How Cannabis Disappeared From Western Medicine
The decline of cannabis medicine.
Cannabis did not leave Western medicine because it had suddenly been proven ineffective. By the time it was effectively removed from clinical practice in the mid-twentieth century, it had established a place in nineteenth-century medicine through roughly a century of documented use, practitioner observation, and published case literature. The question worth asking is not whether cannabis had that foothold. It did. The more consequential question is why it failed to hold. The answer is not a single cause. It is a convergence of forces, each of which would have created pressure on its own, and which together made cannabis increasingly difficult to sustain as a mainstream medicine during the period when Western pharmacology was consolidating its standards for what a reliable drug needed to be.
Cannabis had entered Western practice through tinctures, extracts, and other pharmaceutical preparations derived largely from Indian hemp. It appeared in pharmacopoeias and dispensatories and was prescribed for pain, spasm, sleep disturbance, and other difficult clinical problems. Its place was never as secure as that of simpler or more predictable drugs, but it was established enough that its eventual disappearance requires explanation.
Those standards were not arbitrary. They reflected genuine advances in chemistry, clinical method, and pharmaceutical infrastructure. The problem was not that cannabis was judged unfairly by those standards. The problem was that many of the qualities shaping cannabis's clinical behavior were also the qualities that made it hardest to meet them.
The New Standards
The second half of the nineteenth century brought a reorganization of Western pharmacological practice that is easy to underestimate in retrospect. What was being built was not just a more precise chemistry but a new framework for what medical authority over a substance required. A medicine needed to be identifiable: its constituents describable, its preparation replicable. It needed to be assayable: its strength testable by methods independent of the practitioner's judgment. It needed to be storable: its behavior predictable over the interval between manufacture and use. And it needed to be prescribable with enough confidence that a physician writing an instruction and a pharmacist filling it could share a common expectation of what the patient would receive.
These demands emerged from the same professional project. As medicine reorganized around laboratory science, institutional education, and peer accountability, the evidentiary bar for a drug's reliability rose. A preparation that worked sometimes, for some patients, in ways that were difficult to reproduce or explain became harder to defend professionally. This was not a demand for molecular purity or single-compound isolation. Many drugs of the period were botanical preparations of considerable complexity, and botanical medicines generally faced challenges involving potency variation, raw material quality, and assay. Cannabis was not unique in having these problems. What distinguished it was the degree to which they resisted the corrective tools the period had available, and the particular difficulty of achieving the workable consistency the framework required: that a practitioner could return to a preparation and find it behaving within an expected range.
The newer medicines that followed did not come without complications of their own. Chloral hydrate, aspirin, and the early barbiturates each carried risks and limitations. But their identity, strength, and prescribed dose could be specified more readily than those of cannabis extracts. Physicians working under rising professional accountability had reason to favor medicines they could manage with greater procedural confidence. Cannabis did not disappear because it lost a direct clinical competition. It receded because it accumulated disadvantages at a moment when the profession was becoming less tolerant of them.
The Preparation Problem
What made cannabis difficult to standardize was not a failure of chemistry alone. It was something more fundamental: the plant's therapeutic behavior was shaped at every stage by preparation, and the relationship between preparation and clinical outcome was far more complex than researchers had yet mapped.
Cannabis-derived substances had been isolated by chemists in the nineteenth century, but the principal cannabinoids responsible for cannabis's characteristic effects were not adequately identified and structurally characterized until the twentieth century. Through the Victorian era and into the early twentieth century, practitioners were working with crude extracts of uncertain and shifting composition. Attempts to identify and stabilize the active fraction repeatedly produced materials that behaved differently from the preparations they came from, or that varied significantly between batches. The plant's chemistry was sensitive to temperature, solvent, storage conditions, and the part of the plant used. What reached the patient was not a defined chemical entity but a preparation whose profile depended heavily on how it had been made, and the field lacked the tools to characterize that profile reliably enough to compensate for its variability.
This created a specific clinical problem that went beyond inconsistent strength. Different preparations could vary not only in apparent strength but also in onset, duration, and the character of the reported response. Dosing guidance could not be standardized with confidence because the material being dosed was not yet understood well enough to determine what was varying between preparations. Physicians who reported positive outcomes were not necessarily using preparations comparable to those used by physicians who reported failures or adverse effects. The clinical literature accumulated, but it was harder to interpret than it should have been, because the preparation variable was neither controlled nor well understood.
Traditional systems had managed preparation variability through accumulated conventions and practical experience with specific forms. That body of knowledge was real, but it was too embedded in particular traditions to translate directly into the chemical specifications Western pharmacology needed. That gap was never bridged. What Western medicine inherited was cannabis without a framework for reproducing its effects with the consistency the new pharmaceutical standards required.
Law and Decline
By the early twentieth century, cannabis was already losing ground in Western medical practice, but that process was incomplete and uneven. Prescriptions persisted. It remained in pharmacopoeias. Practitioners in some specialties continued to reach for it. The decline was real but gradual, and it had not yet resolved into elimination. What changed in 1937 was the legal environment.
The Marihuana Tax Act did not directly prohibit medical cannabis, but its practical effects on medical use were significant. The Act imposed registration, taxation, recordkeeping, and order-form requirements on the medical and pharmaceutical handling of cannabis. The American Medical Association opposed the bill at the time, specifically on the grounds that it would obstruct legitimate medical practice. That opposition was overridden. The administrative burdens the Act created, combined with the professional risk of association with a substance that had been publicly linked to moral panic in the campaigns preceding the legislation, gave physicians and pharmacists additional reason to abandon a medicine already losing professional ground.
The legislation and the preceding decline are best understood as interacting rather than sequential. Cannabis had been weakened in professional standing by the accumulated difficulties of the previous decades: the standardization failures, the competition from more manageable medicines, the clinical literature that was harder to interpret and extend than the evidence accumulating for newer drugs. The 1937 Act arrived when cannabis was peripheral, and the burdens it imposed accelerated the profession's withdrawal from its use. Declining use and the additional weight of legislation together produced the outcome that neither would have guaranteed alone.
The removal of cannabis from the United States Pharmacopoeia followed in 1942, amid declining medical use, growing regulatory pressure, and persistent claims that cannabis posed serious psychological and behavioral risks. Other countries moved in similar directions during the same period, under the influence of international drug control frameworks developing in parallel. By mid-century, cannabis had been removed not just from practice but from the institutional infrastructure through which a medicine sustains itself professionally: the pharmacopoeias, formularies, and research pipelines that keep clinical knowledge alive and developing. That institutional removal would prove harder to reverse than the clinical decline that preceded it.
What Was Lost
The century of accumulated documentation that cannabis left behind when it exited Western pharmacopoeias was not a coherent, usable body of knowledge. It was a literature shaped by preparation variability, inconsistent dosing, and an absence of chemical characterization that made it difficult to know, in retrospect, what any given clinical report was actually describing. When cannabis re-entered serious research attention in the latter half of the twentieth century, that literature was less useful than its volume suggested. The preparation variable that earlier clinical work had never controlled was the same variable that made its results so difficult to interpret.
What the disappearance removed most durably was the opportunity to resolve that variable under the conditions where resolution might have been possible: within an active clinical tradition, over time, through the iterative refinement a medicine in continuous use undergoes. Its removal interrupted that possibility. Whether the process would have succeeded cannot be known. Cannabis left before the chemistry needed to characterize its constituents properly had been developed. The endocannabinoid system, the distinction between acid and decarboxylated cannabinoid forms, the chemical diversity of the plant's cannabinoid and non-cannabinoid constituents: none of this was understood in 1937, or in 1942. The tools that could have made cannabis more standardizable arrived after it had already been removed from the contexts where standardization would have mattered most.
The preparation knowledge that traditional medicine had built over centuries did not survive the transition into Western pharmacological practice in a form that institutional medicine could inherit or extend. When cannabis left, it left without a workable preparation science, and the gap was not filled. What returned, decades later, was largely the raw plant and the raw clinical question, without the institutional infrastructure or preparation framework needed to interpret it consistently.
The disappearance was not a verdict on whether cannabis was medicine. It was a record of what happens when a medicine's most important properties are also the properties a changing system is least equipped to handle. Preparation variability that resisted standardization, competition from more tractable medicines, shifting professional norms, and the additional weight of regulation and legislation: these forces did not resolve the question of what cannabis could do. They only ensured that the question would have to be asked again much later, without a continuous preparation framework to build from.
Quick Map — References & Citations
Zuardi, Revista Brasileira de Psiquiatria, 2006 — Historical review of cannabis in medicine.
The review traces cannabis use through its nineteenth-century adoption and twentieth-century decline in Western medicine.
Informs: The article’s historical arc, including cannabis’s established medical foothold, competition from newer drugs, and gradual loss of professional standing.
Peschel, Scientia Pharmaceutica, 2016 — Quality control of traditional cannabis tinctures.
The study examines how plant material, extraction conditions, chemical markers, and storage stability affect cannabis tincture composition.
Informs: The preparation problem is presented as more than inconsistent strength, with solvent, temperature, storage, and raw material capable of changing what a preparation delivers.
Gaoni and Mechoulam, Journal of the American Chemical Society, 1964 — Isolation and structural identification of THC.
The researchers isolated and described the structure of the principal psychoactive constituent of hashish.
Informs: The article’s explanation that the principal cannabinoids responsible for cannabis’s characteristic effects were not adequately identified and structurally characterized until the twentieth century.
Musto, Archives of General Psychiatry, 1972 — Historical analysis of the Marihuana Tax Act of 1937.
The article examines the political, administrative, and professional circumstances surrounding passage of the Act.
Informs: The legislation is treated as an accelerant acting on a medicine already losing ground, rather than as the sole cause of cannabis’s medical decline.
American Medical Association, Council on Science and Public Health, 2009 — Historical review of medicinal cannabis in the United States.
The report summarizes nineteenth-century preparations, competing pharmaceuticals, the AMA’s opposition to the 1937 Act, and cannabis’s removal from the United States Pharmacopoeia in 1942.
Informs: The decline is presented as the result of interacting pressures, including difficult standardization, easier-to-manage alternatives, regulatory burden, and institutional removal.
References
Zuardi, A. W. (2006). History of cannabis as a medicine: A review. Revista Brasileira de Psiquiatria, 28(2), 153–157. doi:10.1590/S1516-44462006000200015
Peschel, W. (2016). Quality control of traditional cannabis tinctures: Pattern, markers, and stability. Scientia Pharmaceutica, 84(3), 567–584. doi:10.3390/scipharm84030567
Gaoni, Y., & Mechoulam, R. (1964). Isolation, structure, and partial synthesis of an active constituent of hashish. Journal of the American Chemical Society, 86(8), 1646–1647. doi:10.1021/ja01062a046
Musto, D. F. (1972). The Marihuana Tax Act of 1937. Archives of General Psychiatry, 26(2), 101–108. doi:10.1001/archpsyc.1972.01750200005002
American Medical Association, Council on Science and Public Health. (2009). Use of cannabis for medicinal purposes. CSAPH Report 3-I-09.
