← Longevity & Supplement Guides
Hans Selye, a Failed Hormone Hunt, and How ‘Stress’ Became a Word
In the mid-1930s, in a small and crowded laboratory at McGill University in Montreal, a young biochemist named Hans Selye was looking for a new hormone. He had been injecting rats with extracts of cow ovary and placenta, and the rats kept showing the same three changes: swollen adrenal glands, a shrunken thymus and lymph tissue, and ulcers in the stomach and gut. A new hormone would have made his career. So he did the obvious control. He injected other extracts, then dilute formalin, which comes from no living tissue at all. The rats showed the same three changes.
There was no new hormone. What Selye had found was a body responding to being harmed, whatever the harm happened to be. A few months later he gave that response a name, and later still he gave it a word he would spend the rest of his life explaining. The word was stress. Almost everyone reading this used it this week. Very few of us use it the way he meant it.
A hormone that wasn’t there
The record of that turn is unusually plain. A biographical profile from Selye’s own lifetime describes it as a failed hormone hunt: the organ changes appeared after ovarian and placental extracts, and then also after formalin and after almost any kind of injury. The first thing he discovered was that his hypothesis was wrong.
What he did next is the useful part. Instead of dropping a messy result, he widened it. He exposed rats to cold, to surgical injury, to spinal shock, to sublethal doses of very different drugs, and to excessive muscular exercise on a treadmill. If the response was the same after all of them, then the response itself was the finding.
Selye later said the idea was older than the rats. As a medical student in Prague in 1925, he had been struck that doctors put all their effort into telling diseases apart, and almost none into what every sick patient had in common: the plain state of being ill. He called it “the syndrome of just being sick”. A decade later the rats gave him a way to measure it.
Seventy-four lines in Nature
He published on 4 July 1936, in a letter to Nature of about 74 lines titled “A Syndrome Produced by Diverse Nocuous Agents.” Selye had written “noxious”; an editor changed it to “nocuous”. The letter listed the agents, described the organ changes, and described a sequence over time: an acute first reaction, a phase in which the animal seemed to adapt and the changes receded, and, if the insult continued for weeks, a final phase in which the adaptation was lost.
Those three phases became the famous ones: alarm, resistance, exhaustion. He called the whole sequence the general adaptation syndrome. “General”, because it did not depend on the cause. “Adaptation”, because he saw it as defence, not damage.
Look at the word he did not use as his headline. The 1936 title is about a syndrome and about agents. The language of the paper is the language of pathology. The single word that would carry the idea around the world was not yet its name.
Selye’s model, redrawn
The general adaptation syndrome: a shape, not a measurement
The first hours
Resistance briefly drops below normal, then the body mobilises. In the rats: adrenals enlarge, thymus shrinks.
Days to weeks
The body adapts and holds resistance above normal. The organ changes recede. This is the phase the system is built for.
If it never stops
Only when the same insult continues without relief does adaptation fail. Duration, not the alarm, is the problem.
Redrawn after the schematic Selye used in his books from the 1950s. The curve shows a shape, not data, so the axis has no units on purpose. In 1936 the animals were rats, and the stressors were cold, injury, drugs and forced exercise. That is a long way from a difficult inbox.
The world before: one cause, one disease
Selye’s idea was hard to accept because it went against the most successful idea in medicine at that time. Since the late nineteenth century, progress had come from specificity: one microbe, one disease; one missing vitamin, one deficiency; one hormone, one gland. A good scientist looked for the specific cause. A “non-specific” response sounded like the noise you remove to see the signal.
There was one strong precedent. The Harvard physiologist Walter Cannon had described the body’s emergency reactions in Bodily Changes in Pain, Hunger, Fear and Rage (1915), the book that gave us “fight or flight”. In The Wisdom of the Body (1932) he made popular the word homeostasis, the body’s ability to keep its internal conditions steady. But Cannon’s emergency was fast and nervous: adrenaline, a racing heart, seconds to minutes. Selye’s was slow and hormonal, and it ran through the adrenal cortex over days and weeks.
Physiologists were not impressed. The historian Russell Viner, who studied how the theory was made, concludes that Selye’s findings were largely rejected by academic physiology until the 1970s. Selye did not win over the physiologists. He went past them. A contemporary profile credits Frederick Banting, the co-discoverer of insulin, with his first grant: a modest $500. The rest came from a much larger audience.
Borrowing a word from engineers
“Stress” was an engineering word before it was a medical one. In the physics of materials, three terms are separate. The load is the weight you put on a beam. The stress is the internal force that load creates in the beam. The strain is how much the beam deforms.
What Selye had found in his rats was a deformation, the body’s change in response to a load. In engineering terms, that is closer to strain. McGill’s own account of his work says he came to regret the borrowed physics word. Several later accounts say he wished he had chosen “strain”. Either way, the confusion was built in from the start, and it was his.
The word also had problems in other languages. When Selye gave a series of lectures at the Collège de France in Paris in 1946, he struggled to say “stress” in French. The French words that came close (agression, tension, détresse) each covered only part of the idea, and he chose to use the English word with a masculine article. German writers translated it as Schaden, damage. Spanish writers used sufrimiento, suffering. In each translation the idea moved a little toward harm.
Thirty years later Selye still had to clean up after the word. In a 1975 paper titled “Confusion and controversy in the stress field” he wrote that people kept using “stress” for both the cause and the response, and that the only fix was to split it: the stressor is what happens to you, stress is what your body does about it. Half a century on, most of us still use the word for the cause.
From the laboratory to the living room
Selye was very productive. He published more than 1,700 articles and 39 books, and a recent review counts ten nominations for the Nobel Prize, none successful. In 1945 he left McGill to direct the Institute of Experimental Medicine and Surgery at the Université de Montréal. In 1950 he published a large treatise on “the physiology and pathology of exposure to stress”. By then the word had settled.
Then he wrote for everyone else. The Stress of Life (1956) took the adrenal gland to the general reader. Its editor compared him to Pasteur. Stress Without Distress (1974) went further: it was a philosophy of how to live, not only a physiology. It made popular a word Selye used for the kind of stress that does you good: eustress, from the Greek prefix for “good”. The title of a 2023 review of his legacy quotes his best-known line about it: stress is “the spice of life”.
Viner’s argument is that Selye succeeded by recruiting a wide audience outside academic physiology. Each group saw in “stress” proof of something it already believed about the mind, the body and industrial life. And that is how a finding about adrenal glands in injured rats became a word for a Monday.
Timeline
How a lab finding became a word on your wrist
Dates from Selye’s 1936 letter, Rochette et al. (2023), Viner (1999), Petticrew & Lee (2011), Mason (1971), Selye (1975) and McEwen (1998). Rows in coral mark the two turns that matter most here; the grey row marks the funding the stress story did not mention for decades.
Who paid for the stress story
This is the part of the story that his popular books leave out. In 2011, Mark Petticrew and Kelley Lee of the London School of Hygiene & Tropical Medicine published an analysis of internal tobacco-industry documents in the American Journal of Public Health. Their finding: Selye, often called “the father of stress”, received extensive tobacco-industry funding, and his work was used in litigation to argue against a causal role for smoking in heart disease and cancer.
The documents, as they report them, show a relationship that grew over time. In 1958 Selye asked a cigarette maker for research money and was refused. In 1959 industry lawyers asked for his help with the argument that a statistical link between smoking and cancer did not prove cause. In 1966, lawyers for two tobacco companies paid him $2,000 for a report. In 1969 the industry’s Council for Tobacco Research gave him a “special project” of US$50,000 a year for three years, and the Canadian industry pledged a similar amount. That June he testified before a Canadian House of Commons committee against health warnings and advertising restrictions on cigarettes.
The logic was simple, and it is why this matters for the word. If disease comes from “stress”, a vague load that modern life puts on everyone, then the specific causes become less visible. Selye also presented smoking as a way to relieve stress. Petticrew and Lee conclude that the history has implications for how we judge the scientific integrity of parts of stress research.
None of this cancels the rats. The 1936 observations were real and they were repeated. But it is useful to know that the idea of stress as a general explanation for illness was, for some years, financially supported by people who wanted that explanation to replace a more specific one.
The myth: “stress” is the thing that happens to you
The popular version of Selye says that stress is harmful, that he proved it, and that the aim of a good life is to have less of it. Each of those three ideas is wrong in a different way.
First, in his own terms, stress was the response, not the event. Second, the response was a defence. His model puts the damage only at the end, in the exhaustion phase, after a load continues without relief. The alarm and the adaptation are the body working correctly. Third, he did not aim for zero. The man who made “eustress” popular thought a life with no demands would be its own kind of problem.
His central claim, that the response is the same whatever the cause, has not survived intact either. In 1971 the American researcher John Mason, who had spent years measuring stress hormones in primates, published a re-evaluation of “non-specificity”. He argued that much of what looked like one response to many physical insults could come from the one thing those insults had in common: they frighten and upset the animal. If so, the common factor is an emotion, not the damage. The meaning of an event, not only its physical force, shapes the response.
Viner’s last observation is the sharpest. The modern, everyday use of “stress”, he writes, contains none of the physiological postulates of Selye’s original findings. We kept the word and lost the theory.
What a wrist means by “stress”
Which brings the story to the number many people now read in the morning. Most wearables, including the estimated stress load on the Agen Band, build their stress figure from heart rate and heart-rate variability. When your nervous system shifts toward arousal, heart rate tends to go up and the beat-to-beat variation tends to go down. The device sees the shift and reports it.
In one way, this is Selye’s non-specific response on your wrist. The sensor cannot tell a hard interval session from an argument, a happy wedding from a work deadline, or excitement from dread. It records the body’s mobilisation, and it is blind to the cause and to what the cause means to you. Mason’s point stands too: the meaning you give an event changes the response the device then records.
So a stress score is honest in a narrow way. It is an estimate of arousal, built from derived signals, and it is most useful as a trend against your own baseline, not as a verdict on your day. It does not diagnose anything. A high number after a hard training session is the alarm phase doing its job. The pattern Selye would have worried about is different: a raised number that stays raised for weeks, with no recovery between.
What the modern evidence actually says
The best modern frame comes from the neuroendocrinologist Bruce McEwen. In a 1998 review in the New England Journal of Medicine, he described the stress mediators (cortisol, adrenaline and their partners) as both protective and damaging. In the short term they help you adapt, a process he called allostasis. When they are turned on too often, or fail to turn off, the cost builds up as “allostatic load”. That is Selye’s exhaustion phase, rewritten with better chemistry: the danger is in duration and recovery, not in the response itself.
The second line of evidence is about belief. Abiola Keller and colleagues linked the 1998 US National Health Interview Survey (28,753 adults) to death records through 2006. People who reported a lot of stress and believed that stress affected their health a lot had a 43% higher risk of dying in that period (HR 1.43, 95% CI 1.2–1.7). This is an association from one survey question. It cannot separate a harmful belief from people who correctly sense that something is already wrong. But it fits the idea that the meaning we give stress matters.
Small laboratory experiments point the same way. In a randomised study by Jeremy Jamieson and colleagues (2012), people told to see their racing heart before a stressful task as useful and adaptive showed more efficient cardiovascular responses (higher cardiac efficiency, lower vascular resistance) than controls. Alia Crum and colleagues (2013) showed that short films could change people’s “stress mindset”, and that a stress-is-enhancing mindset was associated with moderate, not extreme, cortisol reactivity. These are short, controlled studies in volunteers. They show that interpretation can change the response in the moment. They do not show that a mindset changes long-term health.
Load, stress, strain: three words to use carefully
Load is what you carry: the training block, the deadline, the new baby. Stress, in Selye’s later language, is your body’s response to it. It is measurable in part, and it is not good or bad by itself. Strain, in the engineer’s sense, is the change the load leaves behind. It is the word Selye reportedly wished he had used, and recovery exists to undo it. When a day feels heavy, it helps to name which of the three you mean.
What his story is evidence for
I think Selye’s life supports three conclusions, and none of them is “avoid stress”.
One: a measurement can be real while the meaning given to it drifts. The adrenal changes were real. The word that described them moved, over fifty years, from “the body’s defence” to “the thing that is ruining your life”. Most health numbers go through something similar on the way from the laboratory to the app.
Two: who pays for an idea is part of the idea. A general, cause-free explanation of illness was useful to an industry that wanted attention moved away from a specific cause. That does not make the explanation false. It does mean you should ask, of any explanation that conveniently blames everything in general, who is pleased that it is blamed.
Three: the part of Selye’s model that has held up best is the shape, not the word. Mobilise, adapt, and fail only when the load does not end. Training plans guided by heart-rate variability use exactly this logic: add load, watch the response, and back off when recovery does not come back. It is the general adaptation syndrome, used on purpose.
The bottom line
“Stress” began as a rat’s defence against harm and became our word for the harm. Selye would have told you the response is the healthy part, and that the risk is in a load that never stops. This week, when your stress number or your own sense of pressure goes up, ask two questions before you try to make it go down. Did I choose this load? And when did I last recover from it? If the answer to the second is “not for weeks”, that is the signal worth acting on. Rest is part of the answer, and so are a slow, long exhale and a look at your HRV trend. If the pressure stays high, or starts to affect your sleep, health or mood, talk to a doctor. A raised number on its own is just the alarm phase, and the alarm phase is the system working.
