The wounded soldiers at Anzio should have been in agony. Henry Beecher, a Harvard anaesthetist serving with the US Army in Italy during the Second World War, noticed something strange: men with shattered limbs and deep shrapnel wounds often reported surprisingly little pain and turned down morphine — at rates far higher than civilians with comparable injuries back home. Beecher's conclusion was radical for its time. Pain is not a simple readout of tissue damage. Context — in this case, the overwhelming relief of being alive and headed away from the front — can reshape what the body actually feels.
Back at Harvard, Beecher pushed the idea further. In 1955 he published a paper called The Powerful Placebo, arguing that in study after study, roughly a third of patients improved after receiving treatments with no active ingredient at all. The paper had real methodological flaws — later critics pointed out that much of what Beecher counted as placebo response was simply people getting better on their own, or extreme symptoms drifting back towards normal. But its influence was enormous. It helped make the placebo-controlled, double-blind trial the gold standard of modern medicine: if a new drug cannot beat a sugar pill, it does not work.
The word itself is older and stranger than the science. Placebo is Latin for "I shall please", the opening word of a chant from the medieval Office of the Dead. Professional mourners hired to sing it at funerals became known, dismissively, as placebos — flatterers paid to fake grief. By the late eighteenth century, doctors had borrowed the word for medicines given more to please the patient than to cure them.
What is actually happening
For decades the placebo effect was treated as a nuisance, or as proof that the symptom was imaginary. Then researchers started finding real biology. In 1978, a team led by Jon Levine gave dental surgery patients a placebo painkiller, and then quietly administered naloxone — the drug used to reverse opioid overdoses. The placebo's pain relief vanished. The implication was startling: expecting relief had triggered the brain's own opioid system, and blocking that system blocked the effect. Placebo analgesia, in other words, is not "all in the mind" in the dismissive sense. It is in the brain, measurably, in chemistry.
Other conditions tell similar stories. Brain-imaging studies of people with Parkinson's disease have shown that a placebo injection can trigger release of dopamine — precisely the neurotransmitter the disease depletes. Expectation, it turns out, is not a vague mood but a physiological instruction.
The ritual matters as much as the pill. Studies have found that placebo injections tend to outperform placebo pills, that taking more pills beats taking fewer, and that the colour of a tablet nudges its effect — blue pills lean sedative, red and orange lean stimulant. The most theatrical demonstration came in 2002, when the surgeon Bruce Moseley published a trial in the New England Journal of Medicine on arthroscopic knee surgery for osteoarthritis. Some patients got the real operation. Others got anaesthesia, three small skin incisions, and a carefully acted-out pantomime of surgery — then were stitched up with their knee untouched. Two years later, the sham-surgery patients reported just as much pain relief as those who had the genuine procedure.
The honest placebo, and its evil twin
The standard objection to using placebos in the clinic is that they require deception. Ted Kaptchuk's group at Harvard tested that assumption in 2010 with patients suffering from irritable bowel syndrome. They handed out pill bottles boldly labelled "placebo" and told patients, truthfully, that the pills were inert but that placebos had been shown to help. The openly labelled placebos still beat no treatment. "Open-label" placebo research has since grown into a field of its own, suggesting the ritual of care — attention, a plausible treatment, a story about recovery — carries power even without the trick.
Expectation cuts the other way too. The nocebo effect — from the Latin for "I shall harm" — is the placebo's dark mirror: expect side effects, and you are more likely to feel them. In drug trials, participants receiving inert pills routinely report headaches, nausea and fatigue, and some drop out because of side effects from a substance that has none. Reading a long list of possible side effects can, by itself, help produce them.
None of this makes placebos a cure-all. The best evidence suggests they chiefly move symptoms that the brain itself constructs and regulates — pain, nausea, fatigue, anxiety. No sugar pill shrinks a tumour or mends a broken bone. But within that territory, the placebo effect is one of medicine's most reliable phenomena: a demonstration, repeated daily in clinics and trials around the world, that belief is not the opposite of biology. It is part of it.
Quiz nuggets
- Placebo is Latin for "I shall please"; its harmful twin, the nocebo effect, comes from "I shall harm".
- Henry Beecher's 1955 paper The Powerful Placebo helped make placebo-controlled trials the standard test for new drugs.
- In a 2002 New England Journal of Medicine trial, sham knee surgery relieved osteoarthritis pain as well as the real operation.
- Naloxone, the opioid-overdose antidote, can block placebo pain relief — evidence the effect uses the brain's own opioids.
- In a 2010 Harvard study, placebos worked even when the bottle was honestly labelled "placebo".