Kanzi: The Ape at the Brink of the Human Mind

by Sue Savage-Rumbaugh

  Finally I turned to Kanzi and explained that I was unable to open his door because Tamuli would not return my keys. It then occurred to me that perhaps I should solicit his help as he often seemed to be able to communicate with Tamuli far better than I could. “Please tell Tamuli to give me my keys,” I implored. Kanzi climbed to the top of his room where wire mesh separated his area from Tamuli and looked at her while making a small noise. Tamuli approached Kanzi, looking directly at him. Kanzi made several multisyllabic sounds to Tamuli. Tamuli listened, then to my amazement quietly walked over and handed my keys back.

  Did Kanzi tell her to give me my keys? Did she understand him and comply? It certainly seemed so. If such an event were to occur between human siblings, we would call it language, even if it were in a tongue we could not yet recognize or catalogue.

  To further our understanding of animal intelligence we must learn to ask better questions—questions that focus on unusual events, rather than mundane and readily controllable ones. If we were to start with the assumption that animals are conscious and capable of thought, reason, and complex communication, we would find it difficult to come up with evidence that would completely disprove this view. Instead, we start with the premise that they are incapable of such accomplishments and find it difficult to disprove this view.

  We do not realize how deeply our starting assumptions affect the way we go about looking for and interpreting the data we collect. We should recognize that nonhuman organisms need not meet every new definition of human language, tool use, mind, or consciousness in order to have versions of their own that are worthy of serious study. We have set ourselves too much apart, grasping for definitions that will distinguish man from all other life on the planet. We must rejoin the great stream of life from whence we arose and strive to see within it the seeds of all we are and all we may become.

  Our definitions of man, readied anew for each additional discovery of capacities in animals, continue to impede our sense of belonging to the greater whole. In demeaning the capacities of animals, we found it easy to glorify our own. Having invented language, we turned and looked down upon the well-spring of life from which we arose with something akin to disdain. We catalogued our achievements, chronicling in detail how distinct they were from all other creatures, hesitant even to say that a continuum existed between ourselves and them. Not only did we deny animals the potential for thought, we assumed they had no awareness of their own existences.

  Gordon Gallup, a psychologist at the State University of New York, Albany, two decades ago performed what became a classic experiment, as it was the first attempt at a behavioral definition of self-awareness. The experiment was conceptually simple, but not as easy to perform as it might seem. Gallup determined whether chimpanzees were able to recognize their own images in a mirror by placing a red, odorless/tasteless dot on their foreheads while they were anesthetized. Once the chimpanzees awoke they paid no attention to the dot until they happened to glance in the mirror. Then they noticed the dot and immediately set about removing it.

  Chimpanzees, like ourselves, are quite concerned about appearances and show mirror recognition early, sometimes even before human children. Panzee has been the most precocious of our apes in this regard, passing her version of the dot test accidentally at six months of age when she first became interested in her reflection. I noticed that when a mirror was present in the room, every so often she would walk by it and check out her reflection. But I was uncertain as to whether she recognized herself or not. One day, while playing, she nicked her browridge ever so slightly, but enough that a tiny red scrape appeared. At the time, she paid this mild bump no heed whatsoever, continuing her rough and rowdy play as though nothing had happened. About forty-five minutes after this injury, having not once paid attention to the minor scrape in the meantime, Panzee happened to be walking by the mirror and, as usual, stopped to glance at her reflection. This time, however, she paused, sat down, and gazed intently for about forty-five seconds. Then she slowly reached up and touched the red spot on her forehead with her index finger while watching herself in the mirror. After touching the spot, she then looked at her finger, but there was no blood, as the scrape was so slight. She then leaned forward and looked at the spot more closely in the mirror for about fifteen seconds. Then, seemingly having satisfied her curiosity, she moved on to other activities and paid the scrape no more attention at all.

  None of our other apes have displayed mirror recognition at such an early age; however, they have all demonstrated, in a variety of ways, that they not only understand their own image, but they enjoy playing with it and even altering it. Moreover, they recognize themselves not only in mirrors, but on television as well. We first observed this one day when we were testing Austin’s sorting skill while taping the session. Austin was sorting colored blocks and sipping an orange drink at the time he began to notice himself in the live monitor I had set up. We could tell that he was playing with his television image because he kept bobbing up and down and making funny faces while staring at the monitor from only a few inches away. He then positioned his face adjacent to the screen and began to sip his orange drink, carefully observing himself roll the liquid from one side of his mouth to the other.

  Food manipulation with the lips is something that apes are very good at and spend a great deal of time practicing. Since, in the wild, their hands are often occupied holding onto tree limbs, they have to be able to pick and peel fruit with their lips and teeth. They become very adroit with their lips and seem to take great pride in this skill. I once observed Sherman demonstrate for me his ability to roll a dime, standing on edge, from one side of his mouth to the other and back.

  Austin’s initial fascination with his video image engaged his undivided attention for twenty minutes. Later on, he hit upon the idea of using the television to look down his throat. He had found this difficult to do with a mirror, since when he tilted his head back he could not look very far down his throat. He soon realized, however, that the television, unlike the mirror, could show him his throat from a variety of different angles depending on the position of the camera. Consequently, he began to open his mouth extremely wide while looking at the television screen so that he could see what was at the bottom of his throat. We would then point the camera directly down into his mouth while he watched on television. This was Austin’s idea, not mine. Earlier, when he first got a good look at his throat and found it too dark to see, he left to look for a flashlight and then returned to the same position to shine the light down his throat. Such an activity was never suggested to him, nor even demonstrated for him. It emerged from his own desire to see what the bottom of his throat looked like.

  Sherman had no interest in this at all, and though he watched Austin look at his throat on television many times, he never did so himself. Sherman had other uses for the television image. He preferred to use it for brushing his teeth, applying lipstick, and putting large fur shawls around his shoulders while bobbing up and down—apparently to practice the fearsomeness of his display. Even though Austin watched Sherman do all of these things, he elected to do none of them. He looked at his throat and either watched himself eat or, if he had no food, pretended that he was eating out of an empty bowl with an empty spoon, studiously observing his image as he did so.

  These were the things that engaged Austin and Sherman. They were individually specific and their separate interests emerged from within each of them. If we attempted to elicit these behaviors on purpose we were almost never successful. They appeared only when the apes themselves were interested and engaged with the television.

  Kanzi was different yet again. He neither brushed his teeth nor looked down his throat. Instead, he used his image to practice blowing up balloons and blowing bubbles with bubble gum. He wanted to be good at both things and both were difficult for him. None of the other apes used a mirror for these purposes. Panbanisha did not really use the mirror in the sense that the others did. She was content to admire herself in the m
irror, checking her appearance from every angle. When her canines began to come in, she spent long periods of time simply looking at these new large and potent teeth in the mirror. She also, at this time, began to use them.

  Sherman and Austin not only recognized themselves in the mirror and on television, they also differentiated between live and taped images of themselves. When they saw a self-image, they would stick out their tongues, bob their heads, and so on to see if it were a live image. If it was live, the behaviors I described above would begin to appear. If the image did not react, they typically went on about their business, ignoring the television unless something very interesting was happening on the screen.

  Working with Emil Menzel, we were even able to test this differentiation in a controlled way by using a port through which the chimps extended their hand. A television camera was focused on their arm and presented them a picture of what their arm was doing on the other side of the port. Once they put their hand through the port, the only way they could see what it was doing was by watching the television. They were able to perform all sorts of manipulative feats that required visual monitoring of the hand. For example, if presented with two closed clear containers, one with food and one without, they reached for the one with food. Even when the television image was reversed, they still utilized it, albeit a bit awkwardly, to guide their hand movements. When the live image of their hand was replaced, without warning, with a taped image of their hand, they quickly recognized that the movements they were making did not correspond with what they saw on the television (even though it was their own hand) and withdrew their hand from the port, since they preferred not to have it extended if they could not visually monitor what it was doing.

  Most chimpanzees pass Gallup’s dot test easily once they have had sufficient exposure to mirrors, though early isolation rearing can cause them to fail, suggesting some disruption of self-concept as a result of lack of socialization. Monkeys, however, are notorious for failing this test in spite of many attempts, regardless of rearing conditions. The curious thing is that monkeys do learn how mirrors work; when they see someone behind them in the mirror, they turn around to look, rather than attempting to touch the object in the mirror. It has even been found that when lightweight plastic daisy flowers are attached to collars on their necks, monkeys will notice the flowers in the mirror and reach behind their heads to remove them.

  Thus the puzzle remains as to why they do not pass the dot test. Are they unaware that the image in the mirror is really them? Do they a lack a self-concept? Can it be that they understand everything they see in the mirror, except who they are? These results are often used to support such a view by philosophers and scientists who view self-concept as something that either must be present in the form that most human adults experience it, or entirely absent.

  It could be that monkeys simply do not care that they have a dot on their fur. In the wild, chimpanzees and baboons have been observed feeding together in the same tree on sticky-gooey fruit. When wads of the fruit get stuck in the chimpanzees’ hair, they stop eating to remove them when they are clearly visible on thighs or stomachs. Baboons, however, seem oblivious to these sticky wads even though they can clearly see them. Perhaps monkeys are more dependent on other monkeys for self-tidiness than apes, or perhaps their self-concept does not deal as much with appearances as does the self-concept of apes.

  If we are to explore the psychological worlds of other animals we would benefit from accepting the fact that they may have self-concepts that are not equivalent to our own in form or content. In fact, it is far more rewarding to press to understand minds and self-concepts that are not equivalent in all manners to our own.

  To fail to try to understand the world from the point of view of the lion or the bat is to admit that the human existence is so limited that it cannot project itself satisfactorily into the minds of different creatures. Do we really want to accept this limitation when we quite satisfactorily project ourselves into all sorts of invented imaginary creatures, even those with very different sensory systems and value systems than our own? All one need do is to read a few comic books to conclude that the projection abilities of our species are great indeed and that our children, at least, have little difficulty in going beyond their ordinary frameworks of reality.

  Even within our own culture, the way individuals react to their own image depends, to a large extent, on how they feel about themselves. On one occasion I was videotaping some of the children with impairments in Mary Ann’s project, to document their keyboard communication skills. One child noted himself on the live monitor I was using and grew interested in his image. Consequently, I turned the monitor around so that all the children could see themselves and continued to tape. One by one, I focused on the different children. Many of them seemed puzzled, never having seen themselves on live television before. Most of them stared at the television for some time. After about ten minutes several of them winked, stuck out their tongues, or waved their hands to see if that was really them on the television. Others grew very shy and turned away from the image as though bothered every time the camera zoomed in on them. Still others glanced at themselves furtively every so often. The ones that showed clear self-recognition by sticking out their tongues, for example, were also the ones that were the most outgoing—the showoffs of the group. The ones that turned away from their own image were the shyest of the students. Shall we conclude that they did not recognize themselves because they ignored their image?

  Dogs fail Gallup’s dot test also. Must we infer that they also lack a self-concept? If they don’t have a concept of self, what do they have when they look and listen and carry out our commands? What kind of a concept do they have when one dog tells another that he is not to touch his bone, even when he is out of the room? The dot test is an important milestone, but unfortunately it has prevented us from asking other questions about what it is that is happening when an animal looks into a mirror.

  Dogs sometimes stare into mirrors, as though attempting to make sense of the scene there. On one occasion I approached a dog in the process of studiously gazing at its reflection in the mirror. I stood quietly behind the dog, next to an open door that was also reflected in the mirror. The dog continued to gaze at itself with studied intensity. After a few moments I began to move my hands, and immediately, the dog’s gaze shifted from looking at its own mirror image to mine. As it watched me, its ears became erect as if listening for the slightest sound. I moved my hands in larger motions and suddenly the dog began to bark in alarm at my mirror image. When it was clear that the dog was quite upset, I spoke. As soon as I spoke the dog wheeled around and ran straight to me, wagging its whole body in a desire to be patted. I patted the dog until it calmed down.

  As soon as it was calm, it went directly back to the mirror of its own accord and assumed the identical location it had been in before and began to study the images in the mirror once again. I stood in the same place, approximately six feet behind and a foot to the left of the dog. The dog’s eyes looked first at its own image and then looked again for mine, and once it found it, began again to stare intently. As I again stood very still, the dog made no sound. After approximately thirty seconds, I once more moved my hands and immediately the dog began to bark at my image with great distress. A little later I spoke and again the dog turned quickly and rushed to me to be patted.

  The dog returned to the mirror and once again took the same position and this time immediately began looking for my image. I moved and was no longer reflected in the mirror. After gazing in the mirror as though searching for me but not finding me, the dog turned and walked away.

  It was hard to avoid the impression that the dog was attempting to understand how there could be two of me, or how I could appear to be in two places at the same time. At the very least the dog was attempting to puzzle its way through something as it returned twice to search for what it had previously seen in the mirror.

  Much of what we wish to learn about animals and the
ir minds can be learned only when they are already engaged with the question we wish to pursue at the time. We cannot place a mirror in front of a dog, put a dot on its head, and then wait for self-recognition. The dog must be engaged with the mirror and we currently do not understand dogs well enough to bring this engagement to the forefront at will.

  Engagement with a topic of interest is a widely accepted phenomenon in our own species—though one not well understood. There are children who become engaged with chess, for example, and learn all the rules easily, though they may do poorly in school in many subjects. There are children who become engaged with dinosaurs and can tell you the names of a hundred different species though they cannot recall a single date in their history class. There are three-year-olds who become engaged with backhoes and tractors and can tell a John Deere from a Ford tractor, though they cannot remember a nursery rhyme they have heard forty times.

  Engagement, and its relationship to the accumulation and processing of information, is a little-studied phenomenon, representing as it does, individual skills rather than those that can be measured in a group of people.

  Currently, our understanding and measurement of human intellectual capacity is oriented toward group skills and toward activities that can be elicited on command, regardless of the state of engagement. Indeed, being able to engage one’s focus on the questions of the examiner, rather than on one’s own interest, is the primary measure of test-taking ability, and test-taking ability is the primary measure of intelligence. When we find that animals do not do well when compared to people in this way, we must not assume that we have really measured their intellect. Perhaps we have measured only our own limited ability to engage them.