Are We Smart Enough to Know How Smart Animals Are

by Frans de Waal

  To know the friends of your opponents takes experience. It implies that individual A is aware not only of her own relations with B and C but also of the relation between B and C. I dubbed this triadic awareness, since it reflects knowledge of the entire ABC triangle. It is the same with us, when we realize who is married to whom, who is a son of whom, or who is the employer of whom. Human society could not function without triadic awareness.7

  The second example concerns wild chimpanzees. It is well known that there is no obvious connection between a male’s rank and his size—the biggest, meanest male does not automatically reach the top. A small male with the right friends also has a shot at the alpha position. This is why male chimps put so much effort into alliance formation. In an analysis of years of data collected at Gombe, a relatively small alpha male spent far more time grooming others than did larger males in the same position. Apparently, the more a male’s position depends on support from third parties, the more energy he needs to invest into diplomacy, such as grooming.8 In a study in the Mahale Mountains, not far from Gombe, Toshisada Nishida and his team of Japanese scientists observed an alpha male with an exceptionally long tenure of more than a decade. This male developed a “bribery” system, selectively sharing prized monkey meat with his loyal allies, while denying such favors to his rivals.9

  Years after Chimpanzee Politics, these studies confirmed the tit-for-tat deal making that I had implied. But even while I was writing my book, supportive data were being gathered. Unknown to me, Nishida had followed an older male at Mahale, named Kalunde, who had moved himself into a key position by playing off younger, competitive males against one another. These young males sought Kalunde’s support, which he handed out rather erratically, making himself indispensable to the advancement of any one of them. Being the dethroned alpha male, Kalunde made a comeback of sorts, but like Yeroen, he didn’t claim the top position for himself. He rather acted as power behind the scenes. The situation was so eerily similar to the saga I had described that I was thrilled, two decades later, to meet Kalunde in person. Toshi, as the late Nishida was known to his friends, invited me for some fieldwork, which I gladly accepted. He was one of the world’s greatest chimpanzee experts, and it was a treat to follow him around through the jungle.

  Living in the camp near Lake Tanganyika, one realizes that running water, electricity, toilets, and telephones are greatly overrated. It is entirely possible to survive without them. Every day the goal was to get up early, eat a quick breakfast, and get going before the sun rose. The chimps would have to be found, and the camp had several trackers to assist us. Fortunately, chimps are incredibly noisy, which makes them easy to locate. Chimps do not travel all in a single group but are spread out over separately traveling “parties” of just a few individuals each. In an environment with low visibility, they rely heavily on vocalizations to stay in touch. Following an adult male, for example, you continuously see him stop, cock his head, and listen to others in the distance. You see him decide how to respond, by replying with his own calls, silently moving toward the source (sometimes in such a hurry that you are left struggling through tangled vines), or continue on his merry way as if what he just heard lacked any relevance.

  By then Kalunde was the oldest male, only about half the size of a prime adult male. Being around forty, he had shrunk. But despite his advanced age, he was still into political games, frequently accompanying and grooming the beta male until alpha returned from a long period of absence. Alpha had traveled to the fringes of the community territory, escorting a sexually receptive female. High-ranking males may go for weeks on end “on safari” with a female, as it is known, in order to avoid competition. I knew about alpha’s unexpected return only because Toshi told me in the evening, but I had noticed great agitation in the males that I had been following the whole day. They were restless, running up and down the hills, totally exhausting me. Alpha’s characteristic hooting and drumming on empty trees had announced his return, making everyone hypernervous. In the following days, it was fascinating to see Kalunde switching camps. One moment he would be grooming the returning alpha; the next he’d be hanging out with the beta male, as if trying to decide which side he should be on. He offered the perfect illustration of a tactic that Toshi had dubbed “allegiance fickleness.”10

  You can imagine that we had much to talk about, especially comparing wild versus zoo chimps. Obviously, there are major differences, but it is not as simple as some people think, especially those who wonder why one would study captive animals at all. The goals of both types of research are quite different, and we need both. Fieldwork is essential to understanding the natural social life of any animal. For anyone who wants to know how and why their typical behavior evolved, there is no substitute for observing them in their natural habitat. I have visited many field sites, from capuchin monkeys in Costa Rica and woolly spider monkeys in Brazil to orangutans in Sumatra, baboons in Kenya, and Tibetan macaques in China. I find it very informative to see the ecology of wild primates and to hear from colleagues what sort of issues they are fascinated by. Fieldwork is nowadays very systematic and scientific. The days of a few scribbled observations in a notebook are gone. Data collection is continuous and systematic, typed into handheld digital devices, and complemented with fecal and urine samples that allow DNA analysis and hormone assays. All this hard, sweaty work has enormously advanced our understanding of wild animal societies.

  Yet in order to get at behavioral details and the cognition behind them, we need more than fieldwork. No one would try to measure a child’s intelligence by watching him run around in the schoolyard with his friends. Mere observation doesn’t offer much of a peek into the child’s mind. Instead, we bring the child into a room and present him with a coloring task or a computer game, let her stack wooden blocks, ask questions, and so on. This is how we measure human cognition, and it is also the best way to determine how smart apes are. Fieldwork offers hints and suggestions but rarely allows firm conclusions. One may encounter wild chimpanzees who crack nuts with stones, for example, but it is impossible to know how they discovered this technique or how they learn it from one another. For this, we need carefully controlled experiments on naïve chimpanzees who receive nuts and stones for the first time.

  Captive apes under enlightened conditions (such as a sizable group in a spacious outdoor area) have the added advantage of providing a close-up look at naturalistic behavior that one can’t get in the field. Here apes can be watched and videotaped much more fully than is possible in the forest, where primates often disappear into the undergrowth or canopy as soon as things get interesting. Fieldworkers are often left to reconstruct events based on fragmented observations. To do so is an art, and they are very good at it, but it falls short of the behavioral detail routinely collected in captivity. If one studies facial expressions, for example, zoomed-in high-definition videos that can be slowed down are essential, which require well-lit conditions rarely encountered in the field.

  No wonder the study of social behavior and cognition has fostered integration between captive and fieldwork. The two represent different pieces of the same puzzle. Ideally, we use evidence from both sources to support cognitive theories. Observations in the field have often inspired experiments in the lab. Conversely, observations in captivity—such as the discovery that chimpanzees reconcile after fights—have stimulated observations in the field on the same phenomenon. If, on the other hand, experimental outcomes clash with what is known about a species’s behavior in the wild, it may be time to try a new approach.11

  With regard to the question of animal culture, in particular, captive and fieldwork are now often combined. Naturalists document geographic variation in the behavior of a given species, suggesting a local origin and transmission. But they often cannot rule out alternative accounts (such as genetic variation between populations), which is why we need experiments to determine if habits can spread by one individual watching another. Is the species capable of imitation? If so, this
greatly strengthens the case for cultural learning in the field. Nowadays we move back and forth all the time between both sources of evidence.

  But all these interesting developments happened long after my observations at Burgers’ Zoo. Following Kummer’s example, my goal at the time was to spell out what social mechanisms may underlie observed behavior. Apart from triadic awareness, I spoke of divide-and-rule strategies, policing by dominant males, reciprocal deal making, deception, reconciliation after fights, consolation of distressed parties, and so on. I developed such a long list of proposals that I devoted the rest of my career to fleshing them out, at first through detailed observations, but later also experimentally. Proposals take so much less time to make than their verification! The latter can be very instructive, though. One can set up experiments, for example, in which one individual can do another one favors, as we did with our capuchin monkeys, but then add a condition in which the partner can do favors in return. This allows favors to travel in both directions between two parties. We found that monkeys become noticeably more generous if favors can be done mutually than if only one of them has the opportunity.12 I love this kind of manipulation, since it allows far more solid conclusions about reciprocity than any observational account. Observations never quite clinch the deal the way experiments can.13

  Even though Chimpanzee Politics opened a new agenda for research while introducing Machiavelli’s thinking to primatology, I was never quite happy with “Machiavellian intelligence” as a popular label for this field.14 This term implies an end-justifies-the-means manipulation of others, ignoring a vast amount of social knowledge and understanding that has nothing to do with one-upmanship. When a female chimpanzee resolves a fight between two juveniles over a leafy branch by breaking it into two and handing each youngster a piece, or when an adult male chimpanzee helps an injured, limping mother by picking up her offspring to carry it for her, we are dealing with impressive social skills that don’t fit the “Machiavellian” label. This cynical identifier made sense a few decades ago, when all animal (including human) life was customarily depicted as competitive, nasty, and selfish, but over time my own interests have drifted into the opposite direction. I have devoted most of my research to the exploration of empathy and cooperation. The exploitation of others, by using them as “social tools,” remains a great topic and is an undeniable aspect of primate sociality, but it is too narrow a focus for the field of social cognition as a whole. Caring relationships, the maintenance of bonds, and attempts to keep the peace are equally worthy of attention.

  The intelligence required to effectively deal with social networks may explain why the primate order underwent its remarkable brain expansion. Primates have exceptionally large brains. Dubbed the Social Brain Hypothesis by British zoologist Robin Dunbar, the connection with sociality is supported by a relation between a primate’s brain size and its typical group size. Primates that live in larger groups generally have larger brains. I always find it hard, though, to separate social and technical intelligence, since many big-brained species are strong in both domains. Even species that hardly handle any tools in the wild, such as rooks and bonobos, may be quite good at it in captivity. It remains true, though, that social challenges have been neglected for too long in discussions of cognitive evolution, which tend to focus on interactions with the environment. Given how all-important social problem solving is in the lives of our subjects, primatologists have been right to amend this view.15

  Triadic Awareness

  Siamangs—large black members of the gibbon family—swing high up in the tallest trees of the Asian jungle. Every morning, the male and female burst into spectacular duets. Their song begins with a few loud whoops, which gradually build into ever louder, more elaborate sequences. Amplified by balloonlike throat sacs, the sound carries far and wide. I have heard them in Indonesia, where the whole forest echoed with their sound. The siamangs listen to one another during breaks. Whereas most territorial animals need only to know where their boundaries run and how strong and healthy their neighbors are, siamangs face the added complexity that territories are jointly defended by pairs. This means that pair-bonds matter. Troubled pairs will be weak defenders, while bonded pairs will be strong ones. Since the song of a pair reflects their marriage, the more beautiful it is, the more their neighbors realize not to mess with them. A close-harmony duet communicates not only “stay out!” but also “we’re one!” If a pair duets poorly, on the other hand, uttering discordant vocalizations that interrupt one another, neighbors hear an opportunity to move in and exploit the pair’s troubled relationship.16

  To understand how others relate to one another is a basic social skill that is even more important for group-living animals. They deal with a far greater variety than the siamang. In a baboon or macaque troop, for example, a female’s rank in the hierarchy is almost entirely decided by the family from which she hails. Owing to a tight network of friends and kin, no female escapes the rules of the matrilineal order according to which daughters born to high-ranking mothers will themselves become high-ranking, while daughters from families at the bottom will also end up at the bottom. As soon as one female attacks another, third parties move in to defend one or the other so as to reinforce the existing kinship system. The youngest members of the top families know this all too well. Born with a silver spoon in their mouth, they freely provoke fights with everyone around, knowing that even the biggest, meanest female of a lower clan will not be allowed to assert herself against them. The youngster’s screams will mobilize her powerful mother and sisters. In fact, it has been shown that screams sound different depending on the kind of opponent a monkey confronts. Thus, it is immediately clear to the entire troop whether a noisy fight fits or violates the established order.17

  The social knowledge of wild monkeys has been tested by playing the distress calls of a juvenile from a loudspeaker hidden in the bushes at a moment when the juvenile itself is out of sight. Hearing this sound, nearby adults not only look in the direction of the speaker but also peek at the juvenile’s mother. They recognize the juvenile’s voice and seem to connect it with its mother, perhaps wondering what she is going to do about the trouble her offspring is in.18 The same sort of social knowledge can be seen at more spontaneous moments, when a juvenile female picks up an infant that is unsteadily walking about, only to carry it back to its mother, which means that she knows which female the infant belongs to.

  In white-faced capuchin monkeys, the American anthropologist Susan Perry analyzed how individuals form coalitions during fights. Having followed these hyperactive monkeys for over two decades, Susan knows them all by name and life history. During a visit to her field site in Costa Rica, I saw the characteristic coalition stance firsthand. Known as the overlord, two monkeys threaten a third with stares and wide-open mouths, one leaning on top of the other. Their opponent thus faces an intimidating display of two monkeys wrapped into one, with both threatening heads stacked on top of each other. Comparing these coalitions with known social ties, Susan found that capuchins preferentially recruit friends who are dominant over their opponent. This by itself is rather logical, but she also found that instead of seeking the support of their best buddies, they specifically recruit those who are closer to themselves than to their opponent. They seem to realize that there is no point appealing to their opponent’s buddies. This tactic, too, requires triadic awareness.19

  Two white-faced capuchin monkeys adopt an “overlord” position, so that their adversary is confronted by two threatening faces and sets of teeth at once.

  Capuchins solicit support by abruptly jerking their heads back and forth between a potential supporter and their adversary, a behavior known as headflagging, which is also used against danger, such as a snake. In fact, these monkeys threaten everything they don’t like, a tendency sometimes used to manipulate attention. Susan once observed the following deceptive sequence:

  Pursued by a coalition of three higher-ranking males, Guapo suddenly stopped in his track
s and began to produce frantic snake alarm calls while looking at the ground. I was standing by him and could plainly see that there was nothing there but bare ground. He headflagged to Curmudgeon [one of his enemies] for support against the imaginary snake. Guapo’s pursuers stopped short and stood up on their hind legs to see if there was a snake. After cautious inspection, they once again began threatening Guapo. Switching tactics he glanced up at a passing magpie jay (a nonmenacing bird) and did three bird alarms in rapid succession—calls that are usually reserved for large raptors and owls. Guapo’s opponents looked up, saw that it was not a dangerous bird, and again resumed threatening Guapo. He reverted to the snake alarm call tactic once again vehemently bouncing at the bare patch of ground, threatening the “snake” vocally. Although Curmudgeon continued to glare at Guapo for a bit longer, the rest of the group stopped threatening him, and he was able to resume foraging for insects, moving slowly and nonchalantly towards Curmudgeon while occasionally casting a furtive glance in his direction.20

  While such observations suggest but cannot prove high intelligence, there is an urgent need for information on the cognition of wild primates. Fieldworkers are finding ingenious ways to collect it. In Budongo Forest, in Uganda, for example, Katie Slocombe and Klaus Zuberbühler set out to record the screams of chimpanzees under threat or attack. These loud vocalizations serve to recruit aid, which prompted the scientists to see if the acoustics of screams depend on the audience. Given the dispersed lives of wild chimpanzees, only individuals who are within earshot—the audience—are likely to provide aid to a screaming victim. In addition to finding that the intensity of the calls reflected the intensity of the attack, the scientists noted a subtle deception encoded in them. Chimpanzee victims apparently exaggerate their screams (making the attack sound more severe than it truly is), provided their audience includes individuals that outrank their attacker. In other words, whenever the big bosses are around, chimp victims scream bloody murder. Their vocal distortion of the truth suggests precise knowledge of their opponent’s status relative to everyone else.21