Are We Smart Enough to Know How Smart Animals Are

by Frans de Waal

  The case for episodic memory in animals was further strengthened when the American psychologists Stephanie Babb and Jonathon Crystal let rats run around in an eight-armed radial maze. The rodents learned that once they had visited an arm and eaten the food in it, it would be permanently gone, so there would be no point returning to it. There was one exception, though. They occasionally found chocolate-flavored pellets, which would be replenished after long time intervals. The rats formed an expectation about this delicious food based on where and when they had encountered it. They did return to those specific arms, but only after long intervals. In other words, the rodents kept track of the when, what, and where of chocolate surprises.8

  Tulving and a few other scholars were hardly satisfied with these results, however. They fail to tell us—the way Proust did so eloquently—how aware the birds, rats or apes are of their own memories. What kind of consciousness, if any, is involved? Do they view their past as a piece of personal history? Since such questions are unanswerable, some have weakened the terminology by endowing animals only with “episodic-like” memory. I don’t agree with this retreat, however, since it gives weight to an ill-defined aspect of human memory known only through introspection and language. While language is helpful to communicate memories, it is hardly what produces them. My preference would be to turn the burden of proof around, especially when it comes to species close to us. If other primates recall events with equal precision as humans do, the most economic assumption is that they do so in the same way. Those who insist that human memory rests on unique levels of awareness have their work cut out for them to substantiate such a claim.

  It may, literally, be all in our heads.

  The Cat’s Umbrella

  The debate about how animals experience the time dimension heated up even further in relation to the future. Who’d ever heard of them contemplating events that lay ahead? Tulving drew on what he knew about Cashew, his cat. Cashew seems capable of predicting rain, he said, and is good at finding places to take cover, yet “never thinks ahead and packs an umbrella.”9 Generalizing this astute observation to the entire animal kingdom, the eminent scientist explained that while animals adapt to their present environment, they sadly fail to imagine the future.

  Another human uniqueness proponent noted that “there is no obvious evidence that animals have ever agreed on a five-year plan.”10 True, but how many humans have? I associate five-year plans with central government and prefer examples drawn from the way both humans and animals go about their daily business. For example, I may plan to buy groceries on my way home, or decide to surprise my students with a quiz next week. This is the nature of our planning. It is not unlike the story with which I opened this book regarding Franje, the chimpanzee who gathered all the straw from her night cage to build a warm nest outdoors. That she took this precaution while still indoors, before actually feeling the cold outside, is significant because it fits Tulving’s so-called spoon test. In an Estonian children’s story, a girl dreams of a friend’s chocolate pudding party where she can only watch other children eat, because everyone has brought their own spoon, and she has not. To prevent this from happening again, she goes to bed that night clutching a spoon. Tulving proposed two criteria to recognize future planning. First, the behavior should not follow directly from present needs and desires. Second, it should prepare the individual for a future situation in a different context than the current one. The girl needed a spoon not in bed, but at the chocolate pudding party she expected in her dream.11

  When Tulving came up with the spoon test, he wondered if it was perhaps unfair. Wasn’t it too demanding for animals? He proposed this test in 2005, well before most experiments on future planning were conducted, apparently unaware that apes pass the spoon test every day in their spontaneous behavior. Franje did so when she gathered straw in a different location and under different circumstances than where it was needed. At the Yerkes Primate Center, we also have a male chimp, Steward, who never enters our testing room without first looking around outdoors for a stick or branch that he uses to point at the various items in our experiments. Even though we have tried to discourage this behavior, by removing the stick from his hands so that he’ll point with a finger like everyone else, Steward is stubborn. He prefers to point with a stick and will go out of his way to bring one with him, thus anticipating our test and his self-invented need for a tool.

  But perhaps the nicest illustration, out of dozens I could offer, is a bonobo named Lisala, who lives at Lola ya Bonobo, a jungle sanctuary near Kinshasa where we conducted studies of empathy. The observation in question was unrelated to this topic, however, and was made by my coworker Zanna Clay when she unexpectedly saw Lisala pick up an enormous fifteen-pound rock and lift it onto her back. Lisala carried this heavy load on her shoulders while her baby clung to her lower back. It was rather ridiculous, of course, since it impeded her travel and required extra energy. Zanna turned on her video camera and followed the bonobo to see what the rock might be for. Like any true ape expert, she immediately assumed that Lisala had a goal in mind, because, as Köhler had noted, ape behavior is “unwaveringly purposeful.” The same holds for human behavior. If we see a man walking in the street with a ladder, we automatically assume that he wouldn’t be carrying such a heavy tool for no reason.

  Lisala, a bonobo, carries a heavy rock on a long trek toward a place where she knows there are nuts. After collecting the nuts, she continues her trek to the only large slab of rock in the area, where she employs her rock as a hammer to crack the nuts. Picking up a tool so long in advance suggests planning.

  Zanna filmed Lisala’s trek of about half a kilometer. It was interrupted only once when she put down the rock and picked up some items that were hard to identify. Then she put the rock back onto her back and continued her travels. She walked all told almost ten minutes before she reached her destination, which was a large slab of hard rock. She cleared it of debris with a few swipes of her hand, then put down her rock, her infant, and the collected items, which turned out to be a handful of palm nuts. She set out to crack these extremely tough nuts, placing them on the large anvil while banging them with her fifteen-pound rock as a hammer. She spent about fifteen minutes on this activity, then left her tool behind. It is hard to imagine that Lisala had gone through all this trouble without a plan, which she must have had well before she picked up the nuts. She probably knew where to find those, hence planned her route via this location, to end up at a point that she knew had a hard enough surface for successful cracking. In a nutshell, Lisala fulfilled all of Tulving’s criteria. She picked up a tool to be used at a distant location for the processing of food that she could only have imagined.

  Another remarkable instance of future-oriented behavior was documented at a zoo by the Swedish biologist Mathias Osvath, this time involving a male chimpanzee, Santino. Every morning before visitors arrived, Santino would leisurely collect rocks from the moat surrounding his enclosure, stacking them up in neat little piles hidden from view. This way he’d have an arsenal of weapons when the zoo opened its gates. Like so many male chimps, Santino would several times a day rush around with all his hair on end to impress the colony and the public. Throwing stuff around was part of the show, including projectiles aimed at the watching masses. Whereas most chimps find themselves empty-handed at the critical moment, Santino prepared his rock piles for these occasions. He did so at a quiet time of the day, when he was not yet in the adrenaline-filled mood to produce his usual spectacle.12

  Such cases deserve attention since they show that apes do not have to be prompted by experimental conditions concocted by us humans to plan for the future. They do so of their own accord. Their accomplishments are quite different from the way many other animals orient to upcoming events. We all know that squirrels collect nuts in the fall and hide them for retrieval in winter and spring. Their hoarding is triggered by the shortening of day length and the presence of nuts, regardless of whether the animals know what winter is. You
ng squirrels naïve about the seasons do exactly the same. Whereas this activity does serve future needs and requires quite a bit of cognition regarding what nuts to store and how to find them again, the seasonal preparations of squirrels are unlikely to reflect actual planning.13 It is an evolved tendency found in all members of their species and limited to only one context.

  The planning of apes, in contrast, adjusts to the circumstances and is flexibly expressed in myriad ways. That it is based on learning and understanding is hard to prove from observation alone, however. It requires subjecting apes to conditions that they have never met before. What happens, for example, if we create a situation in which clutching a spoon, so to speak, is advantageous later on?

  The first such study was conducted in Germany by Nicholas Mulcahy and Josep Call, who let orangutans and bonobos select a tool that they couldn’t use right away even though the rewards were visible. The apes were moved away to a waiting room to see if they would hold on to their tool for later use even if the right occasion would arise only fourteen hours later. The apes did so, yet it could be (and has been) argued that they might have developed positive associations with certain tools, hence valued them regardless of what they knew about the future.14

  This issue was addressed by a similar experiment in which apes selected tools, but this time the rewards were kept out of sight. The apes preferred a tool they could use in the future over a grape placed right next to it. They suppressed their desire for an immediate benefit to gamble on a future one. Once they had the right tool in hand, however, and got a second presentation of the same set of tools, they did pick the grape. Clearly, they didn’t value the tool over anything else, because if they did, their second choice should have replicated the first. The apes must have realized that once they had the right tool in hand, there was no point having a second one of the same kind, and that the grape was a better choice.15

  These clever experiments were foreshadowed by Tulving’s proposal as well as by Köhler, who was the first to speculate about future planning in animals. There is now even a test in which, instead of presenting apes with actual tools, they are given an opportunity to fabricate them in advance. Apes learned to break a board of soft wood into smaller pieces to produce sticks with which they could reach grapes. Anticipating the need for sticks, they worked hard on having them ready in time.16 Their preparations resembled the behavior of wild apes, which travel long distances with raw materials that they turn on the spot into tools by modifying, sharpening, or fraying them. They sometimes bring more than one type of tool to a task in the forest. Chimps carry toolkits of up to five different sticks and twigs to hunt for underground ants or raid bee nests for honey. It is hard to imagine an ape searching for and traveling with multiple instruments without a plan. Just so, Lisala picked up a heavy rock that by itself was useless and that could serve its purpose only in combination with nuts that she had yet to collect as well as a hard surface located far away. Attempts to explain this kind of behavior without foresight invariably sound cumbersome and far-fetched.

  The question now is whether similar evidence can be produced without reliance on tools such as spoons, umbrellas, or sticks. What if we consider a wider spectrum of behavior? How this might be done was again demonstrated by Clayton’s scrub jays. These birds routinely cache food, and although some scientists complain that this behavior offers a rather narrow window on cognition, it is a window nonetheless and one that differs radically from the one used for primates. It exploits an activity that corvids are particularly good at, just as tool studies exploit specialized primate skills. The outcome has been most remarkable.

  Caroline Raby offered jays an opportunity to store food in two compartments of their cage that would be closed off during the night. The next morning they would get a chance to visit only one of the two compartments. One compartment had become associated with hunger, since the birds had spent mornings there without breakfast. The second compartment, on the other hand, was known as the “breakfast room” because it was stocked with food every morning. Given a chance in the evening to cache pine nuts, the birds put three times as many nuts in the first room as in the second, thus anticipating the hunger they might suffer there. In another experiment, the birds had learned to associate both compartments with different kinds of food. Once they knew what kind to expect, they tended to store a different food in each compartment in the evening. This guaranteed a more varied breakfast if they ended up in one of those compartments next morning. All in all, when scrub jays stash away food, they do not seem guided by their present needs and desires but rather by the ones they anticipate in the future.17

  In thinking of primate examples without tools, the ones that come to mind are social situations in which it helps to be diplomatic. For example, chimpanzees sometimes arrange a secret rendezvous with the opposite sex. Bonobos don’t need to do so, since others rarely interrupt their sexual escapades, but chimpanzees are far less tolerant. High-ranking males don’t allow rivals near females with an attractive genital swelling. Nevertheless, the alpha male cannot always be awake and alert, hence occasions do arise for young males to invite a female to get away to a quiet spot. Typically, the young male spreads his legs to show his erection—a sexual invitation—making sure that his back is turned to the other males or that, with his underarm leaning on his knee, one of his hands loosely dangles right next to his penis so that only the wooed female can see it. After this display, the male nonchalantly wanders off in a given direction and sits down out of view of dominant males. Now it is up to the female, who may or may not follow. So as to give nothing away, she usually takes off in a different direction, only to arrive, via a detour, at the same spot as the young male. What a coincidence! The two of them then engage in a quick copulation, making sure to stay silent. It all gives the impression of a well-planned arrangement.

  Even more striking are the tactics of adult males challenging each other for status. Given that confrontations are almost never decided between two rivals on their own but involve support for one or the other by third parties, it is to their advantage to influence public opinion beforehand. The males commonly groom high-ranking females or one of their male buddies before launching into a display, with all their hair on end, to provoke a rival. The grooming gives the impression of them currying favors in advance, knowing full well what the next step will be. In fact, there has been a systematic study on this issue. At Chester Zoo in the United Kingdom, Nicola Koyama recorded for over two thousand hours who groomed whom in a large chimpanzee colony. She also noted what kinds of conflicts arose among the males, and who allied with whom. When she compared records on both behaviors—grooming and alliances—from one day to the next, she discovered that males received more support from the individuals they had groomed the day before. This is the sort of tit-for-tat that we are used to in chimpanzees. But since this connection held only for the aggressors, and not for their victims, the explanation was not simply that grooming promotes support. Koyama viewed the connection as part of an active strategy. Males know beforehand which confrontations they are going to incite, and they pave the way for them by grooming their friends a day in advance. This way they make sure to have their backing.18 It reminds me of the politics at university departments, where colleagues come to my office in the days leading up to an important faculty meeting to influence my vote.

  Observations are suggestive yet rarely conclusive. They do, however, give an idea under what circumstances future planning might be useful. If naturalistic observations and experiments point in the same direction, we must be on the right track. For example, a recent study suggested that wild orangutans communicate future travel routes. Orangutans are such loners that their encounters in the canopy have been described as ships passing in the night. They often travel on their own, accompanied only by their dependent offspring, and remain visually isolated for long stretches of time. Auditory information about one another’s whereabouts is often all they have.

  Carel van Schaik—
a Dutch primatologist who once was a fellow student of mine and whose field site on Sumatra I visited—followed wild males right before they went to bed in their self-made nests high up in the trees. He recorded over a thousand whooping calls made by these males before nightfall. These loud calls may last for up to four minutes, and all orangs around pay close attention, because the dominant male (the only fully grown male with well-developed cheek pads, or flanges) is a figure to be reckoned with. There is usually only one such male in a given area of the forest.

  Carel found that the direction in which adult males call before going to sleep predicts their travel path the next day. The calls contain this information even if the direction changes from day to day. Females adjust their own routes to the male’s, such that sexually receptive females may approach him, and other females know where to find him in case they are being harassed by adolescent males. (Female orangutans generally prefer the dominant male.) Although Carel recognizes the limitations of a field study, his data imply that orangutans know where they will be going and vocally announce their plan at least twelve hours before its execution.19