Laughter and health: what really holds up

In the nineteen sixties the American journalist Norman Cousins had a projector and a box of comedies wheeled into his hospital room. He lay there with an inflammatory disease of the spine, in pain severe enough that he could not turn onto his side, and by his own account the outlook was poor. Then he noticed something that interested him more than anything else in the ward: ten minutes of proper belly laughter bought him two hours of sleep without pain.

He wrote a book about it. Anatomy of an Illness came out in 1979, became a bestseller, and released into the world a sentence everyone has heard since: laughter is the best medicine. Cousins recovered, his story became the founding legend of a whole field, and the question of what actually holds up in it hung in the air for another forty years.

Why the Cousins story is not evidence

It was one patient describing himself. That alone would justify caution, but there is more. Cousins was also receiving high doses of intravenous vitamin C, he had moved out of the hospital into a hotel room where nobody woke him at six to take his temperature, and for the first time in a long while he felt that his own treatment was his decision. Which of those did the work cannot be separated out afterwards.

Add to that an awkward property of the illness: in some people it settles down on its own and nobody really knows why. Turn such a case into a method and you get exactly the kind of story that can be quoted forever and verified never. Cousins himself never claimed that comedies cure anything. He wrote about his own part in his treatment, about hope, and about a patient being more than a body in a bed. The shortcut "laughter heals" was manufactured by the audience.

A founding legend is a decent way to open a conversation. The answer to what laughter actually does to the body must come from somewhere else.

What goes on in the body while you laugh

Laughter is first of all a breathing event. On a single intake of air the body pushes out a run of short, regular exhalations, and in between you are barely breathing at all. The diaphragm, the muscles between the ribs and the abdominal wall all join in, which is why your stomach aches after a long fit of laughing much as it does after a set of sit-ups.

During the laughing itself, heart rate and blood pressure briefly climb. The more interesting part comes afterwards: in the minutes after it subsides, both drop below where they started and muscle tension lets go. This is measurable and real. It is also short lived, minutes to a few tens of minutes, and the shift is small.

The breath is worth a second look. During a fit of laughing you are not breathing in the usual sense, only squeezing out the last of the air, and when it ends a deep intake has to follow. That forced deep breath with a slow release is the most plausible candidate for the stretch of calm that arrives after laughter. Slow breathing without a single joke manages it too.

Much the same goes for calories, the subject of more nonsense than anything else here. Measurements suggest that ten minutes of laughing costs single digits to a few dozen calories more than sitting quietly. That is not nothing, but it is not an hour in the gym either, and nobody has ever laughed their way to a smaller waist.

A sober summary of the physiology runs like this: to the body, laughter resembles a short, mild exertion followed by a moment of release. If that were all, this article would stop here. Except the most interesting numbers are not in the lungs.

What is documented decently well

The most modest claims stand up best. The immediate effect on mood is repeatable and surprises nobody: after a fit of laughing, people report less tension, and that is measurable again and again. It is not deep and it does not last until the evening. But it holds.

It is worth spelling out what "documented decently well" means here. We are talking about laboratory experiments with a few dozen people, often students, measuring a window of several minutes to an hour after watching something funny. A design like that reliably catches an immediate change. About what happens to a person over five years it says nothing, and nobody should ask it to.

Considerably more interesting is a finding from Robin Dunbar and his team in 2011. People who had laughed together in a group then tolerated a higher level of an unpleasant stimulus than people who had watched something neutral. Their pain threshold, in other words, went up. The explanation points towards endorphins, the body's own damping system, also set off by running or singing in a choir.

None of which was about pharmacology. What interested Dunbar was what binds people into groups now that we have no time to groom each other's fur the way apes do. Shared laughter, on his account, is the cheap substitute: it takes a few seconds, it works on several people at once, and it leaves a good feeling about the other person. The raised pain threshold is a side product, not the point.

The key word in that experiment is "together". This was real, contagious laughter in a group, not a polite smile at a screen. And that brings us to a fact that catches most people out: laughter is roughly thirty times more likely in company than alone, and only a minority of it follows a joke. Robert Provine established this by spending years listening for when people on the street laugh, and wrote it up in a book in 2000. The mechanism is taken apart in more detail in the piece on why we laugh in the first place.

Where the legend outgrew the data

Here is an honest sorting of the most common claims by what is left:

Behind most inflated headlines sits a single trap known as reverse causation. Ask who laughs a lot. Someone who has people to laugh with, who is not in pain, who sleeps and is not sitting alone in a flat. So when data show that people who laugh more are healthier, it can mean that laughter makes health. Or that health makes laughter. Most studies cannot pull those apart.

Then add the way laughter gets measured. The usual questionnaire item runs along the lines of "how often do you laugh", except that people do not count their laughs. They answer from how they remember their mood over the past week. What you measure is closer to a memory of feeling all right than to laughter.

A study that would settle the question would look something like this: take a thousand similar people, assign half by lottery to ten years with plenty of laughter and half to ten years without, then compare. Nobody will run that experiment, because laughter cannot be handed out like a tablet and withholding it for a decade would never be approved. What remains is questionnaires and short laboratory trials, and that is exactly the level of certainty we have to make do with.

Laughter yoga without the halo

Carol goes to the community centre on Thursday evenings for laughter yoga. It is a group activity that started in India in the mid nineties on a simple idea: you can laugh on purpose, without jokes, as a breathing exercise. Eighteen people stand in a circle, clap, breathe and force out "ha ha ha" until it tips into the real thing. Carol says the first ten minutes are excruciating and after that they are not.

Her brother in law Dave took the mickey out of it for half a year, until the evening he went along. He came home reporting it was exactly as strange as he had expected, and he has been going back ever since. That is a fairly good description of what we know about the method so far.

Studies of it tend to be small, short and awkward to build a control group for. It is hard to convince anyone they were not just laughing. What shows up repeatedly is better mood immediately after a session. What does not show up is anything approaching a therapeutic effect.

The likeliest explanation is dull and also the point. The active ingredient is probably not the diaphragm but the fact that eighteen people meet every Thursday at six, talk and walk out to the bus stop together. Laughter therapy works mostly as company therapy. That is no put down: the social component is the best supported thing in this whole subject.

Who with, not what at

Dave watches the same comedy series alone at his laptop and gets through a whole episode without laughing once. In Carol's kitchen the same episode does not survive to the halfway mark, because somebody is constantly taking aim at somebody else. Identical content, incomparable outcome.

The practical consequence is unexpectedly concrete. If you decide you want to laugh more, you usually reach for a list of films. The calendar is more useful. When did you last laugh hard enough that your stomach hurt? And was anyone there, or was it in bed with your phone?

For most people that answer narrows to two or three specific people and one situation that repeats. Some of us only really lose it over Friday drinks. For others it is a phone call with a sister, or a colleague's never ending grievance at the coffee machine. When such a window vanishes from the week, it usually has not vanished because life stopped being funny. It vanished because the meet up, the call or the break dropped out.

How you laugh plays a part as well. Rod Martin and his team described four humor styles in 2003, sorted by whether the joke is warm or sharp and whether it points at others or at you. The bonding registers go with better wellbeing, the self mocking one with worse. These are correlations rather than causes, and no style is a defect. It is closer to a habit, and habits can be nudged. For a rough sense of where your joking tends to land, there is a short humor styles test.

The most useful way to look at laughter is not as a medicine but as a thermometer. If there is a stretch of life in which you barely laugh, that information cures nothing, but it points rather precisely at the state you are in. Putting on comedies then is like breathing on the thermometer for a better reading. Calling Carol and Dave is a different move.