The Mating Mind: How Sexual Choice Shaped the Evolution of Human Nature (54 page)

From the first greeting to the millionth word, much can go wrong. Personalities clash. Arguments go unresolved. Incompatibilities arise. Jokes fall flat. Boredom ensues. Both individuals must clear the million-word hurdle before they contribute to the next generation. When language first evolved, it may have been a ten-word hurdle, or a thousand-word hurdle. But at each step, both individuals were trying to extract, by using the language available to them, as much information as they could. The more talking they did, the more of their minds they revealed. The more verbal courtship revealed, the greater effect sexual selection could have.

This courtship theory has been mocked as the "chat-up theory" of language evolution. It is all too easy to describe in salacious terms. One could write about nimble tongues playing across strong columns of warm air, the syncopated breath of lovers tickling those most sensitive surfaces of the human body—the eardrums—and conversation as minds dancing together in a tango of frenzied cognitive foreplay. But there is no reason to make sexual selection sound so lubricious. Human verbal courtship is the least superficial form of courtship that evolution has ever produced. A million words give a panoramic view of someone's personality, past, plans, hopes, fears, and ideals. It would be misleading to make our verbal courtship sound like second-rate erotica, or to focus on the risible chat-up lines sometimes heard in singles bars. Verbal courtship continues for months after people first meet, and it becomes the bedrock of human intimacy and love.

Public Speech as Covert Courtship

Verbal courtship can be viewed narrowly as face-to-face flirtation, or broadly as anything we say in public that might increase our social status or personal attractiveness in the eyes of potential mates. Sexual flirtation during early courtship accounts for only a small percentage of language use, but it is the percentage with the

most important evolutionary effects. This is the time when the most important reproductive decisions are made, when individuals are accepted or rejected as sexual partners on the basis of what they say. Yet, if language evolved only for face-to-face flirtation, we would talk much less than we do. Why do we bother altruistically giving away information when we are not directly courting a particular individual?

Verbal courtship in the broader sense explains why we compete to say interesting, relevant things in groups. Sexual choice permeates human social life, because anything that raises social status tends to improve mating prospects. If a man gains a reputation as an incisive thinker who consistently clarifies group decision-making and mediates social conflicts, his social status and sexual attractiveness increase. If a woman gains a reputation as a great wit and an inventive storyteller, her status and attractiveness increase as well. Public speaking and debate allow individuals to advertise their knowledge, clear thinking, social tact, good judgment, wit, experience, morality, imagination, and self-confidence. Under Pleistocene conditions, the sexual incentives for advertising such qualities would have persisted throughout adult life, in almost every social situation. Language puts minds on public display, where sexual choice could see them clearly for the first time in evolutionary history.

Form and Content

If language evolved for sexual display, shouldn't we go around trying to say the most difficult possible tongue twisters? Shouldn't human sexual competition follow the style of Cyrano de Bergerac, who demonstrated his physical and mental fitness to the beautiful Roxane by improvising a ballad of rhyming alexandrines, including three eight-line stanzas and one quatrain, while sword-fighting his sexual rival the Vicomte de Valvert, all timed perfectly so that Cyrano's last word coincided with the Vicomte's death? That would be impressive. But it is not what sexual selection demanded.

What we say is generally more important than how we say it.

The formal structure of language evolved principally as a medium for conveying ideas and feelings, which tend to attract sexual partners by revealing our personalities and minds. Sexual selection shapes language's content more than its form. Or rather, the form evolved in the service of the sexually selected content, rather than as a sexual display in its own right, as bird song did. Some of us prefer sexual partners with deep thoughts expressed succinctly to partners with many words but no thoughts. Sexual selection need not favor the superficial chatterbox over the Zen master who utters an enlightening and memorable 17-syllable haiku once a day. If it had, we would all resemble people with Williams syndrome, who tend to produce fluent, grammatical, large-vocabulary streams of relatively trite speech.

Nevertheless, there are some hints of sexual ornamentation in the human voice's pitch and timbre, the size of our vocabularies, the complexity of our grammar, and the narrative conventions of storytelling. For example, adult human males have deeper voices than children or women, which may reflect female choice favoring a low-pitched voice as an indicator of large male body size. (A deep voice does not have to correlate perfectly with large body size in order to work as an indicator.) Female frogs prefer lower-pitched male frog calls, and women generally find the deep, resonant voice of Isaac Hayes more sexually attractive than those of the Vienna Boys' Choir. Even in the television show
South Park
, the sexual charisma of Hayes's voice shows through in his school chef character, who, despite his low job status, credibly says lines indicative of sexual desirability, like "Damn, woman, I just gave you sweet lovin' five minutes ago!" On the other hand, low pitch could also have evolved through male competition as a threat display, as when the actor James Earl Jones provided the terrifying voice of Darth Vader in
Star Wars.

Apart from examples like this, there is not much evidence of sex differences or sexual selection in the details of language form. In analyzing these details, linguistics made reasonable progress by assuming that language evolved as a cooperative system for the transmission of information. The acoustics of speaking and

listening can be modeled fairly well by optimal information-transmission models where it is assumed that speakers and listeners are trying to minimize the joint costs of such transmission. Speakers pronounce words just clearly enough to be understood, but not so clearly that their jaws and tongues get exhausted; listeners work pretty hard to understand what is said, but not so hard that their auditory cortex evolves to enormous size. Likewise, the cooperative model has helped language researchers to understand grammar (syntax), word structure (morphology), and word meaning (semantics). These aspects of language make it look like a system designed for efficient information transmission.

However, the same cooperative model would work reasonably well in analyzing many details of peacock courtship displays. If one assumed that peacock courtship evolved for the efficient, cooperative transmission of iridescence patterns from peacock to peahen, one could successfully describe most of the anatomy of the peacock's tail and the physiology of the peahen's visual system. His tail works pretty hard to produce iridescence, but her eyes work pretty hard to perceive it. Her eyes may be optimally attuned to the wavelengths of light reflected by his tail, just as our ears are optimally attuned to the sound spectra produced by speech. The movement patterns of his tail may be optimally adjusted to produce maximum iridescence-transfer to her eyes under most lighting conditions. And so forth. At the level of signal transmission and reception, peacock courtship may have the appearance of a cooperative system.

The details of signal production and perception cannot usually distinguish cooperative communication from courtship display. The differences emerge more at the level of signal cost, signal content, receiver attitude, and overall pattern of social interaction— aspects of language not typically studied in linguistics. Courtship displays usually have high costs and high degrees of difficulty, taking into account everything relevant to display effectiveness. At first glance, human language looks like a very cheap and easy form of signaling. Once your species has evolved language, and you have learned language, and you are fit and healthy, and you have

something to say, and you have the attention of a potential mate, it does not take much time, energy, or effort to say it. The hard part, of course, is having something interesting to say. The difficulty of effective verbal courtship is not the cost of moving your jaw and tongue, but the cost of thinking of something verbally expressible that will impress another human. This cost depends entirely on the listener's threshold for being excited: intelligent listeners demand intelligent utterances, and these are difficult to produce.

In cooperative communication, the receiver may be mildly skeptical about the information conveyed. In courtship, the receiver is extremely judgmental not only about the information, but about the signaler. When listening, we automatically evaluate whether what is being said makes sense, whether it is congruent with what we know and believe, whether it is novel and interesting, and whether we can draw intriguing inferences from it. But we also use all of these judgments to form an impression of the speaker's intelligence, creativity, knowledge, status, and personality. We assess the information content of utterances, not just to make inferences about the world, but to make attributions about the speaker.

This is why perfectly grammatical, well-spoken, true sentences can fail as conversational gambits. Consider the old English nursery rhyme:

Tommy Snookes and Bessy Brookes Were walking out one Sunday;

Says Tommy Snookes to Bessy Brookes,

"Tomorrow will be Monday."

As a sentence evaluated according to traditional linguistic standards, Tommy's utterance is perfectly successful. It passes the tests of grammaticality. But as a social act of courtship, Bessy will not be impressed. Tommy's comment is too obvious. It is true, but irrelevant. It provokes no further thought or response. Bessy may

suspect Tommy of low intelligence, social laziness, or nervousness.

In real human social life, conversational failures like that of Tommy Snookes are relatively rare. This is not because everyone is good at verbal display, but because those who are not learn to keep relatively quiet. People tend to socialize with friends and sexual partners who show roughly their own verbal ability level—their verbal compatibility has already determined which social relationships were formed. The majority of human conversation occurs between sexual partners and long-term friends. They have already chosen each other as mates or friends precisely because their first few conversations were mutually interesting, evoking mutual respect and attraction. Ordinary talk between old friends and lovers still includes sufficient verbal display to maintain mutual respect, but may not include the same verbal fireworks as the first few conversations did. That is why conspicuous verbal display plays only a minor role in everyday speech. Thus the costs of effective display and the risks of display failure look low. But this is an illusion: meet someone new, and these costs and risks surge back into salience.

Many language researchers remain preoccupied with studying the principles of syntax, by inviting native speakers of a language to tell them which sentences follow the language's grammatical rules and which do not. These decisions are called "grammaticality judgments." From an evolutionary perspective, it seems peculiar for linguistics to focus on this very narrow sort of normative judgment. People often speak ungrammatically in real conversation, but such rule violations are almost always ignored. People are much more interested in normative judgments about whether a speaker is truthful, relevant, interesting, tactful, intelligent, and sympathetic. Traditional linguistics has exiled all such questions to the subdiscipline of "sociolinguistics," which concerns how people use and judge language in real social interactions. Sociolinguistics is the evolutionarily crucial level of analysis, where all the social and sexual pressures that could have shaped language show themselves. But modern sociolinguistics is a small,

underfunded social science that has proved highly skeptical of evolutionary psychology.

We have a quandary: the syntax theorists who study grammaticality judgments dominate the conferences on language evolution, while the sociolinguists who study more evolutionarily important social judgments about speakers will not talk to evolutionary psychologists. Grammaticality judgments are extremely useful scientific data for analyzing the principles of syntax, but social judgments about the intelligence, personality, and attractiveness of speakers are much more potent as selection pressures. (Of course, social judgments of a speaker's intelligence may rely, in part, on judging their grammar, along with what they have to say, their voice quality, their social tact, their verbal self-confidence, and so forth.) We need an evolutionary sociolinguistics that can finally test evolutionary theories of the social and sexual benefits of language against data on the social and sexual uses of language in different cultures. From the standpoint of traditional linguistics, syntax, morphology, and semantics are the core of human language—but from a Darwinian viewpoint they are just the incredibly complex design details of a signaling adaptation centered upon social functions and social content.

Life Stories

Verbal courtship allows individuals to tell their life stories quickly and verifiably. Humans can learn more about each other in an hour than mute animals can in months. Within minutes of boy meets girl, boy and girl typically know each other's names, geographical origins, and occupations. In the first heady hours of chatter, they usually learn about each other's families, past and current sexual relationships, children, friendships, work colleagues, adventures, travels, ideological convictions, hobbies, interests, ambitions, and plans. By the time a sexual relationship has lasted a few months, lovers usually have a pretty good idea of each other's lives from childhood onwards. By contrast, chimpanzees can never gain direct information about one another's past experiences or long-term plans. They can only make a few rough