My mum is a pharmacist. Growing up, that meant our house had a particular attitude toward medicine: precise, evidence-based, and faintly dismissive of anything that couldn’t be traced to a molecule. When someone claimed a remedy had worked, the highest form of scepticism in our household was the quiet verdict: “It’s only a placebo.”
Only. As if that settled the matter.
It took me years to realise that “only a placebo” is one of the most extraordinary sentences in all of science. It isn’t a dismissal. It’s a confession – a confession that the human mind and body, when given sufficient expectation of healing, will sometimes actually heal. No drug required. The placebo doesn’t reveal that the patient was fooled. It reveals that the patient healed themselves.
That reframing changes everything.
The Man Whose Tumours Dissolved (and Came Back)
In 1957, a physician named Bruno Klopfer reported the case of a patient he called Mr. Wright – a man with advanced lymphosarcoma who had been given weeks to live. His tumours were the size of oranges. His body was riddled with fluid. He was, by every reasonable measure, dying.
Wright had heard about a new experimental drug called Krebiozen and pleaded to be included in the trial. His doctors agreed, largely as a matter of mercy. Within days of his first injection, the tumours had shrunk to half their size. Within ten days, he was walking around the ward, chatting with nurses. The tumours had, for all practical purposes, dissolved.
Then the newspapers ran reports suggesting Krebiozen was worthless. Wright read them. He relapsed almost immediately, and his tumours returned.
His physician, unwilling to give up, told him the newspaper reports were wrong – that a new, improved, double-strength version of the drug had arrived and that this time it would work. He injected Wright with plain water.
The tumours dissolved again. Wright recovered again.
Two months later, the American Medical Association issued a definitive statement: Krebiozen was useless. Wright read it. He died within days.
The water had worked. The newspaper had killed him. In neither case was the active ingredient the molecule.
Sham Surgery That Worked
If the Mr. Wright case feels too singular to be convincing, consider what happened in Houston in the late 1990s when a surgeon named Bruce Moseley ran one of the most provocative trials in medical history.
Moseley performed arthroscopic knee surgery on patients with osteoarthritis – a routine, widely performed operation. But for a third of his patients, he performed sham surgery: he made the incisions, went through all the physical theatre of the real procedure, and then sewed the patients back up without doing anything to the knee.
The results, published in the New England Journal of Medicine in 2002, were deeply unsettling for the surgical community. The sham surgery group improved just as much as the real surgery group. Not slightly less. Just as much. Patients who had received no actual intervention reported less pain, better mobility, and higher satisfaction scores.
This was not a fringe result. It was a rigorous randomised controlled trial in one of the most respected medical journals in the world.
The implication was unavoidable: a significant portion of the benefit from knee surgery – in some patients, perhaps all of it – comes not from what the surgeon does to the joint, but from the patient’s belief that something effective has been done.
Placebos Work Even When You Know They’re Placebos
The obvious objection to all of this is that it only works when people are deceived. Tell someone they’re getting a sugar pill, and the effect should vanish. Belief requires ignorance.
That assumption was tested in 2010 by Harvard researcher Ted Kaptchuk in a trial of patients with irritable bowel syndrome. Half received no treatment. The other half were given a bottle clearly labelled “Placebo Pills – made of inert ingredients.” They were told explicitly, in plain language, that the pills contained no active medicine.
The placebo group improved significantly anyway. Not just slightly. Significantly. Significantly more patients in the placebo group reported adequate symptom relief compared to the no-treatment group.
Kaptchuk’s interpretation was careful: the benefit wasn’t coming from belief in a lie. It was coming from the ritual of treatment itself – the act of taking something, the relationship with a physician who was paying attention, the implicit message that something was being done. These things are real. They have real physiological correlates. Deception is not required.
The Hotel Maids Who Lost Weight Without Dieting
The placebo effect is not confined to medicine. In 2007, psychologist Alia Crum (then a graduate student at Harvard, working with Ellen Langer) published a study that has become one of the most cited pieces of research on the relationship between belief and physiology.
Crum and Langer recruited 84 hotel maids – women whose daily work is genuinely physical. They were cleaning rooms, pushing carts, scrubbing bathrooms, changing beds. By any objective measure, they were getting substantial exercise. But most of them, when asked, said they didn’t exercise. They didn’t perceive their work as exercise.
One group of maids was given information explaining that their work counted as exercise and met recommended physical activity guidelines. The other group was told nothing.
After four weeks, without any change in actual behaviour, the informed group showed measurable improvements in weight, blood pressure, body fat, waist-to-hip ratio, and body mass index. The group told nothing showed no change.
The only variable was what they believed about what they were already doing. The belief changed the body.
The Darkness of the Same Coin: The Nocebo Effect
Everything the placebo can build, its shadow – the nocebo – can destroy.
The nocebo effect is what happens when negative expectations produce real harm. It is less discussed, partly because it is less commercially convenient, and partly because it is genuinely troubling. If belief can heal, belief can also kill.
In 1992, the physician Clifton Meador published an account of a man he called Sam Shoeman, who had been diagnosed with end-stage oesophageal cancer and told he had months to live. Shoeman died, on schedule, at roughly the predicted time. The autopsy, however, revealed something disturbing: the tumour was tiny. Far too small to have killed him. There was no clinical explanation for his death consistent with the size of the cancer.
Meador’s interpretation was blunt: Shoeman had, in effect, been killed by a diagnosis. He had believed with complete conviction that he was dying, and his body had obliged.
The Poison Ivy That Only Grew Where Expected
In 1962, Japanese researchers Yujiro Ikemi and Shunji Nakagawa published a study that has never lost its power to disturb.
Thirteen schoolboys who had declared themselves highly sensitive to Japanese lacquer tree (which produces urushiol, the same irritant as poison ivy) were blindfolded and told their right arm was being touched with the lacquer tree and their left arm with a harmless chestnut tree leaf. The experimenters reversed the actual stimuli: the left arm received the irritant, the right arm received the harmless leaf.
Every boy developed a reaction – on the arm where they expected to develop a reaction. Most of them showed no reaction, or a reduced reaction, on the arm that had actually been touched with the irritant.
Their skin was responding to information, not chemistry.
Voodoo Is Real (Just Not in the Way You Think)
In 1942, the physiologist Walter Cannon published a paper titled “Voodoo Death” in the journal American Anthropologist. Cannon had collected documented accounts from multiple cultures – Aboriginal Australia, Africa, the Caribbean – of individuals who had been cursed or hexed by a figure of authority and who subsequently died, sometimes within days, with no identifiable physical cause.
Cannon proposed a physiological mechanism: the activation of the sympathetic nervous system under extreme and prolonged terror, leading to a state of irreversible shock. The curse was real because the belief in the curse was real. The body does not distinguish clearly between a threat it has been told about and a threat it can verify.
This is not magic. It is not superstition. It is the same mechanism, inverted, that allows a sugar pill to shrink tumours.
Hypnosis as a Mega-Placebo
Irving Kirsch – the same psychologist whose meta-analyses punctured much of the received wisdom about antidepressants – has proposed that hypnosis is best understood as a “non-deceptive mega-placebo.” It is, in his formulation, a context specifically designed to amplify expectancy, which turns out to be the strongest single predictor of whether any intervention, medical or psychological, will work.
When a hypnotherapist creates the conditions for deep suggestion, they are not doing something categorically different from what Kaptchuk’s doctors were doing when they handed patients a bottle labelled “Placebo Pills.” They are constructing a set of expectations and a context that gives the mind permission to do what it is already capable of doing.
This is not a limitation of hypnosis. It is hypnosis. The mechanism is the mechanism.
The distinction between a drug that works via its molecule and a placebo that works via belief starts to blur considerably when you look at the data. Antidepressants, for instance, have been shown in Kirsch’s own meta-analyses to perform only marginally better than placebo in most patients – and the placebo response in antidepressant trials is itself enormous, and has been growing over decades as the cultural expectation of antidepressant effectiveness increases.
What this suggests is not that antidepressants don’t work. It suggests that we have been thinking about “the mechanism” too narrowly. Molecules and beliefs both produce measurable, physiological change. The only question is which pathway is being used.
Rewriting “Only”
The case studies above are not anecdotes in the dismissive sense – tales told in the absence of evidence. Most of them come from published, peer-reviewed research. Mr. Wright’s case was reported by a physician. The Moseley knee surgery trial appeared in the New England Journal of Medicine. Kaptchuk’s open-label placebo trial was published in PLOS ONE. Crum and Langer’s hotel maid study ran in Psychological Science. Ikemi and Nakagawa’s poison ivy research was published in a Japanese psychosomatic medicine journal. Sam Shoeman’s case appeared in the Southern Medical Journal.
These are documented findings, not folk tales. And taken together, they point toward something that is simultaneously deeply unsettling and deeply hopeful:
The mind is not a passenger in the body. It is a driver. And it is driving continuously, whether or not we are paying attention to it.
“Only a placebo” means: the body just healed itself based on the expectation that it could. There is nothing “only” about that. There is, in fact, nothing more remarkable in all of medicine.
My mum the pharmacist was right that placebos are not the drug. She was not quite right that they are therefore nothing. What they are is the clearest available demonstration that the boundary between belief and biology is far more permeable than we were taught.
The placebo effect is not a bug in our research methodology. It is a glimpse of something true about what we are.
Sources: Bruno Klopfer, “Psychological Variables in Human Cancer,” Journal of Projective Techniques, 1957. J. Bruce Moseley et al., “A Controlled Trial of Arthroscopic Surgery for Osteoarthritis of the Knee,” New England Journal of Medicine, 2002. Ted Kaptchuk et al., “Placebos without Deception,” PLOS ONE, 2010. Alia Crum and Ellen Langer, “Mind-Set Matters: Exercise and the Placebo Effect,” Psychological Science, 2007. Clifton Meador, “Hex Death: Voodoo Magic or Persuasion?”, Southern Medical Journal, 1992. Yujiro Ikemi and Shunji Nakagawa, “A Psychosomatic Study of Contagious Dermatitis,” Kyushu Journal of Medical Science, 1962. Walter Cannon, “Voodoo Death,” American Anthropologist, 1942. Irving Kirsch, “Clinical Hypnosis as a Nondeceptive Placebo: Empirically Derived Techniques,” American Journal of Clinical Hypnosis, 1994. Irving Kirsch, “Antidepressants and the Placebo Effect,” Zeitschrift für Psychologie, 2014.
