All scientific communities must keep themselves intellectually alive. It is important that we should attempt sometimes to tackle intriguing questions that are, strictly speaking, beyond the reach of robust, achievable, scientific methodologies. The small, flawed and inconclusive literature on the relationship between creativity and mental disorder, which includes contributions from some scientific heavyweights (e.g. Reference AndreasonAndreason, 1987; Reference PostPost, 1994) represents a respectable attempt to use empirical methods to explore one such intriguing question. The fact that these matters are unlikely ever to be resolved does not discredit the effort.

In this issue of the Journal, Wills (Reference Wills2003) has made a significant contribution to this literature. It will be particularly valued by that group of psychiatrists who prefer to frequent jazz clubs rather than conference gala dinners (those who prefer gala dinners are advised to obtain a copy of the CD reissue of Kind of Blue by Miles Davis). The paper raises interesting scientific issues, some of which Wills acknowledges.

WHO IS CREATIVE?

Jazz is a living, developing music based on the interplay between compositional and improvisational elements. Any attempt to identify the most creative players is bound to be difficult. Wills has focused on the period 1945–1960, which by common consent was the great classical period of small-group jazz, during which time the parameters and major genres of ‘modern jazz’ were established. After this came free jazz and a variety of fusion forms, and the boundaries of jazz became increasingly indistinct. Two great contemporary jazz musicians, Wynton Marsalis and Courtney Pine, both initially inspired by the classical period, have arrived at diametrically opposite positions on the value of these post-1960 developments. Wills has wisely confined his study to an uncontroversial period.

The exclusion from the study of musicians who are still alive is appropriate. However, it does create a somewhat idiosyncratic sample, excluding major figures such as Ornette Coleman and Sonny Rollins while including lesser (though very talented) players such as Wardell Gray and Howard McGhee. The inclusion of Charlie Christian is problematic. Although hugely influential in the period 1945–1960, he died at the age of 25 in 1942 (reportedly following a drug and alcohol binge while being treated as in-patient for tuberculosis). However, these are not necessarily fatal methodological flaws. There is no evidence that this introduces bias with respect to the presence or absence of mental disorder. As long as everyone included was very creative, the exclusion of some creative players is irrelevant.

Wills’ list of great musicians of the era is attractive and would command general support among jazz enthusiasts. However, this face validity conceals a problem. Such a list written in 1960 probably would have been quite different. For example, Ornette Coleman was widely regarded as a charlatan in the 1950s, whereas today his late 1950s/early 1960s work remains influential. Dave Brubeck (best known for the ubiquitous ‘Take Five’) was wildly popular at the time but has tended to be judged retrospectively as a ham-fisted pianist who vulgarised the form to the tastes of White middle-class youth (Reference Carr, Fairweather and PriestleyCarr et al, 1995). Historical judgements are not necessarily more accurate than contemporaneous judgements. The two great blues–rock guitarists of the mid-1960s were Eric Clapton and Jimi Hendrix. Clapton is frequently now dismissed as a plagiarist of Black bluesmen such as Freddie King, whereas Hendrix is said to have reinvented the instrument, his enduring influence extending into jazz and modern classical music. However, it was Clapton who made the key innovations, exploring the tonal and harmonic potential of the overdriven valve amplifier and introducing jazz-inflected extended improvisation. Hendrix's reputation has been preserved by his early death, which robbed him of the opportunity to make a truly bad record. Clapton's early ground-breaking work, on the other hand, has been obscured by his post-1970 output of unremittingly easy-listening music.

Within Wills’ sample there is a subgroup of outstanding musicians who changed the nature of jazz, namely John Coltrane, Miles Davis, Charlie Parker, Thelonious Monk, Charles Mingus and Dizzy Gillespie. Even among such undisputed talents, the elusive and heterogeneous nature of creativity is evident. For example, Davis was a technically average player who remorselessly explored new musical territory, whereas Coltrane was an outstanding virtuoso of limited compositional originality.

In the face of these uncertainties we cannot be confident that any methodology can capture a uniform and objectively creative sample.

CAN BIOGRAPHERS BE TRUSTED?

It would be very difficult to study mental disorder and creativity prospectively. Consequently, diagnoses can be derived only retrospectively from biographical and journalistic sources and these are not comparable with clinical sources. They were written in order to sell periodicals and books and there is likely to be an emphasis on the colourful and sensational aspects of the subjects’ lives. Different sources often conflict. Everyone seems to agree that Miles Davis was habitually unpleasant, which may have been due to aloof narcissism, defiance inspired by the civil rights movement or some other factor (Reference Davis and TroupeDavis & Troupe, 1990). Jimi Hendrix, on the other hand, was a timid, shy and anxious man or an aggressive, feckless libertine, depending on which account one prefers (Reference MurrayMurray, 1989).

Biographers taint facts with interpretation. Wills uncritically accepts that Thelonious Monk suffered from a dementing process caused by excessive drug usage. This assertion does not easily withstand critical scrutiny. Similarly, Coltrane is said to have shown signs of obsessive–compulsive disorder. However, ‘excessive practising’ and ‘searching for the perfect mouthpiece’ are not necessarily pathological behaviours in a man who achieved supreme mastery of his instrument.

MORE MENTAL DISORDER THAN WHOM?

If there is an association between mental disorder and creativity, then it is necessary to demonstrate that the rate of mental disorder among creative people is higher than among a valid comparison group. Comparison with rates of mental disorder experienced by the general population is not valid. Unfortunately, comparison with a group of truly talentless jazz musicians would be difficult to achieve. Admittedly the rates of mental disorder described by Wills do seem high. The rate of heroin dependence is very high, although the rate of bipolar affective disorder (frequently cited as the most likely specific diagnosis to be associated with creativity) is closer to the lifetime incidence for the general population than one might have expected. It is possible, however, that in the search for an association with creativity Wills and his predecessors have overlooked a simpler, more robust scientific finding.

IN SEARCH OF A NULL HYPOTHESIS

The most serious flaw in this field of research is that it is essentially rhetorical in nature. Eysenck (Reference Eysenck1995) and Andreason (Reference Andreason1987) both clearly articulated an underlying belief that some types of mental disorder might share a common genetic origin with creativity. Wills explores issues of social environment and personality, and this broader perspective is welcome. What he has not considered is the possibility that life as a jazz musician might actually cause mental disorder. There is a prima facie case. These mainly Black musicians were promulgating a radical and anti-establishment music in the era of McCarthyism and segregation. They were generally poorly paid, constantly toured and kept irregular hours. Drugs, alcohol and casual sex were freely available. Such a lifestyle was hardly compatible with good mental health.

As scientists, we can go further. We should consider a null hypothesis; that there is no association between mental disorder and creativity, neither positive nor negative.

Like jazz, pop/rock music had a great classical period during which the major genres were established, but it was brief, roughly from 1965 to 1970. Biographical material on a number of major figures from that era (e.g. Reference Buckley and EllinghamBuckley & Ellingham, 1996) clearly describe mental illnesses in the subjects consistent with the acute manifestations and long-term course of schizophrenia, although the symptoms are usually ascribed by biographers to the use of LSD, probably erroneously (Reference Poole and BrabbinsPoole & Brabbins, 1996). A few individuals are reported to have been in the throes of acute psychosis when they produced great and enduringly influential work. Although the material supporting this observation is in the public domain, it cannot be set out here, as most of the affected individuals are still alive, and, in some cases, still performing. This is interesting in itself, given the generally very high rate of early death among musicians of the era.

What this observation suggests, and what Wills and some of his predecessors have surely demonstrated, is that even severe mental disorder is not incompatible with creativity, that there is no negative association between the two. This is in contrast to the expected consequences of disorders that are destructive to a wide range of mental functions. The scientific lesson here is that people cannot be suppressed completely by a mental disorder or fully described by a diagnosis. The human spirit can defy all types of adversity and the mentally ill can produce great art that communicates meaningfully to the rest of us. We should feel humbled by such findings.

DECLARATION OF INTEREST

The author is a practising psychiatrist and gigging musician in a jazz–blues band.