One of the most profound lines of inquiry into the roots of crime in America has been to look at the role of lead poisoning. Kevin Drum's 2013 cover story in Mother Jones is a good introduction; Megan Cottrell's 2012 piece for the Reader centers the story in Chicago.

The gist is this: Lead poisoning irreversibly reduces cognitive and behavioral capabilities; if those capabilities are lowered across a population, more people are more likely to commit crimes. On the other hand, if you dramatically reduce rates of lead poisoning, crime will drop. And one very compelling argument for the wholesale drop in violent crime over the past few decades is that we've done that, by eliminating lead paint and leaded gasoline.

It's not that lead poisoning is no longer a problem, not by a long shot. The continuing tragedy in Flint, Michigan, and the ongoing reporting of the Tribune's Michael Hawthorne are a testament to that. But lead poisoning rates have dropped, so has crime, and the evidence suggests a causal connection.

A new working paper by Evan Herrnstadt of the Harvard University Center for the Environment and Erich Muehlegger of the University of California-Davis (h/t Madeleine Thomas at Pacific Standard) lays out an argument for yet another link between environmental pollution and crime. And the path of the argument is interesting: comparing 13 years of crime data in Chicago to the direction of the wind, and in doing so, looking for localized effects. Which they found:

On days when the wind blows orthogonally to the interstate, we find that violent crime increases by 2.2 percent on the downwind side. The effects we find are unique to violent crimes–we find no effect of pollution on the commission of property crime. As further evidence that we are identifying a causal effect of road pollution, we find that the effect of being downwind is attenuated on days when the direction of the wind is not orthogonal to the road and when wind speeds are sufficiently high to disperse pollution.

How could windblown air pollution increase violent crime? The authors suggest two pathways. One is that it makes you angry or impulsive, via neuroinflamation or headaches and head tightness. The other is that it makes you dumber, and they point to several recent findings that there is a correlation. One of these is one of my favorite recent study findings (emphasis mine):

Finally, Stafford (2014) exploits the quasirandom timing of school renovations to link indoor air quality and academic performance. She finds that, while attendance is not affected, performance on standardized tests improve after the renovations. The average mold remediation project led to a 0.15 standard deviation improvement in test scores, while the average ventilation project led to a 0.04-0.09 standard deviation improvement. More recent literature extends to alternative measures of cognitive impairment to test scores: Archsmith (2015) finds evidence that baseball umpires make more mistakes when calling strikes on polluted days.

It's kind of like Do the Right Thing, but with pollution instead of heat. And they suggest that it could parallel lead: As with lead, we've done a good job reducing air pollution in cities, so it follows that its decline could play a role in violent crime's decline.

This is a working paper, and it's plowing new ground. The authors describe it as "the first quasi-experimental evidence that air pollution affects violent criminal activity." Think of it as an interesting frontier.

But it also made me think of another interesting frontier: Chicago's Array of Things, the network of environmental sensors being built around the city. Those sensors are being set up to monitor, among many other things, carbon monoxide, nitrogen dioxide, and ozone—all pollutants touched on in the study. When planning for the Array of Things began, the people behind it had some ideas for what the data gathered by the sensors could do, but part of the point was just to gather it and let other people find uses for it; that's why the data's open. If Herrnstadt and Muehlegger's study opens up a new trail for environmental-justice research, it seems an obvious fit for the AoT.