There is plenty we don't know about criminal behavior. Most crime goes unreported so it is hard to pick out trends from the data, and even reliable sets of statistics can be difficult to compare. But here is one thing we do know; those with a biological predisposition to violent behavior who is brought up in abusive homes are very likely to become lifelong criminals.

Antisocial and criminal behavior tends to run in families, but no one was sure whether this was due mostly to social-environmental factors or biological ones. It turns out both are important, but the effect is most dramatic when they act together. This has been illustrated in several studies over the past six years which found that male victims of child abuse are several time as likely to become criminals and abusers themselves if they were born with a less-active version of a gene for the enzyme monoamine oxidize A (MAO-A), which breaks down neurotransmitters crucial to the regulation of aggression.

Researchers recently made another key observation; kids with this "double whammy" of predisposition and an unfortunate upbringing are likely to show signs of what's to come at a very early age. The risk factors for long-term criminality—attention deficit hyperactivity disorder, low IQ, language difficulties—can be spotted in kindergarten. So given what we now know, shouldn't we be doing everything to protect the children most at risk?

No one is suggesting testing all boys to see which variant of the MAO-A gene they have, but what the science is telling us is that we should redouble efforts to tackle abusive upbringing, and even simple neglect. This will help any child, but especially those whose biology makes them vulnerable. Thankfully there is already considerable enthusiasm in both the US and the UK for converting the latest in behavioral science into parenting and social skills: both governments have schemes in place to improve parenting in families where children are at risk of receiving poor care.

Some people are uncomfortable with the idea of early intervention because it implies our behavior becomes "set" as we grow up, compromising the idea of free will. That view is understandable, but it would be negligent to ignore what the studies are telling us. Indeed, the cost to society of failing to intervene—in terms of criminal damage, dealing with offenders and helping victims of crime—is bound to be greater than the cost of improving parenting. The value to the children is immeasurable.