The Ape And The Sushi Master Reflections Of A Primatologist

by Franz De Waal

  In doing so, I will cover ground that one might think is safe for a scientist of my training, namely the question of animal culture. The ground is like quicksand, however. There is, in fact, so much resistance to the idea of animal culture that one cannot escape the impression that it is an idea whose time has come. The air is filled with claims and counterclaims; everyone has an opinion, and a strong one at that. In this melee, an entirely new field has come into existence: armchair primatology. It is not unusual for scholars barely able to tell a chimpanzee's front from its behind to criticize experts who have studied the species all their lives, or for someone who has never set foot on a particular island to dispute the findings of a team that has worked there for half a century. It must be a sign of its arrival that primatology has become everybody's business!

  The island in question is Koshima, in the extreme south of Japan, where the first evidence for animal culture was gathered. A high point of my trip was a visit to Koshima, where I talked with the elderly but still sharp Mrs. Mito, who has been there from day one. After having heard so much about it, I was delighted to see with my own eyes how the monkeys still wash sweet potatoes in the ocean.

  Litter-Box Culture

  To introduce the topic of animal culture, let me start with an everyday example: the way cats learn to use the litter box. One of our cats visits the box to take a pee while her three kittens follow. Cats lack a sense of privacy, and so the offspring closely watch morn's activities, leaning over the rim of the box. Young kittens don't have particularly good eyesight, so it is unclear what they see. One of them awkwardly climbs into the box and is soon scratching around, moving litter like her mother. Then all of a sudden the little one, too, crouches and pees, with ears folded back. No one told her to do so, and it is hard to believe that cats have an inborn image of a modern invention such as the litter box. The mother's behavior seems to have triggered the youngster's.'

  Social learning is widespread in animals and may continue well beyond the point at which it started. For example, as a student I worked in a laboratory in Utrecht where one scientist regularly caught monkeys out of a large group with a net. At first, the monkeys gave warning calls whenever they saw him approach with his dreadful net, but later they also did so when he only walked by. Still later, years after his research had ceased, I noticed that monkeys too young to have known the threat he once posed alarm-called for this man, and for no one else. They must have deduced from the reaction of their elders that he was not to be trusted. I recently heard that the group kept this alarm-call tradition up for decades, still always aimed at the same person!

  The passing on of a "predator image" has also been observed in the field, in Kenya, by Dorothy Cheney and Robert Seyfarth. Vervet monkeys have different alarm calls for different predators (such as leopard, eagle, and snake), but need to learn to connect the one with the other. The investigators tested the knowledge of their monkeys by playing alarm calls from a concealed speaker. Since different predators require different responses, they divided the reactions of infant mon keys into three types: one was to run to the safety of mother, another was to react in a way that could get them into trouble, and the third was the correct response. For example, the right response to a snake alarm is to stand upright in the grass and look around. This would be suicidal in reaction to a leopard call, which requires monkeys to climb a tree. Cheney and Seyfarth found that both mother-oriented and wrong responses disappeared with age, whereas correct responses increased. This suggests that young monkeys learn how to react to each specific alarm. It would be incredibly costly for them to do so by trial and error: most likely they pick up information from the rest of the group. Indeed, youngsters who watched adults before responding themselves were more likely to show the right response to a calla

  These findings contradict the widespread belief that survival tactics must be hard-wired and instinctive. Not so in the case of vervet alarm calls: monkeys who fail to pay attention to their fellows simply will not make it. What we have, then, is an absolutely critical set of responses transmitted through the observation of others. Instead of relying on genetic information, this is a social, cultural process. In the laboratory, Susan Mineka has demonstrated the same kind of learning by showing snakes to captive-born monkeys who had never seen one in their entire lives. These naive monkeys were unafraid until they saw their wild-born parents react with intense fear to the same snakes. From that day on, the captive-born monkeys, too, showed snake phobia.? And not only in primates do we see a cultural construction of the enemy image: the same has been demonstrated in birds.8

  The other major domain of cultural learning is food. Animals learn from each other what to eat, and what not. Parent crows that fly daily with their offspring to the local garbage dump to look for tasty morsels instill in them a life-long preference for such sites, whereas the crow family that survives on natural foods will have offspring that carry on the same tradition when they get older. Food aversion is similarly transmitted. This was first noticed by a German rodent-control officer who set out poisoned bait, killing wild rats in large numbers. After a while, however, the remaining rats began to avoid the bait, and their offspring would do the same. Without any direct experience with the bait, young rats would eat only safe foods.

  An experimental psychologist, Bennett Galef, tested this in his laboratory by feeding rats two diets of different texture, taste, and smell. He then laced one of the diets with lithium chloride, which makes rats sick. This procedure led the animals to avoid the contaminated diet. The question now was how the rats' offspring would react after removal of the contamination. Both diets were again perfectly okay to eat, but adults fed exclusively on only one diet due to their bad experience with the other. It turned out that the pups acted like their parents. Of 240 pups given a choice of both diets, only one ate any of the food that adults in its colony had learned to avoid.9

  All of these examples-the alarm calling, the snake fear, the food aversion-arouse intense debate among psychologists about the exact learning mechanism involved. One might think it is mere imitation, but increasingly the term "imitation" is being reserved for cases in which a solution to a problem is copied with an understanding of both the problem and the model's intentions. This usage has turned "imitation" into a small, cream-of-the-crop subset of social learning, one that may not apply to rats and cats, perhaps not even to monkeys and apes.

  As soon as individual learning enters the picture-that is, a behavior is acquired partially through trial and error-the suspicion is that we are dealing with something simpler than imitation. A good example is the kittens and the litter box: it is very possible that all they learn from their mother is where to do the deed. Once they have been brought to the right spot, the rest can be construed as regular feline responses to the smell of urine and the feeling of loose gravel under their feet. So, even though the kittens act like Mom, this doesn't necessarily mean that they are following her example, and even less that they understand the box's purpose.

  To dismiss such behavior as having nothing to do with imitation is not altogether fair to animals, though, because we don't apply the same standards to people. If, from watching a soccer game, I develop the habit of kicking a ball, this doesn't mean that all there is to becoming a player is to mimic observed actions. It takes years of practice to control the ball and send it whichever way I want it to go. If one could become a soccer star from merely watching the game played by others, the world would be full of them. All imitation is a combination of a general idea picked up from others and individual practice to refine the skill. Inasmuch as we accept this simple truth for human imitation, why be so picky about other animals? True, they often have only a vague understanding of what others are doing-if they understand it at all-but whatever information they gain from watching is built into a solution developed by themselves, which is the way we imitate much of the time as well.

  The simplest form of social learning is known as "local enhancement," in which one individual
is attracted to a place where another is doing something interesting, such as finding food. The attraction then leads the first individual to explore the same situation and learn the solution on its own. The model thus indicates the where rather than the how of the answer. Our kitten example fits this category.

  Another common possibility is known in anthropology as "stimulus diffusion" and in psychology as "emulation." Here a general idea, outcome, or concept is obtained from others but the specifics are worked out independently. A modern-day example is how Microsoft "borrowed" the windows concept from Macintosh. DOS-based machines now produce approximately the same clickable environment on the screen as the Macintosh, but they do so via a totally different programming architecture. Microsoft thus rightly claims that Windows is not an Apple imitation: it is a mere emulation. Similarly, a bird may learn from another that crabs can be opened and that the inside is edible, but will still need to figure out on its own how to get to these softer parts.

  Whatever the exact process, the critical question before we speak of culture is whether an animal would ever have hit on a particular solution or developed a particular habit without the benefit of social companions. Would my kittens have learned to use the litter box on their own? I am afraid not. Would the captive-born monkeys have come to fear snakes on their own? Yes, but only after having been bitten, which is a far trickier way of getting to know snakes than through the observation of others. Social learning has tremendous advantages. We can debate long and hard what to call the process or how complex it is. All that really matters is that one individual adopts a habit under the influence of another.

  The Sushi Master

  Learning from others is second nature to humans: we do it more readily and precisely than any other animal. Therefore, when a young chimpanzee is raised with a human child, the direction of influence is more likely to be from the ape to the child than the other way around. This was discovered the hard way, in the 1930s, by Winthrop and Luella Kellogg, who were forced to terminate a co-rearing experiment in their home when their son, Donald, began to give guttural food barks like those of the female chimpanzee, Gua, with whom he was being raised. When Donald picked up an orange and ran to his parents while grunting "uhuh, uhuh," it was decided that his aping of the ape had gone far enough:

  The situation in which the two lived together as playmates and associates was much like that of the two-child family in which Gua, because of her greater maturity and agility, played the part of the older child. With the added stimulation thus afforded, the younger child in such situations usually learns more rapidly than would otherwise be the case. It was Gua, in fact, who was almost always the aggressor or leader in finding new toys to play with and new methods of play; while the human was inclined to take up the role of the imitator and follower.10

  Gua, too, was a good imitator. The Kelloggs describe how she became a typist after having seen her foster parents type for many months. One day, a very young Gua climbed on the typewriter stool and sat properly behind the machine, moving her hands simultaneously up and down the keyboard, pounding the keys with her fingers. We can only speculate about the literary heights the chimpanzee might have attained had the experiment continued.

  There are now many studies of the mimetic abilities of apes, such as the one by Deborah Custance at the Yerkes Primate Center in Atlanta. Sitting in front of two juvenile chimpanzees, Scott and Katrina, the investigator would make simple gestures, such as raising a foot, slapping the floor, or wip ing her own face, and reward the apes for copying them. After this training, Custance demonstrated a series of gestures that had never been rewarded before. These included puffing out her cheeks, clapping, jumping, and self-hugging. Scott and Katrina's responses were videotaped and evaluated by observers kept in the dark about what the experimenter had demonstrated. This way, there was an independent assessment of the imitation. The two apes did very well, showing that they had no trouble copying arbitrary body movements.11 We may not realize it-being ourselves masters of imitation-but the translation of perceived into performed action is quite a feat. The tendency to act like behavioral Xerox machines sets apes apart from most other animals and makes them obvious candidates for the evolution of culture.

  Masako Myowa-Yamakoshi and Tetsuro Matsuzawa, at the Primate Research Institute of Kyoto University, conducted a more complex study of imitation involving all kinds of objects. Matsuzawa runs a facility in which chimpanzees live outdoors in a social group but can be called inside for a voluntary experiment. Once the ape sits in front of him, the human demonstrates a simple action. All of the apes in this study were fully adult and hence probably less inclined to imitate than juveniles. Seeing each action only once, the apes rarely copied them. They did so only if the action linked two objects (such as putting a ball in a bowl) rather than linking an object with the body (such as putting a bowl on one's head). Interestingly, connecting different objects is typical of tool use in the field, such as when chimpanzees poke a stick into a termite hill or use chewed leaves as a sponge to extract water from a hole. Could it be that the ape mind is set up to pay special attention to technical solutions so as to better replicate them?12

  Under normal circumstances, apes see the behavior of their group mates numerous times, and so have many opportunities to become familiar with them. They watch others at close range, following each and every move in detail. Perhaps, as suggested by Matsuzawa, they follow the model of the sushimaster apprentice. The apprentice slaves in the shadow of masters of an art requiring rice of the right stickiness, delicately cut ingredients, and the simple, eye-catching arrangements for which Japanese cuisine is known. Anyone who has tried-as have I-to cook rice, mix it with vinegar, and cool it off with a hand-held fan so as to quickly mold fresh rice balls in one's hands, knows what an incredibly complex skill this is, and it is only a small part of the job. Actually, I have been told that the reason one never sees female sushi masters is that a woman's hands are too warm for the task-an explanation to be taken with a grain of salt, given that no one ever complains about the sushi that women prepare at home. Men tend to claim high-status jobs for themselves; the exclusion of women from the sushi domain confirms its central place in Japanese culture.

  To return to the apprentice sushi master: his education seems a matter of passive observation. The young man cleans the dishes, plops the kitchen floor, bows to the clients, fetches ingredients, and in the meantime follows from the corners of his eyes, without ever asking a question, everything that the sushi masters are doing. For no less than three years he watches them without being allowed to make actual sushi for the patrons of the restaurant-an extreme case of exposure without practice. He is waiting for the day on which he will be invited to make his first sushi, which he will do with remarkable dexterity.

  This runs counter to imitation the way I described it before, in which an idea caught from others is supplemented with a great deal of individual practice. But who knows what apprentices do in their spare time? It is entirely possible, for instance, that the older masters-who, like all aging male primates, are more patient with younger males-take the apprentice aside after the closing of the restaurant to show him a few tricks and have him try things out for himself. Whatever the truth about the sushi master's education, Matsuzawa's point is that the watching of skilled models firmly plants action sequences in the head that come in handy, sometimes much later, when the same task needs to be carried out.

  It Takes a Village ...

  Watching others is a favorite activity of young primates. They constantly hang around their elders, absorbing every little detail of what is going on. At the same time that psychologists are debating what young animals do with all of this information, and whether it deserves to be called imitation if they duplicate the actions of others, field-workers have taken an entirely different tack to the issue of animal culture. Much like cultural anthropologists, who document how one human population differs from another, they compare different sites and note how each chimpanzee commu
nity has its own way of doing things. This ethnographic method is also being applied to other animals, most successfully to dolphins and whales. The rapidly growing literature gives the impression that we have only scratched the surface: cultural diversity in the animal kingdom probably takes on vast proportions.

  Whereas these observations are not being contested, not everyone agrees that the term "culture" best describes differences between groups. This obviously depends on one's definition. You would think that scholars dispassionately arrive at a reasonable characterization of a phenomenon, after which they only need to agree on what is and is not covered by it. But definitions are rarely neutral; they mirror entire world views. Behind the ongoing culture wars, the debate is about nothing less than humanity's place in the cosmos. Definitions of culture have become the political football in this larger controversy.

  It is not hard to come up with a definition of culture that rules out all species except our own. Even tools can be defined in such a way that they are found only in our species-for example, by requiring that they fit a symbolic context. Such exclusive definitions tend to focus on the highest human achievements associated with a process, declaring these absolutely essential. This is a legitimate line of thought, inasmuch as it lets scientists comfortably speak of the uniquely humnan capacities for culture, tool use, language, morality, and politics.

  My own bias, however, and that of many fellow primatologists, is quite the opposite. We tend to look beyond the brief evolutionary history of the human race, eyeing a much longer past and a much wider range of animals. All the fancy things that humans do with tools and culture are certainly worthy of attention, but they are best kept out of initial definitions so as to cast the net as widely as possible. This approach is cony monplace in biology. Thus, biologists are comfortable saying that both chickens and people walk bipedally even though it is obvious that they do so in radically different ways (look at which way their "knees" are pointing!). Biologists always define processes-nutrition, locomotion, reproduction-in the broadest possible terms because evolution has produced a multitude of means to achieve them.