Adam's Tongue: How Humans Made Language, How Language Made Humans

by Bickerton, Derek

  According to George Williams, an icon of modern evolutionary biology, “Adaptation is always asymmetrical; organisms adapt to their environment, never vice versa.” On the face of things, this sounds indisputable; how could the environment—rocks and trees, wind and rain and sunlight—adapt itself to you and me? But a consequence of Williams’s position, one widely shared among evolutionists, is that evolution becomes a one-way street. “Adaptation” makes it sound as if organisms are doing something positive, but that’s not what it means. It means that animals, including us, are not agents of their own destiny, but automatically throw off random genetic recombinations and occasional mutations from which the environment selects. That’s natural selection. Nothing in the animals’ actual behavior has any effects or any significant consequences. This is the view of evolution taken to its logical extreme by Richard Dawkins’s “selfish gene, genes are everything” approach.

  Now if what I just described was the whole evolutionary story, there’d be no point in searching through the course of human evolution for some special, unique behavior that could have triggered language. There couldn’t be one. Our ancestors must simply have gone on having sex with one another and recombining their genes and tossing out the odd mutation until one fine day they hit the jackpot with some combination that made language, at least in a very simple form, possible. And then once they were capable of language, it was what the French call an embarras du choix; there were just too many things for which language would obviously be useful. Hunting, toolmaking, social relations, rituals, gossip, scheming for power, attracting mates, controlling children . . . All these and more have been proposed as the original function of language. After all, these activities were all carried on by other primates. And since we were primates with primate genes, and since genes were what determined behavior, there was no point in looking anywhere but at our closest relatives, the great apes (who unlike our immediate ancestors had the advantage of being alive and well and available for study), if we wanted to know how language began.

  Irene Pepperberg, who has shown that at least one species of parrot has as much language potential as apes, called this the “primate-centric” approach to language evolution.

  Let’s look a little more closely at Williams’s dictum. “Organisms adapt to their environment.” Not the environment, note—their environment. The environment as a whole doesn’t select for anything. (The weather in Alaska doesn’t bother Hawaiian finches.) A species is only affected by the environment that immediately surrounds it. But that environment is in its turn changed, sometimes drastically, by the species that inhabit it. Goats cause deforestation. Worms enrich the soil. Beavers flood valleys. Seabirds dumped so much guano on the island of Nauru that, now that the Nauruans have sold it all, there’s hardly any island left. So the selector in natural selection isn’t some generalized, abstract “environment”; it’s a part of the environment that has already been worked over by its inhabitants. What living organisms did to that environment will then select for new traits in those organisms that will enable them to modify their environment still further, which in turn . . .

  Get the idea? It sets up a constant feedback process.

  So evolution is no longer just selfish genes mindlessly replicating themselves. It’s a process in which the things animals do guide their own evolution. This happens to be a much more user-friendly view of evolution, but that’s not why you should accept it. You should accept it because it’s closer to the truth.

  It’s only in the last few years that this view, one known to biologists as niche construction theory, has developed; it’s still hardly known to outsiders. Nobody has yet used it to look at language evolution. I’ll explain what niche construction theory says in chapter 5. All we need here is the radically changed picture of human evolution that it gives us. No longer is human evolution, and the complex culture that human evolution produced, a one-of-a-kind anomaly. What drives it can now be seen as a process operating in many other species—possibly most species.

  Human culture is just the human niche.

  It’s the way we adapt our environment to suit ourselves, in the same way that the complex worlds of ant nests or termite mounds are the way ants and termites adapt the environment to suit them. We do it by learning, they do it by instinct; big deal. We can do it by learning only because we have language, which is by now the fruit of instinct just as much as a termite mound is. And language itself is a prize example of niche construction.

  What this new theory suggests is that people have been seeking the origin of language in all the wrong places. Previous treatments fall into one of two categories. Either language was some exotic gift that fell from on high for no very clear reason, or it was such a simple and obviously useful thing that any of a dozen factors might equally well have selected for it. We’ll meet both kinds of explanation in the pages that follow, and see what’s wrong with each of them.

  From the perspective of niche construction theory, language could only be the logical result—maybe even the inevitable result—of some very specific choices our ancestors made and some very particular actions they performed. To be more precise, they must have started to do something that no species of even remotely similar brain power had attempted, something that could not be accomplished unless they somehow broke through the limitations that restrict almost all other animal communication systems. And of course, once they broke through, once they established a new kind of system, they would have moved into a new niche—the language niche. No matter how crude or how primitive that first system was, it would be subject to the same feedback loop—behavior to genes, genes to behavior, behavior back to genes again—that all forms of niche construction create. Language would change, grow, and develop until it became the infinitely complex, infinitely subtle medium that we all know and use (and take totally for granted!) today and every day of our lives.

  I have two goals in writing this book.

  First, I have a burning desire to convince you that language is the key to what it means to be human, and that without understanding how language evolved, we can never hope to explain or understand ourselves. I’m not saying this because the evolution of language happens to be what I’ve been thinking about for the last couple of decades. It’s the other way around. I’ve been thinking about the evolution of language for the last couple of decades precisely because I’m convinced it’s the key to understanding humanity, and for no other reason. I didn’t have to do it. I don’t need the money, not that there’s much money in it. I could have stretched out on a chaise longue by the pool with a pitcher of margaritas and blown the days away. But the desire I have to convince you merely reflects my own passionate need to know, to understand, what humans really are—a need I’ve had all my life.

  Second, I want to dispose of some of the many confounding factors that have bedeviled the study of language evolution, that have made it a chaos of conflicting theories, extravagant claims, and irreconcilable positions. One of these factors I’ve already noted: the “primate-centric bias” that affects so many in this field, focusing exclusively on our genetic continuity with the great apes and ignoring all the environmental and ecological differences between our ancestors and theirs.

  Another factor, closely allied with this one, is the belief that the communication systems of other species make up some kind of hierarchy, like a ladder or a pyramid with language seated firmly on top. It’s as if the communication systems of other species were no more than a series of botched attempts at language: they did their best, but weren’t quite up to it; only we were smart enough to scale the pinnacle. This is what you might call the “homocentric bias”—folk seldom admit to it, but it’s colored all too many theories. Watch out for people who talk about “precursors” of this or that aspect of language, or who seek “stepping-stones to language” in the communication of other species: these are some of the signs of homocentric bias.

  In reality, the communication system of any species is designed simply
and solely to take care of that species’ evolutionary needs. There’s no evidence anywhere for a cumulative or “progressive” tendency operating across communication as a whole.

  A third factor is assuming that language was originally a target for natural selection. This looks like a no-brainer. Language was what evolved, and evolution proceeds through natural selection, so language had to be selected for, didn’t it? The question then becomes simply, what selected for it? Was it hunting, toolmaking, child care, social competitiveness, sexual display? All these and more have been picked by some experts as the pressure. Not surprisingly, there’s no good reason to choose any one of these over the others; indeed all of them are seriously flawed in one way or another.

  The error here, made even by those who think the earliest language was far simpler than the languages of today, is thinking that language could have been a target at all. How could it be a target, even in its simplest, most basic form, when it couldn’t even exist until some of the necessary bits and pieces had been assembled?

  Instead of asking how language evolved, we should be asking what caused our ancestors to take the first halting steps away from the kind of communication system all other animals had and have. We should be looking at those ancestors’ way of life, what they were trying to do and how they did it, and then asking which of the constraints on animal communication those activities would have forced them to break.

  If we can avoid all of these confounding factors, we may be able to get past the two head-butting alternatives in which the language evolution debate is all too often framed:

  • “All communication systems are on a continuum.”

  • “Language is a totally different kind of communication system.”

  Too often, these contradictory positions are argued on ideological rather than scientific grounds: those who want humans to be just another species take the first position, and those who think humans are something very special take the second. We have to realize that the dichotomy is a false one; the second may be true now, but it certainly wasn’t then, whenever “then” was. We have to look, more closely than anyone has yet done, at how our ancestors could have first cracked the mold of animal communication, and how that first breakthrough, in a species not so distant from our own, could have unleashed a cascade of change that would radically alter not just communication, but the very minds that communicated.

  It’s a long story, a complex story.

  But is it the true, the only real story?

  I can’t guarantee that. Science isn’t faith. What seemed certain yesterday can look like nonsense tomorrow, yet become possible again the day after. Not because scientists can’t make up their minds, but because new knowledge is constantly coming in, because that knowledge inevitably changes (hopefully improves) our picture of reality, and because, not being faith-based observers, we have to ensure that our theories fit that picture.

  What I can guarantee is that, on the basis of what we presently know about humans, evolution, human evolution, biology, and language, what you will read in the following chapters represents the best and best-supported account it’s possible to get today. What we know may change, and it may no longer be the best, but our knowledge would have to change a lot before that happened. For what I think will remain true, regardless of new discoveries, is the idea that we must look for the source of language not in the things apes do today, but things our ancestors did that apes didn’t do.

  But that, as they say, is an empirical question.

  You be the judge. If you enjoy this book half as much as I’ve enjoyed writing it, it will have been more than worthwhile.

  1

  THE SIZE OF THE PROBLEM

  WHY THERE’S NO DR. DOLITTLE

  Almost all animate organisms communicate with one another . . . somehow.

  Fireflies flash. Frogs croak. Crickets, grasshoppers, and the like rub their legs together, or against their wing cases, producing the kinds of sound known as stridulations. Birds perform songs of varying degrees of complexity. Wolves howl. Dolphins emit sonar signals; they whistle too. Some lizards inflate pouches in their necks, or change color. Gibbons engage in bizarre duets that can last for an hour or more. Apes and monkeys have a range of strategies: hoots, barks, gestures, facial expressions. Bees dance. Ants do it with chemistry. The means different species use for communication are so bewilderingly diverse and so different from one another that you might well think something pretty complex is going on.

  It isn’t.

  A decade ago, Marc Hauser published what is still the most thorough and complete study of animal communication systems (ACSs for short; sorry about that, I loathe acronyms with a passion, but if you had to read “animal communication system” as often as I’ll have to write it in the next few chapters, you’d understand—even forgive). He found that all the information conveyed by ACSs falls into three broad categories. There are signals that relate to individual survival, signals that relate to mating and reproduction, and signals that relate to other kinds of interactions among members of the same species—call them social signals. Some signals are hard to fit into a single group. For instance, a signal of appeasement, used in confrontations when your enemy looks to be winning, is on the face of things a social signal, but it could also fall under “survival”—if you don’t make it, you could get killed. But no signal falls outside of those three areas. No ACS can be used to talk about the weather, or the scenery, or your neighbor’s latest doings, let alone to plan for the future or recall the past.

  Of course it would be an enormous benefit for any animal if it could recall the past, noting all the mistakes it had made, and plan for the future, eliminating those mistakes. Such an achievement would maximize an animal’s fitness, which is biology-speak for saying the animal would live longer, have more offspring, and spread its genes more widely. And that’s what evolution’s about: who dies with the most kids wins. So you may well wonder why we alone have language—why we don’t inhabit a Dr. Dolittle world, where we could chat with chimps, converse with cats, debate with dogs, rap with rabbits, and yak with yaks, while all these creatures did the same things with one another.

  The answer is, evolution doesn’t develop things just because they’d be useful for a species to have. Evolution is a minimalist. It doesn’t do a lick more than it has to. And it’s also limited by what it has to work with. What it has to work with are the bodily shapes and mental abilities that exist in any species at any given moment, and the behaviors that those shapes and abilities make possible. Since within any one species those shapes, abilities, and behaviors can’t vary all that much, almost all evolutionary changes are gradual and small. Very occasionally there may be a tipping point, but for the most part, nature doesn’t even look like it makes leaps.

  So the means by which animals communicate—all the flashes, calls, gestures, and so forth I mentioned at the beginning of this chapter—are seldom if ever things that were designed from the beginning to communicate with. Rather they are modifications or stylizations or amplifications of things animals would do anyway, things that when they started out may have had little or nothing to do with communication. This was the conclusion of the earliest ethologists, scholars like Nikolaas Tinbergen and Konrad Lorenz, and though interpretations of the function and significance of ACS units have changed radically since the 1950s, our understanding of where they came from hasn’t changed.

  Over time, through frequent co-occurrence, these original behaviors became associated with certain situations, and hence with the kinds of messages appropriate to those situations. Their users didn’t consciously mean to communicate in the way we do when we want a window closed and say, “Please close the window”: ACSs aren’t just a cheap substitute for language, but something entirely different. Their users, in the process of reacting to situations, provided clues as to how other animals should react in those situations; interpreting such clues correctly improved those animals’ chances of survival. Thus in a confrontational situa
tion between mammals, shrinking postures and high-pitched sounds indicate an intent to appease the aggressor. Among songbirds belonging to species that defend territory, songs of a certain type and intensity suggest willingness to combat an intruder. And so on.

  This already makes language sound unlikely. The simplest and most straightforward source for it would be sounds drawn from the ACS of the last common ancestor of chimpanzees and humans. But if those sounds were anything like the sounds chimps make, the chances of modifying or ritualizing them into words, not to mention sentences, look small. And that’s even before you consider problems of meaning.

  To lift yourself by your bootstraps, you first have to have bootstraps.

  WHY OTHER ANIMALS HAVE SO LITTLE TO TALK ABOUT

  It gets worse. Why do ACS units contain only survival, mating, and social signals? It’s because those areas and only those areas are ones where signals can significantly increase an animal’s fitness.

  Look at survival calls. These include predator warning calls and food calls. A predator warning call doesn’t improve the survival chances of the animal that makes the call. In fact it reduces them; it calls attention to that animal, makes it stand out as a target. But it does improve the survival chances of that animal’s close relatives, who carry many of the same genes. This is what biologists mean when they talk about “inclusive fitness.” You don’t do stuff just to increase your own chance of sending more offspring into the world—the same purpose is served if you increase the chances of your siblings or other close kin doing so.