Grand story of delusions

Would it have been possible for psychiatry to have been built on an entirely different set of symptoms and behaviours? Why were certain observations and concepts chosen out of the confusing plenitude of mental abnormalities? Why, for example, did the notion of "delusion" assume such a central position in the characterisation of mental disorder? Might an entirely different system of description have been possible in the past; or might one become possible in the future, perhaps more closely tied to neurobiological abnormalities that many hope will eventually define the major mental disorders? Or, as the author of this volume seems to suggest, does the apparent stability of descriptive psychopathology over the past 100 years argue for "cognitive categories for the description and organisation of the world which are actually wired into the human brain"? A historical analysis of ideas in psychopathology should help us to answer such questions better.

Writing a history of psychopathological symptoms presents formidable problems. Its study requires expertise and depth of knowledge in the methods of history (including special aspects of the history of science) as well as in psychiatry. Few people are qualified for such a task, but German Berrios, a psychiatrist at Cambridge University who has a training in historical methods as well as philosophy, is one. He has written extensively on the history of psychiatry as well as being a practising clinician and respected researcher of a more traditional kind.

This book concentrates on the 19th century, during which time our concepts of abnormal mental states evolved most rapidly. Berrios describes it as a period of major "rupture and recasting" in discourse about insanity. Most would agree that the psychiatry of the end of the 19th century is essentially the psychiatry of today.

After a disappointingly brief introduction to significant historical movements in psychology, especially associationism and faculty psychology, and of "descriptive psychopathology as a cognitive system" - "a device to organise knowledge of a particular type", Berrios launches directly into 16 chapters dealing with most psychopathological states as we know them today. Interestingly, and without comment by the author as to why this particular schema was chosen, they are grouped under the three basic categories of mental life derived from early 19th-century faculty psychology. First, cognition and consciousness (disorders of perception, delusions, obsessions, cognitive impairment, disorders of memory, disorders of consciousness) - what might have been termed the "intellectual insanities". Second, mood and emotions (anxiety and cognate disorders, affect and its disorders, the anhedonias) - the "emotional insanities". And, third, volition and action (the will and its disorders; feelings of fatigue; catalepsy, catatonia and stupor; and a variety of abnormal movements - tremor, rigidity, akathisia and stereotypy - the "volitional insanities". A final section, "Miscellany", examines personality and its disorders, and self-harm (more accurately suicide). Many of the chapters are based on papers previously published in academic journals.

This book is novel in scope; as far as I know, no one has attempted a history of psychopathological concepts across virtually the range of mental disorders. It is also a book of impressive scholarship. The range of sources investigated is staggering. The "name index" runs to more than 2,000 and the references section, excluding footnotes at the end of each chapter, to almost 80 pages. Berrios is obviously a gifted linguist and gives non-German and non-English writers due attention. A major contribution is his elucidation of the centrality of French thinking in the first half of the 19th century during which, he argues, most of the major advances, nowadays often credited to the German psychopathologists of the second half of the century, were made.

The book can be highly illuminating. The transition from an "intellectualistic" descriptive psychopathology to one that embraced disorders of affect and volition is well treated. It is especially helpful in understanding the tenuous connections between "melancholia" as conceptualised up to the 19th century - primarily as a disorder of thinking and behaviour and depression today - conceived primarily as a disorder of mood. Similarly, the discussion of the emergence of the notion of "consciousness", and hence of its potential disorders, clarifies the historical relationships between key concepts of delire, "confusion" and "delirium" - essentially from a state of excited behaviour accompanied by fever, to a disorder of consciousness, including disturbances of attention, cognition and orientation. Throughout the book, Berrios adopts a sober, critical approach, quoting with approval a statement by the French historian of science Gaston Bachelard that "historical accuracy not be sacrificed to superficial order and dubious progressivist views".

There are, however, some significant weaknesses. The major one is a lack of overall cohesiveness. This derives from two sources. First, the range of the book seems impossibly ambitious. It is difficult to see how virtually the entire field of psychopathology can be covered in depth in a work of this size. Second, the origin of many chapters in separate papers is betrayed by haphazard repetition and poor cross-referencing. Major common themes, manifest in evolving ideas in several areas of psychopathology and treated in this book in separate chapters, are nowhere adequately brought together. Only gradually could this reader grasp them from a style often encyclopaedic in texture. There are also references to clearly significant but relatively unfamiliar writers such as Baillarger, de Condiallac and Chaslin, each at some point called "great", yet it is difficult to gain an overall view of their thinking. Thus the absence of an explicit, easily understood overall schema is frustrating, as are inconsistencies in quality and format across chapters. Not infrequently the book begins to read like an annotated bibliography, albeit an exceptionally good one.

The readership of this book is likely to be quite limited, since it is difficult to imagine that anyone not already well acquainted with the language of descriptive psychopathology will be able to wend their way through the text without feeling overwhelmed. It might have been helpful if each chapter had been preceded by an account of the modern definition of the psychopathological terms to be traced historically. Many technical terms are used without explanation, for example, the present-state examination, a semi-structured interview aimed at providing a comprehensive description of a subject's mental state.

For Berrios the historical analysis of the language of psychopathological description is part of a larger programme. Though thoroughly versed in historical methods, he maintains a keen interest in neuroscience. He believes that when the cultural, social and philosophical "noise" in descriptions of psychopathology is evaluated and accounted for, what will remain will be the invariant "biological signals" that are the core of the psychiatric disorders. Perhaps this is feasible, but given the din of competing descriptions and claims that will undoubtedly be impressed on the reader, one doubts that "signal" will be detected. Not really examined in any detail in this book are ways in which the social construction of abnormal mental states reflect shared assumptions between interviewer and patient and the extent to which these might fundamentally shape the patient's self-description and introspections. We will probably do better to recast pathologies of behaviour in an essentially new language that reflects demonstrated neuropsychological or neurocognitive abnormalities ascertained under specified conditions, and which are linked with theories of how brain systems regulate behaviour.

Taking his programme forward, Berrios is completing a companion volume in which he will suggest a "re-calibration of the language of description" to "reduce the mismatch that now exists between symptom description and the information provided by the new neurobiological research techniques". This is a task even more Herculean than the one attempted here. I look forward to seeing the results.

George Szmukler is consultant psychiatrist, Maudsley Hospital, London.