Le bâillement, du réflexe à la pathologie
Le bâillement : de l'éthologie à la médecine clinique
Le bâillement : phylogenèse, éthologie, nosogénie
 Le bâillement : un comportement universel
La parakinésie brachiale oscitante
Yawning: its cycle, its role
Warum gähnen wir ?
Fetal yawning assessed by 3D and 4D sonography
Le bâillement foetal
Le bâillement, du réflexe à la pathologie
Le bâillement : de l'éthologie à la médecine clinique
Le bâillement : phylogenèse, éthologie, nosogénie
 Le bâillement : un comportement universel
La parakinésie brachiale oscitante
Yawning: its cycle, its role
Warum gähnen wir ?
Fetal yawning assessed by 3D and 4D sonography
Le bâillement foetal

mystery of yawning 





mise à jour du
29 octobre 2015
In Ullrich H. (ed.) Hominid Evolution: Lifestyles and Survival Strategies. Gelsenkirchen-Schwelm: Archaea
1999, pp. 262-268
Disinhibition of anti-cultural behavior:
A means of adaptation in early hominids?
Alexander Kozintsev and Marina Butovskaya


Three human behavior patterns, laughter, weeping and yawning, are discussed. Despite differing with respect to evolutionary age, emotional loading, and semantics, all the three can be regarded as displacement activities and social releasers. All the three are inhibited under normal conditions. When disinhibited, they disrupt speech and inhibit action. It is suggested that their primary common function in early humans was temporary inhibition of speech and action. This was necessary, firstly, because of the informational shock resulting from the emergence of language, and secondly, because language automatically generated culture as a system of norms and taboos. Laughter, weeping and yawning, then, might have evolved primarily as physiological and psychological means of preventing stress caused by language and culture.
This article focuses on the evolutionary and cultural aspects of three stereotypic human displays: laughter, weeping, and yawning. They have received strikingly different amounts of attention. While laughter has always been in the foreground, weeping was far behind, and yawning was almost completely neglected. Although all the three are quite mysterious, they seem to have little in common. Their evolutionary age is widely different, yawning being the most ancient and the other two patterns much younger. Their emotional component ranges from very high in weeping to zero in yawning. Their semantics appears to be very dissimilar since laughter and weeping are regarded as opposites and yawning as neutral. Finally, their functions in modern society are different: laughter is extremely social, weeping mostly individual, and yawning plainly anti-social. However, as we will try to demonstrate, they may prove by far more similar than is traditionally believed.
Laughter: biological and cultural origins
We will begin with laughter and its origins. It has been claimed that before apes have been taught to use sign language they are merely playful but humorless (McGhee 1979). As our observations on chimpanzees seem to demonstrate, captivity per se may give rise to rudimentary practical humor since playful violation of a prohibition in a way strikingly similar to certain archaic forms of festive reverse behavior in humans, known as "symbolic inversion" (Babcock 1978), is accompanied by facial expression obviously suggestive of pleasure (Butovskaya and Kozintsev 1996). We have described these forms of ape behavior as "presymbolic inversion" (Kozintsev and Butovskaya 1994).
It is beyond doubt, however, that the ability to sign automatically generates full-fledged humor based on the production of incongruities (McGhee 1979). The question is, why? One explanation is that playing with incongruity is only possible after symbolic communication has been mastered (McGhee 1979). But possible does not mean necessary. Why should intellectual play be necessary for anyone except intellectuals?
Although dozens of theorists have tried to explain the meaning of laughter and humor, these phenomena remain enigmatic (see Keith-Spiegel 1972, McGhee 1979, Raskin 1985, and Morreall 1987, for reviews). We can probably trace their history back to the archaic festive rites where cultural norms were reversed. However little we know about early forms of symbolic inversion, it is quite evident that their meaning extended far beyond the notion of intellectual play. H. Schurtz described these rites as safety valves (Ventilsitten), outlets for the release of subconscious collective discontent resulting from the assimilation of cultural norms and taboos. Sometimes they are seen as an antidote to tedium. Perhaps one of their objectives was to demonstrate the way things should not be and thus to uphold social order by contrasting it with symbolic mythological chaos (Stanner 1960, Leach 1961, Turner 1974, 1978, Babcock 1978, Abramyan 1983).
At the individual level, reverse behavior was practiced by ritual clowns or similarly marginal persons whose social role in archaic societies was highly contradictory since they were both contemptible fun-makers and awesome spiritual beings surrounded by taboos (Willeford 1969, Makarius 1970). Reverse behavior was modeled by the trickster myths which, like the reversal rites, were powerful laughter-provoking stimuli and may be viewed as the deepest cultural roots of humor (Radin et al. 1956; Belmonte 1989).
Laughter as a displacement activity
Of all the numerous theories of laughter, one that is especially relevant for us at present links laughter with so-called derived, or displacement, activities of animals. The term refers to inadequate behavioral responses which appear in situations of ambivalence wherever the adequate (autochtonous) drives are in conflict or the requisite releasing stimuli are absent. This usually happens either during fighting or during sexual activity (Tinbergen 1952). Leyhausen (1973) was probably the first to suggest that laughter is a displacement activity similar to sand digging in sticklebacks, self-licking in cats, and head scratching in humans. Later, the idea was supported by Russell (1987, 1996), who paralleled laughter with other human "substitute responses" like floor pacing, finger tapping, cold sweat, and vomiting.
Tinbergen (1952) believed that displacement activities are mere substitutes and have no motives of their own. Later, however, it was shown that they are motivated by the same stimuli as they would be in the normal context. The only difference is that these stimuli are weaker than the principal ones, so the respective actions are inhibited under normal conditions but become disinhibited in situations of conflict between stronger drives, frustration, motivational deadlock, or absence of sensory feedback (Andrew 1956, van Iersel and Bol 1958, Hinde 1970; McFarland 1985).
The disinhibition theory was used by Porshnev (1974) who linked it with Pavlov's ideas to hypothesize that human language was originally a form of displacement activity that had emerged by disinhibition and had later itself become an inhibitor (specifically, a means of inhibiting action in general and aggression in particular).
So if laughter is indeed a displacement activity, one must, first, find its own motives, and second, understand the reasons of its inhibition under normal conditions. There is no doubt that laughter is indeed strongly suppressed most of the time. It always "waits" for the moment to seize any opportunity and, when disinhibited, it spontaneously breaks out and, in turn, completely inhibits both speech and action. This is strange, given that the emotional component of laughter is small (Deacon 1992, 1997).
It is evidently not enough, then, to say that laughter is merely an expression of pleasure or that people laugh because humor is associated with intellectual play (McGhee 1979), pleasant psychological shift (Morreall 1987) or disturbance adjustment (Russell 1987, 1996). Moreover, certain kinds of laughter are neither caused by humor nor pleasant (Black 1982, Pfeifer 1994), and if they do not result from organic disorders or direct stimulation of the brain, it would be wrong to discard them as "inappropriate" or "parasitic". The same applies to the laughter of preschool children which appears to be "silly", "groundless" or caused by mere agitation (McGhee 1979).
Laughter as a social releaser
To move a step further, one should recollect that laughter is highly stereotypic, ritualized, and contagious, and these features are typical of displays which have a signal function in animals and are known as social releasers (Tinbergen 1959). According to Tinbergen (1952), some displacement activities have been ritualized and have assumed the function of releasers. He suggested that laughter is a releaser that has an appeasing function and has evolved from the defense or threat display through ritualization (Tinbergen 1959). The same view was expressed by Lorenz (1963). Indeed, ethological evidence suggests that laughter derives from the relaxed open-mouth display, or play face, which is a ritualized bite used by nonhuman primates during quasi-aggressive play (Bolwig 1964, van Hooff 1972). In terms of vocalization, however, human laughter is qualitatively different from that of the chimpanzee (Provine and Bard 1996). Apes know that the play face is a releaser and conceal it when they do not want to play (Tanner and Byrne 1993). Human laughter, too, is highly social (Chapman 1973; Provine 1996), and signals an intention to play (Grammer and Eibl- Eibesfeldt 1990). It is contagious by itself, even in the absense of any humor (Freedman and Perlick 1979; Provine 1991, 1996). Also, it has been experimentally demonstrated that humor and laughter are social lubricants that reduce anger and hostility (Dworkin and Efran 1967, Singer 1968, Landy and Mettee 1969).
But if the primary function of laughter was to prevent aggression, why is it so rudimentary in nonhuman primates and so highly developed in humans? Also, why is it most expressed in young monkeys and apes, who are not yet capable of violence, and is known to decrease with age, just the opposite of what might be expected if the aggression prevention hypothesis were true? Before attempting to answer these questions, we will briefly address weeping and yawning, of which much less is known.
Certain oft-cited anecdotal zoological examples notwithstanding, weeping is uniquely human. According to Darwin (1872:154), "a habit like weeping must have been acquired since the period when man branched off from the common progenitor of the genus Homo and of the non-weeping anthropomorphous apes." Weeping, like laughter, appears to be a releaser since it is very stereotypic and communicative. And, like laughter, it disrupts speech and immobilizes (Deacon 1992, 1997).
Although in modern society weeping is believed to be contagious only in young children (Hoffman 1977; Hatfield 1993) and is suppressed by adults in most situations, it was evidently much more social in the past. Collective weeping rites have been described in Australian aborigines, Andamanese natives, and American Indians, and they were related not only to funerary events but to occasions like reception of guests or reconciliation rites where collective weeping was regarded as an expression of friendship and social cohesion (Mauss 1969; Radcliffe-Brown 1933). Friederici (1907) termed this custom "greeting with tears". In many archaic religious traditions, collective weeping and laughter were performed either simultaneously or in succession since death was associated with eating, coitus, and rebirth (Reinach 1912, Frazer 1923, Hocart 1927, Freidenberg 1997).
It has often been claimed that weeping is adaptive because tears lubricate the eyeball and have bactericidal properties. But again, if this physiological mechanism is so valuable, why don't apes need it? Before this question has been answered we can only say that weeping in humans should be best regarded as a displacement activity and a social releaser.
Yawning, which is phylogenetically very much older than laughter or weeping, also appears to be a displacement activity (Tinbergen 1952). Its physiological meaning has become less clear than ever since the old idea that yawning is caused by the deficit of oxygen and the excess of carbon dioxide in blood was not supported by experimental data (Provine, Tate and Geldmacher 1987). This does not mean that yawning is physiologically useless in all vertebrates. This only means that natural selection might have adjusted an old display, which man had inherited from his remote ancestors, for some new purpose. Indeed, according to the principle formulated by Baerends (1958), displays are evolutionarily more conservative than their motivations.
Yawning is doubtless a social releaser since it is stereotypic and contagious to such an extent that the mere mention of it makes people yawn (Provine 1986). Yawning is certainly associated with sleepiness, but this can not explain its high contagiousness, because observing a sleeping person does not induce yawning in the observers or make them drowsy. Paralinguistically, yawning signals not merely sleepiness (Provine, Hamernik and Curchak 1987), but, more generally, lack of interest. Although we have no evidence of collective yawning rites, yawning was doubtless less individual in the past than it is now. Otherwise it is impossible to explain its extreme contagiousness. As our observations indicate, yawning in monkeys is very collective and means that the animals are about to cluster and fall asleep. In humans, yawning, like laughter and weeping, is suppressed most of the time. When disinhibited, it disrupts speech and inhibits action.
Laughter, weeping, and yawning: common function?
So all the three expressive patterns can be regarded as both displacement activities and social releasers. Moreover, it turns out that they are more similar than they appear at first sight. Maybe they originally had a common function? If they did, what was it? Our hypothesis is very simple. However semantically dissimilar laughter, weeping, and yawning may be now, they have had a primary common function which was precisely the same as it is at present: to inhibit speech and action in as many people as possible. Total incompatibility of speech with laughter or weeping (at least sobbing) has been well demonstrated on the brain level (Deacon 1992, 1997). Clearly, yawning is no less speech-inhibiting. Pleasure or relief that can be individually derived from these displays are epiphenomena. Each of the three can, but none of them must, be pleasant per se. Their primary adaptive value should be sought at the social, not at the individual level.
The idea that culture is a burden is very old. Perhaps its most ardent advocate was Freud (1978). But then language, too, is a burden: first, because it automatically gives rise to culture as a system of norms and taboos imposed on people (Porshnev 1974), and secondly, because of the informational shock resulting from the acquisition of a radically new system of symbolic communication. Porshnev (1974) noted that yawning might have provided early hominids with a means of coping with language shock. The same applies to laughter and weeping. In fact, all the three are signs of refusal. According to Freud (1959), humor is "an attitude by means of which a person refuses to suffer", and the ego "refuses to be distressed by the provocations of reality." The definition is much too narrow, of course, since any reality assimilation is tiresome at the least (McGhee 1979). Humor, then, is a playful refusal to deal with reality in a culturally appropriate way.
In the same line, weeping means a refusal to counter reality because it is too sad or too touching (Plessner 1970), and yawning a refusal to be interested in it. And, because assimilating symbolic information, reacting to the provocations of new cultural reality in an adequate way, and even merely being interested apparently required more mental strength and persistence than might be expected from creatures who had just descended from the apes (known to be highly negativistic), refusal, if only a temporary and imaginary one, to do so meant reversal and, at the same time, liberation (Mindess 1971). Perhaps the rudimentary "laughter" of young monkeys and apes, too, is not so much a means of preventing aggression, as a sign of liberation, and an invitation to join in getting liberated, from the necessity to comply with social norms. This is something that most adult animals either do not need or cannot afford. In this respect, as in many other respects as well, man is neotenic since he retains this juvenile feature, the capacity to laugh, throughout the lifespan.
For apes, as for other mammals, the principal means of preventing stress is sleep. Another means, which is important also for maintaining social integration in primate groups, is grooming. It has been hypothesized that in the course of human evolution, social grooming was replaced by language, which had become the principal factor of within-group integration (Dunbar 1993). If so, language had not merely introduced a new stress factor but had also deprived humans of an old stress-reducing device. In this situation, sleep would evidenty not suffice, so new means had to be urgently developed and/or old ones had to be adjusted ad hoc. 
It appears likely, then, that laughter, weeping, and yawning were psychological defense means that protected early humans from stress caused by language and culture, prevented neuroses, and at the same time, being presymbolic, ensured social cohesion on a deeper level than language could do. Paradoxically, laughter and weeping are often regarded as atavisms (which is to some extent correct), and yet they are almost uniquely human. It is precisely at the early stage of human history that they were especially adaptive. No wonder apes begin to joke as soon as they acquire the ability to sign; they do not begin to weep evidently for physiological reasons. We do not know how often the symbolizing apes yawn or how much they sleep. In humans, humor is indeed known to reduce stress (Dixon 1980; Lefcourt and Martin 1986), and it can be seen as an alternative to neurosis (Elitzur 1990). People who are susceptible to boredom appreciate humor more (Ruch 1988). Although the stress-reducing function of weeping and yawning has not yet been experimentally demonstrated, there is little doubt that it does, or at least did, exist.
However, because all the three patterns are manifestations of negativism and make people mute, weak, and inactive, they can only be used during relatively short and sharply delineated breaks between normal activities, otherwise culture would have been endangered. This explains why they are real displacement activities that are inhibited most of the time. Yet, unlike displacement activities in animals, they are strongly motivated and, when disinhibited, they violently break out and begin to function as social releasers.
Laughter, weeping and yawning are not the only patterns that interfere with speech. Respiratory adjustments such as coughing, sneezing, and nose blowing are speech-inhibiting and are sometimes used as paralinguistic means. However, they are by far less contagious (in a psychological, not medical, sense) and, although sometimes being similar to displacement activities, they can hardly be described as social releasers.