Theories of Offending

(1)

© Banff and Buchan College 1 Date of Production/Revision 12/30/2011

Author Code (S44/Finl1) File :n:\(file ref.)

Theories of Offending

Unit Title: Psychology A: History and Development of

Psychology

Unit No: FK8D 34

Lynn Findlay

(2)

Contents

Psychodynamic Theories of Offending ... 3

Megargee’s ‘overcontrolled’ violent offender ... 4

Bowlby’s ‘affectionless psychopath’ ... 4

Evaluation of the psychodynamic approach... 5

Evaluating psychology using the scientific method ... 6

Physiological Theories of Offending ... 7

Lombroso’s ‘atavistic form’ theory ... 7

Sheldon’s theory of somatotypes ... 8

The ‘Extra Y’ Hypothesis ... 9

Learning theories of offending ... 10

Differential association theory... 10

Social learning theory ... 11

Eysenck’s theory of the criminal personality ... 13

Some items from the EPI ... 16

Biological theories of offending ... 18

The Biological Perspective: Aggression and Free Will ... 21

(3)

Psychodynamic Theories of Offending

Freud’s theory of the psyche

In Freud’s psychodynamic theory, personality (or the psyche) has three distinct components: the id, representing primitive desires and the need for gratification, the ego, representing moral and social constraint and the need for gratification, the superego, representing moral and social constraints, and the ego, representing reality and the ability to delay gratification. Within this framework, it is the role of the ego to strike a balance between the demands of the id and the constraints imposed by the superego. This is primarily achieved through the use of defence mechanisms that allow the id’s desires to be satisfied in ways that the superego finds acceptable. To a Freudian, then, behavior that lies outside what society regards as acceptable – be it ‘abnormal’ or ‘criminal’ – is the result of abnormal development of the psyche. Since, in classical Freudian theory, the structure of the psyche is determined in the first five years of life, it follows that the roots of offending are also found in this period, especially in the relationship between the developing child and its parents. The Freudian framework implies a number of possible causes for later criminal behaviour.

A weak superego

The superego is the moral regulator of behaviour. It develops at the end of the phallic stage (about 5 years) as the child internalises its same-sex parent in order to resolve the Oedipus complex. The superego continues to act as a parent within the psyche. It punishes the ego with anxiety when an immoral act is contemplated and with guilt if the act is carried out. A weak superego, developed as a result of abnormal relationships within the family, would result in a person with few if any of the usual inhibitions against antisocial

behaviour. They would act in ways that gratified their id, regardless of the social restraints on doing so.

A deviant superego

Alternatively, a child might develop a superego in the normal way, but the superego in the normal way, but the superego itself has deviant values. The superego is an internalisation of the same sex parent so as a moral regulator it threatens and punishes those behaviours that the parent would find

unacceptable. Consequently, a son raised normally in family with a criminal father might develop a superego that does not react to criminal acts that the father would engage in.

A strong superego

It seems counterintuitive that a strong superego could increase a person’s risk of offending when the superego is the regulator of moral behaviour but there are at lest two ways in which this might happen. An excessively powerful superego would render a person anxious and guilty much of the time, since every time they acted on the id’s desires – however innocuously – their superego would punish them for it. This could result in a person

(4)

imposed by their own superego. Alternatively, an excessively strong superego might prevent the person from expressing any of the antisocial impulses that inevitably build up in their unconscious. Normally, they would express these impulses harmlessly through defence mechanisms (e.g. by sublimating their aggression into sport). If the superego prevents this from happening, the aggression or sexual desire could build up over time until it becomes strong enough to overwhelm the ego and is expressed suddenly and violently as murder or rape.

Megargee’s ‘overcontrolled’ violent offender

Megargee (1966) documented a series of cases of violence carried out by people who were regarded as passive and harmless. For example, an 11-year-old boy who stabbed his brother 34 times with a steak knife was described as polite and softly spoken with no history of aggression (Gross, 1996). Megargee argued that such cases represent a distinct sub-group of violent offenders whose shared characteristic is an apparent inability to express their anger in normal ways and who eventually ‘explode’ and release all their anger and aggression at once, often in response to a seemingly trivial provocation. Freudian formulations like Megargee’s are unfashionable

nowadays and more research attention is given to the majority of violent offenders, whose problem is generally a lack of inhibition of their anger, rather than too much inhibition. Nonetheless, there is evidence that a subset of violent offenders follow the pattern described by Megargee. Blackburn

(1971), for example, found that people convicted of extremely violent assaults tended to have fewer previous convictions and scored lower on measures of hostility than those convicted of moderately violent assaults. However, the existence of such a group does not in itself show that Megargee was correct about the underlying mechanisms responsible. In particular, Megargee’s approach does not adequately distinguish whether such offenders do not

experience anger normally (as the psychodynamic approach would suggest) or whether they experience it but do not express it (Blackburn, 1993).

Bowlby’s ‘affectionless psychopath’

Another psychodynamically inspired explanation of offending comes from Bowlby (1951) although Bowlby was also influenced by ethology and evolutionary theory. Bowlby proposed that the ability to form meaningful social relationships in adulthood was dependent on a close, warm and continuous relationship with the mother in the first few years. Since this relationship acts as the prototype for all future relationships, its disruption would impair the person’s ability to relate to others. This could result in a condition that Bowlby called ‘affectionless psychopathy’. Bowlby presented evidence that early maternal deprivation was related to later criminal

behaviour, notably through his famous ’44 Thieves’ study in which he

reported that 39 per cent of a group of juvenile delinquents has experienced significant disruption to their attachments, compared to only 5 per cent of a non-delinquent group. Such evidence notwithstanding, it is no longer widely accepted that research has been criticised for unrepresentative sampling and

(5)

poor control group matching. Later theorists, particularly Rutter (1971), have pointed out his failure to distinguish between deprivation, privation and the distortion of attachments, each of which may have different effects. Overall, he is now regarded to have overestimated the impact of early life experiences on later offending.

Evaluation of the psychodynamic approach

Psychodynamic theories of offending are no longer widely accepted by psychologists, for a number of reasons. First, there is the difficulty

associated with testing some of the concepts. Psychodynamic theories rely heavily on concepts like the unconscious mind, whose existence is difficult if not impossible to prove. Second, there is a tendency for psychodynamic theories to be able to explain any behaviour, but only after it has happened. As a result, these theories are regarded by many as unprincipled and, since they are incapable of being proved wrong, unscientific. Third,

psychodynamic researchers rely heavily on qualitative case studies in which the participants’ behaviour is interpreted in symbolic terms. This is a highly subjective process: two different analysts may draw very different conclusions from the same set of observations. This type of evidence makes scientists, who prefer quantifiability and objectivity, rather uneasy. Fourth,

psychodynamic therapies that have attempted to treat offending have not been successful (Howitt, 2009), besides being extremely time consuming. It is important not to overlook the positive contributions that psychodynamic theories have made to criminological psychology. Psychodynamic

researchers have pointed to the importance of childhood experiences and parent-child relationships as an influence on offending (Blackburn, 1993) and have identified many important variables relating to delinquent behaviour in adolescence (Hollin, 1989). So, whilst its theoretical explanations have fallen out of favour, the psychodynamic tradition should be credited with pointing in some useful directions for later researchers to follow.

(6)

Evaluating psychology using the scientific method

You are learning how to… In the context of…

 Critically evaluate psychological theories

 Assess the extent to which psychology is scientific

 Psychodynamic theories of offending

What is science?

In the context of the debate about whether psychology (or any approach to psychology) is scientific, science can mean several things. It is possible to distinguish between science as a type of knowledge and science as a

method for obtaining knowledge. If you wish to use an approach’s

scientific status as a way of evaluating it, you need to distinguish between these two aspects.

As a type of knowledge, scientific theories must…

 Be based on empirical evidence

 Be logically constructed and internally consistent

 Be capable of being proven wrong, so they can be modified as the evidence dictates.

As a method for obtaining knowledge, scientific research must…

 Test hypotheses derived from theory

 Be empirical (i.e. gathered through the senses)

 Be objective (i.e. researchers must agree on what they have observed) and unbiased

 Use rigorous, repeatable methods

 Be conducted in a systematic way

 Be open to public scrutiny of methods and data

Evaluating psychodynamic theories

Students often evaluate psychodynamic theories very sloppily. Frequently they make statements like ‘there is no evidence for Freud’s theories’ or ‘psychodynamic theories are unscientific’. This is poor evaluation in the first case because it’s simply not true and in the second place because without any further elaboration this is an unsubstantiated and rather sweeping statement. When evaluating any theory;

 Explain the nature of your criticism.

 Don’t make sweeping statements, especially if they aren’t true.

 Play the ball, not the man.

Write an evaluation of the psychodynamic approach to offending, using the criteria above. Make sure that each of your points is fully explained and elaborated.

(7)

Physiological Theories of Offending

What do we mean by a physiological theory?

Before the 19th Century, discussion of crime and criminals was conducted entirely in moral and philosophical terms. It was only in 1876 when the Italian anthropologist Cesare Lombroso started a tradition of physiological theories of criminality. These have in common a focus on the person’s physical form as a marker of criminality.

Lombroso’s ‘atavistic form’ theory

Lombroso (1876) claimed that criminality was heritable. He suggested that there was a distinct biological class of people that were prone to criminality. These people exhibited ‘atavistic’ (i.e. primitive) features; Lombroso

suggested that they were ‘throwbacks’ who had biological characteristics from an earlier stage of human development that manifested as a tendency to commit crimes. Lombroso claimed that criminal types were distinguishable from the general population because they looked different. The principle markers of criminality were a strong jaw and a heavy brow. However, he also suggested that different types of criminal had different features, so murderers had bloodshot eyes and curly hair, whilst sex offenders had thick lips and protruding ears.

One hundred years or so later, Lombroso’s theory appears faintly ridiculous to most of us, and there is no doubt that it is deeply flawed. First, Lombroso did not use any non-criminal control groups to

establish whether the ‘atavistic’

www.crimeculture.com

features he identified were confined to the criminal population. Second, his sample is likely to have contained a large number of people with

psychological disorders and chromosomal abnormalities, so he has not distinguished adequately between criminality and pathology. Third, crime is neither a natural nor a homogenous category of behaviour; it is a social construction, which makes the argument that criminal behaviour as such is inherited is hard to sustain. Fourth, our current understanding of genetic influences on behaviour does not support the idea that complex behaviours (like most criminal activities) are controlled by single genes.

(8)

To be fair to Lombroso, he modified his theories quite extensively over the course of his career. He eventually came to believe that only about a third of criminality was directly attributable to atavistic features. The majority of criminal behaviour in his later theories was the result of environmental factors such as poverty and poor education. Although Lombroso’s theories are no longer taken seriously by criminological psychologists, it is important to appreciate that he made several important contributions to the discipline. Specifically, Lombroso (1) shifted the study of criminal behaviour from a moral basis to an empirical one; and (2) argued for the interaction of biological, psychological and social factors in causing criminal behaviour. As a result, he is regarded by many as the ‘father of modern criminology’ (Shafer, 1976).

Sheldon’s theory of somatotypes

Sheldon (1949) advanced a theory that shares with Lombroso’s the idea that criminal behaviour is linked to a person’s physical form. Sheldon

distinguished between three basic types of bodily build: ectomorph (thin), endomorph (fat) and mesomorph (muscular). Sheldon believed that bodily build was linked to personality and temperament so ectomorphs were solitary and restrained, endomorphs relaxed and hedonistic and mesomorphs

energetic and adventurous. Pure somatotypes are rare, and most people represent a blending of different types. Sheldon’s principle claim was that mesomorphs are more prone to criminal activity than the other two types. Consequently, his theory predicts that there should be a relationship between how mesomorphic a person is and their degree of criminality.

Sheldon assessed the somatotypes of samples of college students and delinquents from photographs. Each photo was rated for mesomorphy on a scale from 1(low) to 7 (high). It emerged that the delinquents had a higher mean mesomorphy rating than the college students (4.6 vs. 3.8), supporting Sheldon’s claims about the link between body type and criminality. A reanalysis by Hartl et al. (1982) found that the most seriously delinquent of Sheldon’s sample had a mean mesomorphy rating of 5, adding further support to the theory. Although Sutherland claimed that Sheldon’s method for distinguishing delinquents from non-delinquents was not valid, a number of other studies have confirmed that there is a small association between bodily build and criminality (Putwain & Sammons, 2001). It is not clear why, but several possibilities suggest themselves. It might be that a mesomorphic build reflects high testosterone levels, which may result in higher levels of aggressiveness. Alternatively, it could be that people react to mesomorphs in ways that increase their risk of criminal behaviour. Because of the

stereotypes people hold about mesomorphs, they may be drawn into delinquent activities by their peer groups. Alternatively, the judicial system may treat them more harshly, increasing the likelihood that they will officially be labeled as criminal (Blackburn, 1993).

(9)

The ‘Extra Y’ Hypothesis

A slightly later physiological theory suggested that some crime might be attributable to a chromosomal abnormality. Sex is determined by the pattern of a person’s sex chromosomes: XX in a woman, XY in a man. It is a Y chromosome that makes a person male. It is well known that atypical chromosomal combinations can result in atypical sexual development. For example, in Klinefelter’s Syndrome, the combination XXY results in a male form with some female characteristics. Since an ‘extra X’ appears to feminise men, some theorists speculated that an additional Y chromosome might ‘hyper masculinise’ men who had it. Since men are more aggressive than women, it might be that men who have XYY chromosomes might be more aggressive than other men and hence more likely to commit violent crimes.

The idea was advanced that offender populations in prisons and hospitals would be likely to contain large numbers of XYY men. Some claims were made that high profile, prolific offenders, such as the American serial killer Arthur Shawcross, had the XYY pattern. It was eventually established that XYY men are rare in the general population but more common in the offender population (Howitt, 2009). Whilst this is as expected, the problem is that XYY men tend to commit non violent crime, not violent crime as the XYY hypothesis predicts (Epps, 1995). Why might this be? Testosterone levels amongst XYY men are no different from XY men, and they are no more aggressive than the general population. However, they are at a substantially increased risk of developmental delay and learning difficulties (Graham et al., 2007). There is a small association between learning difficulties and criminal behaviour. IQ scores amongst convicted offenders are marginally lower than the general population (Hollin, 1992) and there is a slightly higher prevalence of mild learning difficulties amongst offender groups (Lund, 1990). It might therefore be the case that the higher than expected number of XYY men in the offender population is a consequence of the learning difficulties

associated with the condition.

Conclusions

Whilst there is evidence that some physiological factors are associated with an increased risk of criminal behaviour, it is clear that there is no one

physiological abnormality that causes people to commit crimes. Given the diversity and complexity of the range of behaviours encompassed by the term ‘crime’, this is not really surprising. Even if we were to focus on one category of criminal behaviour, for example, violent crime, it seems unlikely that a single pathological factor would be able to account for all examples. As Lombroso realised, a satisfactory explanation of a crime is likely to require consideration of biological, psychological, environmental and social factors.

(10)

Learning theories of offending

Is criminality learned?

Learning theories of offending are based on the assumption that offending is a set of behaviours that are learned in the same way as other behaviours. Two examples of learning theories are Sutherland’s (1939) differential

association theory and Bandura’s (1973) social learning theory. These types of theory emphasise the family and the peer group as a potential source of criminal behaviours. Social learning theory also raises the possibility that some types of antisocial behaviour may be learned from media sources like television, films and videogames.

Differential association theory

Sutherland (1939) suggested that there were two prerequisites for a person to develop into an offender. They need to learn a set of values and attitudes that support offending, and the need to learn specific behaviours for

committing crimes. These are all learned within the family and peer group. The people that surround a developing child will demonstrate a range of attitudes towards the law and crime, some favourable and some

unfavourable. Sutherland argued that if the child acquires more attitudes that are favourable to crime than unfavourable ones, the result will be that they regard criminal behaviour as acceptable. They may also learn specific methods for committing crimes from those around them. The types of crime the person then goes on to commit will depend heavily on the precise nature of the deviant attitudes they have learned. For example, they might regard it as unacceptable to rob someone, but acceptable to falsify one’s tax returns.

Evidence for differential association theory

The basic prediction of differential association theory is that people who become offenders will have been socialised in families and groups where there are some pro-criminal norms. There should therefore be evidence of pro-criminal norms and probably criminal activity in the families and peer groups of offenders. This is indeed the case. A certain amount of evidence suggests that criminal behaviour tends to run in families. Whilst this is

frequently offered in support of a genetic contribution to offending some of the evidence is equally consistent with differential association theory. For

example, Osborne and West (1982) found that where the father had a criminal conviction, 40% of sons also acquired one by the age of 18,

compared with only 13% of the sons of non-criminal fathers. A great deal of research suggests that criminality is concentrated in a small number of

families. Walmsley et al found that a third of UK prisoners claimed to have a family member also in prison. Matthews (1968) also found that juvenile delinquents are more likely than non-delinquents to report having peers who engage in criminal activity.

Evaluation of differential association theory

Whilst all of this is consistent with differential association theory, Blackburn (1993) raises two problems. First, this pattern seems confined to petty acts of criminality such as vandalism. Second, because the data are correlational

(11)

it is equally likely that adolescents with deviant tendencies seek out deviant peers. A further problem with differential association theory is that some of its constructs are rather vaguely specified. It is difficult to see, for example, how the number of pro-criminal attitudes a person acquires could be

measured with any precision, and Sutherland does not specify by how much pro-criminal attitudes must outnumber pro-law ones in order for the person to become an offender. Like many general theories of criminality, differential association theory runs into problems when required to explain criminal behaviour on an individual level. For example, it is not clear why some people raised in persistent contact with ‘criminogenic’ influences do not go on to commit offences. A final problem for differential association theory is that it does not adequately explain the developmental pattern of offending.

Criminal behaviour in adolescence is relatively common: 40% of offences are committed by people under 21 years and about half of males and a third of females report having committed at last one offence before the age of 25 (Newburn, 2002).

Social learning theory

Bandura’s social learning theory (SLR) suggests that behaviour of all kinds is learned through the observation of models. Models are selected on the basis of a range of characteristics including attractiveness, status and perceived similarity with the observer. Whether or not a model’s behaviour is imitated depends on the observed consequences of their actions. If the model is observed to be reinforced (and the reinforcement has value for the observer) then imitation becomes likely. If the model is punished then imitation

becomes less likely (although the behaviour may still have been learned, it is its expression that observed punishment inhibits). In SLT, criminal behaviour is regarded as qualitatively no different from any other behaviour. In this respect, SLT shares many ideas with differential association theory. However, it is rather more precisely specified, lacking many of the vaguer concepts of differential association.

Evidence for social learning theory

The most compelling evidence for social learning theory comes from a series of classic laboratory studies carried out by Bandura and his colleagues in the 1960s. These studies focused on children’s acquisition of aggressive

responses from adult models. For example, Bandura et al (1963) showed children an adult model behaving aggressively towards an inflatable ‘bobo’ doll. The model was either reinforced (rewarded with sweets) or punished (told off) for her behaviour. A control group saw the model behave

aggressively but with no consequences, good or bad. When the children were allowed to play in a room that contained a bobo doll, those who had seen the model punished were significantly less likely to imitate her actions. An alternate source of evidence for the social learning of aggression is research into the effects of media aggression on behaviour. A natural experiment by Williams (1986) examined children’s levels of aggression before and after the introduction of television into an isolated community. Williams found that over a two year period, aggression in this community’s children rose steadily whilst in a similar community where there already was

(12)

television there was no increase. One possible interpretation of this is that the children learned to behave aggressively from models in the TV

programmes they watched.

Evaluation of social learning theory

There is a large body of research that shows that children can learn behaviour through observation and that their willingness to imitate these behaviours is affected by the observed consequences of a model’s actions. However, there is a substantial difference between children hitting a bobo doll in a lab and people committing criminal offences in the real world. Social learning theorists have largely neglected naturalistic research (Blackburn, 1993) and this means we should be cautious about assuming that the processes demonstrated in the laboratory apply in the same way outside it.

In the absence of evidence that criminality is 100% genetic it is fairly obvious that learning plays a role in offending. But SLT has little to say about the conditions under which violence and criminality are learned. It also

underplays the role of cognition in criminal behaviour. For example, it is well known in our society that criminals are frequently caught and imprisoned, a fairly salient observed punishment. It is also the case that most people work in legitimate employment to acquire reinforcements such as money and status, another very obvious set of models and observed majority of models and reinforcements should promote non-criminal behaviour (Howitt, 2009). Clearly, there is more to offending than SLT would imply.

(13)

Eysenck’s theory of the criminal personality

What do we mean by ‘personality’?

The term ‘personality’ is generally used to refer to relatively stable characteristics of a person that make their behaviour consistent across situations (but many other definitions are possible, depending on the

approach being taken). Hans Eysenck (1964) put forward a theory of criminal behaviour based on a very influential theory of personality he had earlier devised and which he continued to develop throughout his career. Although this theory is usually referred to as

a personality theory of offending, it is important to appreciate that Eysenck’s theory conceives of criminal

behaviour as the outcome of interactions between processes occurring at several different levels of explanation.

Extraversion, neuroticism and psychoticism

Eysenck originally argued that the great variation between people’s

personalities could be reduced to just two dimensions, which related to the underlying functioning of the individual’s nervous system. A person’s level of extraversion (E), neuroticism (N) can be measured using simple pencil-and- paper questionnaires such as the Eysenck Personality Questionnaire (EPQ). People with High E scores are sociable, active, lively and sensation seeking. E is determined by the overall level of arousal in the person’s CNS and ANS. High E-scorers have a low level of arousal and therefore need more

stimulation from their environment. People with high N scores are anxious, depressed and react very strongly to aversive stimuli. N is determined by the overall level of lability in the person’s CNS. Where N is low, the person has a stable, relatively unreactive nervous system whereas a high N score results in a high degree of instability.

Eysenck later added a third dimension of personality, psychoticism (P). People who score high on P are aggressive, antisocial, cold and egocentric. Eysenck was less clear on how P related to the functioning of the nervous system. According to Eysenck, E, N and P are determined largely by genetics. Each trait is normally distributed in the population. That is, most people have moderate E, N and P scores. Extreme scores are rarer and the more extreme a score, the rarer it is.

Social Psychological Biological Behaviour in situations where criminal behaviour is a possible outcome. Responses to socialisation (reinforcement & punishment) Stable psychological traits Functioning of the nervous system Eysenck’s personality theory

(14)

E, N, P and criminal behaviour

In Eysenck’s theory, personality is linked to criminal behaviour via socialisation processes. Eysenck viewed criminal behaviour as

developmentally immature in that it is selfish and concerned with immediate gratification. The process of socialisation is one in which children are taught to become more able to delay gratification and more socially oriented. This is accomplished primarily through conditioning. When children act in immature ways they are punished. Consequently, they come to associate anxiety with antisocial behaviour. Where this process is successful, even thinking about behaving antisocially produces anxiety, so the person avoids doing it.

Eysenck believed that people with high E and N scores had nervous systems that made them difficult to condition. As a result, they would not learn easily to respond to antisocial impulses with anxiety. Consequently, they would be more likely to act antisocially in situations where the opportunity presented itself.

Evidence for Eysenck’s theory

Eysenck’s theory covers a great deal of ground and there are aspects of it that are not easy to test. However, it does make the basic prediction that compared with non-offenders, offender populations should have higher E, N and P scores. It should be fairly straightforward to test this prediction.

Rushton and Christjohn (1981) compared E, N and P scores with self-reports of delinquency in schoolchildren and students. They found that those who reported higher levels of delinquency also scored higher on E, P and N. These correlations are consistent with Eysenck’s prediction.

However, studies of ‘official’ delinquency (e.g. comparing convicted offenders with non-offenders) do not produce such clear cut results. Farrington et al (1982) reviewed 16 studies of the relationship between E, N and P measures with criminal convictions. They found that in the majority of cases offenders scored higher on P and N but not on E. Hollin (1989) notes a similar pattern of findings: offenders generally show higher P and N scores but not necessarily higher E scores. It is not clear why the relationship between E and offending is so inconsistent. One possibility is that E scales actually measure two

things, sociability and impulsiveness and that criminality is associated with the latter but not the former (Putwain & Sammons, 2002).

Evaluation of Eysenck’s theory

There is some empirical support for Eysenck’s theory, but a number of critics have argued that the data are flawed. Research in this area relies heavily on the self-report measures of personality devised by Eysenck and colleagues. Some (including Farrington et al) have suggested that these scales are subject to response bias. There is also the issue that research has made heavy use of comparisons between convicted offenders and the general population. Such research inevitably excludes those who commit crimes and are not caught and convicted. Consequently, many of these studies may only be telling us about the personality characteristics of ‘unsuccessful’ offenders.

(15)

Leaving aside the issues of sampling and response bias there is an argument that the theory itself is somewhat circular. Take the example of the

psychoticism scale. To measure P, respondents answer ‘yes’ or ‘no’ to a series of statements about whether they act aggressively and selfishly.

Their scores on such scales are then used to ‘explain’ why high P-scorers act aggressively and antisocially. This is rather unsatisfactory and P remains a significant weakness of Eysenck’s theory since it is far from clear what (if anything) it measures.

A further issue arises from the way Eysenck conceived of personality as a set of stable traits which cause people to behave consistently across situations. As noted above, this is just one of a range of possible approaches to

personality. Situationist theories of personality suggest that no such consistency really exists. Mischel (1968), for example, argues that the

apparent consistency in people’s behaviour is an illusion that arises from the fact that we typically observe people in similar situations. If we accept this line of argument then we must question the existence of the stable

personality traits on which Eysenck’s theory rests.

Howitt (2009) explores a number of problems with Eysenck’s theory. Whilst applauding its attempt to integrate different levels of theorising (genetic, biological, psychological and social) Howitt notes that the broad sweep of Eysenck’s theory actually addresses few of the real concerns of forensic psychologists, who are more interested in questions about specific types of crime. Eysenck’s theory tells us that rapists and child abusers are extravert, neurotic and psychotic, but it does not tell us why they rape or abuse children. This criticism could equally be levelled at any of the theories of general

criminality that have been put forward since Lombroso.

On the other hand, Eysenck’s theory may point in some useful directions where it comes to preventing crime. His theory suggests that the underlying tendencies that eventually manifest themselves as criminal behaviour are detectable in childhood and that it may be possible to modify the socialisation experiences of high-risk individuals so that they do not develop into offenders. This could lead to interventions based on parenting or early treatment for delinquency and hence may be of great practical benefit in reducing criminal behaviour.

(16)

Some items from the EPI

Here are some questions about how you think, feel and act. After each question is a space for answering ‘yes’ or ‘no’.

Decide whether ‘yes’ or ‘no’ represents your usual way of thinking, feeling or acting, and then circle the appropriate answer.

Don’t spend too much time on any question; go with your first reaction, rather than thinking for a long time about how you want to respond. This isn’t a test of intelligence or ability, just a measure of how you behave.

Do you like plenty of excitement and bustle around you? Yes No

Do you like working alone? Yes No

When you get annoyed do you need someone friendly to talk to about it?

Yes No

Are you often ‘lost in thought’? Yes No

Do you sometimes feel happy, sometimes sad, for no real reason?

Yes No

Are you moody? Yes No

Have you often lost sleep over your worries? Yes No

Have you sometimes told lies in your life? Yes No

Do you sometimes laugh at a dirty joke? Yes No

E:

N:

L:

NB: this is not a full version of the EPI and does not produce valid psychometric measurements. It has been reproduced for

educational/illustrative purposes only and should not be used for any other purpose.

(17)

Eysenck’s personality theory

You are learning how to… In the context of…

 Comment on psychological evidence

 Evaluate psychological theories

 Eysenck’s theory of the criminal personality

Turn these statements and observations into criticisms

Rushton and Christjohn (1981) compared E, N and P scores with self-reports of delinquency in schoolchildren and students. They found that those who reported higher levels of delinquency also scored higher on E, P and N.

Farington et al (1982) reviewed 16 studies of the relationship between E, N and P measures with criminal convictions. They found that in the majority of cases offenders scored higher on P and N but not on E. Hollin (1989) notes a similar pattern of findings: offenders generally show higher P and N scores but not necessarily higher E scores.

Research in this area relies heavily on the self-report measures of personality devised by Eysenck and colleagues.

Research in this area has made heavy use of comparisons between convicted offenders and the general population.

To measure P, respondents answer ‘yes’ or ‘no’ to a series of statements about whether they act aggressively and selfishly. Their scores on such scales are then used to ‘explain’ why high P – scorers act aggressively and antisocially.

Mischel (1968) argues that the apparent consistency in people’s behaviour is an illusion that arises from the fact that we typically observe people in similar situations.

Eysenck’s theory tells us that rapist and child abusers are extravert, neurotic and psychotic, but it does not tell us why they rape or abuse children.

Eysenck’s theory suggests that the underlying tendencies that eventually manifest themselves as criminal behaviour are detectable in childhood.

(18)

Biological theories of offending

Is there a gene for crime?

There is no gene for crime. Modern behavioural genetics has advanced far beyond the simplistic ‘single defective gene’ ideas associated with Lombroso and others. Any serious attempt to link criminal behaviour with genetic

inheritance will start from the view that the nervous system is the organ that determines our behaviour. Each of as has a nervous system whose structure and functioning determines how we learn from and respond to our

environment. Since the organisation of our nervous system is dependent to some extent on our genetic inheritance it follows that our behaviour, including criminal behaviour, may be influenced by our genes.

This view is a long way from the Lombrosian one and embodies very different ideas about how genes may affect behaviour. First, there is an assumption that any particular behaviour is affected by many different genes (i.e. is polygenetically influenced). Second, it is assumed that genes must interact with the environment. They represent a potential to develop in particular ways but different behavioural potentials will only be fulfilled with the operation of a suitable environmental trigger. Third, there is no assumption that crime is caused by a defective genetic inheritance; rather, the genetic influences which may lead to crime are regarded as part of the normal genetic variability within our species. So genes do not cause crime. Under particular

environmental influences they may give rise to certain ways of behaving that are criminal under some circumstances. Hollin (1992) identifies two important questions in relation to this area: (1) is there evidence for a genetic influence on crime; and (2) if so, how does it operate? The first question may be

addressed by using the usual tools of genetic research in psychology: family history, twin and adoption studies.

Family history studies

Criminal behaviour has a tendency to run in families. Osborne and West (1982) compared the sons of criminal and non-criminal fathers. They found that 13 per cent of the sons of the non criminal fathers had criminal

convictions, compared with 40 per cent of the sons of criminal fathers. This is a reliable finding. Clearly, this is consistent with a genetic influence on

offending but it can hardly be regarded as conclusive evidence for the same. It is equally consistent with the idea that criminal behaviour is learned within the family or that a third variable, such as poverty, accounts for criminal behaviour in both fathers and sons (Hollin, 1989). Even if we accept that the higher rate of criminality in the sons of criminal fathers is attributable to genetics, it is still necessary to explain why 60 per cent of them did not go on to commit crimes and why 13 per cent of the sons of noncriminal fathers did (Ainsworth, 2000).

(19)

Twin studies

Twin studies, in which the similarity of mono- and dizygotic twins in respect to particular traits is compared, go some way to addressing the weaknesses of family history studies. Early twin studies generally found higher concordance for criminality amongst MZ than DZ twins, which would support the

suggestion of a genetic influence. However, much of the early research was flawed by a combination of small samples and poor methods for determining zygocity (i.e. whether a particular twin pair was MZ or DZ). Better, more recent studies have addressed these issues and it is noteworthy that they have generally produced lower estimates for the heritability of criminal tendencies than the early research (Putwain & Sammons, 2002). The most convincing studies come from Denmark, where researchers have access to extensive data on over 3,500 twin pairs.

Christiansen (1977) found MZ concordance rates of 35 per cent, compared with 13 per cent for DZ twins. Dalgard and Kringlen (1976), in Norway, found similar results: MZ 26 per cent, DZ 15 per cent.

As with the family history data, these results are consistent with a genetic influence on criminal behaviour. However, several points should be stressed. First, though the MZ concordances are higher than the DZ they are not

especially high, suggesting that any genetic influence on criminality is likely to be slight. Second, since MZ twins are usually treated more similarly than DZ twins, and because DZ twins may be a boy and a girl, the increased

concordances for MZ twins may still be attributable to environmental influences.

Adoption studies

In an adoption study, the similarities between adopted children and their biological and adoptive offspring are compared. Where the similarity is greatest with the adoptive parents an environmental influence can be

assumed on the characteristic in question. Conversely, where there is greater similarity with the biological parents this may indicate a genetic influence.

Several such studies suggest a genetic influence on criminality. Crowe (1972) found that where the biological mother had a criminal record, the child ran a 50 per cent risk of acquiring one by the age of 18, compared with only a 5 per cent risk where the biological mother did not have a criminal conviction.

Hutchings and Mednick (1975) found that where neither biological nor

adoptive father had a criminal record, the son went on to get one about 10 per cent of the time. This rose to 11 per cent where only the adoptive father had a criminal record, 21 per cent where only the biological father had one and 36 per cent where both had a criminal record. More recently, Mednick et al (1994) found no relationship between the number of criminal convictions of adoptive parents and their adopted children but a significant correlation between the number of criminal convictions of the biological parents and their offspring (although only for property crimes, not violent crimes).

Once again, these findings are consistent with the idea that there is a genetic influence on criminal behaviour. Several limitations should, however, be

(20)

mentioned. First, it may not be criminality per se that is being inherited. For example, there appears to be heritable predisposition towards alcoholism (Bohman et al, 1982). Since alcohol abuse is associated with violent crime, this may be where the genetic link comes from. Second, similarities between biological parents and their offspring need not be genetic. Criminal

convictions are associated with poverty and low socioeconomic status. Both of these are stressful. Maternal stress during pregnancy can lead to

developmental problems in the offspring, including behaviour disorders which might lead to criminal convictions later in life. Third, many adopted children are not adopted at birth and their life experiences prior to adoption may significantly affect their later development.

What is being inherited?

If we accept that the evidence shows a genetic influence on criminal behaviour, there remains the issue of how this influence operates. Hollin (1992) discusses a number of possibilities:

 Abnormalities in the CNS, especially the brain. These might lead to poorer cognitive functioning (e.g. lower IQ) or problems regulating behaviour (e.g. ADHD) which might impair decision making and learning, leading in some way to criminal behaviour.

 Abnormalities in the ANS. An abnormally unresponsive ANS would mean that the individual only reacted to strong stimuli. This might impair learning (as above) or possibly lead to a greater than usual tendency for stimulation seeking.

 Abnormalities in the endocrine system. Atypical hormonal activity might conceivably be connected with certain types of offence. For example, testosterone may play a role in sexual and violent crimes.

The problem, as Hollin points out, is that even considered in isolation each of these systems is enormously complex and the relationship of their functioning with behaviour remains in many respects poorly understood. They also

interact with each other in complex ways. Consequently, whilst there are many studies that have identified biological correlates of offending, none of these can be taken as evidence for biological causes.

(21)

The Biological Perspective: Aggression and Free Will

Is aggression part of human nature?

Cases derived from ‘Mind Machine: The Violent Mind’ (BBC, 1988)

Is Aggression Part of Human Nature?

Aggression is certainly part of human life. Apart from everyday reports of violent crime we are a pretty warlike species. Since the end of World War II there have been about 30 days on which a war was not taking place

somewhere on the planet. Many have argued that the machinery of aggression has been built into our brains by evolutionary processes. It is certainly the case that we are not alone in being an aggressive species. Other animals with which we share a common ancestry, such as

chimpanzees, have also been observed to engage in intergroup aggression that looks a great deal like warfare.

So it could be argued that aggression and violent behaviour are an inevitable fact of life, over which we have no control. However, when we talk about each other and ourselves this idea is rarely given houseroom. Rather, we talk about ourselves as if we had the capacity to choose our actions: free will. Often, when we view examples of aggression and violence we hold the

person that carried them out morally responsible for the harm that arises from their actions.

Free Will, Moral Responsibility and Psychology

The ideas of free will and moral responsibility are built into our legal system. Generally, the law says that, on order for a crime to have been committed there must be actus reus – the guilty act – and mens rea – the guilty mind. In

other words, in law, moral responsibility is generally predicated on intent. At this point, Psychology and the law tend to part company. The law is predicated on the notion of free will. As we have seen over the past few weeks, Psychology frequently is not. Although for different reasons, most schools of psychological thought take the model of the physical sciences, where all effects have physical and mechanistic causes. Consequently, most psychological perspectives deny the existence of free will. For example, the behaviourists believe that our behaviour is determined by our conditioning and history of reinforcement. The Freudians believe that our behaviour is determined by unconscious processes. The biological approach is

deterministic in a similar way. When explaining human behaviour, biological psychologists use biological concepts. Their theories tend to emphasise:

 Genetics, and the possibility that certain behavioural tendencies are inherited.

 The nervous system, and the way that certain behaviours are linked to the functioning of particular parts of the brain.

(22)

 Chemical influences, and the way that substances such as

neurotransmitters and hormones (and their analogues, drugs and toxins) can alter the functioning of the brain.

What these have in common is that they are governed by physical processes that follow the laws of physics. In other words, from the point of view of a biological psychologist, the brain, which causes all of our behaviour, is a machine. An incredibly complex machine, but a machine nonetheless. And machines do not have free will. Their behaviour unfolds in a predictable way according to knowable laws and principles. So, aggression and violence are behaviours produced by the brain, and the brain operates in a mechanistic way. This type of deterministic and reductionist view precludes the idea of choice or agency and, consequently, makes the idea of moral responsibility irrelevant.

Nonetheless, we carry on (biological psychologists included) as if people do make choices about their actions and can be held responsible for the

consequences. However, there are circumstances where we (and the law) accept that biological processes have overridden ‘free will’, and that a person should not be held liable for the consequences of their actions.

Case 1 – Dawn

Dawn has episodes of erratic behaviour, and occasionally has violent fantasies. For example, she reports that on one occasion, whilst preparing dinner for her family, she started to experience great feelings of resentment against her husband. She was using a knife at the time and started to fantasise about stabbing him. On other occasions she has behaved in ways that, whilst not aggressive, are certainly bizarre. For example, she has been known to remove all her clothing and wander off into the garden in the middle of the night.

Discussion question: what immediate assumptions might we make about Dawn’s behaviour? If Dawn had stabbed her husband, would we hold her responsible?

Dawn is diabetic. Under some circumstances, her blood-glucose level can drop dangerously low, and she becomes hypoglycaemic. This state is dangerous for the person, as it can lead to coma and ultimately, death. In Dawn’s case it can also lead to abnormalities of psychological functioning. In her words, ‘It feels as if I’m moving through a tunnel…everything disintegrates but at the same time becomes intensely vivid…it’s lethal…your own body behaves on its own…your mind goes absent’.

As Dawn’s blood-glucose level drops, her cerebral cortex, the part of the brain responsible for planning, reasoning and ‘rational’ behaviour starts to shut down. However, her limbic system, a more primitive part of the brain involved with responding to threats and aggressive behaviour, carries on functioning. Its aggressive impulses can no longer be controlled by the cerebral cortex

(23)

Discussion question: most people agree that, if Dawn did commit a violent act whilst hypoglycaemic, she should not be held morally responsible.

However, consumption of a large amount of alcohol can have very similar effects on the brain to hypoglycaemia: impaired cortical functioning leading to uninhibited behaviour. Much violent crime is alcohol related. If Dawn attacked her husband whilst drunk would we draw the same conclusion? After all, the biological processes are very similar.

Cases 2 and 3 – Arthur and Colin

Arthur has recurring dreams about being attacked. The dreams are accompanied by feelings of terror and great anger. Sometimes he has woken up from such dreams to find that he has assaulted his wife. He has hit and kicked her and once tried to strangle her, apparently whilst asleep. Colin also suffered from recurring violent dreams following a head injury he

sustained in a car accident. He awoke one day to find he actually had killed his wife.

Discussion question: if someone claims to have committed a criminal act, such as homicide, whilst asleep, is it our first instinct to believe them? How could we test their claim?

Frontal cortex (thinking, planning, rational behaviour) is inhibited whilst the person sleeps. At times, sensory and motor cortex (areas associated with movement and sensation) are highly stimulated, leading to the experience of dreaming. The limbic system (emotion) is often activated whilst we dream. The pons paralyses us to stop us from acting out our dreams. Psychologists who examined both Arthur and Colin agreed that both were suffering from sleep disorders. In Arthur’s case, he suffers from night terrors. In this

disorder, people tend to have very strong negative emotions whilst asleep, on which they are inclined to act. They may lash out in their sleep and are sometimes violent, but often have no recollection of this when they wake up. In Colin’s case, it was decided that he was suffering from REM behaviour disorder. This is rarer and often more dangerous than night terrors. If

generally occurs when the pons is damaged. The pons paralyses us when we go into REM sleep, and this stops us from acting out our dreams. If the pons is damaged, then people can engage in complex – and occasionally violent – behaviour, whilst asleep.

In Colin’s case, he attacked and killed his wife. REM behaviour disorder sufferers are not usually dangerous to anyone but themselves. There have been a number of cases where sufferers have injured themselves whilst acting out elements of their dreams. The following case description is fairly typical:

A Patient came to a local sleep laboratory because he was keeping his family awake all night with shouting and acting out his dreams. His wife was forced to sleep in a different room not only so she could get some sleep, but also because she feared for her own safety. The patient managed to fall out of bed on a nightly basis, often injuring himself in the process. He was a war veteran and would often dream he was trying to avoid enemy attack. Thinking that it

(24)

would help, he purchased a hospital bed with railings. Still he managed to climb out over top of the railings and fall to the floor. The patient then had to resort to sleeping on a mattress on the floor. When he was monitored in the sleep lab during overnight testing, and he entered REM sleep, muscle tone activity was detected when it should have been absent. Talking, laughing, shouting, flailing of the arms and kicking of the legs were all observed. He nearly fell head first out of the bed on several occasions. The patient has been prescribed a medication which will make his muscles relax during REM sleep so that he will no longer act out his dreams. To update, the patient is now doing fine and is sleeping in his own bed again. (Orr, 2003)

The court accepted Colin’s defence that he was not acting of his own free will when he killed his wife, and he was acquitted.

Case 4 – David

David was a gardener with a reputation for being a very passive and mild mannered person. Over a very short period of time his behaviour changed markedly. He became very bad tempered and lashed out at family members, something they describe as being completely out of character for him. Whilst working at a client’s house he got into an argument with her and beat her to death. The prosecution argued that he had lost his temper and murdered her in a fit of rage.

The jury believed the prosecution and he was sentenced to life in prison.

Discussion question: how significant is it that David was reported to be acting ‘out of character’ in the days immediately prior to the murder?

At his trial, David’s defence argued that his behaviour had not been under his control. Shortly before his behaviour changed, David had accidentally been exposed to a high concentration of an organophosphate pesticide called

carbaryl. Carbaryl is a substance that can have an effect on the nervous system. In fact, it behaves similarly to many nerve gasses (chemical weapons that exert their effect by impairing the functioning of the NS). Carbaryl works by affecting the way that brain cells communicate with each other. This communication occurs at synapses, little gaps at the junctions of nerve cells. Neurones (nerve cells) release chemicals to send signals across the gap. These chemicals are called neurotransmitters. After it has been released, the neurotransmitter is broken down to prevent it from stimulating the post-synaptic neurone for any longer than necessary. Carbaryl prevents the breakdown of a neurotransmitter called acetylcholine (ACh). Exposure to high doses of carbaryl causes ACh to build up in the brain and other parts of the nervous system.

One of the many parts of the brain that uses ACh is a structure called the hypothalamus. This structure is important in a huge range of behaviours including eating, drinking, sex and aggression. Electrical stimulation of

specific parts of the hypothalamus can cause a rat to show signs of rage and attack behaviour.

(25)

David’s Defence argued that the carbaryl to which he had been exposed had affected the functioning of his hypothalamus. A small provocation would have started the parts of the hypothalamus associated with attack behaviours to start firing. However, because of the carbaryl, they would not have been able to stop, leading to uncontrolled rage and aggression.

Case 5 – Sandie

Sandie came from a very stable background and seemed quite normal whilst growing up. However, around the age of 15 years her behaviour started to become erratic and violent. Her aggressive acts were quite unusual. For example, she once threw a bottle through a police station window, then waited around outside for the police to come and arrest her. Her extremely violent behaviour brought her to the attention of psychiatrists, who declared that she was untreatable.

Eventually, Sandie found herself accused of stabbing a co-worker seventeen times. She had no recollection whatsoever of doing this.

Discussion question: do we view violent women differently to violent men?

Whilst examining her history when preparing her defence, her defence team noticed that her violent outbursts seemed to occur at regular monthly

intervals. Further examination revealed that her aggressive acts almost invariably coincided with her menstruation.

One possible explanation for this concerns the effect of progesterone on the limbic system. Progesterone is a hormone that is released during pregnancy and also just prior to menstruation. As we have already seen, the limbic system plays a role in threat and attack behaviour. Progesterone seems to have a calming effect on the limbic system. It is possible that Sandie’s aggressive behaviour stemmed form the fact that she released too little progesterone during the relevant phase of her menstrual cycle. What is certain is that when she was treated with drugs to increase her progesterone levels, Sandie became markedly less aggressive. Rather than being

imprisoned, she was given a probation order conditional on continued treatment.

Discussion question: psychologists are divided on the question of whether pre-menstrual syndrome is a useful construct. Apart from the ambiguous nature of some of the evidence, some have argued that it paints a picture of women as ‘inherently pathological entities’ who are slaves to their biology. What do you think of this view?

Case 6 – Julie

Julie had a normal childhood, but in adolescence started to experience absence seizures, a form of epileptic fit. These got worse as she got older and she started to have full-blown seizures. Often, these were accompanied by intense feelings of panic and dread. Sometimes, Julie would start running,

(26)

as if to escape from something or someone. In her teens she started to carry a knife, just in case she found herself in a dangerous part of town after one of her ‘running attacks’.

On one occasion she went to see a film with her father, and had a panic seizure during the film. She ran off to the toilet to hide, and whilst in there started to have a hallucinatory episode in which her body became distorted. Unfortunately, in the middle of this attack another woman came into the toilet and accidentally brushed against her. In the grip of a terrible panic, Julie stabbed her in the heart and ran off (the woman survived, thanks to first aid administered by Julie’s father).

Julie was hospitalised rather than arrested, and the medical team treating her thought that her aggressive outburst and her epilepsy might be related. They implanted electrodes in different parts of her brain to try and find out more about her seizures. They discovered that she had an epileptic focus very close to her amygdale, part of the limbic system. Experimentally, they stimulated the area electrically. Julie suddenly entered into an absence seizure. Following on from this her face started to twitch as if she was very angry, and she suddenly flew into a frenzy of attack behaviour.

A surgeon operated on her brain to remove the focus from her amygdale. Since then she has experienced no further aggressive outbursts or attacks of rage.

Discussion question: what are the issues raised by treating aggressive behaviour by operating on people brains?

(27)

Theories of offending and key debates in psychology

Are criminals born or made? And, how can we detect them? This question has baffled psychologists, sociologists and criminologists for many years, and is the very essence of trying to establish the nature of criminality. The born or made argument, known as the “NATURE VERSUS NURTURE” debate, asks whether criminality is due to genetic factors (NATURE), and therefore

unavoidable, or whether it is the product of social situationalism, environmental surroundings and other external factors (NURTURE).

Theories that explain offending in terms of only one factor are

REDUCTIONIST, because they reduce complex human behaviour down to one explanation without considering the interaction of the many factors that can influence a person to become an offender (and more often than not the complex interaction between nature and nurture). In direct contrast, those which consider many factors are HOLISTIC.

Some explanations of offending imply that offenders do not have any choice in becoming such- their behaviour is caused by factors beyond their own control. These theories are criticised on the grounds that they are

DETERMINISTIC, and bring into question the role of personal responsibility; if behaviour is beyond a person’s control, should they be held responsible for their actions? Those theories which suggest that behaviour is a product of the decisions and choices of each individual advocate FREEWILL.

Definition Approaches NATURE NURTURE FREEWILL DETERMINISM REDUCTIONISM HOLISM

(28)

Index ADHD, 20 affectionless psychopathy, 4 ANS, 13, 20 atavistic, 7, 8 Bandura, 10, 11 Bowlby, 4 criminogenic, 11 defence mechanisms, 3, 4 deterministic, 21, 22 differential association, 10, 11 ectomorph, 8 ego, 3, 4 endomorph, 8 extraversion, 13 Eysenck, 13, 14, 15, 17 free will, 21, 22, 24 genetics, 13, 18 id, 3 Klinefelter’s Syndrome, 9 Lombroso, 7, 8, 9, 15, 18 media, 10, 11 Megargee, 4 mesomorph, 8 Mischel, 15, 17 nervous system, 13, 18, 21, 24 neuroticism, 13 Oedipus complex, 3 personality, 3, 8, 13, 14, 15, 17 psychodynamic, 3, 4, 5, 6 psychoticism, 13, 15 Rutter, 5 Sheldon, 8

single defective gene, 18

social learning theory, 10, 11, 12 socialisation, 14, 15

somatotypes, 8 superego, 3

Sutherland, 8, 10, 11 XYY, 9