On Thursday, Gil Kerlikowske, the director of the Office of National Drug Control Policy, announced the results of a study that—at least according to him—demonstrated a link between marijuana use and crime. The study analyzed data collected via the Arrestee Drug Abuse Monitoring program (ADAM II), which took urine samples from arrestees in five cities over a 21-day period last year. “Marijuana remained the drug most often detected in ADAM II arrestees in all five sites in 2012, ranging from 37 percent of ADAM II arrestees testing positive in Atlanta to 58 percent testing positive in Chicago,” the study reported. “In three of the five sites, over half of the adult male arrestees tested positive for marijuana.”
Kerlikowske, who opposes marijuana legalization, said in a speech Thursday that the study showed that America needs to “acknowledge and come to grips with the link between crime and substance use.” But correlation is not causation. Just because a high percentage of arrestees tested positive for marijuana does not mean that smoking marijuana made them commit crimes. Here are other things that over half of the adult male arrestees probably had in common: pants, food in their stomachs, a mother who loves them, an impoverished background, an affinity for one or more of the local sports teams. As Steve Fox of the Marijuana Policy Project told a McClatchy reporter, “We could release a study tomorrow showing that 98 percent of arrestees in the United States drank water in the 48 hours before they engaged in criminal behavior. Does that mean that water causes crime?”
Now, Kerlikowske only said that drug use and crime were linked, not that drug use causes crime. But still, the implications are obvious. Kerlikowske is not a stupid man, and he’s not actually a terrible drug czar. He has argued that drug abuse needs to be treated as a public health issue, not just a matter of criminal justice, and I couldn’t agree more. In his speech, Kerlikowske mentioned the need to move the drug policy reform debate beyond “bumper stickers.” One good way to do that is to move beyond studies that don’t necessarily say anything at all.