Doing violence to genetics

We were going to take it slow this week, but, well, some stories we just can't ignore.  So, a piece in this week's New York Times says that geneticists at the University of Connecticut are planning to genotype Adam Lanza in an effort to figure out what caused him to kill 20 children and 7 adults in Newtown, Connecticut.  Indeed, in an effort to explain the biology of extreme violence. As the story tells it,

They could look at all of Mr. Lanza’s genes, searching for something unusual like gene duplications or deletions or unexpected mutations, or they might determine the sequence of his entire genome, the genes and the vast regions of DNA that are not genes, in an extended search for aberrations that could determine which genes are active and how active they are.

Some -- many -- people believe that behavior such as Lanza's is so extreme that there must be a genetic explanation.  But, if it is that extreme, and there is a genetic explanation -- and we'd say this is very unlikely, but let's just say for the sake of argument that there is -- it will mean nothing about the rest of us.  Every behavioral distribution has extremes, and when genetic associations have been found, they often pertain only to the extremes, and are generally rare, and so do essentially nothing to explain the rest of the distribution.  But, that's assuming Lanza's behavior was extreme in the distribution of violent behavior, and that it was biologically determined.  Was it in fact any more violent than any other premeditated murder, most of which don't make the news? 

Further, Mr. Lanza, like all of us, is more than likely to have numerous "aberrations" or uncommon mutations or regions in his genome.  The scientific question is how to determine what they mean. The suggestion in the NYT story is that any uncommon stretch of DNA might be causal, but we know that's not true, since we all have something uncommon, and most of us are not prone to extreme behaviors. Most people with 'Asperger's' syndrome, if that even is what Lanza was affected by, if that is even properly called a 'disease', do not commit any crimes whatever, and don't go over the edge like he did, either.

Perhaps the geneticists will be looking for mutations that have previously been suggested to be associated with violence, or "risk taking" or whatever else might strike them as explanatory.  But, any such mutations will be common in most populations, will be found in many people who aren't violent or risk takers, and will, as with all genes associated with complex traits, explain a very small fraction of the behavior.  They have very little, if any, predictive power, and certainly can't predict one-off running-amok as Lanza did.

Or, perhaps they'll be looking for genes for 'mental illness' -- again, very few if any genes have been identified that explain very much of the variation in mental illnesses, whose definition is often hopelessly vague anyway, including those such as schizophrenia that seem to run in families.  The vast majority of such traits are polygenic and seem to have an environmental component as well. And, of course very few people with mental illness of any type are mass murderers, so this seems another fruitless avenue.

Of course, there are those geneticists who claim we can't afford not to do this study.  That is, to us, thoroughly wrong.  If anything, we should not waste the money for this, because there are vastly more important issues we could and should address than trying to geneticize a unique event.   And, by the way, that is not so unique in the US, and one might be tempted, therefore, to say that many people here have the genotype 'for' such murderous mayhem.  But then does that suggest in any serious way that people in other countries that have hardly any of this kind of behavior don't have the genotypes?  That would be a real stretch.

So it's hard to believe that the people doing this work aren't doing much more than blatantly and shallowly exploiting a tragedy based on very little evidence that they'll find what they are looking for. They are apparently motivated by the assumption that behavior is genetically determined, or they wouldn't be doing it.  But where does that road take us?  Let's say they find something -- again, that's unlikely -- but what would society do with the answer?  Would it really be predictive?  Would we genotype everyone at birth and quarantine -- or worse -- anyone with what looks like a genetic predisposition to violence?  Or, would we just send them off to military school at a very young age? Did we learn absolutely nothing in the long 20th century, about the morality and ethics of such eugenic measures?

Clearly, violent behavior doesn't have a simple explanation; like most traits, some people blame society, some people genes, and where you come down on this is likely to be based on your prior beliefs about nature vs nurture rather than any ironclad scientific evidence.  Like most complex behaviors, it's likely that violence is a product of genes and environment, and every Adam Lanza will have gotten to their tragic end in their own way. That's unfortunate for the idea that we could avoid such things by anticipating them, but perhaps is a clear signal that we need other approaches -- like gun removal, early counseling for people with disturbed personalities, and so on.  There is far less 'science' in that, but perhaps far more cure.