Madness Explained

by Richard P. Bental

  Axis 2 of DSM-III and its successors attempted to accommodate dimensions of disorder within a categorial system. However, as might be expected of such an attempt, the division of abnormal personalities into separate categories proved problematic from the outset. Research showed that the reliability of these categories was very low (one study found an average kappa value of 0.41)45 and the authors of DSM-III were forced to acknowledge that many patients would meet the criteria for more than one type of personality disorder.

  A psychological approach to psychosis-proneness

  Meehl’s ideas prompted psychologists to search for personality ‘markers’ or indicators of schizotaxia in the hope that these could be used to detect people at risk of schizophrenic breakdowns. As it was believed that the features of schizotaxia were quite subtle, many investigators believed they might be detected by examining unusual responses to conventional personality questionnaires. A common research strategy involved asking the mentally well (but presumably schizotaxic) relatives of schizophrenia patients to complete standard personality tests. Unfortunately, this line of research led nowhere and had been all but abandoned by the end of the 1970s.46

  At about this time it became obvious to some psychologists that a better way of approaching this problem might be to question people directly about psychotic experiences. In the United States, Loren and Jean Chapman at the University of Wisconsin devised a series of questionnaires for this purpose, each of which focused on a different symptom. Their early questionnaires measured physical and social anhedonia (items included ‘I have had very little desire to buy new kinds of foods’, and ‘It’s fun to sing with other people’, to be marked as either ‘true’ or ‘false’) and perceptual aberrations (‘Occasionally I have felt as if my body did not exist’). Later questionnaires measured magical thinking (‘Some people can make me aware of them just by thinking about me’) and hypomania (‘I am so good at controlling others that it sometimes scares me’).47 In Britain, at the Department of Experimental Psychology at Oxford University, Gordon Claridge took a slightly different approach, and devised a single questionnaire based on Robert Spitzer’s DSM-III definition of schizotypal personality disorder.48 These questionnaires have been followed by many others, so that scales of schizotypy or psychosis-proneness, if not quite a dime a dozen, have now become commonplace instruments of psychological research.

  Gordon Claridge has recognized that the apparent continuum from sanity to madness, confirmed by studies that have employed these questionnaires, necessitates a revision of the Kraepelinian model of psychosis.49 However, he does not believe that it is necessary to abandon a biomedical model of madness altogether. He has pointed out that there are widely recognized physical diseases that are extremes of normal variation. An example is hypertension, an abnormal elevation of blood pressure, which leads to an increased risk of strokes and heart attacks. Blood pressures are distributed normally* in the population. Some people have very low pressure, some have very high pressure, but the majority of the population lies somewhere in between these two extremes. The cut-off point between low and high blood pressure is arbitrary, and is determined pragmatically by doctors, who balance the need to identify those most at risk of vascular disease with the hazards of unnecessarily treating a large proportion of the population. Nonetheless, as high blood pressure has disastrous (and sometimes fatal) consequences, few people would doubt that it is a cause for medical concern.

  On this analogy, psychosis might be thought of as the extreme end of a normally distributed spectrum of personality traits on which we impose an arbitrary cut-off point to separate those who are mentally ill from those who are not. Claridge’s approach is slightly more complex than this, in that he assumes that there is a discontinuity between madness and sanity, which, using the blood pressure analogy, might be thought of as equivalent to a stroke. He argues that a person who has psychotic traits may lead a normal life, much as a person with high blood pressure can lead a normal life, unless some kind of adverse environmental event precipitates a breakdown. Claridge suggests that there are many different kinds of events that might lead to such a crisis, including both psychological and physical traumas (see Figure 5.1). This model is similar to Meehl’s, and assumes that individuals differ in their vulnerability to mental illness, so that a relatively minor life event can precipitate a breakdown in someone who is especially at risk, whereas a major crisis is required to precipitate a breakdown in someone who is less vulnerable.

  There is insufficient space here to review the very large volume of research that has been conducted with questionnaire measures of psychosis-proneness. However, two general findings are worth noting. First, there is some evidence that scores on these questionnaires predict future psychotic breakdown, although not with the kind of precision that would be useful to clinicians. In a study carried out by Loren and Jean Chapman, several hundred Wisconsin students were tested on a battery of schizotypy questionnaires and were then divided into high scoring and normally scoring groups. After ten years, fourteen of the high scorers and only one of the low scorers had been admitted to hospital with a psychotic illness.50 While this finding would not justify the mass testing of young adults in the hope of preventing mental illness, it does suggest that there is a real link between schizotypal personality characteristics and outright madness, as supposed by Meehl, Claridge and the Chapmans.

  Second, the same kinds of arguments that have been made about the continuum between normal personality and schizophrenia have also been made about mood disorders. Hagop Akiskal of the University of California at San Diego has suggested that many patients treated by psychiatrists or psychologists for ‘borderline personality disorder’ in fact suffer from subclinical bipolar disorder.51 This type of personality, it will be recalled, has been characterized in terms of extreme emotional reactions, unstable self-esteem, self-destructive behaviour and intense

  Figure 5.1 Claridge’s model showing the relationship between schizotypal personality and schizophrenia, and showing the parallel with systemic diseases such as hypertension (from G. S. Claridge (1990) ‘Can a disease model of schizophrenia survive?’, in R. P. Bentall (ed.), Reconstructing Schizophrenia. London: Routledge). Note the familiar bell-shaped normal distributions for both blood pressure and schizotypal traits. and unstable relationships with others. In addition to the obvious clinical similarities between prototypically bipolar and borderline patients, Akiskal was able to point to evidence of high levels of mood disorder in the close relatives of borderline patients.

  Given the evidence against Kraepelin’s division between dementia praecox and manic depression, which we considered in the previous chapter, it would obviously be interesting to know whether borderline mood states overlap with measures of psychosis-proneness derived from the concept of schizophrenia. Two lines of research suggest that they do. First, in a study by Robert Spitzer, psychiatrists were asked to evaluate their patients on the proposed DSM-III definitions of schizotypal and borderline personality disorder. It was found that about half the patients who met the criteria for one of the diagnoses also met the criteria for the other.52

  The second line of evidence concerns the clustering of psychotic traits in normal individuals. With Peter Slade and Gordon Claridge, I conducted one of the earliest studies in this area. We gave 180 students a range of personality questionnaires, fourteen of which had been designed to measure psychosis-proneness. (The questionnaires included those designed by the Chapmans and by Claridge, along with various others.) Factor analysis of the questionnaire scores revealed that they fell into four separate groups. The first included scales measuring perceptual abnormalities and bizarre ideas, and seemed to correspond (roughly) to delusions and hallucinations, or the positive symptoms of psychosis. The second group included the anhedonia scales together with a measure of introversion, and might be thought to correspond (again roughly) to the negative symptoms of psychosis. A third group seemed to involve the subjective experience of anxiety and cognitive disorganiza
tion, and the final group appeared to measure self-reported anti-social behaviour.53 The first three of these dimensions corresponded reasonably well with the results from factor analytic studies of schizophrenia symptoms, which we considered in the previous chapter.

  Although it would be risky to base a theory entirely on the questionnaire responses of a relatively small number of British university students, these findings have now been repeated by other investigators who have carried out studies with much larger and more representative samples.54 (Indeed, we eventually replicated our original findings with a sample of 1095 people drawn from various walks of life.)55 The findings also seem to have some cross-cultural validity, as a study of over 1200 people living in Mauritius has shown that this pattern of scores on schizotypy measures seems to be unaffected by culture, gender or religious affiliation.56 In general, these studies have found that hypomanic traits (presumably related to mania) correlate very highly with odd beliefs and perceptual aberrations (traditionally thought of as a characteristic of schizophrenia). This finding suggests that schizophrenia and bipolar personality characteristics overlap, and therefore further undermines Kraepelin’s assumption that there are two separate types of psychosis.

  Madness, Creativity and Evolution

  In her biography of Nobel-prize-winning mathematician John Nash Jr, Sylvia Nasar describes the following encounter:

  It was late on a weekday afternoon in the spring of 1959, and, though it was only May, uncomfortably warm. Nash was slumped in an armchair in one corner of the hospital lounge, carelessly dressed in a nylon shirt that hung limply over his unbelted trousers. His powerful frame was slack as a rag doll’s, his finely moulded features expressionless. He had been staring dully at a spot immediately in front of the left foot of Harvard professor George Mackey, hardly moving except to brush his long dark hair away from his forehead in a fitful, repetitive motion. His visitor sat upright, oppressed by the silence, acutely conscious that the doors to the room were locked. Mackey finally could contain himself no longer. His voice was slightly querulous, but he strained to be gentle. ‘How could you,’ began Mackey, ‘how could you, a mathematician, a man devoted to reason and logical proof… how could you believe that extraterrestrials are sending you messages? How could you believe that you are being recruited by aliens from outer space to save the world? How could you?’

  Nash looked up at last and fixed Mackey with an unblinking stare as cool and dispassionate of that of any bird or snake. ‘Because,’ Nash said slowly in his soft, reasonable southern drawl, as if talking to himself, ‘the ideas I had about supernatural beings came to me the same way that my mathematical ideas did. So I took them seriously.’57

  Some authors have argued that the discovery of a continuum (or several continua) linking psychosis and normal functioning may help to explain a feature of madness that has puzzled evolutionary thinkers, namely its persistence within populations over many generations. On the face of it, madness has grim implications for survival and reproduction. People suffering from severe psychotic symptoms often find it difficult to work, are relatively poor, are often socially isolated, and face a high risk of early death from suicide.58 Not surprisingly, research has shown that they enjoy less reproductive success (that is, they have fewer children on average) than their fellow human beings.59

  A paper co-authored by the evolutionary biologists Julian Huxley and Ernst Mayr and psychiatrists Humphrey Osmond and Abram Hoffer, published in Nature in 1964, first pointed out that genes causing vulnerability to psychosis should be selected out over successive generations unless these social and reproductive disadvantages are balanced by advantages. They suggested that some kind of physiological benefit to non-affected relatives, for example enhanced resistance to infection, might compensate for the selective disadvantage of lower survival and reduced fertility experienced by the severely ill.60 However, over thirty years later there is still no evidence to support this theory, and more recent researchers have argued that the benefits associated with psychosis lie in the social rather than the physical domain. Various kinds of social benefits have been postulated. For example, it has been suggested that genes for paranoia encourage a healthy defensiveness in threatening environments,61 or that schizophrenia genes cause just the right degree of social strife to facilitate the splitting of overlarge groups in primitive societies.62 Most of these hypotheses are evolutionary ‘just so’ stories of little merit, and the only substantial research exploring the positive consequences of madness has focused on creativity.

  The idea that madness and creative genius are related predates modern psychiatry, and can certainly be traced back to Aristotle.63 The evolutionary arguments of Huxley and others have given new life to this idea, provoking several lines of investigation. Some researchers have conducted biographical surveys of historically important people in the hope of finding evidence of high rates of psychiatric disorder. Table 5.1 gives a list of historical figures thought to suffer from serious mental illness, compiled from various sources by the American psychologist Dean Simonton.64

  The problems faced by studies of this sort should not be underestimated. It is difficult to decide who should be included in this kind of survey, and even more difficult to make inferences about mental health from biographical data. Inevitably, psychiatric symptoms are inaccurately attributed to some people, and others who were clearly mentally ill are overlooked. (Simonton’s table misses out two of my own favourite examples of disturbed genius, the mathematicians Kurt Gödel and John Nash. Gödel suffered from paranoid delusions and died from malnutrition after refusing to eat.65 As we have already seen, Nash, a Nobel prize-winner and creator of a mathematical theory of rational behaviour, spent much of his life in psychiatric hospitals,

  Table 5.1 Eminent persons with supposed mental illness (adapted from D. K. Simonton (1994) Greatness: Who Makes History and Why. New York: Guilford).

  Schizophrenia

  Scientists: T. Brahe, Cantor, Copernicus, Descartes, Faraday, W. R. Hamilton, Kepler, Lagrange, Linnaeus, Newton, Pascal, Semmelweiss, Weierstrass, H. Wells

  Thinkers: Kant, Nietzsche, Swedenborg

  Writers: Baudelaire, Lewis Carroll, Hawthorne, Hölderlin, S. Johnson, Pound, Rimbaud, Strindberg, Swift

  Artists: Bosch, Cellini, Dürer, Goya, El Greco, Kandinsky, Leonardo da Vinci, Rembrandt, Toulouse-Lautrec.

  Composers: Donizetti, MacDowell, F. Mendelssohn, Rimsky-Korsakov, Saint-Saëns

  Others: de Sade, Goebbels, Herod the Great, Joan of Arc, Nero, Nijinsky, Skaha

  Affective disorders (depression or bipolar disorder)

  Scientists: Boltwood, Boltzmann, Carothers, C. Darwin, L. de Forest, J. F. W. Herschel, Julian Huxley, T. H. Huxley, Jung, Kammerer, J. R. von Mayer, V. Meyer, H. J. Müller, J. P. Müller, B. V. Schmidt, J. B. Watson

  Thinkers: W. James, J. S. Mill, Rousseau, Sabbatai Z’vi, Schopenhauer

  Writers: Balzac, Barrie, Berryman, Blake, Boswell, V. W. Brooks, Byron, Chatterton, J. Clare, Coleridge, William Collins, Conrad, Cowper, H. Crane, Dickens, T. Dreiser, R. Fergusson, F. S. Fitzgerald, Frost, Goethe, G. Greene, Hemingway, Jarrell, Kafka, C. Lamb, J. London, R. Lowell, de Maupassant, E. O’Neill, Plath, Poe, Quroga, Roethke, D. G. Rossetti, Saroyan, Schiller, Sexton, P. B. Shelley, C. Smart, T. Tasso, V. Woolf

  Artists: Michelangelo, Modigliani, Pollock, Raphael, Rothko, R. Soyer, Van Gogh

  Composers: Berlioz, Chopin, Elgar, Gershwin, Handel, Mahler, Rachmaninoff, Rossini, R. Schumann, Scriabin, Smetana, Tchaikovsky, Wolf

  Others: C. Borgia, Clive, O. Cromwell, A. Davis, J. Garland

  Note: Because almost all of these diagnoses are not based on objective clinical assessments, most are highly tentative.

  believing that he was receiving messages from extraterrestrials.)66 These problems notwithstanding, the most carefully conducted biographical surveys have consistently pointed to unusual levels of psychiatric disturbance in creative and influential people.67

  Other investigators have attempted to avoid th
e uncertainties of historical research by studying living people. For example, Nancy Andreasen interviewed a group of creative writers living in Iowa, and found abnormally high levels of mood disorder in both the writers and their relatives.68 American psychologist Kay Redfield Jamison similarly interviewed a group of eminent British writers and artists (chosen on the basis of having won at least one top award in their field) and also found high levels of mood disturbance. Although very few in Jamison’s sample had been treated for bipolar disorder (most having been treated for depression) many described creative episodes in which they had experienced mood and energy changes consistent with hypomania.69

  If psychosis and creativity have common genetic roots, the relatives of mentally ill patients should show evidence of unusual creativity. This prediction was tested in a study conducted in Iceland by psychiatrist John Karlsson, who investigated the occupations of the relatives of schizophrenia patients and found evidence of high levels of creativity.70 In a more recent study carried out in Denmark by Ruth Richards and her colleagues, manic-depressive patients and their relatives were interviewed about their lives, and their responses were evaluated using a standard measure of lifetime creative achievement. The patients and their relatives scored higher than a mentally well control group.71