Humphry Osmond—Abram Hoffer

Humphry Osmond and Abram Hoffer’s collaboration at the Saskatchewan Mental Hospital in Weyburn turned a study of LSD as a possible model of mental illness into an influential investigation of experience as a therapeutic instrument. Working within the Psychotomimetic model, they first asked whether the drug could reproduce aspects of psychosis; they then tested whether a deliberately induced crisis could help people with alcohol dependence. Their work linked clinical experimentation to a new principle: the experience produced by a drug, and the conditions in which it occurs, may matter as much as the chemical itself.

A research partnership in Saskatchewan

Osmond came to Saskatchewan after the authorities at St. George’s Hospital proved unsupportive of his mescaline research. The province was seeking to attract innovative medical work through generous funding and unusual institutional freedom. After answering an advertisement in The Lancet, Osmond accepted an invitation to move his family and research project to Weyburn, a remote prairie community forty-five miles north of the North Dakota border. The Saskatchewan Mental Hospital in Weyburn would become a major center for research into compounds then known as psychotomimetics.

There, Osmond joined Abram Hoffer, a Canadian psychiatrist who shared the prevailing interest in the biochemical basis of mental illness. As research director and colleague, Hoffer worked with Osmond on experiments using LSD-25 supplied by Sandoz. Their partnership was both institutional and intellectual: they began from the same model of the drug as a means of inducing and studying madness, and their accumulating reports would lead them to question that model.

The public first encountered their approach through a 1953 account in the popular Canadian magazine Maclean’s, which presented a journalist’s LSD experience under the headline “My 12 Hours as a Madman.” Sidney Katz, the first civilian participant in one of their Weyburn experiments, had been led to expect madness and described terrifying distortions of faces and surroundings. Yet he also reported visions of extraordinary beauty. His mixed account exposed a tension that would recur in the researchers’ work: the experience could be frightening, but it was not reducible to horror or insanity.

From psychosis research to treatment

Osmond and Hoffer initially administered LSD to colleagues, friends, family members, volunteers, and themselves as a window into the biochemistry of mental illness. Their attention gradually shifted from the drug as a chemical probe to the experience it occasioned, and to the possibility that perceptual disruption might itself have therapeutic value.

A late-night discussion in an Ottawa hotel room in 1953 supplied a specific hypothesis. The researchers noticed parallels between LSD experiences and accounts of delirium tremens, the severe, hallucinatory episode that can accompany alcohol withdrawal. Some recovering alcoholics interpreted the ordeal of delirium tremens as a conversion experience and a source of spiritual awakening. Osmond and Hoffer therefore asked whether a controlled LSD-induced delirium might similarly help alcoholics remain sober.

They tested the idea over the following decade, administering a single high dose in an attempt to simulate the crisis and shock a patient into sobriety. Their reports covered more than seven hundred alcoholics; in roughly half the cases, they said, participants became sober and remained so for at least several months. They considered the results more effective than those of other therapies then available. Hoffer’s account of their guiding principle was that they had considered not only the chemical but the experience a key factor in therapy. That emphasis helped establish a central premise of psychedelic therapy: treatment could depend on what a person underwent, rather than on drug administration alone.

This approach also placed subjective reports at the center of clinical interpretation. At a time when behaviorism privileged observable outcomes and treated the mind as a black box, Osmond and Hoffer’s attention to volunteers’ descriptions gave the interior life of the session evidentiary importance. Their work thereby connected treatment results with questions about consciousness and meaning, rather than limiting evaluation to outward behavior.

Evidence that unsettled their model

The reports did not consistently resemble the delirium-tremens analogy on which the treatment hypothesis rested. Although volunteers sometimes described hallucinations, paranoia, or anxiety, many accounts were markedly positive. Participants reported a feeling of unity with the world, a capacity to observe themselves objectively, heightened sensory experience, new understandings of philosophy or religion, and increased sensitivity to other people’s feelings. For many alcoholics at Weyburn, the central event seemed closer to transcendence or spiritual revelation than temporary psychosis.

These accounts led Osmond and Hoffer to doubt both the delirium-tremens explanation and the broader Psychotomimetic model. Aldous Huxley’s account of his mescaline experience reinforced their doubts: what a clinician might classify as depersonalization, hallucination, or mania could, they considered, be understood as mystical union, vision, or ecstasy. Their collaboration thus changed direction not because they abandoned the therapeutic question, but because the participants’ descriptions challenged the assumptions used to frame it.

They also learned that the session’s surroundings and the conduct of the therapist could strongly affect what participants experienced. An engaged, empathetic therapist—ideally one familiar with the drug through personal experience—could help prevent a difficult session. Osmond and Hoffer suspected that some psychotic reactions might be influenced by the impersonal atmosphere of a white room and white-coated clinician. Though the terminology had not yet become established in this context, they were already recognizing the importance of preparation, setting, and human support to treatment outcomes.

Influence, replication, and limits

By the end of the 1950s, LSD was widely regarded in North America as a promising treatment for alcohol addiction. The Saskatchewan provincial government helped make LSD therapy a standard treatment option for alcoholics in the province. Their work also circulated beyond Weyburn: Al Hubbard visited the hospital and assisted Osmond and Hoffer in their work, while a 1959 therapeutic handbook by Duncan C. Blewett and Nick Chwelos drew heavily on their case reports.

The apparent success did not persuade everyone in the Canadian medical establishment. In the early 1960s, the Addiction Research Foundation in Toronto attempted a more tightly controlled replication, seeking to distinguish the effects of LSD from other influences. Clinicians placed participants in neutral rooms, told staff not to engage with them except to administer an extensive questionnaire, and sometimes restrained or blindfolded them. The results did not match those reported in Saskatchewan, and a number of participants had terrifying experiences.

The discrepancy sharpened rather than settled the debate. Critics could point out that the treatment performed less well under rigorous controls; supporters could argue that those controls removed conditions essential to its success. The contrast made the partnership’s enduring implication clear: a psychedelic treatment cannot be understood simply by recording which drug was given. The setting, the therapist’s presence, and the participant’s experience were part of the intervention—and therefore part of the question the research had to answer.