Are We Smart Enough to Know How Smart Animals Are

by Frans de Waal

  This cat-and-mouse setup shows how much bodies matter. Much of our knowledge about ourselves comes from inside our bodies, and much of what we know about others comes from reading their body language. We are very attuned to the postures, gestures, and facial expressions of others, as are many other animals, such as our pets. This is why Menzel never liked the “theory” language that took over once ToM exploded as a topic as a result of other ape research. The central question became whether apes or children hold a theory about the minds of others.16 I have trouble with this terminology, too, because it makes it sound as if we understand others through a rational evaluation not unlike the way we figure out physical processes, such as how water freezes or how continents drift apart. It sounds far too cerebral and disembodied. I seriously doubt that we, or any other animal, grasp the mental states of someone else at such an abstract level.

  Some even speak of mindreading, a term reminiscent of the telepathic trickery of magicians (“Let me guess what card you have in mind”). The magician, however, operates entirely on the basis of which card he has seen you lay your eyes on, or some other visual cue, because there is no such thing as mindreading. All we can do is figure out what others have seen, heard, or smelled, and deduce from their behavior what their next step may be. Putting all this information together is no minor feat and takes extensive experience, but it is body reading, not mindreading. It allows us to look at a situation from the viewpoint of another, which is why I prefer the term perspective taking. We use this capacity to our own advantage but also to the advantage of others, such as when we respond to someone else’s distress or fulfill the needs of another person. This obviously gets us closer to empathy than ToM.

  Human empathy is a critically important capacity, one that holds entire societies together and connects us with those whom we love and care about. It is far more fundamental to survival, I’d say, than knowing what others know. But since it belongs to the large submerged part of the iceberg—traits that we share with all mammals—it doesn’t garner the same respect. Moreover, empathy sounds emotional, something cognitive science tends to look down upon. Never mind that knowing what others want or need, or how best to please or assist them, is likely the original perspective taking, the kind from which all other kinds derive. It is essential for reproduction, since mammalian mothers need to be sensitive to the emotional states of their offspring, when they are cold, hungry, or in danger. Empathy is a biological imperative.17

  Empathic perspective taking, defined by the father of economics, Adam Smith, as “changing places in fancy with the sufferer,” is well known outside of our species, including dramatic cases of apes, elephants, or dolphins helping one another under dire circumstances.18 Consider how an alpha male chimpanzee at a Swedish zoo saved the life of a juvenile. The juvenile had entangled himself in a rope and was choking to death. The male lifted him up (thus removing the rope’s pressure) and carefully unwrapped the rope from his neck. He thus demonstrated an understanding of the suffocating effect of ropes and knew what to do about it. Had he pulled at the juvenile or the rope, he only would have made things worse.

  I speak of targeted helping, which is assistance based on an appreciation of the other’s precise circumstances. One of the oldest reports in the scientific literature concerns an incident, in 1954, off the coast of Florida. During a capture expedition for a public aquarium, a stick of dynamite was set off under the water surface near a pod of bottlenose dolphins. As soon as one stunned victim surfaced, heavily listing, two other dolphins came to its aid: “One came up from below on each side, and placing the upper lateral part of their heads approximately beneath the pectoral fins of the injured one, they buoyed it to the surface in an apparent effort to allow it to breathe while it remained partially stunned.” The two helpers were submerged, which meant that they couldn’t breathe during the entire effort. The pod remained nearby and waited until their companion recovered, after which they all fled in a hurry, taking tremendous leaps.19

  Two dolphins support a third by taking her between them. They buoy the stunned victim so that her blowhole is above the surface, whereas their own blowholes are submerged. After Siebenaler and Caldwell (1956).

  Another case of targeted helping occurred one day at Burgers’ Zoo. After having cleaned the indoor hall and before releasing the chimps, the keepers hosed out all the rubber tires and hung them one by one on a horizontal log extending from the climbing frame. Upon seeing the tires, female Krom wanted one in which some water remained. The chimps often use tires as vessels to drink from. Unfortunately, this particular tire was at the end of the row, with multiple heavy tires hanging in front of it. Krom pulled and pulled at the one she wanted but was unable to move it. She worked in vain on this problem for over ten minutes, ignored by everyone except Jakie, a seven-year-old that she had taken care of as a juvenile. As soon as Krom gave up and walked away, Jakie approached the scene. Without hesitation he pushed the tires off the log one by one, beginning with the front one, followed by the second in the row, and so on, as any sensible chimp would. When he reached the last tire, he carefully removed it so that no water was spilled and carried it straight to his aunt, placing it upright in front of her. Krom accepted his present without any special acknowledgment and was already scooping up water with her hand when Jakie left.20

  Having gone over numerous incidents of insightful assistance in The Age of Empathy, I am pleased that there are now finally controlled experiments.21 For example, at the PRI where Ayumu lives, two chimps were placed side by side while one had to guess what kind of tool the other needed to reach attractive food. The first chimp had a choice between a range of tools—such as a straw to suck up juice or a rake to move food closer—only one of which would work for her partner. She’d need to look at and judge her partner’s situation before handing her the most useful tool through a window. This is indeed what the chimps did, showing a capacity to grasp the specific needs of others.22

  The next question is, do primates recognize one another’s internal states, such as the difference between a partner who is hungry and one who is sated? Would you give up precious food for someone who has just eaten a big meal right in front of you? This is the question Japanese primatologist Yuko Hattori asked the monkeys in our capuchin colony.

  Capuchins can be quite generous and are great social eaters, often sitting in clusters munching together. When a pregnant female hesitates to descend to the floor to collect her own fruits (being arboreal, these monkeys feel safer higher up), we have seen other monkeys grab more than they need and bring handfuls of food up to her. In the experiment, we separated two monkeys with mesh wide enough to stick their arms through, while one of them received a small bucket with apple slices. Under these circumstances, the provisioned monkey often brings food to its empty-handed partner. They sit next to the mesh partition and let the other one reach through to take food out of their hands or mouth, sometimes actively pushing it in their direction. This is remarkable, because the circumstances allow the possessor to avoid sharing altogether by staying away from the mesh. We found one exception to their generosity, however: if their partner had just eaten, the monkeys became stingy. Of course, this could be due to a sated partner being less interested in food, but the monkeys were stingy only if they had actually seen their partner eat. A partner that had been fed out of sight was treated as generously as any other. Yuko concluded that the monkeys judged the need, or lack thereof, of their companions based on what they had seen them eat.23

  In children, an understanding of needs and desires develops years before they realize what others know. They read “hearts” well before they read minds. This suggests that we are on the wrong track in phrasing all this in terms of abstract thinking and theories about others. At a young age, children recognize, for example, that a child looking for his rabbit will be happy to find it, whereas a child searching for his dog will be indifferent to the rabbit.24 They have an understanding of what others want. Not all humans take advantage of th
is capacity, which is why we have two kinds of gift-givers: those who go out of their way to find a gift that you might like, and those who arrive with what they like. Even birds do better than that. In one of those cognitive ripples typical of our field, empathic perspective taking has been suggested for corvids. Male Eurasian jays court their mates by feeding them delicious tidbits. On the assumption that every male likes to impress, experimenters gave him two foods to choose from: wax moth larvae and mealworms. But before giving the male a chance to feed his mate, they would feed her first with one of those two foods. Seeing this, the male would change his choice. If his mate had just eaten a lot of wax moth larvae, he’d pick mealworms for her instead, and vice versa. He did so, however, only if he had witnessed her being fed by the experimenter. Male birds thus took into account what their mate had just eaten, perhaps assuming that she’d be ready for a change of taste.25 Jays, too, may attribute preferences to others, taking another’s point of view.

  At this point, you may wonder why perspective taking was ever declared uniquely human. For this, we need to look at a series of ingenious experiments in the 1990s in which chimpanzees could gain information about concealed food either from an experimenter who had witnessed the hiding process or from another one who had been put in the corner with a bucket over his head. Obviously, they should ignore the second experimenter, who had no idea, and follow the directions of the first. They made no distinction, however. Or an ape could beg for cookies from an experimenter sitting out of reach with a blindfold over his eyes. Would the chimps understand that there was no point stretching out an open hand to someone who cannot see them? After a great variety of such tests, the conclusion was that chimpanzees fail to understand the knowledge that others have and don’t even realize that knowing requires seeing. It was a most peculiar conclusion, given that the main researcher himself relates how playful apes put buckets or blankets over their heads and walk around until they bump into each other. When he himself put things over his head, however, he immediately became the target of play attacks by these apes, who exploited his obscured vision.26 They knew he couldn’t see them and tried to catch him by surprise.

  I knew a couple of juvenile male chimps who loved to throw rocks at us, practicing their impressive long-range aim, invariably doing so as soon as I moved my camera to my eye, which made me lose visual contact. Such behavior alone tells us that apes know something about the vision of others and that tests with blindfolds must therefore be missing something. But as happens so often among experimentalists, behavior in the testing room was given priority over real-life observations. As a result, human exceptionalism was loudly proclaimed, most dramatically by concluding that apes do not possess “anything remotely resembling a ToM.”27

  This conclusion was warmly greeted and is still being broadcast today even though it has not held up to scrutiny. At my home institution, the Yerkes Primate Center, David Leavens and Bill Hopkins conducted tests in which they placed a banana outside a chimpanzee enclosure where humans regularly walked by. Would the chimps draw attention to get people to hand them the fruit? Would they distinguish between people who could see them and those who could not? If so, this would suggest that they grasped another individual’s visual perspective. The chimps did, because they’d give visual signals to people who looked in their direction, but they’d vocalize and bang on metal if people failed to notice them. They even pointed at the banana to clarify their wishes. One chimpanzee, afraid to be misunderstood, pointed first with her hand at the banana and then with a finger at her own mouth.28

  Intentional signaling is not limited to captive apes, as became clear when scientists put a fake snake on the path of wild chimpanzees. Recording the apes’ alarm calls in a Ugandan forest, they found that calling is not just a reflection of fear, because the chimps vocalize regardless of whether the snake is near or far. It is rather a warning intended for others: they call more when others are present, especially friends who have failed to notice the serpent. Callers look back and forth between nearby chimps and the danger, calling more to companions who are naïve about it than to those who already know. Callers thus specifically inform those who lack knowledge, likely because they realize how knowing requires seeing.29

  A critical test of this connection was conducted by Brian Hare, then a student here at the Yerkes Primate Center. Brian wanted to know if apes exploit information about another’s visual input. A low-ranking individual was enticed to pick up food in front of a high-ranking one. This is a tricky thing to do, and most subordinates shy away from the confrontation. They were offered a choice between pieces of food that the dominant individual had seen being hidden and pieces hidden without him knowing. The subordinate, on the other hand, had watched it all. In an open competition, like an Easter egg hunt, the safest bet for the subordinate would be to pick only those food items that the dominant had no clue about. This is exactly what they did, showing that they understood that if the dominant had not seen the hiding process, he couldn’t know.30 Brian’s study threw the question of animal ToM wide open again. In an unexpected twist, one capuchin monkey at the University of Kyoto and several macaques at a Dutch research center recently passed similar tasks.31 This is why the whole notion that visual perspective taking is limited to our species is now in the trash bin. Each of the above experiments in and of itself may not be entirely watertight, but taken together they come down on the side of perspective-taking abilities in other species.

  It is a testimony to Menzel’s pioneering work that we keep hiding food or snakes, and pitting guessers against knowers. It remains the classical paradigm to assess these capacities both in humans and in other species. Perhaps most telling is an experiment by Menzel’s son Charles. Like his father, Charlie Menzel is a deep thinker, unsatisfied with easy tests or simple answers. At the Language Research Center here in Atlanta, he’d let a female chimp named Panzee watch while he hid food in the pine forest around her outdoor enclosure. Charlie would dig a small hole in the ground to put a bag of M&Ms into it, or place a candy bar in the bushes. Panzee would follow the process from behind bars. Since she could not go where Charlie was, she would need human help to eventually get the hidden food. Sometimes Charlie would hide it after all other people had gone for the day. This meant Panzee could not communicate with anybody about what she knew until the next morning. When the caretakers arrived, they were unaware of the experiment. Panzee first had to get their attention, then provide information to someone who had no clue as to what she was “talking” about.

  During a live demonstration of Panzee’s skills, Charlie told me that caretakers generally have a higher opinion of apes’ mental abilities than does the typical philosopher or psychologist. This high opinion was essential for his experiment, he explained, because it meant that Panzee was dealing with people who took her seriously. All those recruited by Panzee said they were at first surprised by her behavior but soon understood what she wanted them to do. By following her pointing, beckoning, panting, and calling, they had no trouble finding the candies hidden in the forest. Without her instructions, they would never have known where to look. Panzee never pointed in the wrong direction, or to locations that had been used on previous occasions. The result was communication about a past event, present in the ape’s memory, to ignorant members of a different species. If the human followed the instructions correctly and got closer to the food, Panzee would vigorously bob her head in affirmation (like “Yes! Yes!”), and like us, she’d lift her hand up, giving higher points, if the item was farther away. She realized that she knew something that the other didn’t know, and was intelligent enough to recruit humans as willing slaves to obtain the goodies of her desire.32

  Just to illustrate how creative chimps can be in this regard, here is a typical incident at our field station. A young female grunted at me from behind a fence and kept looking at me with shiny eyes (indicating that she knew something exciting) alternated with pointed stares into the grass near my feet. I couldn’t figure out what
she wanted, until she spat. From the trajectory, I noticed a small green grape. When I gave it to her, she ran to another spot and repeated her performance. Having memorized the locations of fruits dropped by the caretakers, she proved an accurate spitter, collecting three rewards this way.

  Clever Hans in Reverse

  So why did we at first reach the wrong conclusion about animal perspective taking, and why has it happened so many times before and since? Claims about absent capacities range from the idea that primates do not care about the welfare of others, do not imitate, or even fail to understand gravity. Imagine this for flightless animals that travel high above the ground! In my own career, I have faced resistance to the notion that primates reconcile after fights or console those who are distressed. Or at least I heard the counterclaim that they do not truly do so—as in, they do not “truly imitate” or “truly console”—which immediately gets one into debates about how to distinguish what looks like consolation or imitation from the real thing. At times, the overwhelming negativity got to me, as an entire literature burgeoned that was more excited about the cognitive deficits of other species than about their actual accomplishments.33 It would be like having a career adviser who all the time tells you that you are too dumb for this or too dumb for that. What a depressing attitude!

  The fundamental problem with all these denials is that it is impossible to prove a negative. This is no minor issue. When anyone claims the absence of a given capacity in other species, and speculates that it must therefore have arisen recently in our lineage, we hardly need to inspect the data to appreciate the shakiness of such a claim. All we can ever conclude with some certainty is that we have failed to find a given skill in the species that we have examined. We cannot go much further than this, and we certainly may not turn it into an affirmation of absence. Scientists do so all the time, though, whenever the human-animal comparison is at stake. The zeal to find out what sets us apart overrides all reasonable caution.