Related topics

Melvyn Bragg, hysteria and the indubitable flatness of the earth

Freud, Charcot and hysteria

Hysteria revisited

Rediscovering the unconscious

Freud's legacy

Freud, Satan and the serpent

Freud's false memories

The bewildered visionary

Flirting with Freud

Lacan goes to the opera

The cult of Lacan

Hysteria, medicine and misdiagnosis


From Why Freud Was Wrong: Sin, Science and Psychoanalysis (Revised edition, Harper Collins 1996).

This essay combines a section from the opening of Chapter 5 of Why Freud Was Wrong with the Appendix ('The Diagnosis of "Hysteria"') which was originally written as a continuation of that section. The book itself also contains two lengthy chapters on Charcot -  'Hypnotism and Hysteria' and 'Charcot's Mistake'.

For an account of the role played by misdiagnosis in the development of psychoanalysis, see the extract from my Freud (Weidenfeld, 2003):
Freud, Charcot and hysteria. For a recent long but uncompleted essay, see Hysteria revisited (2007).

misconceptions about the history of psychoanalysis is the belief that Freud’s early patients came to him because they were suffering from emotional difficulties or because they displayed symptoms which clearly had a psychological origin. The reality was very different. A large proportion of the patients whom Freud treated during his early years in private practice had initially sought medical advice because they were suffering from physical symptoms; they had enlisted the help of a physician for no other reason than that they believed themselves to be ill. Among their symptoms were headaches, muscular pain, neuralgia, gastric pain, tics, vomiting, clonic spasms, petit mal, epileptoid convulsions, and a host of other physical reactions.
[1] It was Freud who, by either making or confirming a diagnosis of hysteria, came to the conclusion that the origin of these symptoms was to be found in his patients’ emotional lives – and specifically in the traumatic events which had supposedly given rise to their illnesses.

This consideration is extremely important in any assessment of the early history of psychoanalysis. For, whether or not Josef Breuer’s case of Anna O. was founded upon a misdiagnosis, it seems likely that some of Freud’s own cases were. Freud, indeed, would be unusual among nineteenth-century nerve specialists if he had not misdiagnosed a considerable number of his patients. This is because he practised at a time when medical science had only just begun to emerge from a long period of extreme diagnostic poverty. Many of the most basic diagnostic techniques which are taken for granted by modern physicians had still to be discovered. The lumbar puncture, which is the only way in which Breuer could have tested his momentary hunch that Anna O. was suffering from meningitis, was not developed until 1891, and was not in general use until the early part of the twentieth century. X-rays, which would eventually become one of the most useful of all diagnostic aids, were discovered only in 1895 – the same year in which Studies on Hysteria was published. The electroencephalogram, which would revolutionise neurology and psychiatry and lead to the final definition of temporal lobe epilepsy, was not invented until 1929, and was not in general use until the 1940s. Many other basic techniques of neurological investigation would not be developed until even later. The computed tomography scan, for example, which uses X-ray transmission readings to generate an image of the brain and which can display some lesions, tumours and other signs of pathology directly, began to be generally used only in the late 1970s. Not only were these diagnostic techniques unavailable to Breuer, Freud and their contemporaries, but neurology and psychiatry were relatively young and under-organised branches of medicine whose stores of knowledge were only just beginning to be built up.

Both medical historians and modern physicians sometimes underestimate the degree of diagnostic darkness to which their nineteenth-century predecessors had become habituated. This is partly because the direct evidence which might lead to a more realistic assessment is not always available. Doctors tend not to advertise their misdiagnoses any more than they are wont to display the corpses of their patients. Frequently, indeed, they are genuinely unaware of their own mistakes. Indirect evidence usually remains, however, and it is intriguing how often this too tends to be ignored. One of the reasons is that many medical historians are themselves physicians and are interested primarily in a view of medicine which portrays it as a continual progress towards the pinnacle of the present day. By such orthodox commentators ‘medicine’ tends to be invisibly re-defined as ‘successful medicine’. The result is that they end up writing a Whig-history of their own profession, concentrating on real medical breakthroughs. The mistakes, misdirections, deceptions and self-deceptions in which the larger part of medical history consists disappear almost completely. [2]

One of the facets of medical history which tends to be obscured in this way is the manner in which disease-syndromes have frequently been brought into existence by doctors not because they correspond to any real clinical entity, but in order to provide a refuge from diagnostic uncertainty. One example of such a ‘syndrome of convenience’ is provided by neurasthenia – which was invented in 1869 by the American physician George M. Beard, and which would eventually play a significant role in psychoanalysis (see below, Chapter 8). The possibility which we must consider, however, is that hysteria itself should be understood as just such a syndrome.

This view has been canvassed by a number of psychiatrists and neurologists ever since the time of Charcot – and sometimes as a direct response to the clinical vagaries of Charcot’s work. In 1908, for example, Steyerthal predicted that: 

Within a few years the concept of hysteria will belong to history ... there is no such disease and there never has been. What Charcot called hysteria is a tissue woven of a thousand threads, a cohort of the most varied diseases, with nothing in common but the so-called stigmata, which in fact may accompany any disease. [3]  

But although agnosticism about the concept of hysteria has received significant support within the psychiatric profession (particularly in the United States), the problem has by no means been completely resolved. In Britain, and in some parts of continental Europe, hysteria is still referred to as though it were a distinct syndrome in a number of psychiatric textbooks, and some neurologists, psychiatrists and physicians still believe that the concept is a useful one. In its current usage the term ‘hysteria’ bears almost no relationship to its original meaning. For it no longer refers to a disorder of the womb. Instead it is used to refer to any symptom or any abnormal pattern of behaviour for which there is no apparent organic pathology and which is therefore believed to be a product of emotional distress, anxiety or some other psychological cause. Those who propose that hysteria might be an entirely unnecessary concept readily accept that it is sometimes difficult to find an organic pathology behind certain physical symptoms. They merely suggest that, since the term ‘hysteria’ does not refer to any specific or definable disease, it is a sham-diagnosis rather than a real one. If all patients who appear to be suffering from physical symptoms but who have no detectable organic pathology are to be dubbed ‘hysterical’ then, they argue, the concept of hysteria becomes so broad and so vague as to be quite meaningless. Hysteria, in effect, ceases to be the very specific disease entity it was always historically considered to be, and becomes merely a negative assertion about the nature of certain symptoms. The adjective ‘hysterical’ is therefore used as though it were a synonym for ‘non-organic’ or ‘psychogenic’. At the same time, however, quite inconsistently, the noun ‘hysteria’ is used as though it referred still to a positive disease-entity and patients are actually said to be ‘suffering from hysteria’. Since, in the current usage of the concept, this is tantamount to claiming that a particular patient is suffering from physical symptoms which cannot be explained, it would be much better, in the view of some thoughtful psychiatrists and neurologists, if the term ‘hysteria’ were abandoned completely.

One of the most damaging effects of the term ‘hysteria’ in the past is that it has encouraged doctors to think they have arrived at a diagnosis of symptoms which, in reality, remain mysterious. This in turn means that it is much easier for doctors to miss real but obscure organic illnesses. The point has been well made by the psychiatrist Eliot Slater: 

The diagnosis of ‘hysteria’ is all too often a way of avoiding a confrontation with our own ignorance. This is especially dangerous when there is an underlying organic pathology, not yet recognised. In this penumbra we find patients who know themselves to be ill but, coming up against the blank faces of doctors who refuse to believe in the reality of their illness, proceed by way of emotional lability, overstatement and demands for attention ... Here is an area where catastrophic errors can be made. In fact it is often possible to recognise the presence though not the nature of the unrecognisable, to know that a man must be ill or in pain when all the tests are negative. But it is only possible to those who come to their task in a spirit of humility.

In the main the diagnosis of ‘hysteria’ applies to a disorder of the doctor–patient relationship. It is evidence of non-communication, of a mutual misunderstanding ... We are, often, unwilling to tell the full truth or to admit to ignorance ... Evasions, even untruths, on the doctor’s side are among the most powerful and frequently used methods he has for bringing about an efflorescence of ‘hysteria’. [4]  

Eliot Slater developed his sceptical attitude towards the diagnosis of ‘hysteria’ only after a great deal of research. This included a meticulous study of eighty-five young or middle-aged patients who had received the diagnosis of ‘hysteria’ at the National Hospital for Nervous Diseases in London during the years 1951, 1953 and 1955. The most important and the most surprising findings of this study were, as he himself put it, ‘the gravity of the after-history and the frequency of misdiagnoses’. During a follow-up period which averaged only nine years, twelve of the eighty-five patients had died, fourteen had become totally disabled and sixteen partially disabled. Most of these cases of death or disability were due to organic illnesses which had been mistaken for ‘hysteria’. Among the conditions which had been misdiagnosed either by neurologists or by psychiatrists – including Eliot Slater himself – were three cases of vascular disease, three of tumour and a number of cases where supposedly hysterical black-outs and fits were subsequently rediagnosed as epileptic. Four of the deaths were due to suicide, but in two of these instances the patient had suffered from organic diseases which had not been diagnosed by doctors at the National Hospital. One was a man suffering from various symptoms, including pain in the legs, unsteadiness of gait and impotence. Although Slater himself had diagnosed ‘hysteria’, the man was later admitted to another hospital and found to be suffering from disseminated sclerosis. In another case a woman who complained of severe headaches and poor vision was held to be suffering from ‘drug addiction and hysteria’. She was transferred to the Maudsley Hospital, from which she discharged herself after two weeks, her illness having been diagnosed as ‘conversion hysteria’. Two years later she died of a brain tumour.

After discussing these and many less serious misdiagnoses and placing them in the context of medical history, Slater comes to the conclusion that the diagnosis of hysteria has no validity whatsoever – a conclusion which he states in even more outspoken terms than in the essay cited earlier: 

Looking back over the long history of ‘hysteria’ we see that the null hypothesis has never been disproved. No evidence has yet been offered that the patients suffering from ‘hysteria’ are in medically significant terms anything more than a random selection. Attempts at rehabilitation of the syndrome, such as those by Carter and by Guze, lead to mutually irreconcilable formulations, each of them determined by their terms of reference. The only thing that hysterical patients can be shown to have in common is that they are all patients. The malady of the wandering womb began as a myth, and as a myth it yet survives. But, like all unwarranted beliefs which still attract credence, it is dangerous. The diagnosis of ‘hysteria’ is a disguise for ignorance and a fertile source of clinical error. It is, in fact, not only a delusion but also a snare. [5]  

Eliot Slater’s views have exercised considerable influence on psychiatrists and neurologists over the past thirty years and the use of the term ‘hysteria’ has declined in consequence. In the United States the diagnosis has, in theory at least, disappeared from mainstream psychiatry. Yet there appears to be a significant gap between theory and practice. If we are to believe the psychiatrist Philip Slavney, writing in 1990, the term still enjoys some currency even in American medical practice: ‘Despite condemnation from physicians and feminists ... the concept of “hysteria” is alive and well in the practice of medicine. No term so vilified is yet so popular; none so near extinction appears in better health.’ [6]

As these words suggest, the questions raised by Slater’s argument are very far from having been resolved. While this is not the place for a complete review of the problem, the continued confusion which surrounds the concept of ‘hysteria’ makes an abbreviated account seem necessary.

It should already be evident that Slater’s position is not new or revolutionary. A small anthology of agnostic reactions to the concept of ‘hysteria’ is contained in Aubrey Lewis’s paper ‘The Survival of Hysteria’ (1975). As early as 1874, W. B. Carpenter objected to the view that hysteria was a specific illness the grounds that ‘there is no ... fixed tendency to irregular action as would indicate any positive disease.’ In 1899 J. A. Ormerod suggested that the objections to ‘hysteria’ were obvious: ‘not only that it has become etymologically meaningless but also that to many minds it has the disagreeable connotation of a certain moral feebleness in the patient, and of unreality in the symptoms’. In 1904 the Swiss psychiatrist Dubois wrote that ‘hysteria’ should not be regarded as a disease entity, and in 1908 Steyerthal pronounced the unequivocal rejection of the idea that hysteria was a disease which has already been quoted (see above, Chapter 6). In 1911 Gaupp summarised the reaction which had by then taken place against Charcot: ‘Nowadays the cry is ever louder: away with the name and concept of hysteria: there is no such thing, and what we call hysteria is either an artificial, iatrogenic product, or a melange of symptoms which can occur in all sorts of illnesses and are not pathognomonic of anything.’

In 1925 Bumke looked back on the history of psychiatry and wrote ‘There was once a disease hysteria, just as there was hypochondria, and neurasthenia. They have disappeared. The syndrome has replaced the disease entity.’ In 1953 Kranz put forward his view that 

hysterical phenomena are only modes of reaction which fundamentally are available to everybody and are not in themselves abnormal, but become so in that they last unduly long, become fixed or are excessive ... It is reasonable to ask that we should at least drop the word ‘hysteria’ in favour of ‘hysterical reaction’, and in the end give up this term to, loaded as it is with moral value judgments: we can make ourselves understood by psychiatrists without it. But in spite of all that ‘hysteria’ will not disappear altogether from psychiatric vocabulary for a long time to come. [7]  

Kranz’s prophecy has proved accurate, especially with regard to the situation in Britain. Although Eliot Slater’s subsequent attempt to dislodge the concept of ‘hysteria’ was probably as influential as any of the earlier interventions, neither his arguments nor the conclusions he drew from his research have been universally accepted. Aubrey Lewis, having anthologised the views quoted above, describes how he conducted his own follow-up inquiry on patients diagnosed as hysterical at the Maudsley Hospital. He reports that he did not find any significant incidence of misdiagnosis, and that therefore his study did not bear out Slater’s conclusions. He notes that a significant divergence between the results of a study based on a neurological hospital, and those of a study made at a psychiatric hospital was only to be expected. Lewis still draws the conclusion, however, that the diagnosis of ‘hysteria’ is legitimate, ‘so long as it is regarded as a reaction’. He ends his paper by observing that ‘the majority of psychiatrists would be hard put to it if they could no longer make a diagnosis of “hysteria” or “hysterical reaction”; and in any case, a tough old word like hysteria dies very hard. It tends to outlive its obituarists.’

Even before Lewis’s reply, the neurologist Sir Francis Walshe, writing in the British Medical Journal in December 1965, sought to rebut Slater’s argument, seeing in it ‘a challenge to neurologists once again to justify the concept of hysteria as a nosological entity in its own right’. In a remarkable paper Walshe passionately restates many of the central doctrines associated with the traditional concept of ‘hysteria’, some of them dating back to Sydenham and beyond. Thus he reiterates the ancient view that ‘hysteria’ often takes the form of ‘a mimesis or ... caricature of disturbances on the physiological and morphological levels’ and goes on to stress, again in traditional terms, that ‘in view of the polymorphic manifestations of hysteria, diagnosis and psychological study present peculiar difficulties’. Although generally commending the views of Babinski, Walshe expresses regret that he declined to see hysteria as compatible with deception and ‘pathological lying’ on the patient’s part – ‘for it has long been acknowledged that the hysteric is a master, or a mistress of this upon occasion, and it may be an integral element in what is essentially a psychical illness. Lhermitte has said that “hysteria is the mother of deceit and trickery.”’

Although Walshe stresses at several points in his article that physicians are fallible and prone to make mistakes, he simultaneously upholds a view of medicine in which it is implicitly regarded as a perfect science. He thus sees as one of the crucial characteristics of hysteria the presence in the patient of patterns of disorder that ‘plainly arise from mental dispositions’ and ‘which are not congruous with nature’s laws as observed in the physical and biological sciences ...’ Walshe’s final words about what he terms ‘the unity of hysteria’ maintain the traditional view expounded throughout his paper:

Whatever the kaleidoscope of its manifestations, I submit that its essential difference from somatic disease is that it constitutes a behaviour disorder, a human act, on the psychological level. An hysterical paraplegia is exactly this, but a compression paraplegia is not this at all.

Apart from the mimesis of somatic disease hysteria may present, the dramatizations, the exaggerations and the pathological lying are also behavioural disorders, part of the total expression of the abnormal psychical state which is hysteria. [8]  

Some of the other contributions to the debate have taken a similarly conservative view, and a number of psychiatrists seem surprisingly untroubled by the possibility that such a confused concept might increase the risk of misdiagnosis which is always faced by patients with obscure cerebral or neurological disorders. In other areas of psychiatry, however, the extraordinary variety of meanings which the word ‘hysteria’ has traditionally been made to bear, and the bewildering array of physical symptoms and mental states it has been invoked to explain, has given rise to concern. One response to such concern has been to attempt to resolve the problem through the adoption of new terminology. It was this approach which was taken by the American Psychiatric Association when the diagnosis of ‘hysteria’ disappeared from their Diagnostic and Statistical Manual of Mental Disorders in 1952. The shift away from traditional terminology has been consolidated in later editions. But although the concept of ‘hysteria’ is conspicuously absent from the list of recognised diagnoses, the manual does give criteria for the diagnosis of three disorders which are clearly derived from the traditional concept – ‘conversion disorder’, ‘somatization disorder’ and ‘histrionic personality disorder’. The research criteria for the diagnosis of ‘conversion disorder’ as given by the third edition (DSM III) in 1980 were as follows: 

A.     The predominant disturbance is a loss of or alteration in physical functioning suggesting a physical disorder. It is involuntary and medically unexplainable... ...

B.     One of the following must also be present:

     (1)     A temporal relationship between symptom onset and some external event of psychological conflict.

     (2)     The symptom allows the individual to avoid unpleasant activity.

     (3)     The symptom provides opportunity for support which may not have been otherwise available. [9]  

There are at least two apparent advantages of this approach. In the first place the disappearance of the label ‘hysterical’, with its pejorative and morally censorious overtones, is a considerable gain. In the second place the insistence that the physical symptom should be involuntary has the effect of separating this putative psychiatric disorder from deliberately feigned or simulated illnesses – a category which the traditional concept of ‘hysteria’ tends confusingly to embrace.

The DSM III definition of conversion order, however, is far from satisfactory. One major problem is that, although it excludes consciously simulated illness, it does not exclude the unconscious simulation of illness. What this means in practice is that patients with imaginary symptoms which have no apparent physiological basis have to be placed in the same category as patients whose symptoms seem real, but are not susceptible to medical explanation. The dangers of this approach should become evident if we consider the subsidiary indicators given for the disorder. Criterion (2) – that the symptom allows the individual to avoid unpleasant activity – is, it will be noted, scarcely specific to emotionally based disorders. Most forms of illness, from broken legs to acute appendicitis, create just such opportunities. Criterion (2) is thus rather like saying that a specific name may be given to a plant providing that its leaves are green. Though the restriction may create the illusion of rigour, the field of definition is not very much reduced. Something similar can be said about the next criterion. For since most illnesses provide an opportunity for seeking support – if only from a physician – criterion (3) is almost as empty as criterion (2). Among the subsidiary criteria this leaves only (1), which demands that there should be some kind of temporal relationship between the onset of the illness and ‘some external event of psychological conflict’. The most fitting response to this is perhaps Slater’s, in the words which are quoted in the main body of my text: ‘Unfortunately we have to recognise that trouble, discord, anxiety and frustration are so prevalent at all stages of life that their mere occurrence near to the time of onset of an illness does not mean very much.’ [10]

In view of the fact that the subsidiary criteria (1), (2) and (3) are objectively empty, or very nearly empty, it would seem that, in DSM III, the diagnosis of ‘conversion disorder’ relies almost entirely on the main condition and that therefore the only strict criterion is that the patient’s symptoms were medically inexplicable.

It is difficult not to draw the conclusion that, in formulating its criteria in this particular instance, the American Psychiatric Association did little more than take an old diagnostic error and give it a new name together with a new aura of respectability. Since the very concept of ‘conversion’ is specifically psychoanalytic, and since it is historically indivisible from Freud’s own idiosyncratic theories of ‘hysteria’, it further seems that the creation of the category ‘conversion disorder’ was a politically astute way of preserving the old concept of ‘hysteria’ in euphemistic disguise.

To say this is not to rule out the possibility that there can be a direct relationship between prolonged stress or severe emotional trauma and some physical symptoms. Many common disorders do seem to be stress-related. In most cases, however, we do not yet understand the precise physiological mechanism of such a relationship. To confer medical respectability on a label originally invented by a nineteenth-century nerve-doctor who put forward as a scientific fact an entirely fictional account of the pathology of ‘hysteria’ seems, on the face of it, an unsatisfactory way of dealing with medical uncertainty. To allow the resulting syndrome, which has supposedly been carefully delimited, to be equally applicable to real physical symptoms and imaginary or spectral ones (providing they are not consciously produced) is merely to compound the original confusion.

Since 1980, DSM III has itself been revised and the definition of conversion disorder has been modified yet again. But the underlying concept has remained unaltered. Meanwhile relatively new terms such as ‘conversion disorder’ and ‘somatization’ have not entirely succeeded in ousting the older terminology. As Aubrey Lewis predicted, the term ‘hysteria’ has outlived its obituarists, and is still sometimes used as a diagnosis.

The dangers of this situation feature prominently in one of the most searching contributions to the entire debate, C. D. Marsden’s paper ‘Hysteria – a Neurologist’s View’, which was published in Psychological Medicine in 1986. After reviewing the concept of ‘hysteria’, Marsden gives careful consideration to the problem of misdiagnosis. He cites Slater’s finding that 58 per cent of his series of ‘hysterical’ patients had an underlying organic illness and then quotes the work of Tissenbaum who, in a paper published in 1951, specifically warned of the danger of misdiagnosing patients with neurological disorders. No less than 53 (13.4 per cent) of a series of 395 patients with organic neurological disorders were originally wrongly diagnosed as suffering from psychiatric illness. This tendency towards misdiagnosis was particularly marked in the field of movement disorders, and among Tissenbaum’s patients as many as 40 per cent of those with Parkinson’s disease were initially diagnosed as suffering from psychiatric disorders. Since about half of all patients who are diagnosed as ‘hysterical’ have some kind of movement disorder as their main symptom, such a high rate of misdiagnosis is extremely significant. Marsden’s own experience, as a neurologist specialising in movement disorders, bears this out. He notes that in a standard diagnostic manual published in 1970 Engel lists a wide range of such disorders as symptoms of hysteria, including spasmodic torticollis, writer’s cramp, blepharospasm and spasmodic dysphonia. [11]

He goes on to observe that all these conditions are now thought to be manifestations of a physical disorder, namely torsion dystonia. In his experience, however, ‘50% or more of such patients are still initially misdiagnosed as hysterics.’ Marsden goes on to endorse one of the most significant of the arguments put forward by Slater: 

There can be little doubt that the term ‘hysterical’ is often applied as a diagnosis to something that the physician does not understand. It is used as a cloak for ignorance. In addition we can still recognise new neurological diseases. Not only can a patient’s symptoms be dismissed as hysterical because the physician makes a mistake out of inexperience, but also because the illness has only recently been identified.

Neurology has never been and is not static. Many neurological diseases are still not widely recognised ... No doubt there are many other neurological conditions still undiscovered. History tells us that there must be illnesses which presently we do not recognise but dismiss as ‘hysterical’ 

At this point in his argument, however, Marsden makes it clear that he does not believe that medical progress is likely to remove altogether the small percentage (1 per cent) of patients who make up the category which has tended in the past to be labelled 'hysteria’. Since there are still likely to be at least some patients who ‘exhibit symptoms and/or signs that cannot be explained by organic or functional disease’ the question which arises is how such patients are to be described. For, as Marsden writes, 

It is essential for communication between doctors and other health workers to have some form of shorthand to explain the state of affairs. Consider the paralysed patient who cannot walk, who may or may not have a mild paraparesis, but whose major problem is weakness or even total paralysis not due to organic or functional disease. How are we to convey this concept? 

Having noted Slater’s plea for the abandonment of the diagnosis of ‘hysteria’ he goes on to observe that neurologists have sometimes fallen into the trap of calling such symptoms ‘functional’ – ‘he had a functional paraplegia’. As Marsden points out, however, this common usage of ‘functional’ is actually a misuse of a word which correctly designates an illness which is presumed to be a real organic disorder but which has no visible pathology. Another alternative sometimes resorted to is ‘psychogenic’, as in ‘he has a psychogenic paraplegia’. But Marsden brusquely, and I believe quite justly, dismisses this usage by referring to the view of Aubrey Lewis. According to Lewis the word ‘psychogenic’ is ‘at the mercy of inconsistent theoretical positions touching on the fundamental problems of causality, dualism and normality. It would be as well at this stage to give it a decent burial, along with some of the fruitless controversies whose fire it has stoked.’ [12]

Next Marsden considers the alternative ‘conversion disorder’ which, as we have seen, has been endorsed by the American Psychiatric Association. He finds this slightly more palatable than ‘psychogenic’ but nevertheless goes on to reject it on grounds very similar to those which I have already given. For as he points out, this ostensibly neutral term presumes a particular pathogenesis which was described by Freud, a pathogenesis which Marsden finds unconvincing.

Having temporarily renounced use of the term ‘hysteria’, and having declined for good reasons to adopt the most readily available substitutes, Marsden now recounts his own quest for an alternative. After experimenting with ‘feigned’ or ‘simulated’, as in ‘a simulated paraplegia’, he eventually rejects this usage on the grounds that the words suggest deceit where none may be meant. ‘Most patients with neurological hysteria are not malingerers, and do not appear to be consciously pretending or trying to deceive.’ He considers and rejects ‘fictitious’ on similar grounds and finally toys with another possibility: ‘A fable is a fictitious tale, so why not “Aesop’s syndrome” – he has an Aesoplegia (Aesopsia, Aesothesia, etc.).’ This suggestion is perhaps not made entirely seriously, and Marsden goes on to suggest that the term ‘hysteria’ should be reconsidered. As he recognises, and as Slater cogently argues, one of the central objections to the medical use of the term is that is liable to be treated as a diagnosis rather than as a description. Marsden refers us back to Slater’s original (1965) paper in which this distinction was discussed by reference to Brain’s (1963) distinction between the adjectival and substantival views of hysteria, where the adjective was seen as implying a description of the symptom and the noun a disease. Marsden goes on to quote the relevant passage from Slater:

I shall endeavour to persuade you that, to use Brain’s terminology, the adjectival view can be maintained with some qualifications, whereas the substantival view cannot ... it would be legitimate, I believe, in a given instance to say that a particular symptom was ‘hysterical’; ... however one should be aware of the possibilities of error. There is no ‘hysterical’ symptom which cannot be produced by well-defined, non hysterical cause ... With such a caveat, then, the adjectival use may be allowed to pass. However, to suppose that one is making a diagnosis when one says that a patient is suffering from ‘hysteria’ is, as I believe, to delude oneself. The justification for accepting ‘hysteria’ as a syndrome is based entirely on tradition and lacks evidential support. No closely definable meaning can be attached to it; and as a diagnosis it is used at peril. Both on theoretical and on practical grounds it is a term to be avoided. [13]  

After making some adventurous detours to explore the possibility of new terminology, Marsden thus returns almost to the place where he first started and arrives at the conclusion that ‘“hysterical” remains the historical and the best choice to describe such symptoms, provided that the term is not used to imply a disease.’

It must be said that, occurring as it does in the course of one of the most thoughtful and constructive contributions to the debate, this view is disappointing. All the more so in view of the fact that, as we shall see, Marsden goes on to make a number of extremely cogent suggestions about how the symptoms he deems ‘hysterical’ should be treated.

The problem with his reversion to ‘hysterical’ is that it is nowhere defended by argument and is supported only by the invocation of Slater. Yet the passage which Marsden quotes smacks of a compromise which Slater makes in order to deflect criticism from an argument which some might consider extreme but which, if maintained consistently, ought to be recognised as moderate and reasonable. The central objection to Slater’s attempt to split the concept of ‘hysteria’ into an illegitimate noun signifying a (non-existent) disease and a legitimate adjective which can be used to describe some symptoms is that it introduces a crucial inconsistency into his argument, while at the same time it defies ordinary language-use. It is rather like licensing the use of the adjective ‘canine’ while denying the existence of dogs. Under such a semantic regime it would, of course, be perfectly possible for somebody who heard what he thought was a dog barking to talk of having heard a canine noise. But it would, strictly speaking, be illegitimate to draw the conclusion that this particular canine noise indicated the existence of a dog. The case of the dog who did not bark in the night is difficult enough. But it must be said that the case of the non-existent dog who repeatedly does bark is even more mysterious and more confusing. [14]

In taking over Slater’s ill-considered linguistic compromise Marsden makes it quite clear that, in his new usage, the word ‘hysterical’ will not mean the same thing as it traditionally meant in the past. The standard dictionary definitions of the adjective will therefore no longer apply. For, rather than implying a positive characterisation of a symptom, or clutch of symptoms, ‘hysterical’ will now be used simply as a way of referring, in medical shorthand, to ‘disturbances of function that cannot be explained fully by organic or functional neurological disease’. No emotional aetiology will initially be presumed and indeed the entire question of aetiology and pathology will remain an open question.

There can be no doubt that if in practice the term ‘hysterical’ could indeed be used in this radically new way a great deal of confusion would be cleared up. The difficulty is that Marsden’s usage depends essentially on his own private redefinition of the word. The fact that he seeks, in his paper, to launch this private usage into the high seas of public medical discourse is likely to make very little difference. For although Marsden has redesigned the word ‘hysterical’ internally and loaded it with new meaning, outwardly it remains identical to older vessels bearing the same name. It is therefore liable to the presumption that it carries the same cargo. Repeated use of the term, however much it has been privately redefined, will tend to strengthen the concept of ‘hysteria’ in all of its diverse traditional meanings – just as repeated use of the word ‘divine’ tends to strengthen the concept of God and, in some contexts at least, implies the real existence of such a being.

The immense difficulties of maintaining the adjectival form ‘hysterical’ while renouncing the substantive from which it is derived are illustrated by Marsden himself. For no sooner has this policy been formulated than a table is introduced entitled ‘The Six Rules of Hysteria’. Elsewhere in his paper Marsden uses the term ‘neurological hysteria’ without any reservations.

In view of this it is perhaps not surprising that Marsden’s attempt to invest the term ‘hysterical’ with a radically new meaning should eventually be invoked by a physician who seeks to endorse one of the old meanings. This is what happens when Marsden’s argument is referred to by the Freudian neuropsychologist Laurence Miller in his study of the neurological dimensions of psychoanalysis, Freud’s Brain: Neuropsychodynamic Foundations of Psychoanalysis. [15]

Miller follows Marsden by presenting to the reader a fascinating series of misdiagnoses in which a wide variety of genuine organic diseases have been misconstrued as hysteria. Some of these case histories are taken directly from Marsden’s paper. Miller also draws on many other sources. A particularly striking aspect of the case histories he presents is the frequency with which epileptic seizures are misdiagnosed as hysterical even though EEGs have been administered between episodes. What often happens is that, because electrodes are applied only to the scalp, deep seizure activity is not registered. In these cases the patient is rescued from the diagnosis of hysteria only by depth electrode recordings which confirm the presence of epileptic seizure activity. [16]

Yet, having presented all the evidence necessary to mount a massively sceptical attack on the concept of ‘hysteria’, Miller declines to submit to this evidence. Updating Freud’s conceptual vocabulary slightly, he puts forward a ‘neuropsychodynamic model’ of hysteria which

asserts that the psychical impetus provided by the person’s personality takes advantage of brain dynamics that are usually only seen in their boldest form in structural organic disorders of the brain but that may occur more transiently, more subtly, and in more complexly organised ways, interwoven with ordinary aspects of behaviour, when expressed in the form of ‘functional,’ or ‘hysterical’ symptoms. [17]  

With considerable daring, Miller now treads even closer to the brink of scepticism only to draw back again at the last moment. ‘If symptoms that were yesterday called hysterical,’ he writes, ‘are today considered to be (at least partly) organic because our modern knowledge of pathophysiology is greater than in the past, might not today’s hysterical symptoms just as naturally become tomorrow’s medical syndromes as our knowledge continues to grow?’ The question, as Miller acknowledges, was originally posed by Marsden. Marsden’s answer is one that he echoes and endorses: ‘Not necessarily ... because the discovery of new diseases probably cannot go on for ever, and such new diseases certainly will not account for many of the one percent (Marsden’s figure) of neurological patients presenting with bona fide hysterical symptoms.’ [18] From these words it should be reasonably clear what has happened to Marsden’s careful attempt to redefine the term ‘hysterical’. Without doing any significant violence to the words which Marsden himself uses, Miller has managed to convert bona fide hysterical symptoms which do not indicate the existence of hysteria into bona fide symptoms which do.

Miller’s appropriation of Marsden’s sceptical argument for his own unsceptical purposes should be set alongside an even more remarkable reading of his argument which is offered in Mark Micale’s survey of recent literature on the subject of ‘hysteria’. Having noted that Slater’s attack on the concept of hysteria was energetically resisted in some quarters, he cites as an example of such resistance ‘a prominent London neurologist’ who, we are told, ‘has reaffirmed the value of the diagnosis in neurological practice.’ In a footnote the neurologist is identified as C. D. Marsden and we are referred to the same article which I have discussed here. [19]

One reason why this whole argument continues to trouble physicians and other interested parties is that the questions of medical ignorance and medical progress raised by Miller are extremely important ones. One of the main problems in this area is that, as the history of medicine eloquently demonstrates, soundings taken by physicians of the depths of their own ignorance are notoriously unreliable. Whenever such soundings are taken it is almost invariably claimed that the waters are already shallow and that the dry land of absolute physiological knowledge will soon be in reach. In reality, however, the ocean of medical ignorance has remained both dark and deep and has concealed numberless shoals of undiscovered pathologies and physiological mechanisms. Writing in 1993 the psychiatrist Graeme Taylor pointed out that the tradition of identifying a disease as organic by the presence of structural lesions has been challenged more and more strongly in recent times ‘as it is now evident that many medical, psychiatric and neurological patients have complex dynamic disorders of function in the brain and/or other physiological systems.’ He goes on to suggest that medical research is likely to reveal many supposed psychogenic conditions as ‘“legitimate” disorders of physiological function’. [20]

Marsden himself is exceptionally alert to the possibilities of future research shedding light on symptoms which today remain unexplained. At the same time, perhaps because of his own specialism, he is also exceptionally aware of the high proportion of confirmed misdiagnoses which are associated with the traditional concept of ‘hysteria’. If we can for a moment disregard the question of terminology it is well worth considering the specific recommendations he makes regarding the treatment of those symptoms he classifies as ‘hysterical’. 

Since Marsden uses the term ‘hysterical’ to signify not a homogeneous class of symptoms but merely those signs which are not currently susceptible to medical explanation, he recognises that patients may manifest them for a variety of quite different reasons. The main purpose of classifying disparate symptoms as ‘hysterical’ is not to profess understanding of their nature but to emphasise that further investigations need to be made. The aim of these investigations, according to the scheme which Marsden offers, will be to come to at least a provisional conclusion as to whether the symptoms fall into any one of a number of different categories.

It may prove, on further investigation, that the initially unexplained symptoms are actually the signs of a recognised physical illness which is little known or whose symptoms are ambiguous. Alternatively they may be real symptoms of a disease which is not recognised. They may also be the product of some underlying psychiatric disorder such as schizophrenia or depression. For all these reasons Marsden emphasises that patients with inexplicable physical symptoms should be given further psychiatric and physical examinations. It is conceivable that these tests may lead the examining physician to conclude that the patient is exhibiting ‘abnormal illness behaviour’. Some patients, Marsden suggests, are driven by a desire to help the doctor make a diagnosis: ‘Their anxiety leads to elaboration or exaggeration of their real deficit.’ Another group of patients may enjoy puzzling or outwitting the doctor, while others benefit from ‘their so-called illness’ in financial, social or personal terms. Such ‘abnormal illness behaviour’ may be motivated ‘by fear of disease or death, or by reward as a result of the advantages of the invalid role, or both’. It may be adopted without any conscious awareness of its real motivation. In some cases, however, as in the case of malingering or simulation, it is acted out at a fully conscious level of the mind. 

Marsden goes on to observe that those who consciously simulate illness, or who exaggerate or elaborate real physical illness because of their fear, clearly employ normal brain mechanisms to produce their signs and symptoms: 

But what of those who appear to believe in their loss or distortion of neurological function, quite unconscious of the fact that their nervous system is operating normally, or at least much better than they think. In what way has their brain managed to dissociate conscious awareness from the mechanisms of sensation, movement, or even memory?

Is there a nervous mechanism that can suppress, for example, the conscious appreciation of sensory experience from the reception of sensory information by the brain, or the will to move from the cerebral mechanisms responsible for generating movements?

He goes on to point out that both sensory appreciation and willed voluntary movement involve consciousness and suggests that it is here that contemporary neurobiology faces a major challenge, for ‘the cerebral mechanisms of consciousness are not understood.’ Having discussed this problem he suggests that future research may eventually illuminate this entire field: ‘Exploitation of advanced neurophysiological techniques in those with hysteria may provide one way of studying the mechanisms involved in the generation of hysterical symptoms.’ 

Marsden thus ends his paper on a genuine note of scientific openness, showing himself refreshingly willing to admit the depths of current neurophysiological ignorance, as well as refreshingly determined that those depths should eventually be plumbed. It must be pointed out, however, that at the very same time that he does this, he inadvertently allows the illegitimate substantival form ‘hysteria’ back into the closing paragraph of the very paper in which he announces its banishment. The moral of this story should be clear: If non-existent dogs are encouraged to bark, it will not be long before they bound back from their quarantine-pen bringing non-existent diseases with them.

Both the pathology and the remedy for this particular outbreak of linguistic confusion can be traced, I believe, if we examine the relevant step in Marsden’s own argument. For it will be recalled that his decision to re-adopt the term ‘hysterical’ is made in response to his quest for suitable medical shorthand to convey the concept of symptoms which have no apparent organic cause. Shorthand should, by common consent, be succinct, objective and unambiguous. ‘Hysterical’ is certainly succinct. But it is neither an objective nor an unambiguous way of suspending judgement on the pathology of puzzling physical symptoms. If such symptoms are indeed to be described as accurately and objectively as possible, then perhaps they should be formally referred to as ‘unexplained physical symptoms’. This description may not add greatly to the scientific self-esteem of those physicians and psychiatrists who are obliged to utter it. But scientific self-esteem is not everything. Those medical practitioners who, suffering from ‘physics-envy’, attempt to invest medicine with more precision and certainty than the current state of medical knowledge allows, do a disservice to their profession and to their patients.

The fact that few doctors and few patients would be likely to rest content with such a formula is a point in its favour. ‘Hysterical’, though offered by Marsden merely as an interim label, sounds far too much like a diagnostic conclusion and might easily discourage further investigation. ‘Unexplained physical symptoms’ is patently not a diagnosis and invites – and indeed almost compels – further efforts towards understanding.
 . . . . x

The fact that few doctors and few patients would be likely to rest content with such a formula is a point in its favour. ‘Hysterical’, though offered by Marsden merely as an interim label, sounds far too much like a diagnostic conclusion and might easily discourage further investigation. ‘Unexplained physical symptoms’ is patently not a diagnosis and invites – and indeed almost compels – further efforts towards understanding.

If further investigation shows that there is strong, irrefutable evidence that the symptom is simulated then this verdict should be perhaps be stated by explicitly calling attention to the conscious process – ‘a consciously simulated paraplegia’ is unambiguous, whereas ‘a simulated paraplegia’ might be construed as unconscious. If the symptom is apparently real to the patient, but cannot be confirmed by medical tests, and therefore seems in some sense unreal, we are confronted by the same problem with which Marsden wrestles unsuccessfully in the course of his paper. One possibility which Marsden does not consider is the one suggested by Molière when he called his play about hypochondria Le Malade Imaginaire. Patients might be described as suffering from ‘an imaginary illness’ or ‘an imaginary symptom’. The problem with this usage is that it does not correspond to the experience of patients who genuinely believe that their symptoms are real. For normally we recognise the products of our imagination as such; a novelist does not usually ask his characters to dinner or invite them to stay for the weekend. It is because the term ‘imaginary’ is deficient in this respect that it might well be worth considering an alternative: ‘spectral’. The advantage of this term is that it does correspond to the experience of many patients, and to the observations of many physicians. A ‘spectral’ symptom is a kind of physiological ghost. Like a ghost it can seem completely real to the person who experiences it, and for this reason it can generate strong emotional reactions, such as fear. But, like a ghost, a ‘spectral’ symptom appears to have no physiological substance. This may be because it is indeed the product of the patient’s imagination. But it might also be a kind of physiological hallucination – a product of exactly the kind of subtle neurophysiological disorder of consciousness on whose existence Marsden speculates at the close of his own paper.

There may well be good reasons for not adopting the term ‘spectral’. But it would be difficult to claim that the adequacy of current medical terminology should be counted among them. For even where the concept of ‘hysteria’ has been discarded as old-fashioned, a great deal of confusion still seems to be associated with the terms which have been adopted in its place. One example of such confusion is provided by the increasingly widespread use of the term ‘somatization’. In 1980 DSM III adopted ‘somatization disorder’ as a recognised psychiatric diagnosis, characterising the disorder as a syndrome of multiple somatic symptoms that cannot be explained medically. The revised edition of DSM III, produced in 1987, requires a history of several years’ duration beginning before the age of thirty. The patient must have at least thirteen symptoms from a list of thirty-five. According to the most recent edition of the most authoritative American psychiatric textbook, Kaplan and Sadock’s Comprehensive Textbook of Psychiatry V, ‘A symptom need only be reported by the patient in order to be counted; it is not necessary to establish that the patient actually had the symptom.’[21]

Among the symptoms included in the list of thirty-five are diarrhoea, nausea, back pain, chest pain, trouble walking (sic), difficulty urinating, sexual indifference, and menstrual periods which are judged by the patient concerned to be more irregular or more painful than is normal.

There can be no doubt that physicians do frequently encounter patients who report multiple physical symptoms which they have imagined or exaggerated because of their anxiety, insecurity or need for attention, and that many such patients believe themselves to be genuinely ill. The problem posed by such patients is an extremely serious one, partly because they can use up a disproportionate amount of a country’s health services, and partly because their tendency to take refuge in illness often masks serious psychological distress. [22]

But describing such patients as ‘somatisers’ or judging that they suffer from ‘somatization disorder’ merely adds another layer of confusion to a situation which is already confused enough. For the term somatization has at least two different, mutually contradictory meanings. In Kaplan and Sadock’s Comprehensive Textbook of Psychiatry V we are offered the following definition of the term in the section devoted to ‘Somatoform Disorders’: Somatization is the tendency to experience, to conceptualise and to communicate mental states and personal distress as bodily complaints and medical symptoms.’ We are told that somatization is a general psychological disposition and that it is not in itself a psychiatric disorder although it can become one in extreme manifestations. We are further told that whereas the concept of a conversion reaction was elaborated in the psychoanalytic tradition, ‘somatization disorder originated in the phenomenological and descriptive approach’. [23]

Yet if we turn from the section on ‘Somatoform Disorders’ to that devoted to classical psychoanalysis, we find that the concept of somatization makes its appearance in a list of ‘Immature Defence Mechanisms’. In somatization, we are told, ‘psychic derivatives are converted into bodily symptoms and there is the tendency to react with somatic rather than psychic manifestations.’ [24]

On this view, then, somatization, far from being distinct from conversion, appears to be cognate with the process of hysterical conversion which Freud himself postulated and which was adopted as a key aetiological assumption in the first edition of the DSM, which defined a conversion reaction as a functional symptom resulting from the conversion of anxiety into bodily sensations. The psychiatrist Z. J. Lipowski, who has had a major influence on popularising the term, actually confirms its origin in psychoanalytic terminology when he writes that the term ‘was introduced by Stekel early in this century to refer to a hypothetical process whereby a “deep-seated” neurosis could cause a bodily disorder.’ As Lipowski notes, the term somatization ‘was thus related to, if not identical with, the concept of conversion’. [25] Having acknowledged its psychoanalytic origins, Lipowski then goes on to use the term in the non-psychoanalytic sense given above.

The confusion as to what somatization actually means, and where the concept comes from, is significant. For while it may well be the case that it has been redefined in terms of phenomenology, it must be suggested that its strongest appeal to psychiatrists, and the reason it has been adopted so widely, springs from the fact that it is both congruent with psychoanalytic assumptions and, ostensibly at least, independent of them. In its ‘strong’ sense, which also coincides with its etymological sense, the word ‘somatization’ refers to a process whereby real physical symptoms are supposedly created by transforming psychological or emotional energy into somatic form. In its ‘weak’ sense the word refers to a process in which patients use a multiplicity of physical symptoms, which may be imaginary or non-existent, in order to mask depression or anxiety or in order to establish a particular kind of relationship with doctors. A major problem stemming from this conceptual double-life is that, as is the case with ‘hysteria’, the widespread use of the ‘weak’ form of the word actually tends to reinforce the psychosomatic fundamentalism of those wedded to the ‘strong’ form of the word and to the psychoanalytic aetiologies associated with it. [26]

The greatest practical danger of this state of affairs is that it encourages physicians to entertain in a somewhat inchoate form the extreme Charcotian or Freudian assumption that almost any physical symptom can be produced psychosomatically. This assumption is sometimes actively encouraged by careless and historically ill-informed discussions of topics like ‘hysteria’ which sometimes find their way into influential medical textbooks. In one of the most highly regarded and commonly used British textbooks on clinical neurology, which was first published in 1989, and from which many future general practitioners and hospital doctors learn the principles of neurological diagnosis, Sir Francis Walshe’s attempt to rebut Eliot Slater is cited in positive terms and we find the following discussion of hysteria: 

Hysteria involves a state of dissociation or conversion, unconsciously determined for emotional gain ... The gain is usually not a simple desire to manipulate others or obtain a financial reward, it is often an attempt to reduce intolerable anxiety ...  

Conversion is a concept whereby anxiety is ‘converted’ to a physical symptom and anxiety is relieved in the process ... Conversion symptoms can be motor, such as disturbance of gait, loss of speech, muscle weakness or paralysis and abnormal movements. Sensory symptoms include pain anaesthesias, blindness and deafness... ...

Hysterical symptoms may mimic almost any medical condition, and the diagnosis is even more difficult when there is an ‘hysterical overlay’ [italics added]. [27]  

The extraordinary claim that ‘hysterical symptoms may mimic almost any medical condition’ derives ultimately not from any body of medical knowledge, but from centuries-old medical lore which, even though it is based on physiological fallacies, has been accepted on trust by generations of physicians. Although the author of the passage which is quoted above goes on to warn his readers of the dangers of misdiagnosis, he seems not to understand that many of these dangers are a direct product of formulations such as the one he has given.

The capacity of such formulations to mislead is perhaps best understood if we place them alongside an extreme version of psychosomatics such as that of Freud’s follower Georg Groddeck. Groddeck believed (or sometimes behaved as though he believed) that illnesses performed a psychological function and that specific illnesses could actually be produced by the unconscious, which he called the ‘It’: 

Sometime or other in the course of the treatment I am accustomed to call my patient’s attention to the fact that from the human semen there is born, not a dog, nor a cat, but a human being, that there is some force within the germ which is able to fashion a nose, a finger, a brain, [and] that accordingly this force, which carries out such marvellous processes, might well produce a headache or diarrhoea or an inflamed throat, that indeed I do not consider it unreasonable to suppose that it can even manufacture pneumonia or gout or cancer. I dare to go so far with my patients as to maintain that the force really does such things, that according to its pleasure it makes people ill for specific ends ... 

In this particular case Groddeck writes that he never worries himself in the least ‘as to whether I believe what I am saying or not’. But he does appear to endorse the view that all diseases have a psychological function:  

May I repeat what I am saying? Illness has a purpose; it has to resolve the conflict, to repress it, or to prevent what is already repressed from entering consciousness; it has to punish a sin against a commandment ... Whoever breaks an arm has either sinned or wished to commit a sin with that arm, perhaps murder, perhaps theft or masturbation; whoever goes blind desires no more to see, has sinned with his eyes or wishes to sin with them; whoever gets hoarse has a secret and dares not tell it aloud. But the sickness is also a symbol, a representation of something going on within, a drama staged by the It, by means of which it announces what it could not say with the tongue. In other words, sickness, every sickness, whether it be called organic or ‘nervous’, and death too, are just as purposeful as playing the piano, striking a match, or crossing one’s legs. They are a declaration from the It, clearer, more effective than speech could be, yes, more than the whole of the conscious life can give. [28]  

It would be easy to dismiss Groddeck’s paeans to the purposefulness of disease as a historical curiosity with no relevance to the present. Yet Groddeck’s views are still taken seriously by many people today, including some mainstream physicians. [29]  

It would seem that one of the reasons his theories continue to exercise an appeal some seventy years after they were first published is that, like Freud’s theories with which they are closely associated, they translate into a persuasive (and highly poetic) register a popular folk-theory of medicine which has a very wide appeal. It is from this perspective, I believe, that we should view the claim that hysteria may ‘mimic almost any medical condition’. When such careless claims are made by experienced physicians in textbooks which credulous medical students are expected to treat with respect, they tend to confer academic respectability on this kind of folk-lore. This, in turn, can all too easily result in dangerous or even fatal misdiagnoses. 

In an article dealing with the tendency of doctors to misdiagnose real organic conditions as psychological disorders, Linda Gamlin relates the case of a woman who, by the time she was taken to hospital, was so ill that she nearly died. ‘For over two weeks she had been feverish and extremely weak, with typical signs of liver disease: yellow skin, dark brown urine, and putty-coloured stools.’ The woman’s general practitioner, however, had diagnosed post-natal depression and had associated her illness with an emotional breakdown which she had suffered seven years earlier. This view was repeated by no less than four other doctors in her group practice. Only when her husband rang a hospital consultant in desperation was the proper diagnosis of viral hepatitis made and the woman rushed to hospital. [30]

Such anecdotes can be multiplied almost indefinitely. A common feature of many of them is the credulous and perhaps not always fully conscious acceptance by some physicians of extreme theories of psychosomatic illness for whose correctness there exists no evidence whatsoever, and which are ultimately derived from ancient medical fallacies about the non-existent disease of hysteria.

The careless use of the term ‘somatization’, and, indeed, the very fact that this medically tendentious word is used at all, almost certainly contributes to sustaining this climate of credulity. It also suggests that modifications of terminology alone will not solve any problems. It is the concept of ‘hysteria’ and not merely the external label which needs to be discarded.

This does not mean that we should deny that emotional experiences can have neurological or other physiological consequences. What we should recognise, however, is that emotional events (such as stress, trauma or shock) are themselves experienced by the human organism as physiological changes. It is in these changes, and not in some putative, purely psychological realm, that we should seek the cause of real symptoms which, after exhaustive investigation, do seem to be correlated with intense emotions. If ‘hysteria’ has indeed functioned for centuries as a diagnostic dustbin into which physicians have tossed a huge and ill-assorted selection of diseases, syndromes, symptoms, and responses, there may well be one or several discrete syndromes within it which do have this kind of complex relationship to the physiology of human emotions. This does not mean, however, that the term ‘hysteria’ should be retained, any more than recognition of the reality of, say, catamenial epilepsy, indicates that ‘lunacy’ ought to be retained as a serious psychiatric category. [31]

In those quite different cases where it can be proved beyond doubt that we are dealing with unreal symptoms, which involve no organic dysfunction, then we are by definition dealing not with a disease but with a behavioural problem. There is therefore no reason why a term which is still associated with a disease-concept, and whose currency is owed almost entirely to the prevalence of misdiagnosis and medical ignorance in the past, should be invoked. 

The crux of the problem is that medicine has, for very many centuries, framed its discussions of symptoms in terms of a creationist ontology. Medical practitioners have, in other words, accepted the dualistic proposition that human beings are made up of two separate but interconnected entities – a physical body and a non-physical mind or soul. Such dualism has actually been institutionalised in the profession. For, originally at least, ‘psychiatry’ was understood as a branch of medicine which was not concerned with diseases of the body or any organic dysfunction, but solely with diseases of the mind or soul. The very complexity which has been traditionally attributed to the soul has sometimes actually encouraged physicians to accept or tolerate an impoverished notion of the body and its extraordinary neurological and biochemical complexity.

It is
the fact that orthodox medicine has tended historically to underestimate the neurophysiological complexity of the human body that has enhanced the credibility not only of therapeutic systems such as psychoanalysis but also of a whole range of ‘alternative’ approaches to medicine. For although many of these therapies may be entirely spurious, there can be little doubt that those who proclaim ‘the power of the soul to heal’ are in some cases dealing with quite genuine physiological phenomena which orthodox medicine has accidentally defined out of its model of the body. [32]

During the ‘medical dark ages’, from which we only began to emerge at the beginning of the twentieth century, and during which physicians remained extraordinarily ignorant about countless aspects of human physiology and human pathology, dualistic models of the human organism were inevitable and perhaps even necessary. In our present post-Darwinian era we will only cause confusion if we persist in using them. For when physicians continue to use terms such as ‘hysteria’, ‘somatization’, ‘psychogenic’ and even ‘psychosomatic’, they merely perpetuate the very kind of creationist dualism which I have tried to analyse in the last part of this book. Such dualism is no more conducive to clear thinking about medicine than it is to clear thinking about any form of human behaviour.




[1] This list of the symptoms reported by Freud’s early patients is compiled for the most part from two sources – the index to the Penguin edition of Studies on Hysteria (entries under ‘hysterical symptoms’), and the list which Freud and Breuer themselves give in their ‘On the Psychical Mechanism of Hysterical Phenomena: Preliminary Communication’ (1893), Penguin edition, p. 54. It is in the latter paper that Freud and Breuer refer to ‘epileptoid convulsions’ and ‘petit mal’. 

[2] In the last twenty years an alternative, relativistic approach to medical history has come into being. Reacting against Whiggish views, and failing to recognise that scepticism about the ideology of ‘progress’ is entirely compatible with acceptance of the reality of ‘medical progress’, adherents of the relativistic approach seem sometimes to be embarrassed by the very possibility that modern physicians might know more than ancient ones. As a result they tend to decline on principle to judge past medical practice against current medical knowledge. See the quotation from Hirschmüller and discussion, Chapter 4, note 35.

My comments about the tendency of medical historians to ignore medical mistakes and misdirections have their origin in a conversation with Elizabeth Thornton. Much more recently, after I had written the words which occur in the main text, Frank Cioffi drew my attention to a paper by the psychiatrist E. H. Hare which contains the following passage:  

It may be argued that historians ought to pay more attention than they have done to scientific hypotheses which proved to be failures. The trouble with the history of science, and of scientific medicine, is that it has too often been presented as one long success story; whereas, in fact, a striking feature of the history of science (particularly where science overlaps with medical and social matters) has been the tenacious persistence of supposedly scientific ideas long after they ought to have been abandoned. I think the historical study of scientific failures is important, not only because it is likely to give us a keener insight into the nature of the scientific process, but also because it may lead us to examine more closely the soundness of some of our own pet ideas’ (E. H. Hare, ‘Medical Astrology and its Relation to Modern Psychiatry’, Proceedings of the Royal Society of Medicine, vol. 70, 1977, pp. 105–10). 

One medical historian who has given an unusual amount of attention to the role of misdiagnoses in the history of psychiatry is Richard Hunter. See Richard A. Hunter, ‘Psychiatry and Neurology: Psychosyndrome or Brain Disease’, Proceedings of the Royal Society of Medicine, vol. 66, April 1973; Richard A. Hunter and Ida Macalpine, Three Hundred Years of Psychiatry, 1535–1860, Oxford University Press, 1983. See also Roy Porter, ‘Ida Macalpine and Richard Hunter’ in Mark S. Micale and Roy Porter (ed.), Discovering the History of Psychiatry, New York: Oxford University Press, 1994, pp. 83–94.

[3] A. Steyerthal, Was ist Hysterie?, 1908, Halle a S., Marhold. Quoted by Aubrey Lewis, ‘The Survival of Hysteria’ in Alec Roy (ed.), Hysteria, Wiley, 1982, p. 22.

An interesting sociological perspective on the creation of spurious diagnostic categories is offered by Susan Leigh Starr in her study, Regions of the Mind: Brain Research and the Quest for Scientific Certainty, Stanford University Press, 1989: 

The creation of ‘garbage categories’ is a process familiar to medical sociologists. When faced with phenomena which do not fit diagnostic or taxonomic classification schemes, doctors often make residual diagnoses. One function of such diagnoses is to shunt unmanageable, incurable or undiagnosable patients into other spheres of care where they will not interfere with the ongoing work. Hysteria, senility and depression, for example, have been criticised as such categories ...

Localizationists created such categories for problems that did not have an identifiable localised referent or the possibility of a physical treatment. These patients were diagnosed as hysteric or neurasthenic. (p. 173)

[4] Eliot Slater, ‘What is Hysteria?’, in A. Roy (ed.), Hysteria, 1982, p. 40.

  [5] Eliot Slater, ‘Diagnosis of “Hysteria”’, British Medical Journal, 29 May 1965, p. 1399.

[6] Phillip R. Slavney, Perspectives on ‘Hysteria’, Johns Hopkins University Press, 1990, p. 3. At the end of his book Slavney puts forward what may turn out to be a more realistic view: 

This could well be the last book with hysteria in its title written by a psychiatrist. Although the word is used daily in the practice of medicine, ‘those who would like to drop it once and for all’ seem to have won the battle for control of psychiatric nomenclature, and the next generation of physicians will no longer find it indispensable when they wish to indicate certain traits and behaviours. Hysteria, hysteric, and hysterical are on the verge of becoming anachronisms (p. 190). 

Slavney, it should be noted, is describing the situation as he sees it in the United States, where he is Director of Resident Education in the Department of Psychiatry at the Johns Hopkins University School of Medicine. In Britain it is probably true to say that the term ‘hysteria’, while clearly waning, still enjoys a degree of official recognition. See below, note 22. 

[7] A. Lewis, ‘The Survival of Hysteria’, Psychological Medicine, 1975, vol. 5, pp. 9–12. This paper is reprinted in Alec Roy (ed.), Hysteria, John Wiley, 1982, pp. 21–6. The quotations here are all taken from Lewis’s paper, where full references are given. 

[8] Sir Francis Walshe, ‘The Diagnosis of Hysteria’, British Medical Journal, 1965, 2, pp. 1451–4. 

[9] American Psychiatric Association, Diagnostic and Statistical Manual of Mental Disorders (DSM), 3rd edition, Washington, DC: APA, 1980. The fourth edition of this manual was published in 1994. For an excellent brief critique of the approach of DSM to the problem of mental ‘illness’, see Carol Tavris, The Mismeasure of Woman, New York: Simon and Schuster, 1992, pp. 176–92. For a sceptical view of the background to DSM, the ‘psychiatrists’ bible’, see Stuart A. Kirk and Herb Kutchins, The Selling of DSM: The Rhetoric of Science in Psychiatry, New York: A. de Gruyter, 1992. This book, of whose salutary existence many workers in the field of ‘mental health’ evidently remain unaware, has been described by Thomas Szasz as ‘a well-documented exposé of the pretence that psychiatric diagnoses are the names of genuine diseases, and of the authentication of this fraud by an unholy alliance of the media, the government and psychiatry.’ In his endorsement of the book Szasz goes on to recommend it ‘to anyone concerned about the catastrophic economic and moral consequences of psychiatrizing the human predicament’. 

[10] Eliot Slater, ‘Diagnosis of “Hysteria”’, British Medical Journal, 29 May 1965, p. 1399. See above, Chapter 6, final paragraph.  

[11] G. L. Engel, ‘Conversion Symptoms’ in C. M. MacBryde and R. S. Blacklow (ed.), Signs and Symptoms, 5th edition, Pitman Medical, 1970, pp. 650–68. Quoted in C. D. Marsden, ‘Hysteria – A Neurologist’s View’, Psychological Medicine, 1986, vol. 16, pp. 277–88. 

[12] A. Lewis, ‘“Psychogenic”: A Word and its Mutations’, Psychological Medicine, 1985, vol. 2, pp. 209–15. 

[13] Slater, British Medical Journal, p. 1396. 

[14] The same objection is made by Walshe, who writes that ‘one cannot accept hysteria adjectivally and deny it substantively’ (p. 1452). 

[15] Laurence Miller, Freud’s Brain: Neuropsychodynamic Foundations of Psychoanalysis, New York: The Guilford Press, 1991. 

[16] A similar point is made by Thornton in relation to the phenomenon of hypnosis. See The Freudian Fallacy, revised edition, Paladin, 1986, pp. 95–6. 

[17] Miller, p. 26. 

[18] Miller, p. 80. 

[19] Mark S. Micale, ‘Hysteria and Its Historiography: The Future Perspective’, History of Psychiatry, vol. 1, 1990, p. 108. 

[20] Graeme J. Taylor, review of Edward Shorter’s From Paralysis to Fatigue: A History of Psychosomatic Illness in the Modern Era, published in Psychosomatic Medicine, vol. 55, no. 1, pp. 88–9.

For one recent contribution to the study of ‘hysteria’ which seems to bear out Taylor’s suggestion, see Peter Eames, ‘Hysteria Following Brain Injury’, Journal of Neurology, Neurosurgery and Psychiatry, 1992, 55, pp. 1046–53. This paper contains an account of work in a unit treating severe behaviour disorders after brain injury. Fifty-four patients in this unit showed clinical features which also occur in some descriptions of ‘hysteria’. It was discovered that the appearance of such symptoms was closely correlated with diffuse insults to the brain, including hypoxia and hypoglycaemia (oxygen starvation and abnormal reduction of sugar levels in the blood). One case of ‘hysteria’ was caused when a patient undergoing a routine operation was accidentally placed on nitrous oxide rather than oxygen, and suffered chronic brain damage as a result. Although Eames’s paper contains a misleading account of Charcot’s research into ‘hysteria’ it remains an interesting and extremely valuable contribution to the subject. 

[21] Harold I. Kaplan and Benjamin J. Sadock, Comprehensive Textbook of Psychiatry V, 5th edition, Baltimore: Williams and Wilkins, 1989, p. 1009. 

[22] Much of the work which has been done on ‘somatization’ is clearly of value. See, for example, Z. J. Lipowski, ‘Somatization: The Concept and Its Clinical Application’, American Journal of Psychiatry, vol. 145:11, November 1988, pp. 1358–68; Wayne Katon, Elizabeth Lin, Michael Von Korff, Joan Russo, Patricia Lipscomb and Terry Bush, ‘Somatization: A Spectrum of Severity’, American Journal of Psychiatry, vol. 148:1, January 1991, pp. 34–40; Donna E. Stewart, ‘The Changing Face of Somatization’, Psychosomatics, vol. 31, no. 2, 1990, pp. 153–8. 

[23] Kaplan and Sadock, p. 1009. 

[24] Kaplan and Sadock, p. 375. 

[25] Lipowski, p. 1359. 

[26] A striking, book-length example of this process at work is provided by Edward Shorter’s From Paralysis to Fatigue: A History of Psychosomatic Medicine. The main authority which Shorter cites in support of his own arguments is the psychiatrist Z. J. Lipowski, who has written a great deal about ‘somatization’. Yet whereas Lipowski promotes a complex and ‘weak’ form of the concept, which eschews psychoanalytic notions of conversion, Shorter invokes his authority in support of his own theories, according to which the unconscious mind freely converts the ‘stress of psychological problems into physical symptoms’. The relationship between Shorter’s ideas and Lipowski’s is discussed acutely by Theodore M. Brown in his review of Shorter in Journal of the History of the Behavioral Sciences, vol. XXIX, 1993, pp. 243–5. 

[27] C. David Marsden and Timothy J. Fowler (ed.), Clinical Neurology, Edward Arnold, 1989, p. 428. The section on hysteria from which I quote is part of the chapter entitled ‘Psychiatric Disorders’ which was written by Paul Bridges, Consultant Psychiatrist at Guy’s and the Brook Hospital. Bridges cites the papers by Eliot Slater and Sir Francis Walshe, but he makes no reference to Marsden’s own contribution to the debate. 

[28] Georg W. Groddeck, The Book of the It (1923), Vision Press, 1979, pp. 100–101. 

[29] In his book Migraine, for example, Oliver Sacks suggests that there is a certain kind of migraine which should be approached not only as a physical event ‘but as a peculiar form of symbolic drama into which the patient has translated important thoughts and feelings’. The symptoms of this kind of migraine can constitute ‘a bodily alphabet or proto-language’ and must be interpreted, he suggests, ‘as if they were palimpsests in which the needs and symbols of the individual are inscribed ...’  One of the main authorities invoked in order to justify this approach is Georg Groddeck. Sacks also explicitly draws a parallel between his own speculative account of ‘situational migraines’ and Freud’s theory of hysteria, which he appears to accept (Oliver Sacks, Migraine: Evolution of a Common Disorder, revised edition, Pan, 1981, pp. 223–4). 

[30] Linda Gamlin, ‘All in Whose Mind’, Guardian, 16 July 1991. This article draws some examples from a paper by Erwin K. Koranyi, ‘Morbidity and Rate of Undiagnosed Physical Illnesses in a Psychiatric Clinic Population’, Archives of General Psychiatry, vol. 36, April 1979, pp. 414–19. I am grateful to Linda Gamlin for supplying me with a copy of this paper. 

[31] The suggestion that ‘hysteria’ should be understood as a ‘residual diagnosis’ – or a diagnostic dustbin – is made by Susan Leigh Starr in her book, Regions of the Mind: Brain Research and the Quest for Scientific Certainty, Stanford University Press, 1989.  

[32]   The implication of my own remarks here is that the entire discipline of psychiatry still bears the marks of its birth out of what I have called the ‘medical dark ages’, and that the misdiagnosis of ‘genuine’ organic illnesses and their construal as psychological syndromes has played a major part in its development. For Susan Leigh Starr’s acute and, I believe, accurate analysis of the historical and practical status of the hysteria diagnosis, and for references to work by Richard Hunter and Mark S. Micale which bears on this problem, see above, notes 3 and 4.


Richard Webster, 2004