Marijuana use inevitably leads to heroin and harder drug use.

In October 1970, as Congress was finalizing the Controlled Substances Act, the question of where marijuana belonged in the new scheduling framework was politically awkward. The legislation, championed by the Nixon administration, would organize all controlled substances into five schedules based on abuse potential and accepted medical use. Heroin was clearly Schedule I material. Cocaine was clearly Schedule II. Marijuana presented a problem. Its pharmacology was poorly understood, its medical history was complex, and emerging research suggested it was significantly less harmful than alcohol. Congress resolved the awkwardness by placing marijuana temporarily in Schedule I and establishing a commission, led by former Pennsylvania Governor Raymond Shafer, to study the question and recommend permanent placement.

The Shafer Commission reported back in 1972. Its conclusion was explicit: marijuana did not merit its Schedule I classification, personal use should be decriminalized, and the evidence for severe health harm was insufficient to justify the current framework. Nixon rejected the report without reading it. Marijuana stayed in Schedule I. The gateway drug theory, the claim that marijuana use led inexorably to heroin and harder drugs, remained the cornerstone of federal drug education for the next three decades.

The gateway framework had logic to recommend it. Studies of heroin and cocaine users consistently found that the overwhelming majority had used marijuana before using harder drugs. The sequence was real. What the sequence did not establish was causation. Virtually every heroin user had also drunk alcohol and smoked cigarettes before using heroin; neither substance was labeled a gateway drug. The question of which, if any, prior substance caused progression to harder drugs required methodological care that the early epidemiological work had not applied. Drug prevention curricula, which needed clear messages for classroom delivery, collapsed the distinction between association and cause.

The federal government's own commissioned review eventually said so. In 1999, the Institute of Medicine, at the request of the Office of National Drug Control Policy, which had hoped for a vindication of current policy, published a comprehensive assessment of the scientific evidence on marijuana. On the gateway question, the IOM was direct: while marijuana use preceded use of other illicit drugs in most cases, the evidence did not support the conclusion that it caused progression to harder drugs. The strongest predictor of hard drug use was not which drug someone had tried first but whether they had a pre-existing vulnerability to substance dependence. The gateway mechanism, the report noted, was more plausibly explained by individual risk factors and social environment than by any pharmacological property of marijuana itself.

Further research strengthened that picture. Studies of countries with higher rates of marijuana use but different patterns of hard drug availability found no consistent relationship between marijuana use and subsequent heroin or cocaine use. The role of social networks, the fact that people who used marijuana were more likely to know people who used harder drugs, offered a social rather than pharmacological explanation for the observed correlation. Most significantly, the two substances that most reliably preceded initiation into hard drug use among American youth were alcohol and tobacco, neither of which attracted the gateway label.

The Shafer Commission had identified the problem clearly in 1972. The issue was not lack of evidence but lack of political will to follow where the evidence led. Drug education had been built around a theory that served rhetorical purposes more reliably than empirical ones, and the classroom versions of the gateway narrative outlasted the scientific consensus against it by a generation.