This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.
From the viewpoint of a diligent bureaucrat trying to find where best to expend the public purse, it may well be said that there are identifiable groups whose offspring are more likely to become alumni of HMP Pentonville than of Peterhouse. Lettice agrees that by using information on risk factors, high risk groups can be identified early. Children who have disadvantaged starts through poverty, single and/or inadequate parenthood or in families with a history of crime tend to do worse in later life (Lettice, 2006). In particular socio-economic factors can be well correlated with future criminality. Research has shown a consistent set of factors predicting the likelihood that individuals will commit crimes (Loeber & Farrington, 1998); fractured households, poor parental monitoring, poor discipline, a family history of criminality, parental conflict, a lack of family cohesion, and low socioeconomic status are general risk variables for criminal behaviour.
However, I propose to take this question more literally than that. The question invites us to consider the individual child placed in front of us and to consider what prospects there are to predict the fate of those individuals. This is a good question, as it is often forgotten that correlations which are strong enough to provide a sound basis for deciding questions at the level of the group may prove poor bases for prediction at the level of the individual.
Let us imagine, then, a latter-day Sherlock Holmes who is blessed with an insight into human nature backed by knowledge of the studies of biological factors that are allegedly linked to criminality but who is not given information beyond what is immediately in front of him about the child's social and economic background. What could he look at to give him a clue about this most intractable of mysteries? Any good detective will start with the facts in front of him and then supplement these observations with the abilities now offered by forensic science.
Any good detective will start with the facts before him and then supplement these with the abilities now offered by forensic science.
Our detective might first therefore turn to a consideration of the child's anatomy in his quest to "spot tomorrow's criminal" from the child's physical appearance.
The equation of goodness and beauty is a hallowed prejudice found throughout literature and even made its way into the law of medieval England, which held that if two people fell under suspicion of a crime, the uglier, or more deformed was to be regarded as more probably guilty (Ellis, 1914). Lombroso's work on body typing of criminals was highly influential in the 19th Century. Lombroso (1876) argued that criminals have more primitive physical characteristics e.g. small brains, heavy jaws, projecting ears and crooked or flat noses. Criminal behaviour could be attributed to biological inferiority and "degeneration" causing the unattractive physical characteristics of criminals. His study showed that criminals were often colour-blind, left-handed and physically weak. These findings were mirrored by Sheldon (1942) and Glueck & Glueck (1959) who developed these ideas to argue that different types of criminal had different physiologies; mesomorphs were more likely to be criminal than ecto- or endomorphs. (One should note in passing here that many of these studies demonstrate a constrained notion of criminality - theft & violence, not embezzlement or insider trading).
In 1939 Earnest A. Hooton conducted a 12-year study comparing 13,873 male prisoners in 10 US states with a haphazard sample of 3,023 men drawn from the general population, and found again that Lombroso's findings were valid; there seems to be a difference in body shape and facial characteristics of those who are convicted of criminal behaviours, suggesting that physical appearance is a good indicator of criminal behaviour. (Hooton, 1939)
More recently, however these findings have been discredited. Lombroso himself had reservations about his theory in later life, and many critics argued that Lombroso's study assumes that correlation equals causation; just because criminals were more likely to have poor physical appearances, does not mean that those with poor physical appearance will necessarily show criminal behaviours (Goring, 1919). It is just as, or perhaps more, likely that a poor environment or diet can explain both factors, as argued by Goring in his classic refutation of Lombroso. However this criticism, would not prevent the correlation (were it well-founded) from being used to predict outcomes. It is common - and acceptable - to use such correlated variables as proxy measures: eligibility for free school meals is a widely-used indicator of social disadvantage.
A meta-analysis of influential body build and criminality studies found that many of these studies show design flaws (Rees, 1973). For example, Hooton's study had an inadequate control group and poor sampling methodology, and it can be argued that Lombroso's study itself was skewed by the large proportion of mentally disturbed in the sample.
Regardless, more recent studies have also suggested that stereotypes of facial appearance of criminals affect their final judgement. It is possible then that attractiveness is a better predictor of a person's propensity to be arrested, charged and convicted, than it is of his inherent propensity to criminality (Bull & McAlpine, 1998). Saladin (1988) showed participants eight photographs of men and asked them how capable they considered them to be of committing two crimes, murder and armed robbery. They found that the most attractive men were seen as less likely to commit either (Saladin, Saper, & Breen, 1988). Results of a study by Downs & Lyons (1991) showed that when the severity of crime was controlled for, attractiveness was negatively correlated with the amount of bail or fine imposed. This area of investigation could suffer a flaw - attractiveness may affect conviction rates, but not propensity to offend. Besides, these studies too may suffer from methodological shortcomings (Downs & Lyons, 1991). A review of this area of research by Bull and McAlpine suggests that studies that do not find links go unreported, creating a biased image of the influence of attractiveness.
It may be suspected however that our latter-day Sherlock Holmes will dismiss this particular avenue of clues with a more commonsensical observation: that it is almost impossible to predict the attractiveness of the adult from his or her attractiveness as a toddler. Attractiveness, or lack of, may be affected by life events, which may also have an effect on an individual's inclination to commit crime; for example, unattractiveness could be associated with social disadvantage such as lack of medical and dental care.
Even a flatfooted Dr Watson will notice one aspect of the child placed before him: gender and this could be considered a valuable predictive factor in spotting the future perpetrator of at least some types of crime. One does not need to be a feminist to observe that rapists are overwhelmingly men. "Ninety per cent of all murders are committed by people with a Y chromosome - males. " to quote Steve Jones, a geneticist at University College London (Feresin, 2009) and there is a weight of evidence to suggest that most criminal offences are committed by men. According to Home Office figures, in 2002 male offenders in England and Wales outnumbered female offenders by more than four to one; between 85 and 95 percent of offenders found guilty of burglary, robbery, drug offences, criminal damage or violence against the person are male (Home Office, 2002/2003). Men outnumber women in all major crime categories. This difference still holds when the criminal careers of males and females in the same families are compared. Farrington & Painter (2004) found that the prevalence of convictions for criminal offences was much higher for brothers (at 44%) than for sisters (12%).
There is evidence to suggest that there are innate biological differences between boys and girls, which may affect a person's capacity for crime. Maccoby and Jacklin (1974) argued that sex-typical behaviours, and specifically aggression, arises from the biological differences between men and women; both hormonal and in brain structure. Males in all human societies for which information is available are more aggressive (and thus arguably more likely to commit violence). This is found from early life, and is also replicated in subhuman primates. 1964 studies by Young, Goy and Phoenix show that administration of hormones can change sex-typical behaviours. XX and XY chromosomes influence gonadal determination and thus the hormones produced in the human body, most importantly, the male hormone testosterone. Goy (1978) showed that manipulating testosterone in early life causes changes in sexual behaviours as well as other behaviours e.g. aggression. Individuals with more testosterone and hormones produced from it (i.e. men) engaged in more rough play and hostility than those without.
Children are often encouraged to play along gender lines, which encourage more violent and criminal-type behaviours in males than females. Boys are given more positive encouragement to play rough and be aggressive and to choose masculine toys, whereas girls are encouraged to communicate with their mothers, and interact in play based around caring roles (Fagot & Hagan, 1991). Parents induce self reliance in boys and social conformity in girls (Barry, Bacon, & Child, 1957). This can become a self-fulfilling prophesy as children tend to conform to gender stereotypes - crime is often seen as a masculine-type behaviour, leading to more men adopting those behaviours. Studies have found that by age 24 months, toddlers showed some knowledge of gender-typical activities; when presented with photographs of men and women performing gender-typical or gender-atypical activities, they attended longer to the gender-atypical photos, suggesting stereotype awareness (Witt, 1997). Furthermore, boys are less strictly supervised than girls, both in choice of friends and returning home and night, giving them more opportunities to become engaged in criminal behaviours (Farrington & Morris, 1983).
Steffenmeister (1980) argues that the difference in offending rates between the sexes can be attributed to women's roles in society. Firstly, women have more to lose, due to their natural role in the family; criminal activities are too risky for women. A study by Bartel (1979) showed a 50% rise in female crime in the 1960's correlated directly with a fall in the number of preschool children. Furthermore, women often have less time to engage in criminal activities. Research by Deutsche Bank found that women in full-time paid employment still spent nearly twice as much time on housework on an average work day than their male counterparts (Schaffnit-Chatterjee, 2009). When women live with a partner, their household work load increases significantly whereas men's decreases. The presence of children widens the gap further. German mothers in full-time employment spend (on average per week day) 1.5 hours more on paid and unpaid work combined than fathers in full-time employment. Although female labour force participation is increasing substantially, family work at large continues to be allocated disproportionately to women within a couple (Schaffnit-Chatterjee, 2009). Women are also less likely to have access to criminal subcultures and underworlds - many crimes/criminals place a premium on physical strength and toughness, therefore women will often not meet the conservative norms of criminal groups; furthermore, women are less likely to drink heavily, gamble or approve of violence, decreasing their entry to "buddy" networks of drinking, like after-hours drinking and gambling groups. Data from the General Household Survey 2006 (2008) show that in 2006, 40% of men reportedly drank more than the recommended units of alcohol in a week compared with only 23% of women (Office for National Statistics, 2008).
Nevertheless, whether these sex differences are attributable to sociological factors or innate biological differences, the evidence suggests that the prospects for spotting tomorrow's criminal using gender are good, at least at the group level.
Ethnicity is also an immediately obvious attribute that may be deduced from appearance. Fox (1978) found that crime rates can be predicted quite accurately using only the non-white, adolescence percentage of the population and the consumer price index. However, later recent studies suggest that racial differences in crime are in prevalence than incidence (Petersilia, 1983). Significantly more blacks than white commit at least one offence; blacks are more likely to carry out casual crime, whereas whites are more likely to be hardened criminals. This would suggest nevertheless, that being black, would greatly increase the likelihood of a child to become a later offender. Although they may not become repeat offenders, the evidence suggests that the majority of males from ethnic minorities will offend, usually in adolescence (Flood-Page et al, 2000).
Although early psychologists have tried to attribute this difference in offending rates to genetic factors, most people agree that any genetic factors cannot explain the massive difference in offending rates, and the bulk of racial differences can be attributed to non-genetic factors (Bellair & McNulty, 2005). It has been argued there are two main reasons that blacks are more likely to commit crimes than whites: the sub-culture of violence and the socio-economic status of blacks.
The sub-culture of violence can be seen to refer to specific social settings within the broader Anglo-American culture, where violence is regarded as a more or less normal response to a range of instigating events which would normally not lead to violent response from members of the Anglo-American society (Wolfgang, 1958). Although this is not a uniquely black trait, sociologists such as Silberman (1978) have explained this greater incidence of violence in blacks in terms of the black experience in America specifically, and the history of black slavery in the West. However, this theory has been widely dismissed; if the higher prevalence of violent crime is attributable to dissatisfaction with the white society, we would expect that most crime would be black on white, whereas most black violence is black on black.
A more credible explanation for the difference in offending rates can be seen to be the fact that blacks are more likely than whites to be born into lower socio-economic situations. Information from the U.S. Census Bureau show that in 2006 6.6% of black households surveyed lived on less than $5000 a year, compared with 2.5% of white households whereas 20.2% of white households lived on over $100,000 a year (only 9.1% of black households). The median income for white households was nearly $19,000 higher than for black households. Furthermore, more than 60% of black infants are born out of wedlock; three to five times the number of black infants grow up without a father figure than white infants. Black children are far more likely to be born to teenage or young parents; 40% of black females have become pregnant by 18, compared with 20% of whites. Therefore it may be that race does not explain the differences in rates for offending, but the differences can be attributable to socio-economic factors (U.S Bureau of the Census, 2006).
Although genetic racial differences have been largely discredited as a predictor of criminality, there is evidence to suggest that certain genetic abnormalities are related to criminal behaviours.
Our diligent detective will not stop at a casual examination of the child and would no doubt observe or test the child for intelligence and personality traits. Intelligence is often seen as a predictor of criminal behaviour. Early studies found a very strong relationship between IQ and criminal behaviours (Goddard, 1921), although later studies such as Hirschi and Hindelang (1977) averaged this out to 8 IQ points. Even when socio-economic status was controlled for, this relationship continued. However, it is arguable that this is merely because individuals with lower intelligence are likelier to be caught. Gath (1972) found that juveniles with IQ's above 115 were less often caught by the authorities and were less likely to be institutionalised when convicted, although West and Farrington (1977) found that there was no difference when IQ was compared with self reported crime and measured crime. Quay (1987) suggests that this may be due to "Yea-saying"; individuals with lower IQ are likelier to admit crime, as it is easier to say yes than admit that you do not understand or cannot read (a point that determines the plot of the recent affecting film "The Reader").
Despite these caveats, there is much evidence to suggest both a link between low IQ and criminality and, specifically, a large disparity in verbal IQ between convicted criminals and those with no convictions. This may be because low verbal IQ leads to poor school performances, owing to low verbal skills causing disaffection with school, leading to engagement with criminal activities. Kandel et al (1988)'s study of men that were at increased risk of crime (who had a father with a serious record of crime) found that the men were more likely to avoid crime if they had a higher IQ. It could be argued that the reinforcement of law avoiding behaviour that comes at school is more important and influential to individuals who enjoy the success at school that is allowed by high IQ. However, the majority of these studies ignore organized and corporate crimes, which are mainly perpetrated by individuals with high IQs.
Personality disorders and traits may also be used to assess future criminality. "Tomorrow's antisocial adults are found among today's antisocial children." (Lynam, 1997). Antisocial behaviour can be seen to continue from childhood, as demonstrated by Robins (1966) study of the adult outcomes of 524 clinic-referred boys. Boys that were found to be antisocial as children were more likely to be arrested, and imprisoned 30 years on, and only those who were antisocial in childhood were later diagnosed with sociopathy in adulthood. However, White, et al (1990) found that of over 200 children predicted to have antisocial outcomes at age 11, 85% did not develop severe antisocial behaviour. Although all chronic offenders have a history of childhood anti-social behaviour, childhood anti-social behaviour is too common to be used to predict future offending.
Studies have suggested that children who are hyperactive and anti-social are at particular risk of later criminality. Hyperactivity refers to a pattern of restless, inattentive, and impulsive behaviour in childhood (Lynam, 1996). Longitudinal studies suggest that children with hyperactive disorders are likelier to become adults who commit crimes (Biederman, 1995). A long running study into the outcomes for 104 children with Hyperactive- Impulsivity-disorder conducted by Weiss, et al (1972) found that boys with HIA had more court appearances and displayed more aggression than the controls, at all ages. After 10 years, 40% of the hyperactive group suffered diagnosed personality disorders, compared with 23% of the controls. Blouin, Bornstein and Trites (1978) compared 23 participants having HIA with 22 children who had problems at school, matched by age, gender and IQ. Boys with HIA had greater alcohol intake and exhibited more criminal behaviours than the controls, strong evidence that personality disorders such as HIA can directly impact criminality; they exhibit more criminal behaviours even when compared with children similarly ostracized from the educational systems.
More general personality traits can have also been seen as indicators of later criminal behaviours. Olweus's (1979) meta-analysis of 12 studies of aggressive behaviour from 6 months to 21 years, found a 0.79 correlation, suggesting considerable consistency between childhood and adulthood behaviours. Farrington's Cambridge Study (1978) found that out of the most aggressive children, as identified at age 8 - 10, 59% were still the most aggressive at 12-14 (compared with 29% of the reminder) and 40% at 16-18 (27% of the remainder). Children identified as severely aggressive at age 8-10 were significantly more likely to become violent offenders (14% against 4.5%) and 70% of those with records for violent crime at 21 were rated more aggressive aged 12 (compared with only 23% of those without records). These studies corroborate the theory that personality as a child will greatly impact later behaviours.
At this point our detective may feel he has garnered all he can from direct observation and may wish to make use of the modern tools of the forensic investigator - DNA testing. Two areas are of particular interest- genetic abnormalities and the argument whether criminal tendencies are heritable.
The enzyme MAO-A (monoamine oxidase A) has come to the forefront of the arguments around genetic influence on criminality, after it was used as evidence in sentencing of a man in Italy convicted of murder. (Ahuja, 2009) Variants in the gene for this enzyme- which affects the production of neurotransmitters such as serotonin and dopamine - are correlated with violence, aggression and gang membership, earning it the nickname the "warrior gene". Studies by Kevin Beaver (2009) have shown a link between antisocial or violent behaviour and low activity MAO-A. Beaver found that men with this variant are twice as likely to join gangs and, once in those gangs, four times as likely to use weapons. Moreover, there is evidence that the effects of the MAO-A gene are differentiated by ethnicity. One study found that former victims of child abuse with high levels of MAOA were less likely to commit violent crimes - but only if they were white (Widom & Brzustowicz, 2006). The effect was not evident in non-white children, suggesting that it may not be the gene itself, but some kind of interactional effects that will affect the likelihood of criminality.
Can we identify tomorrow's criminal from the genetic information given to us from their parents? Many studies do provide evidence for criminal inheritance. Osborn & West (1979) found that 40% of sons born to fathers with a criminal record have a criminal record themselves, compared with only 13% of sons with a non-criminal father. Robins, West & Herjanic (1975) found a high correlation between parental conviction and child conviction in black children. Robins (1966) looked at white children referred to the authorities for anti-social behaviours, finding that 48% of mothers and 23% of fathers were psychologically disturbed or mentally handicapped.
However, not all of these studies are methodologically sound. They do not separate environmental and genetic effects. For example, in Robin's 1966 study, only 36% had both parents at home. Furthermore, it cannot separate neighbourhood modelling influences: criminal parents tend to live in areas which have more crime generally, increasing the likelihood of bad peer influences.
These studies can also be seen to be flawed in that they do not isolate genetic influences. Parental attitudes favourable to violence for example, increase a child's risk of engaging in criminal behaviour. A study of 8,000 middle-school students in a large urban area found the strongest predictor of aggressive behaviour was the perception of parents' feelings about fighting. (Orpinas, Murray, & Kelder, 1999). The students who said that their parents are against fighting were significantly less likely to fight or behave aggressively. Conversely, studies from the USA show that children who learn about the risks of drug use from their parents are 36% less likely to smoke marijuana, 50% less likely to use inhalants, 56% less likely to use cocaine, and 65% less likely to use LSD than children whose parents do not teach them about the dangers of drugs (Resnick, et al., 1997). However, parental attitudes may themselves be influenced by genes.
It is also likely that labelling by the justice system will occur- children from high crime areas, or from families which are known to the authorities as offenders, are more likely to be convicted of crimes.
Adoption and/or twin studies can be used to show "proof" of genetic influences, and thus demonstrate the way criminal inheritance could provide a predictor of the prospects of criminality in a child.
The genetic hypothesis suggests a greater association between criminality of biological parents and adoptive parents. Studies have suggested that there is some inherited criminality. Schulsinger (1975) investigated 5000 children adopted at an early age from the Danish adoption register between 1924- 1947. Of the 5000 children, 57 that had been diagnosed as psychopathic were compared with 57 controls taken randomly from the register. When matched for age, sex, class, age at adoption and "in many cases, neighbourhood of rearing" those children whose biological parents were criminal were 2.5 times more likely to be psychopathic, although it must be noted that psychopathy is not a direct correlate of criminality. Another study by Crowe (1972) studied 52 children given up for adoption by 41 female offenders, compared with a control from the state adoption index. Only two controls, compared with eight index cases had criminal record, but further analysis of this data found that there was increased surveillance of the adopted children that had been born to known offenders, increasing the prospects of detection and conviction.
One convincing study in this area is Mednick, Gabrielli & Hutchings' (1984) study of all non-familial adoptions in Denmark from 1924 - 1947, a total of 14,427 male and female adoptees. They found that for children whose biological parents had criminal records, there was a 20% offending rate, compared with 14.7% of children whose biological parents were not criminal, but their adoptive parents showed criminal behaviours. The study also showed that when biological parents had three or more property convictions (suggesting that they are not accidental criminals) their children were twice as likely to be criminal. This supports the theory of genetic inheritance. However, further analysis of these data by Van Dusen, Mednick & Gabrielli (1983) found the socio-economic status of the adoptive family to have an importance that had previously not been recognized in the analysis.
There is also a generalised flaw in the study of adoption studies: it is difficult to separate the knowledge of the parents' criminal past from the adoption process, which may skew the data. It may be that children with criminal parents are less likely to be placed with families with a higher socio-economic status, or who do not have criminal records themselves.
This problem can be overcome by studying the effect of genetic inheritance on twins. By looking at twins, brought up in the same conditions, we can look at the interaction between biological influences and environmental factors.
Perhaps the most influential twin study in this area is Dalgard and Kiringlen's 1976 study of male twins born in Norway 1921 - 1930. Interviews with the pairs where at least one was convicted found that the 31 monozygotic twins had a concordance rate of 25% compared with a concordance rate of just 14.9% for the dizygotic twins, suggesting a genetic inheritance link with criminality: monozygotic twins with identical DNA are more likely to share criminal behaviours. This was corroborated by Christiansen's 1968 Danish study of 6000 pairs of twins born 1880-1910. Christiansen found a higher concordance for monozygotic twins than dizygotic twins, but this difference was more exaggerated higher in middle class and rural areas, possibly because in urban areas or areas of low socioeconomic status, there are more likely to be environmental pressures.
Although the evidence from twin studies is striking, it cannot be seen as absolute: there are often problems, especially in the earlier studies, in identifying which twins are monozygotic/dizygotic. Furthermore, we cannot control for environmental factors, it may be that there is not a genetic link, but the higher concordance rates for monozygotic twins can merely be attributed to the fact that parents and carers are more likely to treat identical twins as one entity. Finally, the differences in concordance could be due to the differences in conviction rates for men and women; monozygotic twins are always the same sex, whereas dizygotic twins can be different sexes.
So, having wrung as much as possible out of the clues that are placed in front of him and made used of the latest DNA techniques, is our detective ready to name the future criminal? It seems doubtful. Although we have found that gender, race and to some extent genetic and personality factors may correlate with later criminality, these predictors cannot be usefully applied at the level of the individual. This of course is a common theme in literature. The individual, from Dickens' David Copperfield to the savage in Huxley's "Brave New World", can always rise above his origins or early life. As argued in the film "Minority Report" the future cannot be determined - whether we argue in terms of free will or merely acknowledge that the multi-factorial and variable factors that affect an individual's course through life are too numerous and complex to predict.
Lettice argues that, "there is a world of difference between taking action to deal with existing problems .... and predicting that problems will exist, and attempting to head them off via early intervention". Furthermore, the evidence is accumulating against early intervention. Action on Rights for Children says "Far from being, at worst, ineffective, a growing body of research suggests that it can actively do harm." (Woolf, 2006) This view is held also by Jean Hine, who argues that although it is possible to identify 'tell-tale' signs in actual offenders, the presence of these does not necessarily identify future offenders. Start with the real villains and work backwards, and the signs were all obviously there, but studies that start with the signs and work forwards do not end up separating the criminals from the law abiding. (Woolf, 2006)
Therefore, although we can identify risk factors for tomorrow's criminals in today's pushchairs, the prospects for identifying the future criminals themselves is poor. Whilst it may be good to target interventions where they are most effective, there are dangers in pretending we can identify future criminal in early life. These dangers include stigmatisation of a group or individual and indeed the danger of the self-fulfilling prophecy. When we look at a pushchair we should not look to find a future criminal. We should see a child, with all the potential - for evil, but also for good and indeed for greatness - that a child possesses.