91av

WHAT is evil? In today’s largely secular western society we may find it hard to define. But, historically, the question of evil has been a theological one. Generations of apologists have searched for ways to justify God in an imperfect world. In the Christian tradition, Saint Augustine distinguished between two forms of evil: “moral evil”, the evil humans do, by choice, knowing that they are doing wrong; and “natural evil”, the bad things that just happen – the storm, the flood, volcanic eruptions, fatal disease.

Whatever your views on God, the distinction is appealing. In our secular world, where crimes are punished, we need to believe that someone who commits an inexplicable act of violence or physical cruelty, such as torture, child abduction or murder, has freely chosen to do so. We accept that there may be mitigating circumstances, including diminished responsibility or mental illness. What we find most chilling are crimes that are planned, where offenders have a goal in mind and demonstrate control. Like Uma Thurman’s character in Quentin Tarantino’s film Kill Bill, people who commit such crimes are seen as unambiguously morally evil. “It’s mercy, compassion and forgiveness I lack,” she says, “not rationality.”

Thurman’s character “The Bride” can be seen as possessing what is now recognised by psychiatrists as a severe antisocial personality. Such people are not clinically mentally ill, and they usually know when they are doing wrong. But are they really evil? Neuroscience is beginning to tell us that while the cognitive, planning part of the brains of severely antisocial people works normally, other neural centres do not. The findings raise a series of thorny questions. Should courts view evidence of antisocial personality traits as mitigating a violent crime? Should medical treatment supplant punishment? Should psychiatrists be compelled to identify people with these mental abnormalities and incarcerate them for their own good and that of society? And, even, does moral evil exist, or is all evil “natural” according to Saint Augustine’s definition?

Whether violence is impulsive or premeditated, we want to know what makes people do it. Here science can move beyond the theological explanation of free will. Starting in the 1990s, Adrian Raine, a British psychologist at the University of Southern California, Los Angeles, pioneered the use of brain imaging to study people with antisocial personalities. His brain scans of impulsive, reactive killers show that these people have reduced activity in frontal regions of their brains, including an area called the orbitofrontal cortex. This area sits at the head of a chain of connected brain regions, including the amygdala, hypothalamus and peri-aqueductal grey, which are activated when we are threatened. People with damage to this area often display loss of control. They are inappropriately irritable and aggressive when crossed, and they find it hard to alter their response to a situation when circumstances change and it would be in their interests to modify their behaviour.

It makes sense that the brains of impulsive killers show abnormally low activity in the frontal cortex. But not all killers lack control. Raine has found that serial killers – people who commit a series of murders and evade capture, for a time at least – do not have reduced frontal activity. Their executive, planning systems are intact and they do not lose control. So what is it that such calculating killers lack?

According to James Blair, another pioneer in this field, who works at the US National Institute of Mental Health in Bethesda, Maryland, the defect lies in the amygdala, a region of the temporal lobe intimately involved in the immediate, implicit processing of emotional cues. The amygdala is activated when we detect unpleasant emotions such as fear or sadness in others, and it is involved in aversive conditioning, when we change our behaviour as a consequence of punishment. Blair and others have found that people who commit premeditated violent acts fail to detect the fear shown by others and do not learn by punishment. The amygdala also initiates the “fight or flight” response that we experience as a raised heart rate and sweating palms. And, likewise, calculating killers do not react when exposed to situations that would distress most people. What they seem to lack is empathy.

When normal people are shown pictures of people’s faces expressing distress, the amygdala is activated. It sends signals to the hypothalamus and other brain regions that prompt our bodies to react with the characteristic racing heartbeat and sweaty palms. Only later, when the brain’s higher cortical regions are engaged, does the emotional response become conscious. Ray Dolan from the Institute of Neurology in London has found that we respond even when our attention is diverted to other matters and we see pictures of distressed faces only at the periphery of our vision. Empathy, concludes Dolan, is automatic and visceral. It is beyond our control. When we feel a wave of emotion at someone’s suffering, or revulsion at the sight of their wounds, we are communing with them, whether we like it or not.

But violent individuals do not react in this way. Is their lack of empathy a cause or an effect of their behaviour? Do such people start life without empathy, does it fail to develop, or do they lose it through wrongdoing? Opinions vary. Raine and others have found that children who later grow up to be violent have abnormal visceral responses such as a relative lack of palm sweating. Twin and adoption studies point to a substantial contribution of heredity to criminal, antisocial conduct, although the genetic effect is stronger for acquisitive and impulsive crime than for premeditated crimes against other people. But there is also evidence that antisocial behaviour is the result of experiences in early life. Jonathan Pincus, a neurologist from Georgetown University School of Medicine in Washington DC, has found that people accused of murder often suffered abuse as children and many exhibit neurological impairment and mental illness.

Undoubtedly, we have much more to learn about the neuroscience of violent and antisocial behaviour, but what we know already must surely influence our views on evil. As our scientific knowledge expands, it seems that the space for moral evil contracts. If we can find explanations for the evil people do, then are we not replacing moral evil, which is freely undertaken, with natural evil, which is beyond our control? To the volcano and the virus might we now add the dysfunctional amygdala and the abnormal orbitofrontal cortex? In which case, perhaps violent offenders are not, ultimately, responsible for their actions and our legal system should change the way it treats them.

We need to be careful. First, we should ask whether the “causes” of antisocial conduct are so prescriptive that they make violence inevitable. If the psychologically disturbed person has no choice but to commit violent acts then they may be absolved of responsibility. In that case, however, you would expect their behaviour to be rather predictable. Yet, as my colleagues Mike Hunter and Gemma Harpin at the University of Sheffield and I have pointed out, clinical evidence and experimental findings do not support this idea (Current Opinion in Psychiatry, vol 15, p 519). Indeed, a recent meta-analysis by Alec Buchanan and Morven Leese from the Institute of Psychiatry in London found that mathematical models were very poor at predicting the conduct of antisocial offenders.

Then there is the problem of over-generalisation. Some commentators have suggested that the architects of the Holocaust and other genocides must have been abused as children – why else would they have committed such appalling acts? But all the available evidence contradicts this. James Waller at Whitworth College in Spokane, Washington, has trawled the literature on what he calls “extraordinary evil”. He and other social scientists find perpetrators of evil who look remarkably normal in psychiatric terms. If we ascribe mitigating circumstances to all violence we risk losing sight of a truth about human nature: that the potential for evil lives within all of us.

How we create ourselves

Aristotle said: “Men become builders by building and lyre-players by playing the lyre; so too we become just by doing just acts, temperate by doing temperate acts, brave by doing brave acts.” Modern thinkers agree with the idea that, in part, we create ourselves through our actions. The way the antisocial brain is configured may limit a person’s choices from the outset, but, argues forensic psychiatrist Paul Mullen from Monash University in Victoria, Australia, it is by choosing evil that we seal our own fate. Mullen sees the antisocial personality as the outcome of a developmental disorder of ethical action. Someone who does not reflect upon the consequences of their actions while growing up – or who is not provided with a parent who will do so for them – creates their own antisocial destiny.

If antisocial conduct is a pattern that emerges early in life, we may be able to intervene to break the cycle. Raine and his colleagues have recently published the results of ground-breaking research that suggests a way forward (American Journal of Psychiatry, vol 160, p 1627). Working on the Indian Ocean island of Mauritius, they put 83 children through an “enrichment” programme. Between the ages of 3 and 5, these children were given extra physical exercise and education and an improved diet that included milk, fruit juice, a hot meal of fish, chicken or mutton, and a salad each day. The researchers then followed them and a control group of 355 others through childhood and adolescence. They found that by the age of 23 the children in their programme were significantly less likely to have engaged in antisocial conduct on their own, at least on the basis of their self-reports and court records.

Whether convicted criminals can benefit from similar intervention is more debatable, although there is evidence that nutritional supplements may reduce antisocial conduct in prison (91av, 16 November 2002, p 38). Nevertheless, much forensic psychotherapy is predicated on the notion that offenders can change. These people are not immune from feelings such as contempt for themselves and their actions, fear of being found out and pain at being rejected by their families. Given psychiatric help and time in an institution, they may reflect and come to see the damage they have done. But changes in behaviour are hard to assess objectively, particularly while an offender is locked up and has limited scope to commit acts of violence. Many psychiatrists want to believe in the possibility of rehabilitation and redemption for violent offenders but the truth is that we really don’t know whether or not treatment works.

Although it is increasingly apparent that these people need help, we cannot ignore the fact that they have committed terrible crimes for which they should be detained. But what about the man or woman who has not yet offended, but who suffers the same neurological abnormalities as the violent criminal? Should these people also be incarcerated? Should they be removed from society and put in psychiatric “quarantine” to protect themselves and their potential victims? Advances in neuroscience make it increasingly easy to identify such people. And the notion that they might be locked away is not some futuristic Orwellian nightmare; it has already been mooted by the UK government in proposals for the new Mental Health Act. If it becomes law, psychiatrists like myself may be called upon to detain people whom we judge to be at risk of perpetrating violent crimes. That prospect is disturbing, not least because of the difficulty in predicting how much risk such individuals pose.

In their recent meta-analysis, Buchanan and Leese looked at 21 studies that aimed to predict subsequent offending in anything from 53 to 3267 patients (The Lancet, vol 15, p 519). For each study, the researchers calculated two indices: sensitivity, or the ability to correctly predict future offending; and specificity, or the ability to exclude false positives. Values varied considerably, but the mean for all the studies was 0.52 for sensitivity and 0.68 for specificity. A sensitivity of 0.52 means that for every 10 people with personality disorder who went on to reoffend, five would be identified and detained and five would be missed. A specificity of 0.68 means that for every 10 people who did not go on to reoffend, seven would be identified and released and three would be detained unnecessarily.

Buchanan and Leese concluded that to prevent one violent act over a year, six potential perpetrators would need to be detained for the duration. As well as infringing the liberty of innocent people, this would clearly be a waste of public resources. And far from keeping potential killers off the street, the introduction of psychiatric quarantine would demonise the innocent and probably deter those who suffer from neurological abnormalities from seeking medical help.

It is difficult not to conclude that although neuroscience gives us a greater insight into the minds of killers, it only complicates the decisions we must make when dealing with the evil that humans do. Faced with questions of culpability, violence and victimhood, our moral sense works rather like our visual system observing the illusion of the Necker cube. When first we look we see one form, but as we stare for longer it reverses: the corner that was towards us now recedes into the plane. Similarly, when we see the perpetrator of a violent act we see a villain – until we know their story. Then a victim may emerge, and it becomes easier to minimise their crime.

Yet we must not lose sight of the evil act itself. Our responsibility as moral agents is not to look away. It is to look clearly and to hold two thoughts in mind: what this man did was wrong, but he is still a man.