One hundred years ago, in The Interpretation of Dreams, Sigmund Freud unveiled the unconscious, and "the psychological century" was born. It has turned out to be a huge disappointment. The gene and the quantum were conceived at the same time as Freud conceived the unconscious; yet, although they have led to sophisticated technologies, psychology and psychiatry, by most standards, are failures. More people than ever are on anti-depressants; drug abuse is rampant; psychotherapies don't work; our jails are fuller than ever.
What happened? Where did it all go wrong? Jerome Kagan, a professor of psychology at Harvard, thinks he has an answer. In his newly published book, Surprise, Uncertainty and Mental Structures, he argues that we have been ignoring what goes on inside our heads.
Consider the following experiments:
From these results, Kagan concludes that neither words nor brain states can adequately take us to the core of human behaviour. The six-month-old babies recognised the concept of "animals" but they had no language to describe it. The adults knew there were subtle differences between the pictures they were shown, but could not put them into words. People could listen to music that expresses different moods but their brainwave patterns failed to match these moods. In other words, something else is going on in the mind, and science has ignored it, greatly impoverishing our understanding of human experience and behaviour.
What is it? According to Kagan, the brain (he prefers to call it the mind) contains psychological structures, to which he gives the ungainly word "schemata". We hold these schemata in our heads, Kagan says, just as those babies held the concept of "animal" long before they had a word for it. We hold visual experiences, olfactory experiences, taste, touch, auditory and emotional experiences somewhere inside us, which enrich the words we use but are separate from them and individual to each of us. And they cannot be related directly to brain activity.
Take smell. We have few words for smells, and tend to describe them through comparisons: we say something smells like a rose, for example. Yet, as Proust noted, if we actually experience a smell from childhood - of cooking, say, or of an old backstairs - we remember far more about the original circumstances than if these things are simply mentioned in conversation. The original memories are stored somewhere in a place that has nothing to do with language and cannot be retrieved simply by talking about the past.
Similarly, we have few words to describe emotions. Despite the best efforts of poets, emotions are largely beyond words, and this is another reason why, in Kagan's view, psychology and psychiatry have stalled. Talking cures, as in psychotherapy, or having people fill out questionnaires, as psychologists so often do, simply do not get at the internal schemata that drive and colour so much of our behaviour. In fact, words sometimes get in the way, as seems to have happened with the adults who were asked to describe the photographs of faces.
We have one word for "fear" but even in quite primitive animals fear is not simply equivalent with hormonal balance or brain state. Give two male lobsters, facing each other, identical injections of serotonin (one of the substances in Prozac). Their behaviour will differ if, in the past, one crustacean has been defeated in a fight by the other. If anti-depressants cannot cope with the complexities of lobster life, they are unlikely to cope with human beings.
All this is controversial enough; but Kagan becomes even more controversial when he applies his theory to everyday life. Take the moral development of children. Kagan says that, for them, concepts such as "good", "bad" or "truth" are not just verbal ideas; they embody schemata. Children and adults respond more quickly to concepts that are, as he puts it, "rich in schemata". That is why adults are faster to answer "false" to the statement "Carrots are blue" than to the statement "The function of vitamins is social justice": they have a clearer and more vivid experience of carrots than they have of social justice.
So a child will learn the concept of "bad" more fully if it is not just a semantic notion. The parent should not just tell the child what is bad but burn the idea into its experience: through frowning, scolding, withdrawing privileges (Kagan avoids the question of whether one should add smacking). The more schemata are involved, the richer the child's concept of "bad" will be. But equally, experience can persuade children that they are irredeemably bad. Kagan considers Mary Bell, the 11-year-old who killed two toddlers. She had been sexually abused by her mother's clients and knew her father's criminal record. "This girl's categorisation of herself as a 'bad child' made it easier for her to murder without the passion of anger or the desire for material gain."
This is all grist to the traditionalist's mill. But that is not Kagan's main aim, which is to rethink psychology and psychiatry from the bottom up. For instance, he attacks the fundamental notion of "personality". Psychologists identify five personality dimensions: neuroticism (calm v worried), extroversion (sociable v unsociable), openness (conservative v liberal), agreeable (good-natured v irritable) and conscientiousness (reliable v careless). But these dimensions are invariably "discovered" via questionnaires, to which people respond with words. Yet, as Kagan says, no doctor would use only the reports of a patient to diagnose disease without examining that patient's actual anatomy and pathology. Kagan's own experiments show that questionnaires are unreliable predictors of behaviour. When asked to describe their children as "sad" or "happy", the answers most parents gave bore no relation to the child's behaviour in the laboratory. Children, too, were asked whether they thought of themselves as "happy most of the time" and whether they liked to "play with many friends". Again, their answers bore little relation to how often they actually laughed, or to the number of other children they played with.
If psychology cannot measure accurately even as basic a trait as personality, how can it hope to achieve an understanding of internal lives?
Kagan is similarly harsh on psychiatry. The central element in the diagnosis of depressive disorder, for example, includes reports of sadness, apathy, sleeplessness and poor appetite. This, says Kagan, is far too general, and rather like assuming that all stomach aches have the same cause.
If depression is the result of failure to meet some moral standard, then it should be called a guilt reaction. If the same symptoms occur in an adolescent worried about his future, because his family is poor, "his condition is better described as a state of hopelessness". A third depressive may have inherited a neurochemical imbalance. "These three patients should be placed in different diagnostic categories."
Kagan's view is that personality starts with an inherited temperamental bias. This determines how surprised we are when faced with uncertainty (all the experiments quoted are really about uncertainty), but is modified by subsequent life experiences. Kagan divides children into those who are "high-reactive" - who respond to unfamiliar sights and sounds "by thrashing their limbs and crying" - and "low-reactive" - who show minimal responses and no crying. In general, the high- reactive children (about 20 per cent) become shy and fearful toddlers and introvert adolescents; the low-reactive ones become bolder toddlers and more sociable adolescents. But that is not all.
High-reactive infants raised in academically encouraging homes become conscientious students and accomplished adolescents. Raised in less-endowed homes, they become shy loners or, if aggression is not stamped on, they can become violent criminals. What is crucial for this vulnerable minority is a proper education, which by definition vastly reduces the degree of uncertainty in people's lives, and teaches them how to minimise and manage it.
Kagan's point is that high-reactive children cannot be identified by questionnaire - their parents describe them in exactly the same way as parents of low-reactive children describe their offspring. They are programmed by their biology to collect a particular set of schemata which, unless checked by education and a supportive family, will turn them into unhappy and antisocial adults.
Only if psychology learns to accept this view - and to observe behaviour properly rather than relying on words or brain scans - can it finally fulfil its promise to benefit humanity.
Peter Watson's intellectual history of the 20th century, A Terrible Beauty, is published by Phoenix