Like grandfather, like father, like son?

Crime and the family: like grandfather, like father, like son?

Gregory, Nathan


Thephenomenon of criminal behaviour has interested scholars from academic disciplines building on centuries of philosophical debate developed from emerging biological, psychological and social sciences. Criminological theories of causation are embedded in conceptual networks that link political ideology classified by philosophical underpinnings. For example, the late 18th and early 19th century utilitarianism consensus classical theory supported the notion of free will, suggesting that individuals were likely to commit crime if the pay-off were greater than the retribution. In contrast, conflict marxist and radical criminology regarded crime as a function of poverty, reflecting a power imbalance in society. The late 19th century saw the emergence of the positivist school, which argued that factors including genetics, poverty, personality and the family were important in determining criminality. This article concentrates on the transmission of crime through families by reviewing a selection of the main studies in the area.

Introduction to the literature

Twin and adoption studies

Several studies have reviewed the literature to examine evidence for a genetic constitution for the development of criminality. Blackburn (1995), Hollin (1993) and Raine (1993) between them cite seven biological studies of monozygotic and dizygotic twins conducted during the 1930s and 1940s, and highlight mean concordance rates of 75% for monozygotic and 24% for dizygotic twins when examined for criminal behaviour. Three later studies which avoided concordance rate errors due to distinguishing twins on appearance, and adopted fingerprinting and blood types to determine zygocity, found reduced concordance rates of 48% and 20% respectively.

Further prospective longitudinal adoption studies report that when both biological and adoptive parents are criminals, 24% of adopted children display criminal behaviour, compared with 13% whose biological/adoptive parents are non-criminal (Mednick et al, 1984). Further American and Swedish studies agree that there is some genetic disposition to crime, indicating cross-cultural generalisability of these findings (Cadoret et α?, 1987; Sigvardsson et al, 1982). Although genetic studies appear to provide persuasive evidence that criminal behaviour runs in families, methodological weaknesses, including small and rare samples, mean that results have to be treated cautiously.

Family studies

Other authors have investigated families to explore the development of criminality. Three American studies report intergenerational transmission of crime. Glueck and Glueck (1950), in a cross-sectional study, compared 500 relatively well-behaved public school boys with 500 institutionalised delinquents, and discovered that 66% of delinquents had a criminal father, compared with 32% in the control group. Similarly, Robins et al (1975) reported that 57% of delinquent sons had a criminal father, while McCord (1977) in the CambridgeSomerville study reported that 29% of sons with convicted violent fathers were convicted for violence, 40% with convicted thieving fathers were convicted for theft and 50% with convicted drunken fathers were convicted for drunkenness.

In the United Kingdom, both Ferguson (1952) and Wilson (1987) have suggested that a convicted family member was influential in the development of offspring criminality, rather than structural factors such as poor housing or poverty. The Cambridge Study on Delinquent Development (Farrington & West, 1990) reported that having a convicted father was the best predictor of offspring offending and antisocial behaviour. This prospective longitudinal survey traced 411 males reared in 397 working class families living in South London in 1961/2 from the age of eight. The results indicated that 64% of families contained at least one convicted person, six per cent accounted for 48% of all offences, and across both generations 56% of convicted mothers mated with convicted fathers, 75% of these couples producing a convicted child (Farrington et al, 1996).

However, although they produce interesting information, family studies require careful attention. Research findings are limited, in that a correlation between two variables (family criminality and offspring criminality) does not prove a causal relationship; a third variable such as environmental factors may be influential. This was particularly the case in the Cambridge Study, which was unable to disentangle the heredity or environmental factors influencing family criminality.

Theories of criminal behaviour within families

There are several competing theories that focus on the management styles of criminal families, influenced initially by the psychoanalytical concept of attachment theory (Bowlby, 1944). A study of 44 delinquent juveniles matched to a control group reported that 39% of delinquents were separated from their mother before the age of five compared to five per cent of the control group. Although criticised for an unrepresentative sample and unreliable assessment methodology, Bowlby concluded that maternal deprivation through rejection was linked to childhood antisocial behaviour and later delinquency. Similarly, McCord (1983) argued that parental rejection was the most powerful instigator of criminality, finding that 56% of rejected boys were involved in serious crime compared with 23% of loved children.

In contrast, social learning theory maintains that behaviour is learnt through observation and modelling of others, primarily within the family (Bandura, 1977). For example, Patterson and colleagues (1984), in an American study of 206 schoolboys, discovered that families teach children delinquent behaviour through a coercive process involving punishment, humiliation, aggression or non-enforced threats. Deviant families were characterised by enmeshed or lax management styles, resulting in poor child supervision and leading to high rates of child stealers and police contact (Patterson et al, 1989).

The size of families appears to correlate with delinquency. Hirschi (1969), in a control theory, reasoned that larger families were harder to discipline and likely to receive less parental attention. For example, Ferguson (1952) reported that 44% of delinquents had three or more convicted family members, while the Cambridge Study discovered that delinquents were likely to come from families with four or more children (Farrington et al, 1996). Further demographic variables, including socioeconomic status, also appear to be indicative of antisocial behaviour, although the empirical findings are inconsistent. Low family income was a relatively independent predictor of later delinquency in the Cambridge Study, suggesting a strain theory that lower class males were unable to achieve middle class status, leading to boredom, frustration and delinquency (Farrington & West, 1990).

Child abuse and neglect

Other authors have examined extensively parental child-rearing practices and the development of delinquency. Haapsalo and Pokela (1998) reviewed six major American longitudinal studies of children and examined the relationship between offspring criminality and exposure to negative child-rearing practices. All six studies reported parental punitiveness as crucial to the development of offspring antisocial behaviour, while four considered lack of parental love, and three laxness, deviant parents and family disruption as important factors. The research indicated that styles of parenting including corporal punishment, power assertion, rejection, physical abuse and neglect were prime predictors of future criminal and antisocial offspring behaviour. Other important factors were poor parental supervision and parent-child interaction, parental criminality and socioeconomic difficulties.

Similarly, Widom (1989a) reviewed thoroughly the relationship between child abuse, neglect and delinquency, noting that tentative conclusions could be drawn from the three prospective and nine retrospective studies examined. For example, prospective data on child abuse and neglect revealed offspring delinquency rates of 10%-17% compared with 8%-26% where research was conducted retrospectively. Furthermore, serious methodological weaknesses were noted in these studies. For example, retrospective bias may lead to inaccuracies, as respondents might forget their behaviour, while conservative bias may underestimate the actual occurrence of abuse. Lack of control groups and failure to consider statistical base rates may confound reports of child abuse, leading to difficulties in assessing the independent effects of abuse and neglect. Nevertheless, child abuse and neglect lead to an increased risk of later violent crime and criminal behaviour, although not all of those abused become criminals.

Cycle of violence

The ‘cycle of violence’ or Victim to offender’ hypothesis has also received attention. Widom (1989b; 1989c) reported that 15% of child physical abuse victims had been arrested for violent crimes compared with 12% of neglected children and eight per cent in a group that had not been maltreated. Lewis et Λ/(1985), in a small-scale American prospective study of nine murderers, reported that 78% had experienced severe parental abuse, while in the United Kingdom, Falshaw and Browne (1997) discovered that 53% of young inmates at the Glenthorne secure unit had been physically abused, 32% neglected, 28% severely emotionally abused and 14% sexually abused, and that 49% reported family violence.

In relation to familial domestic violence, a study of street children in the United Kingdom revealed that 78% of runaway adolescents had been physically or sexually abused and 75% had escaped family violence, 50% resorting to crime for finance (Browne & Falshaw, 1998). Studies considering the impact on children of witnessing parental violence indicate that 53% of violent offenders had observed their parents engage in violence (Bach-y-Rita & Veno, 1974), while Lewis et al (1985) reported that 79% of more violent incarcerated boys had witnessed domestic violence. However, although these studies suggest an unequivocal progression from victim to offender, it is important to remember that the majority of abused children do not become delinquent, and abuse alone does not predict a career of adult crime (Falshaw et al, 1996).


If authors generally agree that there is an intergenerational transmission of crime concentrated in a small minority of families, then identification of effective interventions appears crucial in preventing offspring delinquency. Studies appear to suggest that over time antisocial behaviour is a relatively stable characteristic, and therefore identifiable childhood antecedents for crime have the potential for modification.

However, Browne and Herbert (1997) report that, although at primary level media publicity has enabled greater awareness of the implications of childhood abuse and neglect, the growth of child telephone helplines appears to have limited efficacy. Similarly, Blackburn (1995) reports mixed conclusions on secondary level intervention strategies for delinquent school children that are based on attachment theory and parental skills training. Authors further report that tertiary level interventions have a poor effect on the intergenerational cycle of offending, particularly as implementation occurs when delinquents are already firmly entrenched in their behaviour patterns. For example, Lewis (1985), in a nine-year follow-up of violent boys, stated that those displaying the most violent offences had on average nine institutional placements, while Falshaw and Browne (1995) report that conduct-disordered inmates had experienced five institutional and three foster home placements.

Other researchers strongly advocate that to break the Victim to offender’ cycle, prevention at the level of family relationships is crucial, by screening potential high-risk families (where parental chemical dependence or mental illness, for example, is present) by trained health visitors and social workers. Subsequently, Browne and Herbert (1997) argued that social skills and anger management training coupled with positive childrearing strategies could be implemented to support families. This type of intervention initiative, including Head Start in American and its British equivalent Sure Start, provide educational stimulation targeted at high-risk families at pre-school age. They are now beginning to report evidence of improved social competence and reduced aggression in children.


Although arguments about the theoretical significance of the association between family criminality and later offspring delinquency have advanced, the strength of the association appears modest. Factors including parental criminality, child neglect, physical abuse and exposure to domestic violence appear to be good indicators of later offspring delinquency, but it is still unclear whether this is through genetic or social-familial variables. Studies consistently report that individuals exposed to disruption or family discord have avoided delinquency because of other protective factors that reduced the criminal effects of these families. The interaction between protective factors and risk variables seems a more important issue for future studies.

Finally, further research should be concentrated on the provision and efficacy of early intervention strategies targeted at high-risk families to enable the prediction and prevention of future child delinquency and the development of suitable treatment programmes.


Bach-y-Rita G & Veno A (1974) Habitual violence: a profile of 62 men. American Journal of Psychiatry 131 1,015-7.

Bandura A (1977) Social Learning Theory. New Jersey: Prentice-Hall.

Blackburn R (1995) The Psychology of Criminal Conduct. Chichester: John Wiley.

Bowlburn (1944) Forty-four juvenile thieves. International Journal of Psychology 25 1-57.

Browne K & Falshaw L (1998) Street children and crime in the United Kingdom: a case of abuse and neglect. Child Abuse Review 71-13.

Browne K, Falshaw L & Hamilton C (1995) The characteristics of young persons resident at the Glenthorne Centre during the first half of 1995. Youth Treatment Service Journal 11 (2) 52-71.

Browne K & Herbet M (1997) Preventing Family Violence. Chichester: Wiley.

Cadoret R, Troughton E & O’Gorman T (1987) Genetic and environmental factors in alcohol abuse and antisocial personality. Journal of Studies of Alcohol 48 1-8.

Falshaw L & Browne E (1997) Adverse childhood experiences and violent acts of young people in secure accommodation. Journal of Mental Health 6 (5) 443-55.

Falshaw L, Browne D & Hollin C (1996) Victim to offender: a review. Aggression and Violent Behaviour 1 (4) 389-404.

Farrington D, Barnes G & Lambert S (1996) The concentration of offending in families. Legal and Criminological Psychology 1 47-63.

Farrington & West D (1990) The Cambridge Study in Delinquent Development: a long-term follow up of 411 London males. In: H-J Kerner & G Kaiser (Eds) Criminality; Personality, Behaviour and Life History. Berlin: Springer-Verlag.

Ferguson T (1952) The Young Delinquent in His Social Setting. London: Oxford University Press.

Glueck S & Glueck ET (1950) Unravelling Juvenile Delinquency. Cambridge, MA: Harvard University Press.

Haapasalo J & Pokela E (1998) Child rearing and child abuse antecedents of criminality. Aggression and Violent Behaviour 4(1) 107-27.

Hirschi T (1969) Causes of Delinquency. California: Sage.

Hollin CR (1993) Psychology and Crime: An Introduction to Crimmological Psychology. Routledge: London.

Lewis D, Moy E, Jackson L et al (1985) Bio-psychological characteristics of children who later murder. A prospective study. American Journal of Psychiatry 142 1,161-7.

McCord J (1977) A comparative study of two generations of Native Americans. In: RF Meier (Ed) Theory in Criminology. California: Sage.

McCord J (1983) A forty-year perspective on effects of child abuse and neglect. Child and Abuse and Neglect 7 265-70.

Mednick B, Gabrill W & Hutchings B (1984) Genetic influences in criminal convictions. Evidence from an adoption cohort. Science 244 981-4.

Patterson GR, DeBaryshe BD & Ramsey E (1989) A developmental perspective on antisocial behaviour. American Psychologist44 (2) 329-35.

Patterson G, Dishion T & Bank L (1984) Family interaction: a process method of deviancy training. Aggressive Behaviour 10 253-67.

Raine A (1993) The Psychopathology of Crime. Criminal Behaviour is a Clinical Disorder. London: Academic Press.

Robins LN, West PA & Herjanic BL (1975) Arrests and delinquency in two generations: a study of black urban families and their children. Journal of Child Psychology and Psychiatry 16 125-40.

Sigvardsson S, Clonmger R, Bohman M & von Knorring A (1982) Predisposition to petty criminality in Swedish adoptees. III. Sex differences and validation of the male typology. Archives of General Psychiatry 39 1,248-53.

Widom C (1989a) Child abuse, neglect, and violent criminal behaviour. Criminology’27 251-71.

Widom C (1989b) Does violence beget violence? A critical examination of the literature. Psychological Bulletin 106 3-28.

Widom C (1989c) The cycle of violence. Science 244 160-6.

Wilson H (1987) Parental supervision re-examined. British Journal of Criminology 27 275-301.

Nathan Gregory


Address for correspondence

Nathan Gregory, Team Leader/Specialist Practitioner, Gloucester Assertive Community Treatment Team, Gloucestershire Partnership NHS Trust, Burleigh House, Nettleton Road, Gloucester GL1 1PZ.

Copyright Pavilion Publishing (Brighton) Ltd. Dec 2004

Provided by ProQuest Information and Learning Company. All rights Reserved