Kanzi: The Ape at the Brink of the Human Mind

by Sue Savage-Rumbaugh

  Just as Bev, Connie, and Ruth had become more sociable through their exposure to language training, so too did the children in the Clayton County study. Rose recalls that the mentally retarded children in the study at one school, although they were mainstreamed for lunch times, regularly used to segregate themselves as a group, eating together at one table. “Within a few months of the beginning of the study, our kids were everywhere,” recounts Rose. “They were talking to their regular education peers, ordering food in line like everybody else. What had happened? They now had a way of communication among themselves and with others.”18

  One day she observed Bob, who was part of the study, talking to his peer tutor, Ralph, using the SAL. A friend of Ralph’s came up and started talking to Ralph while ignoring Bob. Ralph said that if his friend wanted to join in the conversation, he would have to use the SAL. Hesitantly, the friend began to do so, and soon said to Bob, “Let’s go outside to the playground.” Bob said yes and the three boys went out as a normal social trio, to play. “These findings highlight the role the SAL may play in mediating or advancing interactive skills between peers,” comment Mary Ann and Rose. “Speech output communication devices may be one important means of enhancing social interactions with nondisabled as well as disabled peer communication partners.”19

  Stories such as those of Bob and his friend are common among the Clayton County children in the study, and give substance to the once unthinkable idea that such children can indeed function in society, and can even contribute in the workplace. To this end, Mary Ann and her colleagues have established Project FACTT (Facilitating Augmentative Communication Through Technology), a collaborative effort between Georgia State University and Clayton County Public Schools. Soon, children who not long ago would have been considered as beyond hope of help, will be doing useful jobs and communicating with fellow employees and their employer. That represents enormous progress from the idea that took Duane “just a few seconds of thought” to come up with in the winter of 1970. “Our understanding of language through the study of chimps has overshot our expectations by a thousand percent,” Duane now says. “I had no idea that chimps would one day be shown to acquire language spontaneously, as human children do, nor that the ape work would inform so powerfully what can be achieved with mentally retarded children.”20

  There is still a great deal to be learned about how language interacts with the cognitive development of mentally retarded children, but all the evidence points to the importance of engendering language skills in them. And if Mary Ann can achieve her goal of getting ever younger children—even toddlers—into SAL programs, further progress is surely inevitable. The mother of one of the Clayton County children once said to her: “You’ve done so much for my son, but, oh how I wish you’d had him when he was two. Who knows how much you could have done for him by now?”

  *The names of the individuals who took part in the research experiments discussed in this chapter have been changed to protect their identities.

  8

  Pan, the Tool-Maker

  At the Wenner-Gren conference we were finding that one by one the great “Definitions of Man” were falling. The belief that man alone can transmit cultural acts had vanished with the early observation of potato washing in Japanese macaques by Junichiro Itani; once this custom originated, mothers passed it on to their offspring. The belief that man alone can make tools had gone by the wayside in the first years of Jane Goodall’s observations in the field. The belief that man alone has language has taken a little longer to fall and is more controversial, but my fellow scientists at “the castle” were beginning to recognize that this, too, was to become a controversy of the past. Nick Toth, the resident expert on stone-tool construction, was wondering if another monolith might fall. Are humans really the only creatures that can use tools to make other tools, he wondered?

  When Nick initially posed this question to me, I was uncertain as to whether he really wanted to find out, or whether he was trying, like so many before him, to shore up yet another concise description of man that could neatly distinguish humans from other creatures in a completely definitive manner. It was the end of a long day at the conference, and we had all eaten dinner at a restaurant in the nearby town. A good measure of wine had been drunk, light-hearted banter had mixed with shop talk, and some members of the group had entertained us with a bout of spontaneous music making. As well as being a fine archeologist, Nick turned out to be a gifted and enthusiastic musician. On our return to the conference hotel, I was sitting near the rear of the bus and Nick was in the very back, legs stretched out, arms folded across his chest, eyes closed, apparently asleep. Suddenly, he opened an eye and beckoned me to join him. “I have something I want to ask you,” he said. “Do you think Kanzi could learn to make stone tools, the way early humans did?”

  I hadn’t talked much to Nick by that point in the conference, and I had formed the impression that he was an archeologist who didn’t like psychology. His question seemed to come out of the blue, and was something I had never thought of trying. From my long experience with chimpanzees, I had gained a great respect for their abilities. I knew that common chimps and bonobos were skilled manipulators of objects in experimental situations, and of course was aware of the many kinds of tool behavior that common chimps display in the wild. My already keen respect for their abilities was increased by the conference presentations of Bill McGrew and Christophe Boesch. Bill pointed out that not only do chimpanzees make tools, but they essentially have a tool kit, meaning that they use similar tools for different ends (for example, they use sticks for termite fishing and as weapons) and they accomplish similar ends with different tools (for example, one chimpanzee was observed to use four different tools in the process of extracting honey from a bees’ nest). Christophe described the complexity of tool-use for nut cracking in the Ivory Coast chimpanzees. These chimps use both a hammer and an anvil, often for as long as two hours per day. Hammers can be either suitable wooden clubs or rounded stones. The anvils are generally indentations on stones or tree limbs. Both hammers and nuts must be carried to these anvils, which themselves are part of a surface that cannot be transported. The tools and nuts are regularly carried over a hundred meters to anvils. The wooden hammers are constructed by the apes themselves and the stone hammers are carefully selected and in high demand. Christophe has even observed instances in which mothers appeared to be teaching their infants nut-cracking techniques.

  But making stone tools is quite another matter. It seemed light years beyond what apes were currently doing. No one had ever seen a chimp intentionally make a flake. Some of Christophers chimps had accidentally broken off stone fragments while attempting to crack nuts, but no chimps had ever been seen to use a serendipitously produced flake as a tool. Indeed, even I could not make a worthwhile stone tool, and I had had Nick there to teach me.

  “That sounds like a fascinating idea,” I replied to Nick. “But isn’t stone tool-making a bit advanced for apes?” I had never seen Kanzi, or even the more dexterous Sherman and Austin, attempt to make stone flakes. And, like most people, I assumed stone tool-making was a human activity requiring human skills. Nevertheless, I was intrigued. “What do you have in mind?” I asked. Nick sat up, and quickly explained.

  In 1949, the British anthropologist Kenneth P. Oakley published a classic book, Man the Tool-Maker. This short volume encapsulated what was widely held to set humans apart as unique: “Possession of a great capacity for conceptual thought, in contrast to the mainly perceptual thinking of apes and other primates, is now generally regarded by comparative psychologists as distinctive of man,” he wrote. “The systematic making of tools of varied types required not only for immediate use but for future use, implies a marked capacity for conceptual thought.”1 The notion of man the tool-maker struck a receptive chord in the minds of anthropologists, and in society in general: Alone among the world’s species, tool-making Homo sapiens fashions an elaborate culture and manufactures a powerful technology
, through which the world has been forever changed.

  When Jane Goodall in the early 1960s reported her observations of chimpanzees making simple tools for harvesting ants and termites, human uniqueness appeared to be threatened. In their tool-making, chimps broke twigs from trees, stripped off the leaves, and then inserted the “fishing stick” into termite mounts or onto ant trails. Tool-using of various kinds has been observed in many creatures, from sea otters, to sand digger wasps, to Galapagos finches. But Goodall’s chimps were doing more than using tools; they were making them. So perhaps Homo sapiens isn’t unique, after all? The anthropologists’ response was, Ah, but humans use tools to make tools, while chimps just use their hands and teeth. Human uniqueness was thereby preserved.

  The ability to make and use tools has long been considered part of the evolutionary package that transformed an apelike creature into the human species. Charles Darwin argued that stone tools were vital as weapons and provided a substitute for the teeth and claws possessed by other large predators. “The free use of the arms and hands, partly the cause and partly the result of man’s erect position, appears to have led in an indirect way to other modifications of structure,” he wrote in his 1871 book, The Descent of Man. “The early male progenitors of man were … probably furnished with great canine teeth; but as they gradually acquired the habit of using stones, clubs, or other weapons, for fighting with their enemies, they would have used their jaws and teeth less and less.”2

  A century later, when the human prehistoric record was much richer than in Darwin’s time, anthropologists continued to view tools as an integral part of the evolutionary wedge that separated humans from apes, specifically in arming humans as hunters. “In a very real sense our intellect, interests, emotions, and basic social life—all are evolutionary products of the success of the hunting adaptation,” said two prominent anthropologists at a landmark scientific meeting in 1966. “Human hunting is made possible by tools.”3

  The shift to becoming a tool-maker—and specifically, a maker of stone tools—has been seen as central to what differentiated humans from apes in an evolutionary sense. By definition, therefore, the very first members of the human family must have been tool-makers. This assumption has been challenged in the past several decades by the many discoveries of ancient human fossils and artifacts from East Africa, and by molecular biological information on modern humans and apes. The first members of the human family are now known to have evolved at least five million years ago, perhaps as many as eight million. And yet, the first recognizable stone artifacts date only to two and a half million years ago. The arrival of these stone tools in the record coincides with the first appearance of the genus Homo, which eventually gave rise to modern humans. The obvious assumption is that this new, large-brained member of the human family was the tool-maker; and that tool-making and increased brain size were linked in some way.

  Whatever the truth is of that assumption—and it is difficult to imagine how it might firmly be substantiated—it raises an important question for archeologists and psychologists about the earliest tool-makers: Were they doing something that was beyond the cognitive capability of apes? Or were they merely bipedal apes who were applying their apelike cognitive skills to non-apelike activities?

  Nick told me that he had been musing over this question for a long time, and had once approached the Gorilla Foundation with a proposal for an experiment, to see if gorillas could learn to make stone tools. The foundation was busy with other projects, and turned down Nick. It was, he said, an idea in search of a collaborator. His proposal was to motivate Kanzi to make stone flakes, not teach him with structured lessons. “We want to avoid the criticism of classical conditioning,” he said. He suggested we would need a box with a transparent lid. Something enticing would be put in the box, and the lid would be secured with a length of string. Kanzi could be shown by example how to make flakes—by knocking two rocks together, for which the proper archeological term is hard-hammer percussion—but there would be no active teaching, no shaping of his hands, no breaking the task down into component parts.

  A little more than a decade ago Richard Wright, of the Bristol Zoo in England, had taught an orangutan to make stone flakes and to use them to cut string to reach a food reward. But Wright had taught the orangutan each component of the task separately. He had also arranged it so that one of the rocks was secured to a plank of wood, not held in the hand as tool-makers did, and Kanzi would. Although Wright’s experiment stands as an important part of primatological research, what Nick had in mind would go further. Wright had taught the orangutan the behavioral components of the task, while our approach would be to impart the conceptual components.

  I made some suggestions to Nick about how the design of the food box, or tool site as we came to call it, could be improved. Nick had underestimated Kanzi’s inclination and ability to tear flimsy objects apart, especially if there is food inside. We talked a little about how to carry out tool-making activities so that Kanzi might want to emulate us. And Nick promised to get in touch with me after we returned to the United States. This he did within a couple of weeks, and I told him that we had made a tool site to his specifications. A week later he arrived at the Language Research Center with fellow archeologist Kathy Schick, their truck laden with a thousand pounds of rock.

  In 1989, Tom Wynn, an archeologist, and Bill McGrew, a primatologist, published a paper called “An Ape’s View of the Oldowan.” The word Oldowan is the name applied to the earliest known stone-tool assemblages, which were found in Africa and date back to two and a half million years ago. The question Tom and Bill addressed in their paper—“When in human evolution did our ancestors cease behaving like apes?”—was essentially the same as Nick’s.4 In other words, given the right circumstances, could apes make Oldowan tools? Tom and Bill’s approach to the question was not experimental, as Nick planned, but instead involved examining the skills required to make Oldowan tools, and then seeking signs of such skills in aspects of chimpanzees’ lives.

  The artifacts that make up Oldowan assemblages were produced from small cobbles, and they include about half a dozen forms of so-called core tools, such as hammerstones, choppers, and scrapers, and small, sharp flakes. The tool-makers were assumed to have had mental templates of these various tool types, which they used to produce tools for a range of different functions. The tools are often found in association with broken animal bones, which sometimes show signs of butchery. The clear inference is that, beginning about two and a half million years ago, our human ancestors began exploiting their environment in a non-apelike way, by using stone tools as a means of including significant amounts of meat in their diet.

  Until quite recently, archeologists argued that the earliest tool-makers lived lives analogous to those of contemporary hunter-gatherers. In other words, they organized themselves into small, mobile bands, established temporary home bases, and divided the labor of hunting and gathering between male and female members of the band. This was a very humanlike way of life, albeit in primitive form, and most definitely unlike that of an ape.

  In recent years, however, a reexamination of the archeological evidence has changed this picture dramatically, making it much less humanlike and more apelike. There is no question that these early members of the human family made tools and butchered animal carcasses. But there is considerable debate over the extent to which they were active hunters as opposed to opportunistic scavengers. And the notion of home bases and a division of labor between the sexes has been abandoned as untenable. The earliest tool-makers are now viewed as bipedal apes who lived and foraged in social groups in a woodland/savannah environment, as baboons and chimpanzees do, and whose repeated use of specific locales for butchery activities created the archeological sites that are being uncovered today.

  An equally important shift of perspective has taken place regarding the tool assemblages themselves. In her pioneering work in East Africa in the 1940s, 1950s, and 1960s, Mary Leakey identified what she
assumed were intentionally produced tool types. But when Nick Toth began a program of experimental archeology in the 1970s, in which he became a proficient maker of Oldowan artifacts himself, he came to a very different view. “My experimental findings suggest that far too much emphasis has been put on cores at the expense of flakes,” he wrote. “It seems possible that the traditional relationship might be reversed: the flakes may have been the primary tools and the cores often (although not always) simply the byproduct of manufacture… . Thus the shape of many early cores may have been incidental to the process of manufacture and therefore indicative of neither the maker’s purpose nor the artifact’s function.”5

  Nick’s reassessment of the Oldowan artifacts revolutionized African archeology, and further changed the perception of the humanness of the earliest tool-makers. According to this new theory, the half dozen different tool types in Oldowan assemblages were not the product of mental templates in the minds of sophisticated tool-makers; instead, they were the results largely of the shape of the cobble from which the flakes were struck, and the nature of the raw material. The only skill required by the earliest tool-makers, therefore, was that of striking flakes off a core using a hammerstone. How cognitively demanding is such a task?

  “One of the more direct ways to assess the cognitive ability employed in tool-use is through examination of spatial concepts,” wrote Tom and Bill in their 1989 paper. “The spatial concepts required for Oldowan tools are primitive. The maker need not have paid any attention to the overall shape of the tool; instead, his focus appears to have been exclusively on the configuration of edges.”6 Tom and Bill identified three spatial concepts in relation to tool-making: proximity, boundary, and order. Simplest of the three is proximity, which refers to the ability to land hammer blows repeatedly in the same or more or less the same position on the core cobble. A more complex notion is that of boundary in space, which has to do with the division of a spatial field into different realms. This refers to the process of striking flakes off both sides of a cobble, yielding a bifacial chopper. The most sophisticated spatial notion is that of order, which concerns control over the location of a sequence of hammer blows.