Kanzi: The Ape at the Brink of the Human Mind

by Sue Savage-Rumbaugh

  Language arose in a particular manner in human prehistory, and this process must have impressed itself in a consistent way in the evidence available today. Read correctly, these lines of evidence, together with those from modern neurology and psychology, should therefore tell the same evolutionary story. Yet two extreme views of language origins have emerged from this data. The differences that Iain and I shared were symptomatic of the disparate opinions regarding the role of language in human evolution.

  The existence of such divergent views on the origin of language suggests that, in some way, the archeological and fossil evidence is being misread. Although my initial motivation for working with apes was to help in the search for a greater understanding of human behavior, I have been unquestionably affected by the data unfolding over the past decades. These recent studies have become increasingly relevant to the unresolved question of the evolution of human language. The data provided by the abilities of Sherman and Austin, coupled with the data of Kanzi and Panbanisha, provide an independent means to evaluate the two models of language origins.

  While what we have learned about the linguistic and tool-constructing capacities of apes does not tell us how hominids became human, it does tell us a great deal about the common substrate of mind shared by apes and our hominid ancestors. Our modern minds evolved from these ancestral minds, which must have shared many characteristics with apes not too different from those alive today. Thus Kanzi and other apes offer a glimpse of the starting point in man’s evolutionary transformation from a state of nature to the modern human condition.

  In his second most famous book, The Descent of Man, Charles Darwin noted the great paradox of the gap between human achievement and that of the rest of the animal world. “[T]here is no fundamental difference between man and the higher mammals in their mental faculties,” he observed. And yet, he acknowledged, “Of the high importance of the intellectual faculties there can be no doubt, for man mainly owes to them his predominant position in the world.”18

  What precisely are these higher “mental faculties”? Every standard text on human evolution cites culture and language. Culture is defined as the process of passing knowledge from one generation to the next, and language is defined as the vehicle by which that knowledge is passed. But what of that period between one hundred thousand years ago, when Homo sapiens appeared with his large brain and his fully upright posture, and forty thousand years ago, when the “cultural explosion” took place?

  Were we passing knowledge down from generation to generation during this period? If so, regardless of whether or not we had invented language, the knowledge we were transmitting must have been very different from the kind of political, ritual, and technical knowledge that we pass along to new generations today. Certainly the tool manufacturing techniques were simple. Language was neither necessary nor probably helpful in passing along the methods of tool construction. Modern tool-makers tend to pass their knowledge along by example, a point made convincingly by both Patricia Greenfield and Tom Wynn at the Wenner-Gren conference. Patricia studied the cultural transmission of weaving in a South American culture and found very little use of language in this activity. Tom, reviewing the literature on the transmission of tool construction and use, concluded that “tool behavior is learned largely through apprenticeship—each actor constructs his own constellations. They are not shared.”

  Indeed, the same can be said of Kanzi. Nick did not utilize language in showing Kanzi how to flake stone, but rather treated him as an apprentice. Kanzi observed, but constructed his own constellation of techniques as well. Thus it would not seem to be the case that we need language to pass along major dimensions of culture. In fact, a moment’s reflection regarding the kinds of behavior that travel easily from culture to culture suggests that a great deal of information passes from culture to culture without language. In Japan, baseball is extremely popular, jeans sell for outrageous prices, and music has taken on a distinctly Western flavor. All of the changes permitted by modern technology travel rapidly around the world, yet without a common language. Try to watch a Japanese baseball game and figure out what the letters on the screen mean or what the announcer is saying. Languages are frozen bits of culture compared to most other sorts of behavior. Languages are clung to, much like other behaviors that fall under the label of “ritual.” When people move from one culture to another they bring with them their language, their religion, their marriage ceremonies, their rites of passage, their child-rearing patterns and their kinship patterns. They acquire and readily adapt to the new forms of shelter, the new foods, the new sports, and the new tools. Thus we see that one of the main uses of language is to transmit the practice of language itself.

  Surely archaic Homo sapiens of two hundred and fifty thousand years ago were capable of passing along techniques of tool use and construction, artistic rendering, shelter construction, and so on, if they knew how to do such things. The fact that they left no mark of such knowledge in the archeological record suggests not they were incapable of passing along these capacities to the next generation, with or without language, but that they had not yet developed these skills. Clearly, early Homo sapiens led an existence so different from that of even the most primitive hunting and gathering societies today, that it is not easy to contemplate what that existence was like.

  I therefore agree with Iain, and others, who suggest that language might have been a cultural invention, as agriculture was. The organizational and technological activities associated with food production and, later, the building of city states, far outdistance those required for a hunting and gathering society. And yet it was the brain of a hunting and gathering species—Homo sapiens—that mastered these new skills. No genetic mutation occurred that allowed people, ten thousand years ago, to do what their forebears did not. Similarly with language, it is quite conceivable that the cognitive apparatus that underlies language might have evolved for other, related purposes, such as planning future actions, tool construction, or social negotiation. In any case, the fact that language-mediating centers are scattered throughout much of the prefrontal region of the brain, and that their location may differ from individual to individual, indicates that a large learning element is present in language acquisition, perhaps exclusively so.

  I do not agree with Iain in his contention that a capacity for symbolic communication must necessarily postdate image making. The central message of the ape-language work is clear. Not only can apes acquire a capacity for symbolic communication through the structured experience we developed with Sherman and Austin, but also they can acquire it spontaneously through casual exposure to language, just as human infants do. It has been easy for Kanzi to comprehend language; the key was in his early exposure to speech and the pairing of the speech with symbols. His comprehension skills are so extensive as to make it implausible to offer a conditioning account. His younger sisters, Mulika and Panbanisha, have followed in his footsteps, revealing that it is the rearing experiences Kanzi received, not Kanzi himself, that permitted him to understand simple language. If Kanzi can acquire language so readily, we must conclude that the ape brain is capable of a primitive language. Either bonobos are utilizing this capability in the wild in ways we have not yet grasped, or exposure to the invention of language is all that Kanzi and his family needed to bootstrap their way into language.

  As we saw earlier, spoken language owes its unprecedented information-carrying capacity not just to the large range of sounds produced by the unique human vocal apparatus, but also to the nature of some of those sounds. The ability to produce consonants in association with vowels, therefore, led the way to the production of an extremely large number of discernible sounds. Kanzi’s comprehension of human speech suggests that he is able to decode consonants. If he could produce consonants as well as vowels, and if he had the requisite degree of neurological control, I have little doubt that Kanzi would be able to speak.

  The ability to produce spoken, symbolic language depended, therefore, on t
he appropriate development of the vocal tract in early human ancestors, not on the evolution of the requisite cognitive capacity. Even in primitive form, such a system of communication would have had considerable survival advantages; and the sophistication of the system would have increased through time as natural selection honed those advantages. Why this occurred in the Homo lineage and not in the australopithecines—as indicated by brain size and organization, and vocal tract anatomy—is an important question. Almost certainly this capacity for spoken language was associated with a more intense social nexus, connected with a more complex subsistence strategy—the beginnings of what eventually blossomed into a hunting and gathering way of life.

  Our work with Sherman and Austin, and particularly with Kanzi, has offered an independent way of judging the interpretation of language origins based on more traditional lines of evidence. The unexpected cognitive underpinnings of symbolic language possessed by modern apes suggests that the early members of the human family were equipped with a greater intellectual capacity for language than is usually assumed.

  If early Homo sapiens were indeed without language, each of us must feel compelled to ask: What were we like, we large-brained, sensitive, intelligent creatures producing limited technology and no art? We couldn’t have been even remotely similar to any living people extant today. For example, even a cursory look at the Sambia stone-age people of New Guinea reveals that language is critical to the entire structure of their culture, their initiation rites, their sexual behavior, their diet, everything they do. Not that they use language to teach these activities; rather, they employ language to proscribe them, to set the rules for when they occur, how they occur, why they occur, and what will happen if they do not occur in the proper way at the proper time.

  Young males, for example, enter into exclusive homosexuality with older males prior to the onset of adolescence. They maintain these relationships, without the knowledge of the female members of the group, until well into their twenties, when they take a wife and become exclusively heterosexual. It is language that imposes, guides, and directs this channeling of sexual orientation. It is also language that defines, guides, and directs the development of fighting skills. That is, language is not used to teach a boy how to fight, but it does define with whom he should fight and why, as well as the level of prowess to be obtained in interaction with enemies. Indeed, it is language that defines the identity of the enemy and that carries grudges against the enemy across one generation to the next, even though younger members of the group may never have experienced in person a reason to define a particular group as friend or foe.

  Without language would we carry such vendettas across generations? Without language would we feel a need to define possessions as belonging to one person or another? Without language, how far ahead would we understand cause-effect relationships? Would we understand the relationship between sexuality and childbirth? Without language would we construct elaborate kinship structures and the obligations that accompany them? Without language would we have recognized ourselves as independent beings, responsible for our own actions? Would we have had any need of moral codes of right and wrong? Would we have had need of clothing for the sake of privacy? Would we have had any understanding of death?

  10

  At the Brink of the Human Mind

  In an essay he wrote almost two decades ago, the Rockefeller University philosopher Thomas Nagel posed the question, “What is it like to be a bat?” The point of the essay was to explore the problem of gaining insight into another individual’s mind, and in particular the mind of another species. Nagel was prompted into the exercise by the work of fellow Rockefeller University scholar Donald Griffin, who, against the accepted scientific wisdom of our time, argued that animals other than humans engaged in reasoning about their world; in other words, said Griffin, humans are not the only species to experience the phenomenon of mind.

  Nagel’s answer to this question was, “We can never know.” He maintained that the bat’s perceptual world is so different from ours, based as it is on echolocation of high-frequency sounds beyond our experience, that we can never comprehend the bat’s mental world. The Austrian philosopher Ludwig Wittgenstein framed the conundrum in a different way, but with the same conclusion. “If a lion could talk,” he said, “we would not understand him.” Customs and cultural conventions create barriers of understanding between different human societies, observed Wittgenstein, so imagine how much more difficult it must be to penetrate the mental life of another species.

  I suspect that the problem has been exaggerated, for several reasons, not the least of which is the uncertainty over the phenomenon of mind itself. We saw in the first chapter of this book that Western philosophy and science have grappled with the issue, from Aristotle, who viewed humans as uniquely endowed with a rational soul, through Descartes, to whom only humans possessed a rational mind, to the modern behaviorists, who, like Descartes, restrict mind solely to Homo sapiens. Despite Griffin’s persistence in encouraging biologists and psychologists to buck the behaviorists’ hegemony and consider the existence of cognitive processes in the nonhuman domain, there remains a hesitancy to accept the idea of minds in species other than our own.

  As noted in Chapter 1, the concept of mind as we human beings experience it—that is, mediated by symbolic language and carrying a cogent essence of self-awareness—has for many people come to represent an unbreachable boundary between humans and nonhumans. Man stands safely on his side of the wall, declaring his separateness from the rest of nature. In part he has done so because the societies he has constructed have isolated animals from his daily existence. Animals no longer share the forest or the plains with us and only through the medium of television can we be impressed by the ways in which they cope with life. In most zoos and laboratories, it is difficult for animals to engage in behavior that displays intelligence, as their food is provided for them and their social groups are temporary, making it difficult, if not impossible, for them to establish traditions.

  When we take animals such as dogs into our homes, they often seem to do things that appear intelligent, but scientists are taught, early in their career, to beware of any interpretations that smack of anthropomorphism—the fear is that the intelligence may be in the eyes of the observer rather than in the mind of the animal. By labeling most interesting descriptions of animal behavior as anthropomorphic anecdotes, we unwittingly eliminate the need for serious scientific attempts to understand such behavior. That is, by assuming that the complex and interesting behavior is in the mind of the observer rather than the animal, it becomes the human observer rather than the animal that we seek to understand.

  But perhaps the deeper reason we so readily declare our uniqueness from animals is to assure ourselves that we are indeed reasoning creatures with a culture created by our own hand and mind. By setting ourselves apart from animals, we experience some small measure of safety. If we Homo sapiens are truly different by virtue of reason, we can look to reason to protect us from falling into the trap of reacting instinctively and losing the evolutionary game. When we look at animals we do not see cities or villages or agriculture or possessions—and their way of life does not look like that to which we would aspire, no matter how much we enjoy watching movies of wildlife. Therefore, it is comforting to assume that we have reason and culture by nature and that these abilities will always keep us from returning to any sort of animal state.

  But at the expense of gaining some comfort in our ability to plan our future, we risk alienating ourselves psychologically from all of the other creatures on this planet. After all, we have all evolved together. As a species, we are just beginning to understand that it is our view of separateness that has led us blindly to exploit the world of nature, be it by destroying natural habitats or performing experiments on animals with insufficient concern for the effect of these actions either upon the animals or upon ourselves as moral beings.

  It is becoming increasingly evident, not only from w
ork with apes, but from studies of species as varied as dolphins, parrots, sea lions, elephants, and wolves, that man has deluded himself by focusing on this separateness. As we come to understand other animals better, our current notion of human uniqueness will likely change and we will realize that future generations may view us as having looked at animals through a distorted lens, much as we now look back at the early explorers who thought that different races of man reflected different levels of evolution.

  To recognize the connection between our intellect and mind and those of other creatures on our common planet, we must permit ourselves to ask questions in a careful and serious way about behaviors that reflect animal intelligence in complex situations. As we begin to do so, we will have to alter our focus from the test paradigms that closet our thoughts. These closet paradigms are experimentally correct, but their very structure limits the behavioral options animals may exhibit to an array of simple responses. Such limitations, carried out for the purpose of upholding the scientific method and preventing self-deception, have unwittingly forced the egregiously erroneous conclusion that animals show little capacity for what we call inference, insight, reasoning, or thought.

  These closet paradigms came about as post-Darwinian inquiries into behavior sought to follow the models set by physics and chemistry. But behavior can never be successfully understood using these models. Living organisms, as subjects of treatments, are not the same from one test to another. Each action upon a living organism alters it in some manner, making it impossible to recreate the former conditions. Researchers have tried to get around this fact by repeating the stimulus presentation until the behavior is stable. Unfortunately, this technique serves only to create the illusion of control over behavior. In fact, it manufactures an artificial situation that limits the options available to the animal and thereby causes it to appear to be under the control of the stimulus. A more fundamental problem arises when researchers attempt to follow the models of physics and chemistry by concentrating on the antecedents of one or two behaviors or even a set of replicable behaviors. Such attempts assume that behaviors can be mixed and matched something like a set of chemical elements, and that an orderly and predictable reaction will follow. Unlike chemicals, behaviors cannot be reasonably separated from the entire context in which they occur. That context encompasses both the actions of the animal across time and the events within the environment across time. Indeed, behavior is fundamentally a time-based phenomenon.