Science journalists like me often agonize over whether to write about crummy scientific, or pseudoscientific, ideas. We fret that our criticism might promote the meme we're trying to stamp out, following the principle that there's no such thing as bad publicity. This concern leads some of my fellow science communicators to ignore, say, creationism and global-warming denial in the hopes that they'll go away. I prefer to confront lousy ideas, especially ones as persistent as the notion that biology can explain why some people become criminals.
The crime-is-in-our-genes notion has popped up in the news lately. Last month, in the same issue in which I defended free will, The Chronicle profiled the psychologist Adrian Raine, who seeks "biological markers that plant the bad seed in the brain." On June 21, the National Institute of Justice sponsored a panel discussion on genetics, leading Patricia Cohen, of The New York Times, to comment that scientists "are cautiously returning to the subject" of the genetic underpinnings of crime.
Critics of this research are often accused of being politically correct, bleeding-heart liberals who attribute all wrongdoing to nasty nurturing. I am, I confess, a liberal, and almost 40 years ago I spent a weekend in a Florida jail after being busted for hitchhiking and resisting arrest (I stupidly tried to run away from the cops). But my objections to the born-to-be-bad meme are based not on my criminality or ideology but on the track record of scientific attempts to trace crime to biology. Claim after claim has been discredited, and so I'm skeptical when researchers insist that now they're finally getting things right.
One of the first biocriminologists was Cesare Lombroso, who in the late 19th century linked male criminality to what he called "atavistic" physiological features. These included a prominent brow, jutting jaw, and "handle-shaped ears," which are also possessed by "savages and apes." In the early 20th century, Charles Davenport, founder of the prestigious Cold Spring Harbor Laboratory, expressed concerned that immigrants from southeastern Europe were "given to crimes of larceny, kidnapping, assault, murder, rape and sex immorality." Italians, Davenport asserted, were prone to "crimes of personal violence," and "Hebrews" to "offenses against chastity."
Scientific enthusiasm for such nonsense faded in the aftermath of Nazism. But in the mid-1960s, the bad-seed theme resurfaced, when British scientists claimed that men born with an extra Y chromosome are prone to violent outbursts. One in a thousand men have the extra Y chromosome. But the British study was based on nine men incarcerated in a mental hospital for violent patients. Other researchers, also focusing on institutionalized patients and criminals, quickly claimed to have found evidence that XYY men are hyperaggressive "supermales," at risk of becoming violent criminals and even serial killers.
After erroneous reports in 1968 that the mass murderer Richard Speck had an extra Y chromosome, the XYY-supermale claim went viral. It was propagated by the Times and other mainstream media, enshrined in textbooks, and written into plots for films, novels, and television shows (one of which, an episode of CSI: Miami titled "Born to Kill," aired just four years ago). Meanwhile, follow-up studies of noninstitutionalized XYY men failed to corroborate claims that they are prone to violence. A 1993 report by the National Research Council, "Understanding and Preventing Violence," concluded that there is no correlation between the XYY syndrome and violent behavior.
By then, scientists had fingered another bad-seed marker, a gene that produces the enzyme monoamine oxidase A (MAOA), which regulates the function of neurotransmitters such as serotonin and dopamine. The correlation emerged from studies of a large Dutch family whose male members were mildly retarded and extremely violent. One tried to stab the warden of a mental hospital with a pitchfork. The men lacked monoamine oxidase A, suggesting that they possessed a defective version of the gene.
Many groups tried and failed to find this flawed version of the MAOA gene in other populations. In a 2002 report in Science, however, researchers led by the husband-and-wife team Avshalom Caspi and Terrie Moffitt linked violent aggression to a variant of the MAOA gene, called MAOA-L, that produces low levels of the MAOA enzyme. The correlation held only if carriers had endured abusive treatment during childhood. At least two follow-up studies, both by researchers at the University of Colorado, failed to confirm the finding of Caspi and Moffitt.
Undeterred, in 2007 geneticists in New Zealand reported that 56 percent of Maori males were carriers of MAOA-L, which scientists and the media were now calling "the warrior gene." "It is well recognized," the researchers commented in The New Zealand Medical Journal, "that historically Maori were fearless warriors." This racial profiling was based on a study of 46 men, who needed to have only one Maori parent to be defined as Maori. MAOA-L is reportedly less common among Caucasians (34 percent) and Hispanics (29 percent), but more common among Africans (59 percent) and Chinese (77 percent).
The warrior gene made headlines again in 2009, when Rose McDermott, a political scientist at Brown University, and four colleagues reported that MAOA-L carriers were more likely than noncarriers to respond with "behavioral aggression" toward someone they thought had cheated them out of money. "Behavioral aggression" was defined as making the "cheater" consume hot sauce. The Dr. Phil talk show, ABC News, and other media hailed the McDermott experiment as evidence that some people are "born to be violent," as a recent National Geographic broadcast put it.
Even granting that inflicting hot sauce on someone counts as "physical aggression," the McDermott study provided little to no evidence for the warrior gene, since the difference between carriers and noncarriers was minuscule. The researchers examined 70 subjects, half of whom carried the warrior gene. Seventy-five percent of those carriers "meted out aggression" when cheated—but so did 62 percent of the noncarriers. When subjects were cheated out of smaller amounts of money, there was no difference between the two groups.
Now, before I get accused of being a blank-slater, let me give biology its due. Crime, like everything we do and think, is underpinned by biology, including genes that we inherit from our parents and previous ancestors. Animal-breeding experiments have established that physical aggression is in part a variable, heritable trait. Anyone who has watched toddlers playing and squabbling has surely reached the same conclusion about humans. And of course most criminals, especially violent ones, carry a Y chromosome.
Also, there does seem to be tentative evidence for a genetic predisposition to psychopathy, which is suddenly the disorder du jour (perhaps in part because of the terrific television series Dexter, about a psychopathic killer of serial killers). Psychopathy is characterized by aggression combined with abnormal lack of empathy for others. Psychopaths are not merely callous; they may actually have difficulty recognizing fearful or sad expressions in others, according to a 2010 article in Scientific American, "Inside the Mind of a Psychopath," by the neuroscientists Kent Kiehl and Joshua Buckholtz. They estimated that as many as 1 percent of all males and a smaller percentage of women—and from 15 to 35 percent of U.S. prisoners—may be psychopaths.
Evidence that psychopathy emerges quite early in life and may be partially inherited has emerged from studies of young twins by, among others, Adrian Raine. If one identical twin scores high in psychopathy, the other is likely to as well, more so than if the twins are fraternal. Raine also claims to have found congenital brain features, such as abnormalities of the amygdala, associated with childhood psychopathy. He recommends testing children for these neuro-markers and providing pre-emptive treatments, such as psychotherapy, so they don't become criminals.
I think labeling children as bad seeds will do more harm than good and has little scientific justification. Raine acknowledges that psychopathy stems from nurture as well as nature. Also, psychopathy is not a discrete syndrome, which you either have or don't have; it comes in degrees. Far from being compulsively violent criminals, many people with psychopathic traits can be quite successful, especially in fields that reward ruthlessness, such as business, law, and politics (but not science journalism; we're all pussycats). Most psychopaths are not criminals, and most criminals are not psychopaths.
Finally, no genetic syndrome can account for major fluctuations in crime rates across time and distance. For example, the U.S. homicide rate is less than a tenth that of El Salvador and almost 10 times higher than that of Japan. U.S. homicide rates—and other major crimes—have fallen by more than 50 percent since peaking in 1991, dropping even through the recent economic downturn. Maybe a few monsters, like Richard Speck or Ted Bundy, are born to be bad, but most crimes clearly stem from mutable environmental factors, and most criminals can redeem themselves.
I've kept out of jail since my lost weekend in Florida. Maybe reductionist criminology can make good, too.