Shrinks wrapped

Is it better to be a doctor and an amateur historian, or the other way round? Can someone who has never been responsible for patients really understand the practical dilemmas of everyday practice, particularly in some earlier period? Or can a practising doctor ever approach the history of their own profession with truly academic detachment? These possibly unanswerable questions emerge most often in the case of psychiatry because social factors are so central to the management of mental illness, and because of its inevitable involvement with government. But if the skills of the two professions are combined, do these demarcation problems then disappear?

A History of Clinical Psychiatry is the joint editorial product of a social historian of medicine (Roy Porter) and a neuropsychiatrist (German Berrios) - each distinguished in his own field. They are no strangers to each other, since they are also the joint editors of the journal History of Psychiatry. Their dyadic relationship is reproduced throughout this very substantial work, with nearly every chapter having both a clinical and a social section. In fact, several of the social sections have medical authors, which might be taken by some as evidence of doctors' imperialism, though that would not be justified here. The international cast of contributors includes such well-known names as Edward Shorter and Harold Merskey.

Through their "clinical" title, Porter and Berrios have separated this volume from most of the considerable recent output of historical work on psychiatry, which has focused mainly on its institutional and legal aspects. By contrast, this is a syndrome-based account, covering 26 disorders that are either "mental" or "neuropsychiatric" ("neurological conditions that include clear psychiatric symptoms"). Hence the rather surprising inclusion of Parkinson's disease and multiple sclerosis, both usually seen as the province of the neurologist. But the section on functional psychoses relates three chapters to individuals - Emil Kraepelin, Carl Wernicke, and Karl Leonhard - a choice which shows the pivotal role of German-speaking clinicians in creating the basis of modern psychiatry. The explanation for this arrang- ement is that it would be "impossible to grasp" the evolution of the psychoses without knowing something about these pioneers, but it is not clear why the same method has not been used with other disorders.

Nosology, the rules by which symptoms are clustered together into diseases, can be understood through two main metaphors: either as the identification of pre-existing clinical phenomena (as a gardener catalogues species), or else as the creation of clinical entities out of the unorganised characteristics of patients (as a sculptor produces a statue). Michael Clark suggests that to varying extents, all "clinical" pictures are "highly selective abstractions from pre-existing social, psychological, and cultural activities". In historical terms, this raises two key questions - has the use of words such as "mania" or "melancholia" changed over time, and have their clinical features, which are believed to be both biologically and socially determined, also changed? If in fact these latter remained constant for long periods, as has happened with some disorders, it would seem to indicate that biological factors are the stronger in those cases.

Probably the most important controversy that arises out of such questions is whether or not schizophrenia affected significant numbers before the 19th century. It has been argued that its apparent "newness" then was simply an artefact of the aggregation of people through industrialisation. Surprisingly, John Hoenig and Trevor Turner, who deal with schizophrenia here, have decided not to attempt an answer, but keep to the modern period.

Because psychiatry is so much concerned with abnormal behaviour, which may have many different causes, the question has often arisen whether it is dealing with a disorder (as understood medically) or with a "social construction". One of the conditions where this dilemma most often arises is substance abuse. Edward Brown describes the attempted institutional management of alcoholism in the 19th century as "both unpalatable and unworkable", pointing out that the most successful form of "treatment" subsequently - Alcoholics Anonymous - "continues to rely on an unstable mix of the language of self-control and the language of disease". Possibly current work on the genetics and neurochemistry of alcoholism will legitimise one of these two approaches.

If German writers dominate the middle period of this account, with the French in second place and Britons making an influential contribution now and then, the later aspects of most disorders reflect the power of American psychiatry. Its DSM diagnostic system has become almost unavoidable in research, though as P. Hoff points out, these operational categories are not "definite or natural entities, but . . . conventions which need further clinical verification".

The section on panic disorder follows what has become the received wisdom: that it was identified in America within the last 20 years. But this does less than justice to the Newcastle group headed by Martin Roth who described the 'phobic-anxiety depersonalisation syndrome in the 1950s; mainly because no effective treatment was available then, it was an idea whose time had not yet come. It is rightly pointed out, though, that it has only been the reduced background anxiety about daily life in industrialised countries which has allowed the experience of intense or persistent anxiety to be seen as pathological.

Though the claim that this is "designed to be the standard reference work on the subject" is a rather bold one, it is unlikely that its detailed history could be surpassed for many years. Yet the volume of detail sometimes makes it difficult to see the wood for the trees, and there is still room for some comprehensive, stellar treatment which would provide an integrated narrative of the whole subject. To make the same historical journey many times - for each disorder separately - has disadvantages, though it suits the reader who wants to know about one condition at a time. Others may be interested in the way clinical psychiatry evolved, say in the 1920s, but that will take some considerable delving in these 684 pages.

Hugh Freeman is former editor of the British Journal of Psychiatry and honorary visiting fellow, Green College, Oxford.