"In short, by these analogies, and coincidences between the phenomena of genius and mental aberration, it seems as though nature had intended to teach us respect for the supreme misfortunes of insanity; and also to preserve us from bang dazzled by the brilliancy of those men of genius who might well be compared, not to the planets which keep their appointed orbits, but to falling stars, lost and dispersed over the crust of the earth."
- Cesare Lombroso1
People have wondered for centuries whether creativity should be dreaded as well as prized. The mantle of prophet, poet, seer, or leader has rested heavily on the shoulders of its wearers. In both literature and life, genius has been viewed with ambivalence, seen as both a curse and a blessing, because it has so frequently been attended by physical and mental suffering. Tiresias was a lonely, crippled outcast as well as a revered sage. Columbus heard a voice condemning him when he reached the shores of Jamaica: "What happens to you today is a deserved punishment for having served the masters of the world and not God." Newton's later years were troubled by recurrent persecutory delusions. Samuel Johnson, Abraham Lincoln, and a host of others experienced terrible bouts of melancholy. Throughout the centuries, the threat of suffering and insanity has hovered over the heads of the eminent and creative.
Although the possible association between "genius and insanity" was noted as early as the classical period, it was not sytematically studied until the 19th century, when a veritable flood of books appeared. Lombroso's The Man of Genius, ' the first lengthy discussion of the subject, was responsible for arousing intense and continuing interest in the relationship between creativity and illness. Like Lombroso, most Victorians were both attracted and repelled by the concept of "genius" and the concept of "madness." The association between the two could be observed in a long series of 19thcentury geniuses: van Gogh, Baudelaire, Schumann, Wilde, Coleridge, Byron, Keats, Mill, De Quincey, Poe, Tolstoy, Dostoevski, Donizetti, Nietzsche, Cowper, Collins, Schopenhauer, and Gogol. Lombroso's major contribution was to suggest that genius was a "degenerative psychosis" and that both mental illness and creativity tended to run together within families. Lombroso illustrated his ideas with many anecdotal case histories of the lives and behaviors of well-known geniuses.
Lombroso's work was quickly supplemented by another monumental study, completed by Galton, titled Hereditary Genius.2 Galton also hypothesized that genius was often inherited and certainly tended to be familial, and he too observed some familial association between mental illness and creativity. Galton amassed a large number of pedigrees that illustrated his argument, most of them supported by anecdote. A number of lesser books followed the precedent of Lombroso and Galton, and a large literature on "genius and insanity" soon accumulated.3"6
These early anecdotal approaches make fascinating reading, but they have clear scientific limitations. Because they begin by identifying cases that illustrate an association between creativity and illness, they do not test a null hypothesis or even indicate how common or uncommon the association may be. They raise a tantalizing series of questions without adequately answering them: Is there really more psychiatric illness among creative people than in the general population? If there is more, how common is it? Is there a tendency for creative people to have a particular type of psychiatric illness, or do they have a wide range of types? If there is an association between psychiatric illness and creativity, is it with specific types of creativity (such as writing or painting) or does it occur in all types of creative people? If there is an association, does it tend to be familial? If the association is familial, is it genetic, nongenetic, or multifactorial?
In an attempt to answer at least some of these questions, the earlier Victorian studies have been supplemented by a series of simpler and better-designed 20th-century studies that approach the subject in a more spare and systematic way. Three basic approaches have been used. Most of them begin with persons recognized for their creativity and assess its possible association with illness. Some investigators have begun with carefully identified creative subjects, most of them at the height of their careers, and studied them intensively through interviews and psychometric tests in order to assess the psychiatric health of the creative person. A second approach has been to examine systematically identified creative persons, and their families as well, in order to evaluate the familial aspects of the association between creativity and illness; these studies rely on interviewing, record study, and anecdote. A third approach has been to employ ingenious experimental designs developed to examine genetic relationships, such as the study of the adopted persons in "creative" occupations and their biologic and adoptive families. Although we are still far from being able to answer all the relevant questions concerning the relationship between creativity and psychiatric illness, each of these approaches has put a few pieces of the puzzle into place.
INTENSIVE STUDIES OF CREATIVE PEOPLE
The researcher in creativity must begin with a difficult question: What is it? How are creativity and intelligence different, or are they? Several investigations have indicated that they are indeed somewhat distinct cognitive qualities. Terman's monumental investigation7 of children identified early in life as having high IQs has taught us a great deal about that type of genius. Terman began with a group of children in the California schools who scored at least 140 on IQ tests. He and his co-workers then studied these children and their families intensively in a variety of ways, including the intellectual success of their relatives, anthropometric measurements, medical history and examinations, educational history and achievement, personality testing, and evaluation of their intellectual and recreational interests. He and his coworkers then followed these children at regular intervals for the rest of their lives, assessing their physical and emotional development, occupational achievements, marital adjustment, and emotional and physical health. Terman's "geniuses" have now reached old age, and the stereotype of the very bright child or adult as scrawny, emotionally fragile, and socially inept or withdrawn has been convincingly demolished.
Terman's "geniuses" have typically been above population norms in almost all ways: they are bigger, stronger, physically and emotionally healthier, and more successful both economically and socially. They have less psychiatric illness and less divorce, although they have had a higher suicide rate than population norms. Some have made notable and creative contributions (Oppenhemier is said to have been a "Terman genius"), but generally the group does not contain a large number of successful writers, musicians, or artists or of innovative scientists or mathematicians. Thus, while Terman's study seems to indicate that there is a negative correlation between genius defined as high IQ and psychiatric illness, it has also indicated that people with high IQs are not necessarily artistically or scientifically creative.
Other investigators have approached the issue by defining creativity operationally, and this appears to be a very logical tactic. Using this approach, a creative person is defined as a person who has been productively creative. In order to identify creative persons for study, respected leaders in a particular field are usually asked to nominate a set number of their peers who are notable for the creativity of their contributions. English professors at Harvard and Stanford might be asked to nominate living writers, for example, or professors of architecture or mathematics might be asked to name the most creative living architects and mathematicians. The people receiving the most nominations are then invited to participate in a study of creativity, usually residen tially. Subjects are given a variety of psychometric tests, such as the Rorschach or IQ tests, and usually interviewed about their intellectual and emotional experiences and methods of working.
This and similar approaches have been used in a number of famous studies of various types of creativity - MacKinnon's study of architects,8 Anne Roe's study of scientists,9 and the studies by Drevdahl and Cattell10 and Barron" ,4 of both artists and scientists, to mention only a few. Although particular studies have used different measuring instruments, conclusions tend to be surprisingly consistent from one study to another. The creative person is characterized by a complex constellation of traits, some of which even appear superficially incompatible. He is intelligent, independent, high on ego strength, fond of challenges, nonconforming, adventuresome, sensitive, introspective, socially detached, and self-assertive. MacKinnon's work with architects has further reaffirmed the notion that intelligence and creativity are different.8 When architects were divided into three groups on the basis of creativity (highly creative, somewhat creative, not creative), all three earned nearly identical scores on several different IQ tests, with a mean of around 120 for each of the three groups. Intelligence thus appears to be a necessary but not sufficient cause of creative ability.
Unfortunately, most studies using the method of personal interviews and testing have not included any careful psychiatric evaluations. Although they contain some suggestive data, they do not deal with it systematically. In studying writers, Barron observed them to have elevated MMPI scores on the D, Sc, Hy, and Pd scales.14 More than half also described visual hallucinations, mystical experiences, and a periodic sense of overwhelming desolation. Drevdahl and Cattell observed of the artists and writers whom they studied that they were "not well adjusted" but, rather, manifested "divine discontent."10 The observations suggest the possibility that depression, and perhaps even psychosis, may occur in creative persons (or at least writers), and they suggest the need for more careful study of the question.
STUDIES OF PSYCHOPATHOLOGY AND FAMILIAL FACTORS
Whereas the studies just described have focused primarily on personality and cognitive factors in creative people, others have directly addressed the "genius and insanity" question by assessing the incidence of psychiatric diagnoses both in creative or eminent persons and in their immediate families. Three major studies of this subject have been completed: Ellis' studies of eminent persons in Great Britain,15 Juda's study of German artists and scientists,16 and Andreasen and Canter's examination of writers at the Iowa Writers' Workshop. 17,18 Results from these three studies are summarized in Table 1. All three suggest that persons whose creativity is expressed in an artistic or scientific way may have a high incidence of psychopathology.
Like the Terman study, the Ellis study examined an aspect of "genius" that may be related to creativity, rather than creativity in pure culture. The subjects whom Ellis studied were notable for their eminence; nearly all were successful, a few notorious, and at least some creative. Using the British Dictionary of National Biography, Ellis selected 1,020 persons of eminence in a variety of fields, including politics as well as the arts and sciences.15 He was able to identify 4.2 per cent as "insane," most of whom were probably schizophrenic but some of whom were probably paretic. Further, 8 per cent were "melancholic" and 16 per cent imprisoned, and 5 per cent had traits suggestive of a personality disorder. Some of Ellis' findings, such as the high rate of imprisonment, are clearly related to his sampling of famous rather than purely creative persons, and the mixed nature of the sample obscures the implications of the other results, such as the relationship between insanity or melancholy and creativity or eminence.
PREVALENCE OF PSYCHIATRIC ILLNESS IN CREATIVE AND EMINENT PERSONS: THREE STUDIES COMPARED
Juda, on the other hand, examined a group more narrowly defined in terms of creativity.16 Although she relied more on historical records than on direct interviewing, she limited her study to gifted artists and scientists. Her sample of 113 artists included 12 architects, 18 sculptors, 37 poets, 20 painters, and 26 composers. Her 181 scientists included 51 theoretical scientists, 112 natural scientists, nine technical applied scientists, and nine statesmen. In both groups personality disorder was the most common diagnosis, being applied nearly twice as often to the artists as to those in the other groups. Juda's concept of personality disorder was Germanic: her "psychopaths" included such categories as "schizoid eccentric," "emotionally unstable," "weak character," "excitable and high-strung," and "hysterical." Although the suicide rate was nearly equal for both groups, the artists had more alcoholism and schizophrenia, while the scientists had more affective disorder.
Andreasen and Canter's investigation of successful writers from the University of Iowa Writers' Workshop has provided somewhat different insights concerning the prevalence and nature of psychiatric illness among writers.17,18 As Table 1 indicates, the rate of illness is quite high. Sixty per cent of the writers had personality disorders, usually cyclothymic personality, and 67 per cent had experienced affective disorder (13 per cent bipolar, 54 per cent unipolar). Forty per cent had drinking problems severe enough for a diagnosis of alcoholism. This rate of illness is considerably higher than that noted in previous studies, but it is the only one based on a direct, structured interview concerning symptoms. Nearly all the writers had been treated for their depressive symptoms, and 27 per cent had been hospitalized. The writers were compared with controls matched for age, educational background, and sex. Although evaluated with the same interview and diagnostic criteria, the controls had a significantly lower lifetime prevalence of psychopathology. Some of the higher prevalence may be accounted for by the fact that seeking psychiatric help has been fashionable in the 20th century and by the fact that writers tend to be extremely introspective and sensitive to internal psychic changes, but even with these qualifications it is clear that writers have a high rate of affective disorder. The results cannot be generalized to other creative groups, however, and consequently similar diagnostic studies using criteria and structured interviews should be done with creative subjects in other fields to determine whether the findings are specific to writers or characteristic of all creative persons.
PSYCHIATRIC ILLNESS IN THE RELATIVES OF CREATIVE OR EMINENT PERSONS
ARE CREATIVITY AND PSYCHOPATHOLOGY HEREDfTARY?
Each in its own way, all the studies cited have implied an association between artistic and scientific creativity and psychiatric illness. The Victorian theorists also suggested, however, that the "genius and insanity" association tended to be hereditary. Several modern studies have also applied scientific sampling approaches to test this hypothesis.
Both Juda and Andreasen and Canter studied their subjects in terms of their family history of psychiatric illness, and the results are summarized in Table 2. Juda found a much higher prevalence of personality disorders among the relatives of her artists and scientists, with the scientists generally having a lower familial prevalence of illness. In the Writers' Workshop study the writers have a much higher familial prevalence of affective disorder than do the controls, although not nearly as high as in the writers themselves, and there is also an increased familial prevalence of creativity among the writers.
Although these data may imply that psychiatric illness and creativity tend to run together in families, they do not necessarily indicate that the association is genetic or inherited. In order to test a genetic hypothesis, more sophisticated experimental designs must be applied. The adopted-offspring design, pioneered by Heston,19 has been attempted by McNeil.20·21 In doing his original study of the adopted offspring of schizophrenic mothers, Heston noted that an unusually large number of the normal offspring seemed to have creative interests or hobbies and proposed the hypothesis that creativity and schizophrenia might be associated, Karlsson22 attempted to test Heston' s hypothesis by using a family-study design. He began with patients suffering from either manic-depressive illness or schizophrenia and examined the percentage of their relatives listed in Iceland's Who's Who. Patients with affective disorder had 23 per cent of their relatives listed, while 13 per cent of the relatives of schizophrenics were listed. Because it is a study of eminence, as opposed to creativity, the Karlsson study has certain inherent limitations.
McNeil20 attempted to test Heston's hypothesis by using an adopted-offspring design similar to that of Heston. He began with adopted persons who were either highly creative (had achieved national recognition) or somewhat creative (pursued such creative occupations as journalism or teaching music). He then studied the prevalence of illness in the adopted persons, their biologic parents, and their adoptive parents. He found that 30 per cent of the highly creative probands were psychiatrically ill, as were 28 per cent of their biologic parents, while only 5 per cent of the adoptive parents were ill. The rates were somewhat lower for the less creative subjects and their relatives. Following European diagnostic nomenclature, McNeil found that about half of the illness observed was affective disorder or "reactive psychosis" and about half was "constitutional psychopathy." Thus McNeil's study gives some additional support to a familial association between affective disorder and creativity and implies that it may in part be genetic.
A variety of evidence has been converging on the familial association between creativity and psychiatric illness. Although all the research completed to date has flaws in sampling, definition of terms, or basic design, most of it leads to the conclusion that creativity and illness may have familial and even genetic associations, and most implies a specific association with affective disorder rather than schizophrenia. This association deserves further testing.
SUMMARY AND CONCLUSION
Beginning with the efforts of Lombroso and Galton, a variety of evidence has been accumulated concerning the association between psychiatric illness and creativity. Despite the numerous studies, however, the results are obscured by a variety of methodologie problems. Some studies have relied only on historical or anecdotal data, and consequently they are likely to provide a serious underestimate concerning the prevalence of illness. Some studies have begun with subjects known to illustrate the association and thus reflect a built-in bias towards the closeness of the association. Others have denned terms rather generally and have equated genius with eminence or high intelligence; the results of these studies cannot necessarily be generalized to people who are creative only. Still others have used psychiatric terminology without careful definition of diagnostic categories, so it is difficult to tell whether or not the results of various studies are comparable with one another.
In spite of these limitations, some tentative conclusions can be reached concerning the nature of the creative person and the creative process. First, the creative person is highly intelligent, although intelligence alone is not enough to produce creativity, and he combines a complex combination of personality traits, such as both boldness and sensitivity. Second, whatever type of diagnostic system is used, the creative person whose talent is expressed through artistic or scientific achievements, especially when notable recognition is attained, seems to have more psychopathology than would be expected from population norms. Most of the symptoms are consistent either with affective disorder or with personality disorder, and considerable evidence indicates that writers in particular are likely to suffer from affective disorder. Third, the families of creative persons may have a higher prevalence both of creativity and of psychiatric illness than occurs in the general population, and there is some suggestion that this pattern may be explained in part on the basis of genetic factors.
Many interesting questions remain to be answered and hypotheses to be tested. Will the association between creativity and affective disorder occur in other creative people, such as musicians or scientists, besides writers? To what extent and in what way do genetic and environmental factors interact to produce creativity and psychiatric illness or health in the creative? Are there any physiologic, neurochemical, or psychologic variables that can act as "markers" for the creative person in general or for the creative person predisposed to developing an illness? What role should the psychiatrist play in working with creative persons? What effect does the treatment of psychiatric problems in creative persons have on the creative process? Does it limit or enhance creativity? Such questions as these are too compelling to be ignored. They demand that the careful scientific approaches to the study of creativity already undertaken by a number of investigators be continued with increasing refinement.
1. Lombroso, C. The Man of Genius. London: Walter Scott, 1891.
2. Galton, F. Hereditary Genius. London: Macmillan and Company, 1892.
3. Hirsch, W. Genius and Degeneration. New York: D. Appleton and Company, 1896.
4. Hyslop, T. B. The Great Abnormats. New York: George H. Doran, 1925.
5. Lange-Eichbaum. W. The Problem of Genius. London: Kegan Paul, Trench, and Trubner, 1931 .
6. Nisbet. J. F. The Insanity of Genius. New York: Charles Scribner's Sons, 1900.
7. Terman, L. M. Genetic Studies of Genius, 6 vols. Stanford, Calif.: Stanford University Press. 1925-1959.
8. MacKinnon, D. W. Personality and the realization of creative potential. Am. Psychol. 20 (1965), 273-281 .
9. Roe, A. The personality of artists. Ed. Psychol. Measurement 6 (1946). 401-408.
10. Drevdahl, J. E., and Cattell, R. B. Personality and creativity in artists and writers. J. Clin. Psychol. 14 (1958). 107-111.
11. Taybr, C. W., and Barron, F. Scientific Creativity: Its Recognition and Development. New York: John Wiiey & Sons, 1 963.
12. Barron, R Creativity and Psychological Health. Princeton, N.J.: D. Van Nostrand Company, 1963.
13. Barron, F, Creative Person and Creative Process. New York: Holt, Rinehart and Winston. 1969.
14. Barron, F. Artists in the Making. New York: Seminar Press, 1972.
15. Ellis, H. A. A Study of British Genius. Boston: Houghton Mifflin Company, 1926.
16. Juda, A. The relationship between high mental capacity and psychic abnormalities. Am. J. Psychiatry 106 (1949). 296-307,
17. Andreasen, N. C, and Canter. A, The creative writer: Psychiatric symptoms and family history. Compr. Psychiatry 75 (1974), 123131.
18. Andreasen, N. C., and Canter. A. "Genius and insanity" revisited: Psychiatric symptoms and family history in creative writers. In Wirt, R1, Winokur, G.. and Roth, M. (eds). Ufe History Research in Psychopathology, Volume 4. Minneapolis: University of Minnesota Press. 1975.
19. Heston, L. L. Psychiatric disorders in foster home reared children of schizophrenic mothers. Br. J. Psychiatry 112 (1966), 819825.
20. McNeil, T. F. The Relationship Between Creative Ability and Recorded Mental Illness. Unpublished doctoral dissertation, University of Michigan, Ann Arbor, 1969.
21. McNeil, T. F. Prebirth and postbirth influence on the relationship between creative ability and recorded mental illness. J. Pers. 39 (1971), 391-406.
22. Karlsson, J. L. Genetic association of giftedness and creativity with schizophrenia. Hereditas 66 (1970), 177-182.
PREVALENCE OF PSYCHIATRIC ILLNESS IN CREATIVE AND EMINENT PERSONS: THREE STUDIES COMPARED
PSYCHIATRIC ILLNESS IN THE RELATIVES OF CREATIVE OR EMINENT PERSONS