The largest survey of its kind ever conducted has revealed a strong genetic component to sexual offending, which appears to hold a significant influence on the risk of close male relatives going on to commit a sex crime.
So proclaimed the article in last week’s NZ Herald. Reporting on research published in the International Journal of Epidemiology, the piece stressed that the researchers “said the findings should not be used to excuse sex offending, to restrict the freedom of the male relatives of sex offenders, or to suggest that there are genes for rape or paedophilia.”
These particular findings may justify none of those measures – as Emily Underwood explains, methodological limitations mean that we need to be careful of leaping to any such conclusions based on this study. But can we be confident that genetic evidence will never be able to fill any such roles?
In a recent publication (pre-review version here: Genetic defense article, web version) Amir Bastani and I explored the possibilities for a genetic-based defence (full or partial). In particular, we considered the possibility of a volitional defence, based on some impairment to the defendant’s ability to control his urges. (The article, I should say, isn’t really an argument for such a defence, so much as an argument against a certain kind of argument against it. If you can follow that…)
In the article, we take as a case study the MAOA gene. This has – somewhat controversially – been linked to certain forms of antisocial and criminal behaviour. Although the precise causal mechanism by which the gene variant affects behavior is not certain, various theories have been advanced as to how it might work, including by affecting levels of certain neurotransmitters like dopamine and serotonin, or the volume of certain brain structures like the amygdala.
The most credible claims about MAOA are that it has some bearing on crimes of what are sometimes called “hot” aggression – unplanned, impulsive actions that inevitably result in the offender being caught and punished, as opposed to premeditated crimes from which the offender hopes to gain. Though it is still far from certain, evidence is emerging that there may be a genetic component in our ability to exercise self restraint, control our temper, or resist our urges.
We all have urges all the time, but the law – and common morality – quite reasonably expects us to control them, and punishes us when we act on urges that involve criminality. The mere fact that someone had deviant or violent urges would not, in itself, provide a defence. Rather, the defence would only apply to those whose capacity for self-restraint is significantly impaired, such that they are rendered incapable of resisting those urges. It’s a central tenet of criminal law that we don’t punish people for that which they could not avoid. Hence, we might think, punishing people for genuinely uncontrollable behaviour is fundamentally unjust.
It may not be a major surprise to learn that this line of argument is not particularly popular, whether with politicians, the general public, or many academics. For one thing, it can sound like a mechanism for “getting away with” horrible crimes. It wasn’t me, Your Honour, my genes made me do it.
While any defence can be exploited by cynical opportunists, though, it isn’t clear why we should be more suspicious of volition-based defences than of, say, self-defence or the sort of insanity defence (“cognitive” insanity) that we currently allow. So why are volition-based defences not presently accepted in New Zealand, either in the form of a volitional limb to our insanity defence, or a diminished responsibility partial defence?
One of the suggested reasons for rejecting such defences is the claimed impossibility of distinguishing those who genuinely could not control their urges, from those who merely chose not to do so. It’s certainly true that a jury considering such a defence would be denied direct access to the mental state of the accused. This, however, is true of any of the mental states that concern the criminal law. As Lord Bridge said in one famous English murder case: ‘you cannot take the top of a man’s head off and look into his mind and actually see what his intent was at any given moment.’
Faced with this epistemic challenge, the law doesn’t simply abandon attempts to determine what was known, or understood, or intended by the accused. Rather, it sets about a process of inquiry that involves drawing inferences from external evidence – how the accused behaved, what he said, his background – and piecing together the most plausible account of his behaviour.
In the article, Amir and I considered a couple of examples of this kind of inferential thinking. Basically, the court is being asked: given all of the evidence, what is the most likely explanation for the defendant’s conduct? That may not sound like a particularly scientific process, but it can be informed to some extent by scientific evidence. And we see no reason why genetic evidence couldn’t be considered in the mix, long with psychiatric/psychological and all of the other sorts of evidence routinely heard in criminal trials. The judge or jury would be free to accept it or reject it – there’s no requirement to accept the defendant’s version of events at face value – but it would be on the table to be considered.
The Swedish study doesn’t seem to offer any suggested causal pathway by which any particular gene might influence offending. It doesn’t, for instance, claim that these offenders had impaired volitional capacities. It just notes that genetic relatives of offenders are themselves more likely to offend. That in itself wouldn’t be enough to make out a “genetic defence.” But the prospect of behavioural genetics informing the way we deal with criminals is not entirely the stuff of science fiction.
Of course, someone excused on the basis of volitional impairment wouldn’t necessarily just be set free to offend again. Just like those found not guilty by reason of insanity under the current rules, provisions would have to exist for detention and treatment of those who still pose a danger to the public. The very factor that means we shouldn’t blame them is also the very factor that might give us good reason to detain them in the interests of public safety.
Which brings me to the second potentially alarming possibility raised by the Swedish research. Is it possible that the presence of such a gene could provide the justification for, say, turning down parole applications, for preventive detention orders, or even – in the scenario that really worries fans of dystopian SF – for detaining people who haven’t committed any offence yet at all?
Again, the Swedish study offers little support for such a conclusion. The research found that close genetic relatives of sex offenders were 4 or 5 times more likely to commit a sex offence than unrelated men. But that still provides a likelihood of just 2.5%, which would presumably be far too low to justify any sort of preventive intervention. That this study doesn’t provide anything more than very weakly predictive, though, doesn’t preclude the possibility that such a thing will one day be discovered. Precisely what we would do with such information is really something worth thinking about.