Does marijuana use really cause psychotic disorders?
By Carl Hart and Charles Ksir
20 January 2029
Does marijuana cause psychotic disorders such as schizophrenia, and do associated symptoms like paranoia lead to violent crimes? That’s what writer Alex Berenson is claiming. As part of his new book promotion, Berenson published a New York Times op-ed that also blames the drug for “sharp increases in murders and aggravated assaults” purportedly observed in some states that allow adult recreational marijuana use.
As scientists with a combined 70-plus years of drug education and research on psychoactive substances, we find Berenson’s assertions to be misinformed and reckless.
It is true that people diagnosed with psychosis are more likely to report current or prior use of marijuana than people without psychosis. The easy conclusion to draw from that is that marijuana use caused an increased risk of psychosis, and it is that easy answer that Berenson has seized upon. However, this ignores evidence that psychotic behavior is also associated with higher rates of tobacco use, and with the use of stimulants and opioids. Do all these things “cause” psychosis, or is there another, more likely answer? In our many decades of college teaching, one of the most important things we have tried to impart to our students is the distinction between correlation (two things are statistically associated) and causation (one thing causes another). For example, the wearing of light clothing is more likely during the same months as higher sales of ice-cream, but we do not believe that either causes the other.
In our extensive 2016 review of the literature we concluded that those individuals who are susceptible to developing psychosis (which usually does not appear until around the age of 20) are also susceptible to other forms of problem behavior, including poor school performance, lying, stealing and early and heavy use of various substances, including marijuana. Many of these behaviors appear earlier in development, but the fact that one thing occurs before another also is not proof of causation. (One of the standard logical fallacies taught in logic classes: after this, therefore because of this.) It is also worth noting that 10-fold increases in marijuana use in the UK from the 1970s to the 2000s were not associated with an increase in rates of psychosis over this same period, further evidence that changes in cannabis use in the general population are unlikely to contribute to changes in psychosis.
Evidence from research tells us that aggression and violence are highly unlikely outcomes of marijuana use. Based on our own laboratory research, during which we have given thousands of doses of marijuana to people – carefully studying their brain, behavioral, cognitive and social responses – we have never seen a research participant become violent or aggressive while under the influence of the drug, as Berenson alleges. The main effects of smoking marijuana are contentment, relaxation, sedation, euphoria and increased hunger. Still, very high THC concentrations can cause mild paranoia, visual and/or auditory distortions, but even these effects are rare and usually seen only in very inexperienced users.
There is a broader point that needs to be made. In the 1930s, numerous media reports exaggerated the connection between marijuana use by black people and violent crimes. During congressional hearings concerning regulation of the drug, Harry J Anslinger, commissioner of the Federal Bureau of Narcotics, declared: “Marijuana is the most violence-causing drug in the history of mankind.” He was compelling. But unfortunately, these fabrications were used to justify racial discrimination and to facilitate passage of the Marijuana Tax Act in 1937, which essentially banned the drug. As we see, the reefer madness rhetoric of the past has not just evaporated; it continued and has evolved, reinventing itself perhaps even more powerfully today.
There have been several recent cases during which police officers cited the fictitious dangers posed by cannabis to justify their deadly actions. Philando Castile, of St Paul, Minnesota, in 2016; Michael Brown, of Ferguson, Missouri, in 2014; and Keith Lamont Scott, of Charlotte, North Carolina, in 2016 were all killed by police who used some version of this bogus defense.
Ramarley Graham, Trayvon Martin, Rumain Brisbon and Sandra Bland all also had their lives cut short as a result of an interaction with law enforcement (or a proxy) initiated under the pretense of marijuana use suspicion.
Back in the 1930s, when there were virtually no scientific data on marijuana, ignorant and racist officials publicized exaggerated anecdotal accounts of its harms and were believed. Almost 90 years and hundreds of studies later, there is no excuse for these exaggerations or the inappropriate conclusions drawn by Berenson. Neither account has any place in serious discussions of science or public policy – which means Berenson doesn’t, either.
Carl L Hart is the chairman and Ziff professor of psychology and psychiatry at Columbia University and author of High Price: A Neuroscientist’s Journey of Self-Discovery that Challenges Everything You Know About Drugs and Society. Charles Ksir is professor emeritus of psychology and Neuroscience at the University of Wyoming and author of Drugs, Society and Human Behavior