Mind Reading
There are three classes of intellects: one which comprehends by itself; another which appreciates what others comprehend; and a third which neither comprehends by itself nor by the showing of others; the first is the most excellent, the second is good, and the third is useless.
—Niccolò Machiavelli, The Prince
Many people believe that thoughts can be transferred from one person to another by means other than the senses. This is known as telepathy. In 1882, the Society for Psychical Research was established in London to investigate telepathy and other so-called psychic phenomena, such as ghosts, trance states, levitations, mediums, and communication with the dead. Its first president was Henry Sidgwick, later professor of moral philosophy at Trinity College, Cambridge, and other distinguished members included the experimental physicist Lord Rayleigh, the philosopher Arthur Balfour, who became prime minister of England from 1902 to 1905, and Sir Arthur Conan Doyle, author of the Sherlock Holmes stories. The Society attracted the interest of famous psychologists such as Sigmund Freud and Carl Jung, and the pioneering American psychologist William James was so impressed that he established the American Society for Psychical Research shortly afterwards.
These societies, and many others dedicated to psychical research, remain active to this day. Laboratories for the study of psychic phenomena emerged in a number of universities, with Stanford University leading the way in 1911. Under the guidance of the famous psychologist William McDougall, Duke University set up an influential laboratory in 1930, and the publication of Joseph B. Rhine’s book New Frontiers of the Mind in 1937 brought the laboratory’s findings to the attention of the general public. Rhine and McDougall coined the term “parapsychology” to refer to psychic phenomenon, and the Journal of Parapsychology, which flourishes still, was established in 1937. In 1983, the noted author Arthur Koestler and his wife Cynthia provided in their wills for the establishment of a chair in parapsychology at a British university. The University of Edinburgh took advantage of the opportunity, and in 1984 the chair was duly established there, with Robert Morris as its first incumbent. Morris died in 2005, and at the time of writing has not been replaced, although the Koestler Parapsychology Unit remains active.
As this brief introduction illustrates, parapsychology has been associated with some distinguished individuals and reputable universities. The tenuous nature of claims of psychic phenomena has also meant that fraudsters have been quick to cash in. Uri Geller, an Israeli-British stage performer, rose to fame in the 1970s for television shows on which he claimed to demonstrate psychic powers. He is perhaps best known for his prowess at bending spoons, apparently through the power of thought—a phenomenon that, if true, is an example of psychokinesis. Geller’s demonstrations are easily duplicated without resort to psychic powers by stage magicians, including James Randi, who wrote a book entitled The Magic of Uri Geller—it was later called The Truth about Geller.1 Geller’s exploits were also unmasked, leaving no spoon unbent, in New Zealand by two psychologists, David Marks and Richard Kammann, who were able to repeat Geller’s demonstrations on television, again without any claim to psychic powers. They too wrote a book exposing the field of psychic phenomena, and Geller in particular, entitled The Psychology of the Psychic.2 These books are highly recommended, but alas do not have the selling power of books that proclaim the existence of the psychic.
Just why people should be so ready to believe in the power of the mind to transcend physical laws is unclear. Perhaps it is a kind of wish fulfillment, allowing us to believe that we can influence people at a distance or communicate with loved ones who have died. Perhaps it is an offshoot of mind-body dualism, attributed to the seventeenth-century French philosopher René Descartes, who proposed that the human mind was not governed solely by physical laws. Most religions depend on the idea of a spirit or god that goes beyond the merely physical. Some devotees have tried to give scientific respectability to psychic forces through labeling; parapsychologists have called it psi (pronounced “sigh”), which to more mainstream scientists, albeit old-fashioned ones, means “pounds per square inch,” while a more ambitious term, coined by the biologist Rupert Sheldrake, is morphic resonance. Sheldrake is the fellow we met in chapter 3 who recently claimed that dogs can use morphic resonance to sense when their owners are coming home unexpectedly.3
Theory of Mind
Part of the reason that the belief in psychic phenomena persists, though, is that people are actually very good at reading minds. We do this through means that are entirely naturalistic, and there is no good reason to invoke supernatural powers or nonphysical channels of communication. The ability to understand, or at least surmise, what is happening in the minds of others is known as theory of mind. It is recursive, in the sense that it involves the insertion of what you believe to be someone else’s state of mind into your own. Let’s consider, then, some of the naturalistic ways in which we can do this.
Emotion is perhaps the simplest mental state to read. We can readily tell whether another person is happy, angry, sad, or in pain, through their facial expressions, bodily postures, or vocalizations. In his book The Expression of the Emotions in Man and Animals, Charles Darwin vividly described outward signs of emotions, and our ability to read the emotions of others is by no means confined to our own species. Darwin was equivocal, though, about whether this was learned or instinctive; as is usually the case, he is worth quoting in detail:
As most of the movements of expression must have been gradually acquired, afterwards becoming instinctive, there seems to be some degree of a priori probability that their recognition would likewise have become instinctive. There is, at least, no greater difficulty in believing this than in admitting that, when a female quadruped first bears young, she knows the cry of distress of her offspring, or than in admitting that many animals instinctively recognize and fear their enemies; and of both these statements there can be no reasonable doubt. It is however extremely difficult to prove that our children instinctively recognize any expression. I attended to this point in my first-born infant, who could not have learnt anything by associating with other children, and I was convinced that he understood a smile and received pleasure from seeing one, answering it by another, at much too early an age to have learnt anything by experience. When this child was about four months old, I made in his presence many odd noises and strange grimaces, and tried to look savage; but the noises, if not too loud, as well as the grimaces, were all taken as good jokes; and I attributed this at the time to their being preceded or accompanied by smiles. When five months old, he seemed to understand a compassionate expression and tone of voice. When a few days over six months old, his nurse pretended to cry, and I saw that his face instantly assumed a melancholy expression, with the corners of the mouth strongly depressed; now this child could rarely have seen any other child crying, and never a grown-up person crying, and I should doubt whether at so early an age he could have reasoned on the subject. Therefore it seems to me that an innate feeling must have told him that the pretended crying of his nurse expressed grief; and this through the instinct of sympathy excited grief in him.4
Whether instinctive or learned, the human ability to infer the mental states of others goes well beyond the detection of emotion. To take another simple and seemingly obvious example, we can understand what another individual can see. This is again an example of recursion, since we can insert that individual’s experience into our own. It is by no means a trivial feat, since it requires the mental rotation and transformation of visual scenes to match what the other person can see, and the construction of visual scenes that are not immediately visible. For example, if you are talking to someone face-to-face, you know that she can see what is behind you, though you can’t. Someone standing in a different location necessarily sees the world from a different angle, and to understand that person’s view requires an act of mental rotation and translation.
To test this ability in children, the Swiss psychologist Jean Piaget developed what is known as the Three Mountains Test, in which the children looked at an arrangement of three model mountains from a particular location in the room. They were then shown photographs of the scene taken from different locations, and asked to select which one showed how the scene would look from some particular location. Piaget found that up to the age of nine or ten, children were unable to solve the problem when the photograph represented the view from a location other than their own.5 This particular task seems to have been unusually difficult, though, and children as young as three or four have been shown to be able to solve a comparable task in which the mountains are replaced by more familiar objects.6
More complex still is the capacity to infer what other people believe, often on the basis of observation and reasoning. This is nicely illustrated by the Sally-Anne test, which is a test of children’s ability to infer false beliefs. The child is shown a scene involving two dolls, one called Sally and one called Anne. Sally has a basket and Anne has a box. Sally then puts a marble in her basket and leaves the scene. While Sally is away Anne takes the marble out of the basket and puts it in her box. Sally then comes back, and the child is asked where she will look for her marble. Children under the age of four typically say she will look in the box, which is where the marble actually is. Older children will understand that Sally did not see the marble being shifted, and will correctly say that Sally will look in the basket. They understand that Sally has a false belief.
The Sally-Anne test is something of a classic in developmental psychology, but may overestimate the age at which children develop theory of mind. The test requires not only that the child remember a series of events, but also that the child understand the question put to her. A recent study suggests that babies may actually understand false belief by the age of two. Children aged 25 months were shown a movie of an actor placing a ball in a box. The actor then looks away and the ball is removed. When the actor returned, 17 of the 20 infants looked at the box where the ball had been placed, evidently expecting the actor to mistakenly look there for the ball. These young infants did seem to understand that the actor would have a false belief.7
Theory of mind evolved because we live complex social lives, where we deal as much with people as with objects—or we did, until computers appeared on our desks. Survival during the Pleistocene, when our forebears competed with dangerous carnivores on the African savanna, required cooperation and social intelligence. The story of the Pleistocene in shaping the human mind will be told in more detail in chapter 11, but suffice to say that reproductive success in humans is driven more by social success than by physical attributes.8
There may be a dark side to social intelligence, though, since some unscrupulous individuals may take advantage of the cooperative efforts of others, without themselves contributing. These individuals are known as freeloaders. In order to counteract their behavior, we have evolved ways of detecting them. Evolutionary psychologists refer to a “cheater-detection module” in the brain that enables us to detect these imposters, but they in turn have developed more sophisticated techniques to escape detection.9 This recursive sequence of cheater detection and cheater-detection detection has led to what has been called a “cognitive arms race,” perhaps first identified by the British evolutionary theorist Robert Trivers,10 and later amplified by evolutionary psychologists.11 The ability to take advantage of others through such recursive thinking has been termed Machiavellian intelligence,12 whereby we use social strategies not merely to cooperate with our fellows, but also to outwit and deceive them. In the sixteenth century, in his celebrated work The Prince, Niccolò Machiavelli gave sage advice:
[It] is useful, for example, to appear merciful, trustworthy, blameless, religious—and to be so—yet to be in such measure prepared in mind that if you need to be not so, you can and do change to the contrary.
The philosopher Daniel Dennett referred to mind reading as the intentional stance, which means that we tend to treat people as having intentional states.13 The notion of intentional state is here used rather broadly, and not just the intention to act in a particular way. It includes other subjective states such as beliefs, desires, thoughts, hopes, fears, and so forth. According to the intentional stance, we interact with people according to what we think is going on in their minds, rather than in terms of their physical attributes—although there is a bit of that too, as I recall from my early days on the rugby field. When you meet a stranger on a dark night, your behavior may be guided partly by the intentional stance, based perhaps on facial expression, but perhaps also by what might be termed the physical stance, based on just how big a hunk the stranger is.
From the point of view of this book, the important aspect of theory of mind is that it is recursive. This is captured by the different orders of intentionality proposed by Dennett. Zero-order intentionality refers to actions or behaviors that imply no subjective state, as in reflex or automatic acts. First-order intentionality involves a single subjective term, as in Alice wants Fred to go away. Second-order intentionality would involve two such terms, as in Ted thinks Alice wants Fred to go away. It is at this level that theory of mind begins. And so to third order: Alice believes that Fred thinks she wants him to go away. Recursion kicks in once we get beyond first order, and our social life is replete with such examples. There seems some reason to believe, though, that we lose track at about the fifth or sixth order,14 perhaps because of limited working-memory capacity rather than any intrinsic limit on recursion itself. We can perhaps just wrap our minds around propositions like: Ted suspects that Alice believes that he does indeed suspect that Fred thinks that she wants him (Fred) to go away. That’s fifth order, as you can tell by counting the words in bold type. You could make it sixth order by adding George imagines that … at the beginning.
According to Robin Dunbar, it is through theory of mind that people may have come to know God, as it were. The notion of a God who is kind, who watches over us, who punishes, who admits us to Heaven if we are suitably virtuous, depends on the understanding that other beings—in this case a supposedly supernatural one—can have human-like thoughts and emotions. Indeed Dunbar argues that several orders of intentionality may be required, since religion is a social activity, dependent on shared beliefs. The recursive loops that are necessary run something like this: I suppose that you think that I believe there are gods who intend to influence our futures because they understand our desires.15 This is fifth-order intentionality.16 Dunbar himself must have achieved sixth-order intentionality if he supposes all this, and if you suppose that he does then you have achieved seventh-order. Call it Seventh Heaven, if you like.
If God depends on theory of mind, so too, perhaps, does the concept of self. This returns us to the opening paragraph of this book, and Descartes’s famous syllogism “I think, therefore I am.” Since he was appealing to his own thought about thinking, this is secondorder intentionality. Of course we also understand the self to continue through time, which requires the (recursive) understanding that our consciousness also transcends the present.
People and Things
A few individuals appear to lack the ability to read minds, and Simon Baron-Cohen has argued that this deficit, which he calls mindblindness, underlies the condition known as autism.17 People with this condition may be intelligent in other respects, but are peculiarly unresponsive to other people. One remarkable case is a woman called Temple Grandin, who has a Ph.D. in agricultural science and works as a teacher and researcher at Colorado State University. Clearly intelligent, she has written several books, three of which describe her own condition and the manner in which she has dealt with it.18 Her plight was also vividly described by Oliver Sacks in his book An Anthropologist on Mars.19 She has had to laboriously teach herself how people act in different circumstances, so that she knows how to act appropriately in social situations. One bonus arising from this strategy is that her habit of detailed observations of behavior has provided insights into the behavior of animals, as revealed in her most recent book, Animals in Translation: Using the Mysteries of Autism to Decode Animal Behavior.20 This book prompted a documentary, broadcast on the BBC on 8 June 2006, rather unkindly entitled “The Woman Who Thinks Like a Cow.” Curiously, though, she is sensitive to emotion in others, and is herself emotionally sensitive. The very specific nature of this disability has led some evolutionary psychologists to propose that theory of mind is a module, independent of other aspects of the human mind.
High-functioning autism, as evident in people like Temple Grandin, is known as Asperger’s syndrome. People with this condition often can pass false-belief tests, such as the Sally-Anne test, but they apparently do so only through verbal reasoning and explicit instructions about the task. As I noted earlier, normal infants seem to instinctively demonstrate an understanding of false belief well before they can demonstrate it verbally, since they will look to where an actor mistakenly believes an object to be hidden. People with Asperger’s syndrome do not do this, suggesting that the spontaneous understanding of false belief is lacking.21
At the opposite end of the spectrum to autism, it has been suggested, lies psychosis.22 At least some aspects of psychosis seem to reflect a hypermentalism. The more florid symptoms of schizophrenia, for example, include hallucinations, delusions, and paranoia. It’s as though schizophrenics read too much into the minds of others, to the point that they think there are plots to exterminate them, or that their minds are controlled by some sinister external agency. The Scottish psychiatrist R. D. Laing23 was particularly adept at expressing the recursive mentality that can cause social relationships to go wrong, and psychosis to develop. Here are excerpts from his aptly titled book Knots:
JILL: I’m upset you are upset
JACK: I’m not upset
JILL: I’m upset that you’re not upset that I’m upset that you’re upset. JACK: I’m upset that you’re upset that I’m not upset that you’re upset that I’m upset, when I’m not.
JILL: You put me in the wrong
JACK: I am not putting you in the wrong
JILL: You put me in the wrong for thinking you put me in the wrong.
JILL: No
JACK: I’ll never forgive you for not forgiving me
The extremes of autism and psychosis may lie on a continuum that governs normal as well as abnormal behavior. Some personality theorists write of the schizotypal personality, with a tendency toward paranoia and magical thinking, but not classifiable as psychotic. Toward the other end of the continuum, some individuals show autistic tendencies, which may include specific language impairments and obsessive-compulsive behaviors. Think of computer geeks. (Perhaps you are even one yourself, although I suppose if you are, you mightn’t know it.) The continuum may also be considered one of mechanism versus mentalism. We live in a complex world of things and people, and nature seems to have provided us with sufficient flexibility to deal with both. As the evolutionary biologist William D. Hamilton put it, “There are people people and things people.”24
Things people may tend to adopt what I have called the physical stance, in which people are regarded more as moving lumps of meat than as mental beings. Perhaps radical behaviorists, and rugby players, tend to see people in this way. Conversely, people people may tend to treat things as people. Inanimate objects, such as cars, ships, or computers, are often ascribed human-like properties. Throughout history, and perhaps human prehistory, people have personified inanimate objects, such as the stars and planets, and have bestowed human properties on nonhuman animals. Personified animals dwell especially comfortably in children’s books, as we saw in chapter 3. And then there’s God, encountered earlier, who appears to be personification without material substance—the ultimate triumph, perhaps, of theory of mind.
Sex and the Brain
To a degree at least, the autistic-psychotic spectrum seems to underlie differences between men and women. Simon Baron-Cohen has described autism as an extreme of male behavior, and men in general seem more concerned with things than with people25—and indeed may tend to treat women as things rather than as people. Despite the efforts of feminists, young boys seem to prefer toy tractors or spaceships, while young girls go for dolls and their mothers’ makeup. The positive symptoms of schizophrenia are more common in females than in males, although negative symptoms and schizophrenia as a whole are somewhat more common in males. And women tend to be more religious—as we have seen, this can be considered a rather complex manifestation of theory of mind. Students of psychology, the science of the mind, are predominantly female—at least in the departments of psychology I am familiar with—although in the old days of behaviorism there were more men about. Men may be more drawn to the sciences of the physical world.
The biological basis of the autistic-psychotic spectrum may depend more on maternal and paternal genes than on the biological sex of the offspring. The battle of the sexes begins in the womb, through the phenomenon of imprinting. Chromosomes come in pairs, one from the mother and one from the father, and imprinting means that one or other can dominate. The parents have different interests in the fate of the offspring, and this can be expressed in the relative influence of maternal and paternal genes. In mammalian species, the only obligatory contribution of the male to the offspring is the sperm, and the father relies primarily on his genes to influence the offspring to behave in ways that support his biological interest. Paternal genes should therefore favor self-interested behavior in the offspring, drawing on the mother’s resources and preventing her from using resources on offspring that might have been sired by other fathers. The mother, on the other hand, has continuing investment in the child both before birth, in terms of nutrient from her own body, and after birth, in terms of breast milk and care-giving. Maternal genes should therefore operate to conserve her resources, favoring sociability and educability26—nice kids, who go to school and do what they’re told.
Maternal genes are expressed most strongly in the cortex, representing theory of mind, language, and social competence, whereas paternal genes tend to be expressed more in the limbic system, which deals with resource-demanding basic drives, such as aggression, appetites, and emotion. Autism, then, can be regarded as the extreme expression of paternal genes, schizophrenia as the extreme expression of maternal genes. Many of the characteristics linked to the autistic and psychotic spectra are physical, and can be readily understood in terms of the struggle for maternal resources. The autistic spectrum is associated with overgrowth of the placenta, larger brain size, higher levels of growth factors, and the psychotic spectrum with placental undergrowth, smaller brain size, and slow growth.27
The relative dominance of maternal and paternal genes should not be confused with the effects of the sex chromosomes themselves. Not all males are aggressive, overgrown hunks who can’t read books and don’t eat quiche, and not all females are polite blue-stocking vegetarians who devour the novels of Jane Austen. Nevertheless there does seem to be a correlation between imprinting and biological sex, with males leaning toward the autistic end of the continuum and females toward the schizophrenic end. This may arise because the X chromosome, along with chromosomes 2, 3, 5, 7, 9, 10, 15, 16, and 17, is one of the chromosomes subject to differential imprinting,28 although presumably only in females since males receive only one copy of the X chromosome through the mother. Given the multitude of chromosomes involved, the effects of imprinting are likely to be diverse, and not exclusively associated with one or other sex.
Imprinting may have played a major role in human evolution. One suggestion is that evolution of the human brain was driven by the progressive influence of maternal genes, leading to expansion of the neocortex and the emergence of recursive cognition, including language and theory of mind. The persisting influence of paternal genes, though, may have preserved the overall balance between people people and things people, while also permitting a degree of difference. Simon Baron-Cohen has suggested that the dimension can also be understood along an axis of empathizers versus systematizers.29 People people tend to empathize with others, through adopting the intentional stance and the ability to take the perspective of others. Things people may excel at synthesizing, through obsessive attention to detail and compulsive extraction of rules.30
Although one might suppose that a balance of maternal and paternal genes is optimal, it is perhaps the departures from genetic balance that add innovation and creativity to our lives. The pioneering British psychiatrist Henry Maudsley noted that insanity was often accompanied by creative genius. One example was the eighteenth-century English essayist Charles Lamb, who was prone to periods of mental illness. Maudsley commented as follows
His [Lamb’s] case, too, may show that the insane temperament is compatible with, and indeed not seldom coexists with, considerable genius.31
Recent evidence suggests that a particular polymorphism on a gene known to be related to the risk of psychosis is also related to creativity in people with high intellectual achievement.32
The tendency to schizophrenia or bipolar disorder may underlie creativity in the arts, as exemplified by musicians such as Béla Bartók, Ludwig van Beethoven, Maurice Ravel, or Peter Warlock, artists such as Amedeo Clemente Modigliani, Maurice Utrillo, or Vincent van Gogh, and writers such as Jack Kerouac, D. H. Lawrence, Eugene O’Neill, or Marcel Proust. The esteemed mathematician John Forbes Nash, subject of the Hollywood movie A Beautiful Mind, is another example. The late David Horrobin went so far as to argue that people with schizophrenia were regarded as the visionaries who shaped human destiny itself, and it was only with the Industrial Revolution, and a change in diet, that schizophrenics were seen as mentally ill.33
The tendency to autism, especially high-functioning autism (Asperger’s syndrome) may lead to genius of a different kind. This includes savants, individuals with extraordinary ability in some narrow domain, such as calculation, music, or even language, but in other respects subnormal. In chapter 5 we encountered the savant and calculation wizard Kim Peek. Savantism may arise from the narrow focusing on one activity at the expense of others. The tendency to autism, though, may also underlie high-level achievement in mathematics, physical sciences, or engineering, coupled with obsessional thinking. Isaac Newton may be a prototypical example. Another is the Nobel Prize–winning physicist Paul Dirac, regarded by some as being in the same league as Albert Einstein, but also described by his fellow physicist Niels Bohr as “the strangest man.”34 Things people may have their uses after all, since they help us understand an increasingly complex physical world. And we should not deny them the capacity to think recursively.
Recursion, then, is not the exclusive preserve of social interaction. Our mechanical world is as recursively complex as is the social world. There are wheels within wheels, engines within engines, computers within computers. Cities are containers built of containers within containers, going right down, I suppose, to handbags and pockets within our clothing. Recursive routines are a commonplace in computer programming, and it is mathematics that gives us the clearest idea of what recursion is all about. But recursion may well have stemmed from runaway theory of mind, and been later released into the mechanical world. I explore this further in chapter 12.
Do Animals Have Theory of Mind?
As Darwin recognized in the extract quoted earlier, there is little doubt that other species can read emotion in others—life often depends on it. In all animals in which reproduction depends on feeding, cleaning, warming, and protecting the young, parents are sensitive to the emotional state of the infant, and act to relieve distress or satisfy hunger or thirst. Many animals also respond with sympathy to perceived distress in others. In one study, for example, mice perceiving pain in other mice intensified their own reaction to pain,35 and in another study monkeys refused to pull a chain to receive food if doing so caused a shock to be delivered to another monkey.36 Frans de Waal documents these and other examples, and notes that chimpanzees, but not monkeys, often give consolation to others in distress. His ready camera records a juvenile chimpanzee putting a consoling arm around a screaming adult who has just been defeated in a fight.37
More problematic, though, are the more cognitive aspects of theory of mind, such as knowing what another individual sees or believes. It is natural to look first in our closest primate relative, the chimpanzee. The question was first put by David Premack and Guy Woodruff in 1978, in a classic paper entitled “Does the Chimpanzee Have a Theory of Mind?” One of their techniques was to show videos of a human grappling with some problem, and then offer the chimpanzee Sarah a choice of photographs, one of which depicted a solution to the problem. One such test showed a woman trying to escape from a locked cage, and one of the photos showed a key, while others showed objects irrelevant to the task. Sarah performed quite well at choosing the appropriate photo, although as Premack and Woodruff recognized, this need not show that she appreciated what was going on in the mind of the person depicted. For example, the key might have been selected through simple association with the cage.
One researcher who has taken up Premack and Woodruff’s challenge is Daniel Povinelli, who has tried further tests to determine whether a chimpanzee can understand what is going on in the mind of a person.38 The results have been largely negative. Chimpanzees readily approach humans to beg for food, and this provided an opportunity to check whether in so doing they are influenced by whether the person can see or not. But when offered the choice of two individuals to beg from, one with a blindfold over her eyes, the animals did not systematically choose the one who could see. The same was true when one of the people had a bucket over her head, or covered her eyes with her hands. Only when one of the people was actually facing the other way did the chimpanzees easily choose the one facing toward themselves. Young children, on the other hand, quickly recognize that they should approach the person who can see them, and understand that this depends on the eyes. The failure of the chimpanzee to appreciate this does not arise from failure to observe the eyes, since they readily follow the gaze of a person confronting them. Chimpanzees may eventually choose the person who can see them, but the behavior is more simply explained by associative learning, and not on the understanding that eyes are for seeing.
Another test depends on the chimpanzee’s apparent understanding of pointing. If a person sits in front of a chimpanzee and points to one of two boxes to left or right, the chimpanzee understands readily enough that if it wants food, it should go to the box that the person is pointing to. But the choice breaks down if the person points from some distance away, and is systematically reversed if the person sits closer to the box that does not contain the food and points to the other one. It seems that chimpanzees respond on the basis of how close the pointing hand is to the box containing the food, and not on the basis of where the hand is actually pointing.
Yet chimpanzees do follow eye gaze,39 just as humans do, and this may suggest that they have at least some understanding that others can see. Povinelli argues, though, that behaviors like following eye gaze have the same instinctive basis in humans as in other primates, but that we “reinterpret” these behaviors as being more sophisticated than they really are.40 For example, we may spontaneously follow the gaze of someone who seems to be gazing at something in the sky without going through an intellectual (and presumably conscious) exercise along the lines of That fellow must be able to see something up there that’s interesting. Gaze following may simply be an adaptive response that alerts other animals to danger or reward, but we humans have intellectualized it, often after the fact. Remember too that even infants gaze in anticipation of where an actor will falsely believe an object to be hidden, well before they can intellectualize the reason for their own behavior.
Other work suggests, though, that the work of Povinelli and his colleagues underestimates the social intelligence of the chimpanzee. Brian Hare and colleagues have shown that a chimpanzee will approach food when a more dominant chimpanzee cannot see the food, but will be reluctant to do so when they can see that the food is visible to the dominant chimpanzee.41 Chimpanzees also seem able to pass the equivalent of the Sally-Anne test, referred to earlier. Subordinate chimpanzees will retrieve hidden food if a dominant chimpanzee was not watching while the food was hidden, or if the food was moved to another location while the dominant chimp wasn’t watching. That is, the subordinate chimps appear to have knowledge of what the dominant knows. The subordinates will also retrieve food if a dominant chimp watched it being hidden, but is then replaced by another dominant chimp who hadn’t watched, suggesting the subordinates can keep track of who knows what. The subordinates failed on another test, though, in which a dominant chimp watched one piece of food being hidden but did not watch another piece being hidden; they failed to consistently choose the food that the dominant chimp hadn’t watched.42 Even so, the chimps in this study seemed to have at least some knowledge of what another chimp knew.
Chimps have also been observed to hide things from each other. In one study, chimps contested access to food with a human experimenter, and chose to approach the food via a route hidden from the experimenter’s view.43 Sometimes this route was a circuitous one. Brian Hare has also shown that dogs, unlike Povinelli’s chimps, can choose food sources according to where either a person or another dog is looking or pointing.44 It is not clear why Hare’s chimpanzees and dogs seem more blessed with theory of mind than were the chimpanzees studied by Povinelli.45
A difficulty in these studies is that they need not imply an intentional stance.46 The animals might simply be responding on the basis of learned cues, without any sense that the action is based on an understanding of what is going on in another animal’s mind—much as people with Asperger’s syndrome can solve theory-of-mind problems without true understanding. If you are a young chimpanzee, the mere sight of a dominant male in the vicinity may act as a simple, learned signal not to behave in certain ways, just as a human child may wait until his mother’s back is turned before punching his younger sister. But a measure of intentionality may be introduced when she later cries even if not hit, so that her older brother will be punished. In order to infer genuine theory of mind, then, we need some evidence that an act has been improvised, rather than based on trial-and-error learning. Richard Byrne, in his book The Thinking Ape, gives an example that may qualify:
A young baboon, coming across an adult baboon who had just exhumed a root to eat, lets out a scream. This scream alerted the young baboon’s mother, who outranked the adult in status. The mother rushed over, took in the scene of the adult, the root and the seemingly distressed young animal and chased the adult off. Meanwhile the young baboon ate the root.
Behaviors like this appear to be intentionally deceptive, and are referred to as tactical deception. Deception itself is widespread in nature, whether in the camouflage of a butterfly wing or the uncanny ability of the Australian lyre bird to imitate the sounds of other species—as mentioned in chapter 3, this includes the sound of a beer can being opened. Tactical deception, however, is that in which the deception is based on an appreciation of what the deceived animal is actually thinking, or what it can see—that is, it implies an intentional stance. The most obvious form of tactical deception in humans is telling lies, which we do in the hope that the recipient will believe what we say, but in nonverbal animals we must seek evidence from their actions. Andrew Whiten and Richard Byrne asked primate researchers for anecdotes from their field studies involving deceptive behavior, and carefully screened out cases in which the animals might have learned the behavior through trial and error.47 They concluded that only the four species of ape occasionally showed evidence of having deceived on the basis of an understanding of what the deceived animal could see or know. Even so, there were relatively few instances—out of 253 observations, only 21 met Whiten and Byrne’s criteria—12 from common chimpanzees and three each from bonobos, gorillas, and orangutans. These numbers are perhaps too small to convincingly demonstrate that great apes can truly “read the minds” of others.
It is sometimes suggested that there is a sort of intermediate between the physical stance and the intentional stance—a biological stance, if you will. Thus young children and apes may be able to attribute goal-directedness and self-generated motion to others, without being able to attribute full-blown intentional states, such as believing, desiring, seeing, or remembering.48 This does not seem to involve recursion; attributing self-generated motion to another creature carries no implication that the other creature can also attribute self-generated motion. There seems no reason at present to suppose that the great apes can advance much beyond this level of attribution.49
Self-Awareness in Other Species?
What, then, of the ability to know one’s own mind? Could a chimpanzee emulate Descartes and infer its own existence? One proposed way to answer this question is the so-called mirror test, in which a mark is applied to the animal’s face, and the animal is then allowed to see itself in a mirror. The question is whether the animal will understand that the mark is on its own face, rather than on that of some stranger in the mirror. It is common practice to anesthetize the animal before applying the mark, so the animal is unaware of the mark being placed. Chimpanzees with prior experience of mirrors reach up and touch the marks on their faces, while those without experience with mirrors react as though there is another chimpanzee in the mirror, as do rhesus monkeys even after many years of experience.50 It appears that the great apes pass the test, although not consistently, whereas lesser apes and monkeys do not, and there is weak evidence that an elephant and a dolphin may also pass. Children typically pass the test by two years of age.51
But even this test need not imply more than a physical stance; that is, the animal may recognize itself as a physical object, without any necessary understanding that that object has desires, beliefs, emotions, and the like. Even we humans may use mirrors to reflect ourselves as objects, rather than as sentient beings—a face to be shaved, or rendered more beautiful. The reflection in the mirror, after all, is not someone with whom one can have a meaningful conversation.
Where Are We?
The question of whether chimps and other species have theory of mind remains highly contested. Some authors continue to reject all claims of theory of mind in nonhuman species, and indeed it may always be possible to explain apparently human-like intentional behavior in terms of behavioristic principles. Derek C. Penn, Keith J. Holyoak, and Daniel C. Povinelli, in an article headed “Darwin’s Mistake,” conclude as follows:
Our most important claim … is simply that whatever “good trick” … was responsible for the advent of human beings’ ability to reinterpret the world in a symbolic-relational fashion, it evolved in only one lineage—ours. Nonhuman animals didn’t (and still don’t) get it.52
In this view, activities in nonhuman animals that seem to imply human-like cognition, such as theory of mind, are reinterpreted in symbolic terms. They dub their theory the “relational reinterpretation” theory.53
There remains the persisting question, though, of what explains the discontinuity. Despite referring to “Darwin’s mistake,” Penn and colleagues continue to use the term “evolved” in the above extract, but reference to a “good trick” gets us little further than reference to God or some miraculous mutation. Moreover, the appeal to “reinterpretation” is unparsimonious, implying an unnecessary step between what we do and how we interpret it. If a chimpanzee hides from prying human eyes, there seems no good reason to suppose that its behavior is any different in principle from that of the naughty child who hides from an angry parent. It smacks of arrogance to dismiss human-like behavior in animals on the grounds that humans have somehow reinterpreted such behavior in our own species to take it beyond the ken of the merely animal.
My own tentative view is that chimpanzees may indeed have some capacity to discern what other individuals can feel, see, and perhaps know. This is first-order recursion, at best. What they may lack, though, is the extension to higher-order recursion—my knowledge that another individual knows what I can see, know, or feel, or even that the other knows that I know what she’s thinking. Whether or not this can explain the full repertoire of distinctively human activities, it does suggest that the difference is more one of degree than one of kind—that small step that proved such a giant leap for humankind.
Runaway theory of mind might be attributed to the highly social lives that we humans live, whether in the nuclear family, the office, the hunting party, or the rugby scrum. In such settings we must constantly monitor our actions in accord with how others think, believe, and feel. The conditions that led to the evolution of complex social behavior are discussed further in chapter 11.
Let’s now return to language, since it too, in its predatory way, depends on theory of mind, and provides further insight into its recursive structure.