-4-01



Kevin Brewer (2000), Psychology and Crime, Heinemann, ISBN 0-435-80653-X.  Read pages 12-25.

Diana Dwyer (2001), Angles on Criminal Psychology, nelson thornes, ISBN 0-7487-5977-8.  Read pages 15-41.

Julie Harrower (1998), Applying Psychology to Crime, Hodder & Stoughton, ISBN 0-340-70556-6.

Clive R. Hollin  (1989), Psychology and Crime, Routledge, ISBN 0-415-01807-2.

Fiona Lintern, Merv Stapleton & Lynne Williams (2004), Study Guide for OCR Psychology: A2 Level, Hodder & Stoughton, ISBN 0-340-81626-0

John J. Macionsis and Ken Plummer (1997), Sociology a global introduction, Prentice Hall, ISBN 0-13-664533-X