GIST: "The New Mexico Supreme Court ruled last month on an extraordinarily important question: Should a criminal defendant be allowed to argue that a specific gene rendered him unable to control his violent behavior? The court concluded the answer was no, in this instance: It upheld the conviction for second-degree murder of Anthony Blas Yepez, who killed a man in 2012. Yepez had sought at trial to introduce evidence that he had what’s been called the “warrior gene” — a version of a gene known as MAOA, which has been linked to violence in some studies. The district court of Sante Fe County excluded expert testimony on that subject; later, the New Mexico appellate court ruled that the exclusion was a mistake but that it did not affect the outcome of the trial. The New Mexico Supreme Court has now found that the district court’s original rejection of the evidence was reasonable.
The state Supreme Court, however, missed an important opportunity: It did not go nearly far enough in batting down the scientifically suspect claim that there is a gene for violence. More broadly, it failed to relay the view of contemporary experts in the field that the very idea of a “warrior gene” is based on obsolete science. Such a ruling might have influenced other courts that either are, or soon will be, forced to confront this issue. It would greatly benefit our criminal justice system — and justice itself — if this outmoded idea about criminality is stamped out.
The Dutch geneticist Han Brunner and colleagues first identified a possible link between violence and a variant of the MAOA gene in an article in Science in 1993: They analyzed the genes of surviving members of a large Dutch family who had sought genetic counseling because the males, over decades, had demonstrated a proclivity for aggressive outbursts (fights, rape, arson). Brunner’s team implicated MAOA, which encodes an enzyme that breaks down brain chemicals that had previously been associated with the regulation of aggression in animals and humans. A low-acting version of MAOA was associated with violence, the study concluded.
Another much-cited study, published in 2002, concluded — by looking at a cohort of more than 1,000 people tracked from age 3 to age 26 — that not everyone with low activity of MAOA was prone to violence. It was only those who had suffered from childhood mistreatment. Scientists hailed this as a milestone study showing a “genotype by environment interaction,” in which variants of a gene can be affected differently by certain aspects of a person’s upbringing and experiences.
Lawyers soon glommed on to such studies. As this work became more widely known, several criminal defendants — including Yepez, who testified that he was abused as a child — have sought to have experts testify on their behalf that their actions are better understood as arising from their bad genes and environment rather than bad choices.
We have identified at least 11 instances in which a defense team sought to introduce “warrior gene” evidence either to make a “not guilty” verdict more likely or to push for conviction for a lesser crime to mitigate punishment. In at least one of these cases, the defendant was convicted on lesser charges than the prosecutor sought, and another person was spared the death penalty. Of course, in jury trials we can’t say for certain if it was the MAOA argument that was decisive. We’ve found at least another 50 in which genetic arguments of other sorts have been made by defense lawyers and seriously grappled with in court — including claims about serotonin levels, family history of violence and the like.
DNA testing is radically reshaping the definition of family
The genetic argument can cut both ways: In a death penalty appeals case in Texas involving Amos Joseph Wells III, who killed his pregnant girlfriend and two other people, Wells’s lawyers note that prosecutors turned the “warrior gene” defense against him. Prosecutors argued that the genetic evidence offered by the defense showed, in fact, that he was hard-wired to kill. The role this dubious scientific claim played in his conviction is one argument his appellate counsel has raised for reversing the outcome.
In Yepez’s case, the New Mexico Supreme Court expressed reasonable caution about the use of genetic information in a murder trial. “Mere genetic susceptibility to a given mental condition is not relevant on the issue of deliberate intent,” the justices wrote. And even if Yepez had a genetic tendency toward aggression, that does not mean he was “incapable of deliberating and planning” a crime, they said. The court also expressed concern about the “stigmatization associated with simplistic interpretations” of the associations between genes and behavior.
That might have been enough had the whole paradigm of behavioral genetic research not shifted since the original MAOA study. That study predates the sequencing of the first human genome and was conducted when scientists could focus only on a few genes at a time. To be sure, many diseases have been found to be affected by single “candidate genes”; rare variants of the CALM2 gene, for instance, increase the likelihood of sudden cardiac death in children. But the attempt to link variation in any single gene to variation in complex behavioral traits like violence is widely perceived to have failed.
The first clues came from attempts to replicate the original MAOA findings, which were very mixed. A few studies succeeded in doing so, but others — including a 2005 study of 774 White men, many of whom had a history of maltreatment and victimization — found no connection. One challenge is that the trait of aggression can be measured in many potentially incompatible ways and that definitions of maltreatment in childhood also vary widely, increasing the possibility that positive correlations are the result of chance.
But the true death knell for this kind of work came with the introduction of whole-genome sequencing and analysis. As scientists started to develop and use more sophisticated computational methods to look for associations between gene variants and variation in human behavior across the entire genome, few to none of the earlier findings from the “candidate gene” era held up.
These “genome-wide association studies” — with sample sizes in the tens to hundreds of thousands — revealed an entirely new picture about the relationship between genetic variation and behavior. They showed that thousands of gene variants each contribute a very small amount to explain why one person differs from the next for a particular behavior. There is seldom one master gene shaping the trait. The operative measure these days is a “polygenic risk score,” which sums the contributions of each gene variant to explain the variance between individuals. Even this measure explains only a small part of the variation in behaviors between people. All of this is to say that we still don’t know a lot about how genetic variations contribute to behavioral variations — but we do know that for any given behavior, there is no one master gene that can explain those differences.
The rise and fall of the science underlying the warrior gene defense occurred in less than two decades. That’s not surprising, given the pace of genomic science. Nor is it anything for scientists to be embarrassed about; it’s how science works. But the law has not kept pace with these seismic changes in our understanding of the genetic contributions to behavior. And that is dangerous, since zombie ideas continue to affect legal decisions.
State v. Yepez offered the New Mexico Supreme Court a chance to set the legal record straight, but it failed to do so. It declined the chance to dig into the research and warn other courts across the country about the speciousness of the warrior gene theory. (One of us, Farahany, served as counsel representing dozen of scholars on an amicus brief explaining the state of warrior gene science in that case; the court declined to admit the brief, even though it was unopposed by both parties to the case).
The court’s failure puts more pressure on scientists and scientifically informed lawyers to spread the word that there is no gene — or combination of genes — yet identified that predisposes people to violence. If such genes are eventually discovered, that will create thorny ethical (and judicial) questions about the limits of personal responsibility. But for now, it’s enough to say that our understanding of the complex relationship between genes and behavior is at a very early stage. We can’t be punishing people, or excusing behavior, based on studies that simply have not held up."
The entire story can be read at:
