Outrage at violence is the easy way out

We call a random machete attack 'evil' but we should think scientifical ly, argues Colin Tudge

The machete attack in Wolverhampton is shocking because it happened - but also because nobody seems to have been able to say anything enlightening about it; and because we are unlikely to learn from this incident any more than we have from comparable episodes in the past. To judge from immediate responses, our understanding of our own species, and of what we are capable, seems hardly to have improved since the Middle Ages; it is as if bewilderment were the only acceptable stance. There has been much talk of "evil" - a concept that seems expressly intended not to throw light but to frighten, invoking images of dark and unfathomable forces. But it matters that we do not understand - and are not, apparently, improving our understanding. The world is growing more crowded, tense, fluid and heterogeneous; the forces that induce violence can hardly be growing less. Understanding will not by itself lead to solutions; but nothing else can possibly do so.

Yet there is a clutch of disciplines whose specific task it is to explain ourselves to ourselves: why in general we behave as we do; why some of us sometimes behave outrageously; what we are capable of. Psychology deals with us each individually; sociology with our interactions; criminology with our perceived misdemeanours. Why don't they tell us more, when things go as awry as this? Perhaps because, these past decades, the conventional disciplines have been missing some important tricks. Perhaps it is time to try a different approach altogether. And perhaps - unpopular though such a suggestion will be on both sides of the Atlantic - the general approach that ought to be given a try is that of bona fide science, which means the testing of testable hypotheses; and perhaps the hypotheses that demand most urgently to be tried are those of modern biology.

It is a fashionable and reasonable generalisation that we are shaped by our "nature"- which effectively means our genetic inheritance - and by our "nurture", which means upbringing. Common sense and observation suggest that both are significant, each interacting with the other. But, like Swift's Big-endians and Little-endians, intellectuals over the past few decades have divided themselves tidily and often acrimoniously into two camps: those who have implied that our attitudes and abilities derive more or less exclusively from our evolutionary inheritance, and those who argue that our minds - our "cultural overlay" - must overwhelm all influences that are merely genetic. Both in their extreme manifestations have been grotesque. Many otherwise right-thinking British intellectuals, H G Wells and the Webbs among them, argued earlier this century in favour of eugenics, sterilising people who don't come up to scratch so as to strengthen the "race", although it was left to Hitler to illustrate the shortcomings. But other intellectuals who are at least equally right-thinking have argued in recent decades, for example, that there are no intrinsic psychological differences even between males and females and that all is due to upbringing - a conceit that commonly dies a death when their little girls, dutifully decked out in Arsenal strip, feed their dollies just the same. Both schools perhaps managed to demonstrate that there is nowt as daft as intellectuals.

In the past few decades, however, the school which feels that it is at least sensible to look at the evolutionary underpinnings of human behaviour has advanced beyond recognition. The agenda now is not simply to ask whether genes influence what we do and how we feel, for that point is established well beyond argument. It is to ask why we have the particular proclivities that we have. The question is posed in Darwinian terms - "Why should natural selection have favoured this particular attitude or predilection?". By contrast, the "all-is-culture" school seems to be going nowhere, trotting out the same preconceptions year after year. The difference in the end lies between science and non-science: those who are exploring human "nature" are framing testable hypotheses; and those who feel that all is culture, on the whole are not.

In 1979 Richard Gelles and Murray Straus, two of America's leading criminologists, declared that "the family is the most frequent single locus of all types of violence from slaps to beatings, to torture to murder". This has become a dogma of criminology and a much-loved media soundbite: that people are in greatest danger from their own families. But is it true, and if so, what does it mean? At first sight it abnegates the powerful modern Darwinian notion of kin selection, which says that close relatives might be expected to protect each other, since copies of the gene that says "Look after your close relatives!" would also, probably, be contained within the very individuals who were being looked after, so that the gene in question simply has a pact with facsimiles of itself.

And, indeed, when you look more closely at the statistics you find - surprise surprise - that brothers do not generally murder brothers, and parents do not commonly murder their own children, as Martin Daly and Margo Wilson point out in Homicide (Aldine de Gruyter, New York, 1988). In fact, people may murder members of their own family more than they murder strangers, but only in special circumstances do they murder blood relatives. Furthermore, these circumstances are predictable in Darwinian terms; for example, brothers might kill brothers when social norms dictate that the father's legacy is split, but that neither beneficiary inherits enough to survive. Husbands do kill wives - but a common criterion for selecting spouses is that they are not "kin", in the true genetic sense. In-laws kill in-laws - frequently. But in-laws are not blood relatives. Step-parents - which almost invariably means step-fathers - kill step- children 100 times more often than they kill their own children; and (so closer studies show) they kill their own children mainly when they feel they have reason to doubt their own paternity. In practice, the detailed patterns of murder within families precisely follow Darwinian predictions. The generalisation - "families are dangerous" - is vindicated only if we leave one supremely significant factor out of account: that we are genetically related to some members of our "family", but no more closely related to others genetically than we are to any stranger.

We may feel instinctively that constant contact between people should alone increase the chances of violence between them, if only because with no contact there can be no violence at all. But the contact within families is at least as great between genetic kin as with non-genetic "kin". Yet violence flares up mainly between people who are in contact and do not share common genealogy.

Statistics to do with violence in families are supposed to be useful in helping us to predict who is liable to behave in what ways towards whom, and what should be guarded against. That is why sociologists and criminologists study them. Wetake it as self-evident that such information is of value. Yet the data is extremely deceptive unless we differentiate between blood relatives and non-blood relatives. That conventional criminology did not perceive the difference between violence to biological children and to step-children is shocking. But until Darwinian biologists suggested the difference could be significant, the original data stood, and so did the dogma that sprang from it.

You do not have to be a Darwinian, though, to see that people might differentiate between their blood relatives and their non-blood relatives. Our literature and mythology have many battered step- children. So why did the sociologists who collected the data not take "blood relationship" into account? Because of preconception. Their dogma told them that all is culture - and indeed that it was politically suspect to suppose otherwise and, therefore, look for genetic influence, even in such an obvious case.

WHAT can Darwinian thinking tell us about events such as Wolverhampton or Dunblane - and, in particular, how to anticipate such tragedies? For the time being, perhaps not much. To explain aberrant behaviour we must first understand what is normal, and evolutionary studies even of normal behaviour are still novel.

But Darwinian theorists are beginning to look at pathologies, and are throwing light on what before was quite opaque. Thus Simon Baron-Cohen at Cambridge University describes in his book Mindblindness how autism, that most mysterious condition, is a special case of a person's inability to look at the world through another's eyes - an evolved ability vital to social cohesion. From this insight has developed a simple, reliable and early method of diagnosis, based on whether or not the child will follow the gaze of another, to look at what others are looking at. Similarly, Myrna Gopnik of McGill University is now showing how specific, evolved disabilities lead to various language difficulties; again with direct relevance to diagnosis and treatment. Pathologies and inabilities of all kinds are beginning to come under Darwinian scrutiny with results that are already bearing fruit.

Finally, too, some Darwinian theorists are at least beginning to look at psychopathologies, of the kind that that generate violence. It will be very surprising, in the light of progress so far, if they do not prove enlightening and helpful. It is an encouraging thought that an understanding of ourselves which could help us to shape a safer and more humane society is on the agenda. But up to now, for the most part, we have been tackling the problems in the wrong way.