If responses to my last post are any guide—including a diss from one of my own students!—many readers reject gun controls as a way to reduce shootings like the recent massacre in Arizona and other gun-related homicides.

Common sense tells me that unbalanced people like Jared Loughner, the Arizona shooter, should have a harder time getting a semiautomatic weapon and high-capacity clips. Moreover, some researchers have found a correlation between levels of household gun ownership and homicide. But I realize that the causal link between shootings and gun controls—and gun ownership—is complex, more so than my previous post implied.

Restrictions on gun ownership in New Jersey, Washington, D.C., and other regions were followed by a surge in firearm-related crimes, as this 2008 article in The New York Times noted. Moreover, a comparison of Wikipedia rankings of nations by gun ownership and by firearm-related deaths shows that these variables are not closely correlated. Yes, the U.S. homicide rate is much higher than in England, Japan and other nations that severely restrict civilian ownership of firearms. But Colombia, with a gun ownership rate less than one tenth that of the U.S., has a firearm-related homicide rate seven times higher. Brazil, which also has less than one tenth as many guns per capita than the U.S., has 50 percent more firearm-linked homicides.

As much as I hate to admit it, these statistics support the slogan that guns don't kill; people do. The link between homicides and easy access to guns—like the link between real violence and media violence—is tenuous. You can make the cause for or against a causal relation, depending on what society or time period you examine. Complexities like these lead to complaints that "social science" is an oxymoron.

But even gun-lovers want fewer homicides, right? So let me suggest another possible way to achieve that goal. The idea was inspired by the evolutionary psychologists Margo Wilson (who died in 2009) and her husband Martin Daly, both of McMaster University in Ontario. In their 1988 book Homicide, often upheld as the gold standard in applying Darwinian theory to social problems, Daly and Wilson pointed out that males have always committed the vast majority of homicides. The reason, the psychologists contended, is that our male ancestors fought fiercely for "control over the reproductive capacities of women," which resulted in an innate male tendency toward violent aggression.

Although today lethal aggression can (often) lead to imprisonment or execution—both of which hamper reproduction—it would have promoted genetic fitness in societies predating the rule of law, according to Daly and Wilson. As evidence of their evolutionary thesis, Daly and Wilson noted that modern men kill blood relatives much less often than they kill unrelated females out of sexual jealousy as well as male rivals and even the children of other men. (One of Daly and Wilson's best-known findings is that stepfathers are many times more likely than biological fathers to kill their children.)

Males, and especially young males with few prospects, also kill nonrelatives to achieve status and "resources"—by committing armed robbery, for example, or shooting a rival drug dealer. Like other evolutionary psychologists, Daly and Wilson struggled to explain variations in behavior among individuals and societies. For example, the homicide rate of their homeland, Canada, is only about a third that of its neighbor, the U.S. Rates of homicide also vary widely from region to region within each country. Why?

I heard Daly and Wilson propose an answer to this puzzle at a 2009 meeting on aggression that I reported on for Scientific American; they also presented the hypothesis in this 2001 paper (pdf). The best predictor of high homicide rates in a region, they asserted, is income inequality. As a measure of such inequality, Daly and Wilson employed the so-called Gini index (named after its originator, the Italian statistician Corrado Gini), which ranks inequality on a scale ranging from 0.0 to 1.0. A region in which everyone has exactly the same income would have a Gini score of 0.0, whereas a region in which one person makes all the money has a score of 1.0.

Daly and Wilson found a strong correlation between high Gini scores and high homicide rates in Canadian provinces and U.S. counties. High Gini scores predicted homicides better than low average income, high unemployment and simpler measures. Basically, Daly and Wilson were blaming homicides not on poverty per se but on the collision of poverty and affluence, the ancient tug-of-war between haves and have-nots. The income-inequality hypothesis, Daly and Wilson asserted, can account for the "radically different national homicide rates" of the U.S. and Canada, the latter of which has more generous social-welfare programs (including universal health care) and hence fewer economic disparities.

Naturally, some researchers have reported data that fail to support the income-inequality theory of homicide. But I find it persuasive, especially because it points toward an attractive solution to high homicide rates: a more equitable economic system, perhaps with higher taxes for the wealthy and more generous welfare programs for the poor. In short, socialism. I hope that opponents of gun control will consider this modest, alternative proposal for reducing lethal shootings.