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

by Bickerton, Derek

  Alex actually talked (I use the past tense because, unfortunately, he died in September 2007). Sure, you say, so what? Lots of parrots talk. But this one talked sense. If you asked him what he wanted, he’d tell you, “I want nut!” Give him a grape instead and he’d say indignantly, “No! I want nut!” He knew and could appropriately use some fifty words, counted up to six, identified seven colors, could do match-to-sample tests (he knew what “same” and “different” meant), and could pick, from an array of objects, the one that is green and three-cornered. Griffin, a younger parrot, has a productive capacity perhaps even equivalent to Kanzi’s: “want grape,” “go chair,” “green birdie,” “go back chair” (as with the apes, his combinations were spontaneous and untrained, but like theirs, they were limited to two or three units).

  Are these just tricks, artifacts of clever animal training? Skeptics who insist that they are totally miss the point. It wouldn’t matter even if they all were. You can’t train animals to do things that their neural infrastructure won’t allow them to do. If their neural infrastructure allows them to do those things, it can only be because that’s the way the animal’s genes built its neural infrastructure. In other words, if the things an animal can be trained to do amount to a form of protolanguage, then that animal is only selection-limited—it already has the range of genetic variation on which a strong-enough pressure could work to produce protolanguage.

  So Irene has posed the jaw-dropping question: Must human language seek its precursors exclusively in the behavior of apes?

  Most people have simply assumed that it must—that language must be a homologous rather than a merely analogous trait.

  Whenever a biologist finds a trait that’s shared by two or more species, his first thought is likely to be, is this a homology or an analogy? A homology is some trait or feature that the species share as a result of their common ancestry. Sometimes an ancestor with a similar feature is known to exist, but even if not, it’s a fair assumption that anything shared by closely related species comes directly from their genetic inheritance. However, similar features are sometimes found in animals with only very remote genetic connections—animals that have many closer relatives lacking the feature in question. Here, the explanation is likelier to be an analogy—the shared feature represents a similar response to a particular environment. The classic example is that of dolphins, sharks, and ichthyosaurs, far from one another in the great bush of life, but each growing a similar dorsal fin in response to the need for rapid maneuvering in deep water.

  Homology is commoner than analogy. Evolution seldom throws stuff away. It works, in Darwin’s phrase, through “descent by modification,” so any feature of a common ancestor is likely (not certain, by any means!) to show up in some form or other in species that descend from that ancestor. But analogy can never be ruled out, least of all where the same capacity shows up in species that are only very distantly related. Like the capacity for learning a protolanguage, shared not just by the more advanced primates but by dolphins, sea lions, and parrots.

  Analogies represent similar solutions to similar problems. Bear this in mind; it will become very important later on.

  If readiness for protolanguage came from a homology, and if that readiness had a range that included both apes and parrots, we’d have to look for a common ancestor around 300 million years ago. And we would then have to explain why so few of the countless millions of descendants of that ancestor have shown any proto-protolanguage capacities.

  If that’s indeed a fact, not just an artifact.

  Because, to the best of my knowledge, nobody has seriously attempted to “teach language” to more than seven species—the four great apes, Californian sea lions, Atlantic bottlenose dolphins, and African gray parrots. I did once, years ago, read a news item about an Italian countess who taught her dog to type; one day, after the mutt crapped on her carpet, he went to his typewriter and tapped out “Bad dog!” (or more likely “Cane cattivo”). But I suspect that story was the brainchild of some cheeky cub reporter on a slow news day (if, however, it’s true, please contact me immediately!).

  Seriously, though, nobody knows how far down the phyla one could go and still get the kind of results obtained from our seven species. But whatever the answer, the odds are that we have an analogy rather than a homology here, and the trigger this time is not a particular type of physical environment—it’s the attainment of a certain level of cognitive capacity.

  It’s tempting at this point to talk about “higher” animals, but I’ll resist the temptation. In his diary Darwin wrote to himself the admonition: never say “higher” or “lower”; what that usually turns out to mean is more, or less, like humans. However, one can hardly refrain from talking about “more” or “less” complex: here we have objective measures, numbers of genes, numbers of cells in the brain. But what exactly do we mean, in this case, by “cognitive capacity”?

  I don’t actually see the level of cognition required for protolanguage being particularly high, at least not in the sense of some elaborate and convoluted mental structure. I think it has more to do with having the aural and/or visual capacity to sense the world as divisible into a large number of separate and distinct categories, and the brain space to sort and systematically store all the features that distinguish the various categories. (Note how careful I am to say that they have categories, not concepts—the distinction between these, one a lot of people don’t even bother to make, will turn out to be crucial in chapter 10.) Once such categories are in place, linking them to signals is always a possibility. As I said at the beginning of this chapter—but it’s high time for a reminder—such signals are not true words, though they share referential properties with words. It’s far from clear that these signals, for the animals concerned, need to be truly symbolic. Such signals are likeliest to be used, like ACS calls, when their referents are physically present. I don’t think that matters, though. I think we’ll find, when we come to human ancestors, that words got going in a very similar way, and indeed had to get going in a very similar way, since you can’t leap from icons and indices to symbols in a single bound.

  It’s certainly a fact that vision is well developed in apes and parrots, and hearing in dolphins (I’m not so sure about sea lions). Whether that’s a necessary or a sufficient precondition for protolanguage or just an accident we won’t know until more people take more species and subject them to the same training that the Magnificent Seven experienced. And I think that it should be done. Go down to gibbons, even macaques. Try prairie dogs to check if brain size really has anything to do with it. Or crows, to see if the parrot’s the only bird that can speak. And we should have some way of brain-scanning all of these animals, including the original seven, while they perform, to see if what happens bears any relation to what happens in the brain when humans use either a natural language or a pidgin. (That will be tricky when we get to birds—their brains are configured quite differently from those of mammals.)

  Once we stop asking the stupid question “Can animals acquire language?” (the short but unhelpful answer is no!) and start asking the sensible question, “What kind(s) of neural substrate are both necessary and sufficient for protolinguistic behavior?,” we can open up whole new fields of research and start to get somewhere.

  It’s already pretty clear, though, that the prerequisites for something approaching language are by no means limited to our immediate relatives. And if, under instruction, a whole range of species can learn some kind of protolanguage, this suggests that, in any species within that range, protolanguage is selection-limited, not variation-limited. In other words, no special changes, magic mutations, “language organs,” or dedicated circuits were needed for language to start.

  Just a large enough brain, a wide enough range of categories, and, most important of all: the right selective pressure.

  PRO BONOBO PUBLICO

  Since we now know we don’t need to be descended from apes to have gotten language, and since there’s
no reason to think our ape heritage, other than giving us protolanguage-readiness and a highly social brain, necessarily contributed much to our getting it, we can bid farewell to our cousins and go looking for language in what, at first sight, might seem all the wrong places.

  But before we do that, there’s one personal experience I’d like to share with you, one that for me shed considerable light on why language didn’t evolve in any species other than ours.

  Not so long ago, Sue Savage-Rumbaugh kindly invited me to visit her spanking new, ten-million-dollar ape research facility, located a few miles south of Des Moines. I met with Kanzi, face to face. Right away he struck me as a personality to be reckoned with. He exuded the kind of serene confidence in his own entitlement that you seldom find outside pop stars, politicians, and the very rich. Under that was a sharp intelligence, both wise and crafty. If you met him at a scientific conference, you’d watch your arguments (or in an inner city, your wallet).

  Kanzi ruled like a pasha over a bunch of bonobos; what he wanted, he got, whether it was the tastiest tidbit, the most desirable female, or the attention of his keepers. The strength of the bond between him and Sue was unmistakable. But what did it spring from? Genuine affection, the Stockholm syndrome, or some mixture of these?

  For after all, Kanzi and his cohorts were prisoners. Granted, their jail was the opposite extreme to the pharmaceutical Abu Ghraibs in which some less fortunate animals languish; their keepers loved them, and all sorts of diversions were laid on for them. But they were still prisoners, not free to go when and where they chose, under the absolute control of another species. (Imagine how you’d feel if aliens from space housed you in one of their “facilities” and “studied” you, however amiably they did it.)

  Under those circumstances the bonobos did just what you’d expect them to—what slaves did back on the old plantation, what any of us would do under similar conditions. They shuffled and sang, shucked and jived—yes massa, sure massa, I’ll say anything you want me to say. For the rest of the time they just got on with their bonobo social lives.

  Recall the conventional wisdom of today: language arose through social intelligence, to deal with the ever-growing complexity and sophistication of primate lives. To the best of my knowledge, not one of the dozens or by now probably hundreds who have endorsed this notion has ever provided a concrete example of one specific problem in social life that you can’t solve without language but can solve with it. But absence of evidence seldom slows the spread of fashionable ideas.

  In fact, the bonobo—according to Frans de Waal, who has studied them as closely as anyone—already “shows an unparalleled social organization” and devotes at least as much time and energy to its social interactions as any other primate, employing a “mental capacity” that “has the power of revolutionizing social relationships.” I’m prepared to bet that the social life of bonobos is at least as rich and complex as the social life of the human ancestors who made the first breakthrough into language, and quite likely more so—we’ll see why in chapter 6. And here bonobos were, being given language for free, so to speak. If the language-from-social-intelligence theory was correct, you’d expect them to seize on it eagerly, exploit it in their daily exchanges. Indeed, you might be baffled to explain why, given so high a level of social intelligence, they hadn’t already discovered language for themselves.

  I watched them over a long weekend, and they showed no sign of any real interest in language. They’d cooperate when Sue asked them to. They’d press one or two of the three hundred or so lexigrams—a set of arbitrary, abstract symbols standing for human words—on a computer screen, and a voice synthesizer would sound out the equivalent English. Then, as soon as they decently could, they’d get back to their games.

  My most vivid memory of my visit concerns the plastic sheets. To encourage the bonobos to learn and use lexigrams, Sue had had them printed on sheets with thick, transparent plastic covers, about the size of a Rand McNally atlas or a fancy restaurant menu, and these sheets were scattered here and there in the apes’ enclosures so they would be available any time a bonobo felt like using them spontaneously. Which was hardly ever. Far from treating them as keys to a new, richer world, the bonobos totally ignored them, except when cajoled by their keepers. For the rest of the time, these plastic dictionaries, piss-stained, fouled with the dirt that accumulates, no matter how often and diligently you clean up, wherever animals (including us) are trapped in confined spaces, were trampled underfoot—kicked around, as the Irish say, like snuff at a wake. The bonobos didn’t want them, didn’t need them. Bonobos just want to have fun. All the lexigrams did was get in their way.

  You and I have a purely species-specific view of language. To us it’s the ultimate adaptation, the core of what we are. We can’t imagine a species that wouldn’t be delighted to have it—that once it got it, wouldn’t cling to it and exploit it as thoroughly as it could.

  But that’s just because we’re us, because we can’t imagine a life without language. For that matter, I even find it hard nowadays to imagine a life without computers, without e-mail, without word processing software. Yet I lived that life, pecked away at my Olivetti with my Wite-Out by my side, tore up and laboriously rewrote draft after draft, handwrote personal correspondence and schlepped it to the P.O., never for one moment imagining or even wishing that some whiz kid would sell me some electronic dingus that would put paid to all that labor-intensive stuff. It was just the way things were.

  Other species have accepted the way things were, have lived without language and prospered from time immemorial. Bonobos handle their complex social lives quite happily without it, so why wouldn’t our ancestors have done the same?

  Well, because they took the road less traveled by, and that made all the difference. They opened up an ecological niche that no animal of their size and complexity had ever entered before.

  5

  NICHES AREN’T EVERYTHING

  (THEY’RE THE ONLY THING)

  THE THEORY OF NICHE CONSTRUCTION

  I guess there are still some people who think our ancestors got cleverer and cleverer until one fine day they just up and invented language, right off the top of their clever little heads.

  But if one species and one only has a host of complex and highly developed languages and no other species, unaided, has anything you could call language at all, then language must somehow form part of the biology of our species, just as much as walking upright does. Exactly how it’s embedded in human biology . . . Well, that’s the question we’re all trying to answer. But no serious scholar nowadays doubts that language is, at bottom, biological rather than cultural, and therefore was not created, but somehow evolved.

  Theodosius Dobzhansky said it best: “Nothing in biology makes sense except in the light of evolution.” But exactly what kind of evolution?

  A decade ago almost everyone would have said, “What a stupid question! There’s only one kind.” That kind was the neo-Darwinian consensus, epitomized in the words of another icon, George Williams: “Adaptation is always asymmetrical; organisms adapt to their environment, never vice versa.” True, some biologists favored a more nuanced version, but for the majority, the organism was impotent, the environment all-powerful, and any interaction between the two ran along a strictly one-way street.

  All that is changing now under the impact of what is known as niche construction theory—a theory that gives animals themselves a vital role to play in their own evolution. Among its many virtues, this theory can explain both the rapid cascades of change that gave rise to Stephen Jay Gould’s theory of punctuated equilibrium, and the emergence, from time to time, of things that look at first like total novelties (language is only one of many examples).

  If we’re going to learn about niche construction theory, there’s no better place to start than with beavers.

  THE BEAVER’S TALE (TAIL?)

  Everyone knows about beavers.

  Beavers build their homes, known as “lodges,” i
n places where no sudden rise in water level will sweep them away, and where no sudden fall will expose them and their kits to predators. Their favorite environment is therefore a marsh or pond. If there isn’t one there already, they make one. They dam fast-running streams by chewing through stems of saplings and bushes, piling the resulting brushwood in the path of the current, and patching the gaps with mud. Sooner or later the dam holds, the water backs up, the land behind the dam is flooded.

  Beavers are what ecologists call a keystone species. They create wetlands (not, alas, as quickly as we are destroying them) that serve as homes and breeding grounds for an immense variety of species—fish, crustaceans, algae, waterfowl. Beavers are a keystone species because, if you took beavers away, many other species would collapse, just as an arch collapses if you knock the keystone out. In making an environment for themselves, they made one for others too. But that’s not all they made. They made themselves.

  One of the things that’s struck everyone interested in nature is the way in which species fit into their particular habitats and ways of life as precisely as a key fits a lock. (If you’ve ever had a new key cut and found that the slightest roughness made it unusable, you know what I mean.) Beavers fit their habitat in just this way. Their teeth are massive, chisel-like objects, just fine for ripping through the toughest bark. Their mouths are shaped so that, if the stems they are chewing through are underwater, the water doesn’t get into their throats and choke them. Glands under their skin pump out oils that effectively waterproof their thick fur. Their feet are webbed for stronger swimming, their lungs disproportionately large so they can work underwater for long periods, their eyelids transparent to protect their eyes while still allowing them to see clearly enough beneath the surface. And their tails are long and flat, driving and steering them in the water, radiating heat when they cross dry land on hot days, storing fat against lean seasons.