Are We Smart Enough to Know How Smart Animals Are

by Frans de Waal

  In short, elephants make sophisticated distinctions regarding potential enemies to the point that they classify our own species based on language, age, and gender. How they do so is not entirely clear, but studies like these are beginning to scratch the surface of one of the most enigmatic minds on the planet.

  The Magpie in the Mirror

  The ability to recognize oneself in the mirror is often viewed in absolute terms. According to Gallup, the pioneer of this field, a species either passes the mirror mark test and is self-aware, or it doesn’t and isn’t.9 Very few species do. For the longest time, only humans and the great apes passed, and not even all those. Gorillas used to flunk the mark test, leading to theories about why the poor things might have lost their self-awareness.10

  Evolutionary science, however, is uncomfortable with black-and-white distinctions. It is hard to imagine that among any set of related species, some are self-aware whereas others, for lack of a better term, remain unaware. Every animal needs to set its body apart from its surroundings and to have a sense of agency (awareness that it controls its own actions).11 You wouldn’t want to be a monkey up in a tree without awareness of how your own body will impact a lower branch on which you intend to land. And you wouldn’t want to engage in rough-and-tumble play with a fellow monkey, with all your combined arms, legs, and tails intertwined, while stupidly gnawing on your own foot or tail! Monkeys never make this mistake and gnaw exclusively on their partner’s foot or tail in such a tangle. They have a well-developed body ownership and self-other distinction.

  In fact, experiments on the sense of agency show that species without mirror self-recognition are very well capable of distinguishing their own actions from those produced by others. Tested in front of a computer screen, they have no trouble telling the difference between a cursor that they themselves control with a joystick and a cursor that moves by itself.12 Self-agency is part of every action that an animal—any animal—undertakes. In addition, some species may possess their own unusual kind of self-recognition, such as bats and dolphins that pick out the echoes of their own vocalizations from among the sounds made by others.

  Cognitive psychology doesn’t like absolute differences either, but for a different reason. The problem with the mirror test was that it introduced the wrong absolute difference. Instead of sharply dividing humans from and all other animals—which, as we have seen, is a staple of the field—Gallup’s mirror test moved the Rubicon slightly to annex a few more species. Lumping humans in with the apes so as to elevate the hominoids, as a group, to a different mental level than the rest of the animal kingdom, didn’t go over well. It diluted humanity’s special status. Still today, claims about self-awareness outside our own species cause consternation, and debates about mirror responses turn acrimonious. Moreover, many specialists have felt the need to conduct mirror tests on the animals in their care, usually with disappointing results. These debates have led me to the sarcastic conclusion that mirror self-recognition is considered a big deal only by scientists working on the handful of species capable of it, whereas all others poo-poo the phenomenon.

  Since I study animals that both do and don’t recognize themselves in the mirror, and have a high opinion of them all, I feel torn. I do think that spontaneous self-recognition means something. It may signal a stronger self-identity, such as is also reflected in perspective taking and targeted helping. These capacities are most marked in animals that pass the mirror test as well as in children who have reached the age, around two, when they do. This is also the age when they can’t stop referring to themselves, as in “Mama, look at me!” Their sharpened self-other distinction is said to help them adopt another’s viewpoint.13 Still, I can’t believe that a sense of self is absent either in other species or in younger children. Rather obviously, animals that fail to link their mirror image to their own body vary greatly in what they understand. Small songbirds and fighting fish, for example, never get over their mirror image and keep courting or attacking it. During the spring, when they are most territorial, tits and bluebirds will respond this way to the sideview mirror of a car and stop their hostilities only when the car drives off. This is absolutely not what monkeys do, nor many other animals. We would not be able to have mirrors in our homes if cats and dogs reacted the same way. These animals may not recognize themselves, but they are also not totally baffled by the mirror, at least not for long. They learn to ignore their reflection.

  Some species go further in that they understand mirror basics. Monkeys, for example, may not recognize themselves but are able to use the mirror as a tool. If you hide food that can be found only by using a mirror to look around a corner, the monkey will have no trouble reaching for it. Many a dog can do the same: holding up a cookie behind them while they watch you in a mirror makes them turn around. Curiously, it is specifically the relation with their own body, their own self in the mirror, that they fail to grasp. But even then, rhesus monkeys can be taught to do so. It requires adding a physical sensation. They need a mark that they can both see in the mirror and feel on their body, such as a laser light that irritates the skin or a cap fastened to their head. Instead of a traditional mark test, this is better described as a felt mark test. Only under these circumstances can monkeys learn to connect their reflection with their own body.14 This is obviously not the same as what apes do spontaneously relying on vision alone, but it does suggest that some of the underlying cognition is shared.

  Even though capuchin monkeys fail the visual mark test, we decided to study them in a way that, surprisingly, no one had ever tried before. Our goal was to see if these monkeys truly mistake their reflection for a “stranger,” as is commonly implied. Capuchins were placed in front of a Plexiglas panel, behind which they faced either a member of their own group, a stranger of their species, or a mirror. It quickly became evident that the mirror was special. They treated their reflection quite differently from a real monkey. They didn’t need any time to decide what they saw, and reacted within seconds. They turned their backs to strangers, barely glancing at them, yet made prolonged eye contact with their own reflections as if thrilled to see themselves. They showed absolutely none of the timidity toward the mirror image that one would expect if they mistook it for a stranger. Mothers, for example, let their infants freely play in front of the mirror yet held them close in case of a stranger. But the monkeys also never inspected themselves in the mirror, the way apes do all the time, or the way Pepsi the elephant had done. They never opened their mouth to peek inside. Thus, while capuchins fail to recognize themselves, they also don’t mix up their reflection with someone else.

  As a result, I have become a gradualist.15 There are many stages of mirror understanding, running all the way from utter confusion to a full appreciation of the specular image. These stages are also recognizable in human infants, which are curious about their mirror image well before passing the mark test. Self-awareness develops like an onion, building layer upon layer, rather than appearing out of the blue at a given age.16 For this reason, we should stop looking at the mark test as the litmus test of self-awareness. It is only one of many ways to find out about the conscious self.

  Nevertheless, it remains fascinating how few species pass this test without a helping hand. After the hominoids, spontaneous self-recognition was observed only in elephants and dolphins. When bottlenose dolphins at the New York Aquarium were marked by Diana Reiss and Lori Marino with painted dots, they spent more time in front of a mirror than if they had been left unmarked. They would race from the spot where the marking took place to a mirror in another pool, at quite a distance, only to spin around seemingly to get a good look at themselves. The dolphins spent more time near the mirror checking out their bodies when they had been marked than without a visible mark.17

  It was unavoidable that the mirror test would be tried on birds. While most species have thus far failed, we have one exception: the Eurasian magpie. It is an interesting species to put in front of a reflective surface. As a child, I learn
ed never to leave small shiny objects, such as teaspoons, unattended outdoors as these raucous birds will steal anything they can put their beaks on. This folklore inspired a Rossini opera, La gazza ladra (The Thieving Magpie). Nowadays, this view has been replaced with a more ecologically sensitive one that depicts magpies as murderous robbers of the nests of innocent songbirds. Either way, they are considered black-and-white gangsters.

  But no one has ever accused a magpie of being stupid. The bird belongs to the corvid family that has begun to challenge the cognitive supremacy of primates. The German psychologist Helmut Prior subjected magpies to a mirror test that was at least as well controlled as any conducted on apes and children. Placed on their black bib (throat feathers), the mark—a tiny yellow sticker—stood out but was visible only with help of the mirror. The birds were untrained, which is a critical difference with the highly coached pigeons employed long ago to discredit mirror research. Put in front of a mirror, the magpies kept scratching with their foot until the mark was gone. They never did the same amount of frantic scratching if there was no mirror to see themselves in, and they ignored a “sham” mark—a black sticker on their black bib. As a result, the self-recognition elite has now been expanded with its first feathered member. Others may follow.18

  Suma, an orangutan at a German zoo, loved to decorate herself in front of the mirror. Here she puts a leaf of lettuce onto her head like a hat.

  The next frontier will be to see if animals care about their mirror image to the point of embellishing themselves, the way we do with makeup, hair care, earrings, and the like. Does the mirror induce vanity? Would any species other than ours be prone to take selfies, if they could? This possibility was first hinted at by observations in the 1970s of a female orangutan at the Osnabrück Zoo, in Germany. Jürgen Lethmate and Gerti Dücker described Suma’s narcissistic ways:

  She gathered salad and cabbage leaves, shook each leaf and piled them up. Eventually, she placed one leaf on her head and walked straight to the mirror with it. She sat down directly in front of it, contemplated her headcover in the mirror, straightened it a bit with her hand, squashed it with a fist, then put the leaf on her forehead and began to bob up and down. Later, Suma arrived holding a salad leaf in her hand at the bars [where the mirror stood] to lay it on her head once she could see herself in the mirror.19

  The Mollusk Mind

  As a biology student, my favorite textbook was Animals Without Backbones. It may seem an odd choice given my current interests, but I was awestruck by all the exotic life-forms that I had never heard about or could scarcely imagine, some of which so tiny that you needed a microscope to see them. The book went into great detail of all invertebrates—from protozoans and sponges to worms, mollusks, and insects—which together make up 97 percent of the animal kingdom.20 Whereas cognition research focuses almost entirely on the tiny vertebrate minority, it is not as if the rest doesn’t move, eat, mate, fight, and cooperate. Obviously, some invertebrates show more complex behavior than others, but they all need to pay attention to their surroundings and solve problems that present themselves. In the same way that almost all these animals have reproductive organs and digestive tracts, they can’t survive without a degree of cognition.

  The brainiest of the bunch is the octopus, which is a soft-bodied cephalopod, or “head-footed” animal. This is an apt name, since their squishy bodies consist of a head that directly joins eight limbs, while the body (the mantle) is positioned behind the head. The cephalopods are an ancient class that arose well before there were land vertebrates around, but the group to which the octopus belongs is a fairly modern offshoot. We seem to have almost nothing in common with them, both anatomically and mentally. Yet they have been reported to open a pill bottle protected by a childproof cap. Since this requires the cap to be pushed down and twisted at the same time, it takes skill, intelligence, and persistence. Some public aquariums show off octopus intelligence by locking the animal in a glass jar that they close with a screw top. Like a true Houdini, the octopus takes less than a minute to grab the cover from within with its suckers and unscrew it so as to escape.

  The octopus has a most remarkable nervous system that allows it to solve challenging problems, such as how to escape from a glass jar closed with a screw top.

  However, when octopuses were given a transparent jar that contained a live crayfish, they failed to do anything. This greatly puzzled the scientists, because the delicacy was clearly visible and moving about. Do octopuses perhaps have trouble unscrewing a lid from the outside? It turned out to be one of those human misjudgments. Despite having excellent eyes, octopuses rarely rely on vision to catch prey. They use mainly touch and chemical information and fail to recognize prey without those cues. As soon as the jar was smeared on the outside with herring “slime,” making it taste like fish, the octopus swung into action and started manipulating it until the top came off. It quickly removed the crayfish and ate it. With further skill development, the process became routine.21

  In captivity, octopuses react to us in ways that we find hard not to anthropomorphize. One octopus was fond of raw chicken eggs—each day it would accept an egg and break it to suck out its contents. One day, however, this octopus accidentally received a rotten egg. Upon noticing, it shot the egg’s smelly remains over the edge of its tank back at the surprised human from whom it had received it.22 Given how well they distinguish people, octopuses probably remember encounters like these. In a recognition test, an octopus was exposed to two different persons, one of whom consistently fed it, whereas the other mildly poked it with a bristle on a stick. Initially, the animal made no distinction, but after several days it began doing so despite the fact that both humans wore identical blue overalls. Seeing the loathsome person, the octopus would withdraw, emit jets of water with its funnel, and show a dark bar through its eyes—a color change associated with threat and irritation. It would approach the nice person, on the other hand, without making any attempt at drenching her.23

  The octopus brain is the largest and most complex of all invertebrates, but the explanation of its extraordinary skills may lie elsewhere. These animals literally think outside the box. Each octopus has nearly two thousand suckers, every single one equipped with its own ganglion with half a million neurons. That amounts to a lot of neurons on top of a 65-million-neuron brain. In addition, it has a chain of ganglia along its arms. The brain connects with all these “mini brains,” which are also joined among themselves. Instead of a single central command, as in our species, the cephalopod nervous system is more like the Internet: there is extensive local control. A severed arm may crawl on its own and even pick up food. Similarly, a shrimp or small crab can be handed from one sucker to the next, as if on a conveyer belt, in the direction of the octopus’s mouth. When these animals change skin color in self-defense, the decision may come from central command, but perhaps the skin is involved as well, since cephalopod skin may detect light. It sounds rather unbelievable: an organism with seeing skin and eight independently thinking arms!24

  This realization has led to a bit of hype: that the octopus is the most intelligent organism in the ocean, a sentient being that we should stop eating. We shouldn’t overlook dolphins and orcas, though, which have vastly larger brains. Even if the octopus stands out among invertebrates, its tool use is rather limited, and its reaction to a mirror is as perplexed as that of a small songbird. It remains unclear whether an octopus is smarter than most fish, but let me hasten to add that such comparisons barely make any sense. Instead of turning the study of cognition into a contest, we should avoid putting apples next to oranges. The octopus’s senses and anatomy, including its decentralized nervous system, make it unparalleled.

  If superlatives of uniqueness were allowed, the octopus might be the most unique species of them all. They defy comparison with any other group, unlike our own species, which derives from a long line of land vertebrates with structurally similar body plans and brains.

  Octopuses have an odd life cyc
le. Most live only one or two years, which is unusual for an animal with their brainpower. They grow fast while trying to stay away from predators until they have a chance to mate and reproduce, after which they die. They stop eating, lose weight, and go into senescence.25 This is the stage about which Aristotle observed: “after giving birth . . . [they] become stupid, and are not aware of being tossed about in the water, but it is easy to dive and catch them by hand.”26

  These short-lived loners have no social organization to speak of. Given their biology, they have no reason to pay attention to one another, except as rivals, mates, predators, and prey. They are certainly not friends or partners. There is no evidence that they learn from others or spread behavioral traditions, the way many vertebrates, including fish, do. The absence of social bonds and cooperation, and their cannibalistic ways, make cephalopods quite alien to us.

  Their main worry is predation, because apart from their own kind, they are eaten by almost everything around, from marine mammals, diving birds, sharks, and other fish to humans. When they get larger, they become formidable predators themselves, as the Seattle Aquarium accidentally found out. Worried about their giant Pacific octopus in a tank full of sharks, staff were hoping that the animal would know how to hide. But then they noticed one dogfish (a small shark) after another disappearing from the tank—and found to their astonishment that the octopus had turned the tables. The octopus may be the only playful invertebrate. I say may since play behavior is almost impossible to define, but the octopus appears to go beyond mere manipulation and checking out of novel objects. The Canadian biologist Jennifer Mather found that given a new toy, the animal will move from exploration (“What is this?”) to repeated lively movements and tossing around (“What can I do with it?”). With their funnel, they blow jets of water at a floating plastic bottle, for example, to move it from one side of their tank to another, or to have it tossed back at them by the water flow of the filter, which makes them look as if they were bouncing a ball. Such manipulations, which serve no obvious purpose and are repeated over and over, have been taken as indications of play.27