Psychologist who created the modern scientific theory of personality


Mention ‘science of personality’ and you may be accused of oxymoron. Yet this science exists, is important and flourishes. It is largely the creation of one man — Hans Eysenck — who died on 4 September, aged 81.

Eysenck was born in Berlin, but left in 1934 because he abhorred Nazism. He came to London University with the intention of reading physics, but his previous studies failed to meet the admission criteria. So he registered for psychology as the nearest thing to science that they would allow.

When Eysenck began his work in the 1940s the field was, frankly, a mess. Thousands of personality traits had been identified, labelled and measured, mainly by short questionnaires. Attempts to simplify this profusion were based on multivariate statistical analyses. These aimed to reduce the surface pattern of correlations between all possible pairs of measurements to a smaller number of underlying factors which could then, statistically speaking, reproduce the main features of the surface pattern. But there was no agreement on the correct form of analysis, although everyone except Eysenck agreed that a purely statistical solution would be forthcoming (if only it were possible to persuade the others that one's own was the best!).

Eysenck cut through the verbiage with a first, cardinal insight — no solution can be established on purely statistical criteria, and any suggested solution can serve only as a hypothesis which then needs to be put to the standard scientific tests of prediction, test and verification. He showed that all of the solutions on offer were mathematically equivalent and intertranslatable. So the argument was not about the maths, but about which version of the maths best applied to the true ‘structure of personality’.

If there is such a thing as a structure to personality, what gives rise to it? Eysenck's second great insight was that the source could lie only in the organization of the brain: whatever genetic, environmental or social influences contribute to a person's individuality, they must have acted by shaping the way in which the brain functions. However, certainly in the 1940s and 1950s, it was impossible to study the relevant features of brain function directly, so inferences needed to be made from experimental studies of behaviour. In this strategy, as he acknowledged, Eysenck followed in the footsteps of Ivan Pavlov. But whereas Pavlov started from observations of behaviour in animals and speculated about human personality, Eysenck started with the measurement of individual differences in human behaviour and sought explanations in the mammalian brain on the basis of animal behaviour. Because he had to draw on what was known in these more basic, relatively underdeveloped fields, the details of his extraordinarily all-encompassing theory of personality were bound to be, in many respects, wrong. Yet he sketched the basic logical shape that such a theory should take in a way that, I believe, will be lasting.

At the time, almost everything that Eysenck wrote on the topic was the subject of hot dispute. Today, there is widespread agreement that his essential postulates were correct. For example, in the best traditions of scientific parsimony, he held that there are only a few (three, in his view) fundamental and independent dimensions that define the space in which human personality can vary. He doggedly defended this position against an opposition that has retreated from an initial minimum of 16 to a modal five at present, two of which (extraversion and neuroticism) are identical to his. He also held that much of the variance along his three main dimensions of personality

(psychoticism being the third) reflects the cumulative action of additive polygenes, and that psychiatric disorder is often due to extreme positions on the resulting continuous distributions of vulnerability. This view — which was initially rejected out of hand by clinicians and social scientists alike — is now so generally accepted that it has sparked an international race to identify the genes that determine high levels of neuroticism and, thus, disorders of anxiety and depression.

Undoubtedly, the construction of a scientific theory of personality was Eysenck's most important and enduring achievement. For this alone he deserved the honours which, in his own country, he (scandalously) never received. In part, the reason for this neglect lies in his activities as advocate and polemicist. His first battle was against psychoanalysis. In 1946, Eysenck was appointed to the Institute of Psychiatry (where he spent the rest of his career) by the psychiatrist Aubrey Lewis. One of his objectives was to start up a training course for clinical psychologists. Having scoured the United States (which was then dominated by psychoanalysis), he spurned all of the examples that he found there and established his own new training model, based on evidence from basic and clinical experimental research.

This model has since been copied throughout the world, and it has contributed much to what has become the most effective method of treating a range of neurotic disorders — cognitive- behavioural therapy. Although Eysenck made little direct contribution to these therapeutic advances, his advocacy was very important in spreading the word. But in the process he alienated a whole generation of psychodynamic psychiatrists. To these he soon added other enemies, as he espoused a variety of politically incorrect causes: he argued that Communists and Fascists have similar personalities; that one cannot exclude the possibility that race- or sex-differences in IQ are, at least in part, genetically based; and that the correlation between smoking and disease may be mediated by a common factor, such as personality type.

At a personal level, Eysenck was the kindest and most courteous of men. In print, however, he loved a fight, using weapons of the intellect but with no holds barred. For this pugnacity his reputation undoubtedly suffered; but his position as the most highly cited psychologist of his generation will, I believe, be matched in the history books.