Does lead lead to everything bad?

I'm skeptical of monocausal explanations for almost anything.

So I was not prepared to be impressed after reading an essay, "America's Criminal Element: Lead", from Mother Jones' Kevin Drum. He posits that a growing mountain of evidence suggests that one factor, and one factor alone, might well account for crime patterns in the United States since the middle of the last century. For incarceration rates. For abortion rates. For lots of bad things. It accounts even for variation of said patterns within neighborhoods, among races and genders and between classes.

That would be the lead, in the form of a gasoline additive.

Subscribe to The Week

Escape your echo chamber. Get the facts behind the news, plus analysis from multiple perspectives.

SUBSCRIBE & SAVE

Sign up for The Week's Free Newsletters

From our morning news briefing to a weekly Good News Newsletter, get the best of The Week delivered directly to your inbox.

From our morning news briefing to a weekly Good News Newsletter, get the best of The Week delivered directly to your inbox.

Sign up

Most stories about crime rates begin and end with correlations. A zig-zaggy line on a chart represents crime, or maybe another social pathology. Another line, hugging the first, represents a technological or social change of some sort. My favorite explanation for the drop in crime in the 1990s was the availability of cell phones.

But with lead, you have to think in three dimensions.

Dimension one is biology: So far as we know, lead kills brain cells. But importantly, lead gums up the channels that help cells communicate with one and other. It seems to affect the way that brains grow and settle into normal patterns of adulthood. As Drum notes, lead also seems to fudge with the prefontal cortex, where the brain's conscious decision-maker function seems to reside. There is very strong (i.e., casual) evidence that links lead concentration in the blood with varying levels of emotional regulation and impulse control.

Dimension two is time: The effects of lead exposure on brains don't generally manifest themselves until decades after said exposure.

Dimension three is pre-existing social capital. Since everyone was exposed to lead in gasoline and not everyone commits a crime, the stuff you bring to the table — generally, where you live, the level of stress you and your family suffer, discrimination based on your race or gender, other illnesses — acts as a sort of a threshold.

Combine these axes together and a coherent story about the rise and fall of several significant social pathologies emerges. When lead was added to gas in order to increase its octane rating, exposure shot up. Millions inhaled the neurotoxins. Brains changed. Behavior changed on massive scales. Then, beginning in 1986, lead was phased out of gas. Unleaded gasoline became the norm. But residual lead remained concentrated in the infrastructure of poor neighborhoods, in particular. Social pathologies declined or stabilized.

What makes this story a coherent one is that the lead effect correlates so remarkably well on levels large and miniscule. Within cities, and within neighborhoods, scientists found that they could retroactively predict crime rates simply by looking at lead levels in year zero and crime levels five years after that. These aren't simple coincidences; study after study of city after city where there's data for lead exposure and crime rates down to the level of neighborhoods confirms a correlation.

Even the strongest correlations aren't sufficient to prove, beyond a shadow of a doubt, that lead is the culprit. But once other factors have been screened away and accounted for by statisticians, if it's lead that remains, and it is lead that remains, then the body of evidence ought to be enough to effect public policy. Lead is out of gas, but it's ubiquitous in home fixtures, soil, and in paints.

The best criticism of the lead hypothesis comes from Jim Manzi. The heart of his problem with Drum's account is that there is a lot more variation in what might account for the coincidence of lead levels and crime rates than Drum admits; changes in consumption habits, regional differences in terms of preference for cars, different regulatory schemes and so on.

But Manzi's gloss applies to almost any claim. And the lead claim is supported by strong evidence at all levels of the bio-psycho-social model. The mere fact that something is inherently complex, as crime rates and other social ills undoubtedly are, does not itself mean that some factors aren't more important than others, and that science cannot tell us whether one factor is truly more explanatory than another.

That lead is bad for you is not a new hypothesis. It's kind of a proven one, actually, and there are some restrictions on lead use in law. But if lead is as bad as it seems, everyone ought to pay a lot more attention to it. Further regulation might be difficult, but taking seriously one's own exposure to lead should not be.