Normal brain vs. murderer's brain
© Public DomainScans of a normal brain, left, beside that of murderer Antonio Bustamante, who was spared the death penalty after a jury was shown these pictures.
In 1987, Adrian Raine, who describes himself as a neurocriminologist, moved from Britain to the US. His emigration was prompted by two things. The first was a sense of banging his head against a wall. Raine, who grew up in Darlington and is now a professor at the University of Pennsylvania, was a researcher of the biological basis for criminal behaviour, which, with its echoes of Nazi eugenics, was perhaps the most taboo of all academic disciplines.

In Britain, the causes of crime were allowed to be exclusively social and environmental, the result of disturbed or impoverished nurture, rather than fated and genetic nature. To suggest otherwise, as Raine felt compelled to, having studied under Richard Dawkins and been persuaded of the "all-embracing influence of evolution on behaviour", was to doom yourself to an absence of funding. In America, there seemed more open-mindedness on the question and, as a result, more money to explore it. There was also another good reason why Raine headed initially to California: there were more murderers to study than there were at home.

When Raine started doing brain scans of murderers in American prisons, he was among the first researchers to apply the evolving science of brain imaging to violent criminality. His most comprehensive study, in 1994, was still, necessarily, a small sample. He conducted PET [positron emission tomography] scans of 41 convicted killers and paired them with a "normal" control group of 41 people of similar age and profile. However limited the control, the colour images, which showed metabolic activity in different parts of the brain, appeared striking in comparison. In particular, the murderers' brains showed what appeared to be a significant reduction in the development of the prefrontal cortex, "the executive function" of the brain, compared with the control group.

The advancing understanding of neuroscience suggested that such a deficiency would result in an increased likelihood of a number of behaviours: less control over the limbic system that generates primal emotions such as anger and rage; a greater addiction to risk; a reduction in self-control; and poor problem-solving skills, all traits that might predispose a person to violence.

Even two decades ago, these were difficult findings to publish, however. When Raine presented a far less controversial paper in 1994 to a peer group, one that showed a combination of birth complications and early maternal rejection in babies had significant correlation with individuals becoming violent offenders 18 years later, it was denounced as "racist and ideologically motivated" and, according to Nature magazine, was simply further strong evidence that "the uproar surrounding attempts to find biological causes for social problems will continue". Similarly, when, 15 years ago, at the urging of his friend Jonathan Kellerman, the child psychologist and crime writer, Raine put together a proposal for a book on some of his scientific findings, no publisher would touch it. That book, The Anatomy of Violence, a clear-headed, evidence-based and carefully provocative account of Raine's 35 years of study, has only now appeared.

The reason for this delay seems mired in ideological enmities. For all Raine's rigour, his discipline of "neurocriminology" still remains tarnished, for some, by association with 19th-century phrenology, the belief that criminal behaviour stemmed from defective brain organisation as evidenced in the shape of the skull. The idea was first proposed by the infamous Franz Joseph Gall, who claimed to have identified over- or underdeveloped brain "organs" that gave rise to specific character: the organ of destructiveness, of covetousness and so on, which were recognisable to the phrenologist by bumps on the head. Phrenology was widely influential in criminal law in both the United States and Europe in the middle of the 1800s, and often used to support crude racial and class-based stereotypes of criminal behaviour.

The divisive thinking was developed further in 1876 by Cesare Lombroso, an Italian surgeon, after he conducted a postmortem on a serial murderer and rapist. Lombroso discovered a hollow part of the killer's brain, where the cerebellum would be, from which he proposed that violent criminals were throwbacks to less evolved human types, again identifiable by ape-like physical characteristics. The political manipulation of such hypotheses in the eugenics movement eventually saw them wholly outlawed and discredited.

As one result, after the second world war, crime became attributable to economic and political factors, or psychological disturbances, but not to biology. Prompted by advances in genetics and neuroscience, however, that consensus is increasingly fragile, and the implications of those scientific advances for law - and for concepts such as culpability and responsibility - are only now being tested. He draws on a number of studies that show the links between brain development, in particular - and brain injury and impairment by extension - and criminal violence. Already legal defence teams, particularly in the US, are using brain scans and neuroscience as mitigating evidence in the trials of violent criminals and sex offenders. In this sense, Raine believes a proper public debate on the implications of his science is long overdue.

Raine was in part drawn to his discipline by his own background. In the course of scanning his murderers, Raine also examined his own PET profile and found, somewhat to his alarm, that the structure of his brain seemed to share more characteristics with the psychopathic murderers than with the control group.

He laughs quickly when I ask how that discovery felt. "When you have a brain scan that looks like a serial killer's it does give you pause," he says. And there were other factors: he has always had a markedly low heart rate (which his research has shown to be a truer indicator of a capacity for violence than, say, smoking is as a cause of lung cancer). He was plagued by cracked lips as a child, evidence of riboflavin deficiency (another marker); he was born at home; he was a blue baby, all factors in the kind of developmental difficulties that might set his own researcher's alarm bells ringing.

"So," he says, "I was on the spectrum. And in fact I did have some issues. I was taken to hospital aged five to have my stomach pumped because I had drunk a lot of alcohol. From age nine to 11 I was pretty antisocial, in a gang, smoking, letting car tyres down, setting fire to mailboxes, and fighting a lot, even though I was quite small. But at that age I burnt out of that somehow. At 11, I changed schools, got more interested in studying and really became a different sort of kid. Still, when I was graduating and thinking 'what shall I research?', I looked back on the essays I'd written and one of the best was on the biology of psychopaths; I was fascinated by that, partly, I think, because I had always wondered about that early behaviour in myself."

As Raine began to explore the subject more, he began to look at the reasons he became a researcher of violent criminality, rather than a violent criminal. (Recent studies suggest his biology might equally have propelled him towards other careers - bomb disposal expert, corporate executive or journalist - that tend to attract individuals with those "psychopathic" traits.) Despite his unusual brain structure, he didn't have the low IQ that is often apparent in killers, or any cognitive dysfunction. Still, as he worked for four years interviewing people in prison, a lot of the time he was thinking: what stopped me being on their side of the bars?

Raine's biography, then, was a good corrective to the seductive idea that our biology is our fate and that a brain scan can tell us who we are. Even as he piles up evidence to show that people are not the free-thinking, rational agents they like to imagine themselves to be - entirely liberated from the limitations set by our inherited genes and our particular neuroanatomy - he never forgets that lesson. The question remains, however, that if these "biomarkers" do exist and exert an influence - and you begin to see the evidence as incontrovertible - then what should we do about them?

Perhaps we should do nothing, simply ignore them, assume, when it comes to crime, that every individual has much the same brain, the same capacity to make moral choices, as we tend to do now. As Raine suggests: "The sociologist would say if we concentrate on these biological things, or even acknowledge them, we are immediately taking our eyes off other causes of criminal behaviour - poverty, bad neighbourhoods, poor nutrition, lack of education and so on. All things that need to change. And that concern is correct. It is why social scientists have fought this science for so long."

The implication of neurocriminology, though - where it differs from the crude labelling of phrenology, say - is that the choice it presents is not an either/or between nurture and nature, but a more complex understanding of how our biology reacts with its environment. Reading Raine's account of the most recent research into these reactions, it still seems to me quite new and surprising that environmental factors change the physical structure of the brain. We tend to talk about a child's development in terms of more esoteric ideas of mind rather than material brain structures, but the more you look at the data the clearer the evidence that abuse or neglect or poor nutrition or prenatal smoking and drinking have a real effect on whether or not those healthy neural connections - which lead to behaviour associated with maturity, self-control and empathy - are made. The science of this is called epigenetics, the way our environment regulates the expression of our innate genetic code.

One result of epigenetics might be, Raine suggests, that "social scientists can actually win from this. I mean, if a child experiences a murder in his or her neighbourhood, we have found that their test scores on a range of measures go down. There is something happening in the brain as a result of that experience of violence to affect cognition. So social scientists can have their cake and eat it. They can say look, we can prove that these environmental social factors are causing brain impairment, which leads to some real, measurable problems."

One difficulty of embracing this "epigenetical" idea of crime is the degree to which such factors should be taken into account in courts of law. There have been several landmark cases in recent years in which particular neurological disorders caused by blows to the skull or undetected tumours have resulted in arguable changes in character and behaviour - and the violent or sexual crime is blamed on the disorder, not the individual. In most of these cases, it has been argued by the prosecution that brain imaging is prejudicial, that the brightly coloured pictures are too compelling to a jury and more emotional than scientific. But if neural scanning becomes more routine, and neuroscience more precise, will there not come a point where most violent behaviour - that of the Boston bombers, say, or the Newtown killer - is argued away in court as an illness, rather than a crime?

Raine believes that there might well be. He even likens such a shift to our change in perception of cancer, until fairly recently often deemed the "fault" of the sufferer because of some repressive character trait. "If we buy into the argument that for some people factors beyond their control, factors in their biology, greatly raise the risk of them becoming offenders, can we justly turn a blind eye to that?" Raine asks. "Is it really the fault of the innocent baby whose mother smoked heavily in pregnancy that he went on to commit crimes? Or if he was battered from pillar to post, or even if he was born with a, abnormally low resting heart rate, how harshly should we punish him? How much should we say he is responsible? There is, and increasingly will be, an argument that he is not fully responsible and therefore, when we come to think of punishment, should we be thinking of more benign institutions than prison?"

But then there is a further thought, that if you start to see criminality as a biological illness, where does a sense of retributive justice stand?

Raine himself was forced to face this dilemma when he became a victim of violent crime. As he recounts in his book, while on holiday in Turkey several years ago, a burglar entered his bedroom and in the struggle that followed tried to cut Raine's throat with a knife. He fought the attacker off, but when the following morning he was presented with two possible suspects by police, he admits to not only choosing the one who looked most like a thug [the man later admitted the crime, under duress], but also to wanting to visit on him the terror he had felt himself.

"I wasn't proud to discover I was a bit Jekyll and Hyde - perhaps we all are in that situation," Raine says when I ask him about his response. "The rational Dr Jekyll knew that if I took this man's brain scan and found he had prefrontal dysfunction, low resting heart rate, a background of neglect, then of course I should cut him some slack. With understanding comes mercy.

But the Mr Hyde, the emotional voice in my head, was saying nothing of the sort: he was saying, he cut my throat, I want to cut his. That event changed me from someone dead set against the death penalty to someone who wouldn't be ruled out of a jury on a capital case in America. I think now my mind will always go backwards and forwards on this, the scientific understanding of the causes of crime versus being a human in society with all these gut reactions to people who commit awful crimes."


If the neuroscience raises as many questions as it answers about culpability after a crime has been committed, what about its role in crime prevention? Here, the questions seem no less fraught.

One of them was posed a couple of years ago by the arch-inquisitor Jeremy Paxman of Shami Chakrabarti, director of Liberty, on Newsnight. "If science could predict with 100% certainty who was going to commit a violent crime, would it be legitimate to act before they commit that crime?"

Chakrabarti was in no doubt: "I would have to say that in a liberal society of human beings, and not animals, my answer to your question would be 'no'."

But if such intervention could prevent Newtown, you wonder, or Dunblane, would any of us be quite so certain? The fact is that the reality will always be a much greyer area because even the most nuanced neuroscience will never produce a perfect prediction of human behaviour. But is there a point at which the science - in identifying the possibility of repeat offending, for example - will be accurate enough to warrant routine scanning of those on the sexual offenders' register?

"The fact is," Raine says, "parole boards are making exactly these kind of predictive decisions every day about which prisoner or young offender we are going to release early, often with crummy evidence. At the moment, the predictors are social and behavioural factors, marital status, your past record. What is not used are biological measures. But I believe that if we added those things even now into the equation, we could only improve the prediction."

Raine cites two very recent brain-imaging studies to back this up. One is a study in New Mexico in which prisoners are scanned on release. "What they are discovering is that if the functioning of the anterior cingulate, part of the limbic system, is lower than normal before release, they are twice as likely to be reconvicted in the next three years. And that marker is more accurate a guide than all other social factors," Raine says. A second study apparently shows if a released prisoner has a significantly smaller volume in the amygdala, the almond-shaped part of the brain crucial for processing memory and emotion, he or she is three times more likely to reoffend." Now, this is only two studies, but what they are beginning to show is proof of concept, that if we added neurological factors into the equation we could do a better job at predicting future behaviour."

At the end of his book, Raine suggests various possible Orwellian futures of such science, an ethical "slippery slope" of interventions that ultimately imagines a society that assesses the biological risk of all individuals - a wide-scale version of We Need to Talk About Kevin - and preemptively locks up those at the extreme end of the curve (a sort of evidence-based Guantรกnamo). He by no means advocates any of it, though when I ask if he would have his own children, two boys of 11, scanned, he suggests he probably would.

"If there was the opportunity for screening at school or through a GP programme, would I do it? Well, if my kids had problems, as a parent I would want to know about them and I would want to know how I might deal with them. If you brought in such things as emotion regulation and impulse control, which we know are risk factors for behaviour, then to me, as a parent, I would sort of want to know what could be done to help with those."

It is perhaps not too wildly far-fetched to imagine that such scans will one day be as routine as immunisation programmes; the bigger question then will be how we begin to react to the results. Raine rather likes the idea of public health programmes as crime prevention: "The teenage brain is still very malleable. There is good evidence from randomised control testing that omega-3 [fish oil] has a positive effect on young offenders, and even mindfulness seems to improve behaviour and brain structures."

You can't help thinking: if only it were as simple as that.

- This article was amended on Sunday 12 May 2013. Adrian Raine is a professor at the University of Pennsylvania, not Philadelphia as we mistakenly said in the original article. This has been corrected.

- This article was amended on Monday 13 May. A paragraph that misrepresented the views of the neuroscientist David Eagleman has been removed. The paragraph implied that Eagleman believed that the possession of particular genes resulted in criminal behaviour. This is not his belief, in his words, "Genes are part of the story, but they're not the whole story. We are likewise influenced by the environments in which we grow up".