Isaac Newton is reported to have laughed precisely once in his life--when a friend asked him what use he saw in Euclid's Elements. Joseph Stalin, too, seems to have been somewhat agelastic. "Seldom did anyone see Stalin laugh," Marshal Zhukov reports in his reminiscences. "When he did, it was more like a chuckle." Other famous agelasts include William Gladstone and Ruth Bader Ginsburg. Did Jesus laugh? That was a core question of Umberto Eco's The Name of the Rose. Because I was unable to finish the novel, I'm not sure what the answer is; in any event, I do not recall many guffaws in the Gospels. ("The total absence of humour from the Bible," Alfred North Whitehead once observed, "is one of the most singular things in all literature.")

Rabelais declared, "Man is the laughing animal." Are we to infer, then, that Newton, Stalin, and the rest were somehow inhuman? Not necessarily. Rabelais was only echoing Aristotle, who in his Parts of Animals asserted not that all humans are laughers, but that "all laughers are humans." Some have contested even the latter claim. Thomas Mann and Isak Dinesen, for example, both insisted that their pet pooches were capable of a kind of inward laughter (though this would be hard to verify). What is certain is that purely spiritual entities--angels, gods--have no place on the laugher curve. That is because, whatever else it is, laughter is physical. You need to have a body to do it.

It is the sheer physicality of laughter that has made it so mysterious to philosophers and scientists. "Those who know why the kind of joy that kindles laughter should draw the zygomatic muscles back toward the ears are very knowing indeed," Voltaire wrote. Actually, the laughter reflex involves the contraction of some fifteen facial muscles, along with the simultaneous stimulation of the muscles of inspiration and those of expiration. Oddly, all these contortions prove very healthy for homo ridens: Robust laughter triples the amount of oxygen in the blood, reduces stress hormones, and bolsters the immune system by heightening T-cell activity.

To understand the nature of laughter, theorists have traditionally looked not to its effects but to its causes: the comic, the risible, the humorous. The easiest way to define the humorous is as that which elicits laughter. This, however, makes "humor causes laughter" a barren tautology. The proposition becomes useful only when humor is characterized independently of laughter.

There are three moldy old theories that purport to do this. The superiority theory--propounded in various forms by Plato, Hobbes, and Bergson--locates the essence of humor in the "sudden glory" (Hobbes) we feel when, say, we see Bill Gates smashed in the face with a custard pie. The incongruity theory--held by Pascal, Kant, and Schopenhauer--says that humor arises when the seemly and logical abruptly dissolves into the low and absurd. "Do you believe in clubs for small children?" W.C. Fields is asked. "Only when kindness fails," he replies.

These theories do not address, however, why feelings of superiority or incongruity should call forth a bout of cackling and chest heaving. An advantage of the relief theory--proposed by Freud--is that it at least tries to explain the causal link between humor and laughter. In this view, the laughable (ideally, a naughty joke) liberates the laugher from inhibitions about forbidden thoughts and feelings. The result is a discharge of nervous energy that distracts the inner censor from what is going on.

If none of these theories has ever been tested experimentally, the reason is simple: It is impossible to produce laughter under laboratory conditions. It is not, however, difficult to go out into the world and see what makes people laugh. Yet for all the centuries of speculation about laughter, no one had bothered to do this until Robert Provine came along. A psychologist who teaches at the University of Maryland's Baltimore County campus, Provine sent assistants to college campuses and shopping malls to eavesdrop on conversations and see what triggered volleys of laughter.

The results were dispiriting. The domain of the risible turned out to consist largely of such howlers as "Got to go now!," "What's that supposed to mean?," and "It must be nice." Only 10 percent of the laugh lines could be rated humorous even by the most charitable standards ("You don't have to drink--just buy us drinks"). These findings, along with the well-known contagiousness of laughter (which in its extreme form has caused epidemics of hysterical giggling among convent girls in Africa), have led evolutionary psychologists like MIT's Steven Pinker to conjecture that the primary Darwinian function of laughter is to serve as a social glue.

Could there really be a common denominator to a phenomenon that encompasses so many varieties? Diabolical laughter, laughter as a joyous awareness of our finitude, as an escape from gravity into levity, as a tool of mockery.

Perhaps the most delightful form of laughter, for the crafty don at least, is the risus sophisticus. According to the Oxford Companion to Philosophy, this is an ancient rhetorical counterploy identified by Gorgias of Leontini as "destroying one's adversaries' seriousness by laughter and their laughter by seriousness." Such tricks are "characteristically employed," the Companion adds, "by an aged philosopher commenting upon a paper of unfollowable complexity by a young post-doctoral Fellow."

JIM HOLT