Freudian Slip?: How far can we trust the father of psychoanalysis? Genius, teacher, healer? Opportunist liar, manipulator? Such questions have resurfaced with a vengeance. Jerome Burne reports

Woody Allen's recent sexual imbroglio did psychoanalysis no favours - 20 years on the couch and he ends up like that? But there's worse. Sigmund Freud, big daddy of 'talking therapies', is now reeling from a sustained and brutal attack in the pages of the New York Review of Books. 'Outraged and horrified' was one of the milder descriptions of the feelings of the psychoanalytic community, which has bombarded the magazine with one of the biggest postbags it has ever seen.

The article comes from Frederick Crews, professor of English at the University of California, Berkeley, who summarised the recent, devastating work of various Freudian revisionists. The figure that emerges from Crews's composite portrait is not the pioneering geographer of the unconscious but an altogether darker figure. A cultivated and sophisticated man, with great literary powers to be sure, but also a wilful opportunist, lacking an empirical approach and ethical scruples who did not hesitate to manipulate the evidence to suit his own ends.

A startling example of this, not found in any of the official literature but recently constructed from the rediscovery of some letters, is the little-known case of the Twenties analyst Horace Frink. Frink was married but, like many of Freud's circle, was having an affair with one of his patients - an heiress and married woman called Angelika Bijur. Freud's advice to Frink was that he and the heiress should divorce their partners and marry, lest his latent homosexuality become overt. Freud's interest was plain: 'You are not yet aware of your fantasy of making me a rich man,' he wrote to Frink, going on to discuss a large contribution to the Psychoanalytic Funds.

The couple took his advice and before long both their devastated and abandoned spouses were dead. Soon afterwards, Bijur sued for divorce from Frink, who declined into a psychotic depression and made repeated suicide bids. There is no sign that Freud felt any remorse about the advice he had given. Crews's point is that such ruthless self-interest was no aberration; that a similar focus on victory at all costs vitiates Freud's most famous cases, including those that provide the foundations of psychoanalytic thought.

There was, for example, the unfortunate Emma Eckstein, who was diagnosed as 'bleeding for love' of Freud himself when she was actually bleeding because 18 inches of gauze had been left in the remains of her nose after a botched operation by Freud's friend and long-term correspondent Wilhelm Fliess. Or the case of 'Dora', later identified as Ida Bauer, which for a long time was regarded in psychoanalytic circles as a 'classical analysis of the structure and genesis of hysteria'.

Bauer, a surprisingly independent 18-year-old, had a family background that by today's standards would have justified the intervention of social services. She had a syphilitic father who was having an affair with the wife of a family friend, Herr K, who in turn had made it plain that he wanted to have sex with Bauer. Her father was not averse to the prospect and when Bauer declared the whole situation intolerable he sent her off to Freud to be cured of tics, suicidal thoughts and insubordination.

Freud's response was to suggest that the reason for her hysteria and distress was that she was a latent homosexual who had fantasies of pregnancy and oral sex and memories of childhood masturbation. She was really in love with Herr K, he said, and in a phrase that was to ring through a million male enclaves for the rest of the century, declared that when she slapped Herr K's face, it 'was by no means a final No'.

But despite their failures, deception and immorality, these cases do not really threaten the principles of psychoanalysis. They could be dismissed by saying: 'OK, Freud was not a very good clinician, but his theory and insights into the human mind were sound.' And this is where Crews's account, drawing heavily on a recent book, Seductive Mirage: An exploration of the work of Sigmund Freud by Allen Esterson, breaks new ground.

According to the official version, two ideas lie at the heart of the psychoanalytic process. First of all, Freud's patients told him 'lurid details' of infantile sexual scenes which he believed to be true - the seduction theory. Later he was forced to conclude reluctantly that he had been told a 'collection of fairy-tales' but that those fairy-tales exist in all of us and are a valuable key to human behaviour.

This interpretation was dramatically challenged almost 15 years ago by a one-time Freudian insider, Jeffrey Masson. Masson claimed that Freud knew that those stories of early sexual abuse, or seduction, were true but that he could not cope with their implications. According to Masson, Freud betrayed his patients by claiming that they had simply wished to have incestuous relationships with their fathers.

This was potentially damaging to the Freudian structure, but Crews's and Esterson's contention is fatal. 'Almost all my women patients told me that they had been seduced by their father,' Freud wrote in 1914. This, claims Esterston, was a cover-up; there were never any stories of early sexual abuse. Esterson's painstaking examination of Freud's papers makes it clear that Freud's technique was to guess at the existence of the primal sexual scenes and then to press his interpretation on to the patient. So Freud founded psychoanalysis on the bizarre principle that his mistakes about 'seductions' must have originated in his patients' thoughts about their parents when they were babies.

Just how much latitude Freud allowed himself in spotting signs of sexual scenes can be seen from the case of Sergei Pankeev, otherwise known as the Wolf Man.

He got his name from a dream that Freud analysed as revealing that Pankeev had witnessed a primal sexual scene. The dream was simply of six or seven white wolves (actually dogs, as Freud later admitted) sitting in a tree outside his window. The interpretation went like this: the whiteness equals bedclothes, stillness represents its opposite, coital motion, the big tails reveal castration and daylight symbolises night. Such imaginative methods enabled Freud to insist that at the age of one, Pankeev had watched his father and mother copulating doggy-style, three times in succession, while he watched from his crib and soiled himself in protest.

Pankeev never did recall such an event. Later he remarked that his wealthy social class meant that his crib would never have been in his parents' room. What's more, Freud himself observed that Pankeev's mother did not enjoy sex, so three times a night doggy-style seems unlikely. But Freud was undeterred. 'We must not be led astray by initial denials,' he said. 'If we keep firmly to what we have inferred we shall in the end conquer every resistance by emphasising the unshakeable nature of our convictions.'

If, say Crews and Esterson, no infantile sexual material existed except what was forcefully put to the patients by Freud himself, then there is no need for a theory of infantile fantasies of incest, no need for an Oedipal complex, no need for repression and for the rest of Freudian theory.

Most of the published letters from psychoanalysts responding to Crews's article make little attempt to engage with him over matters of detail. One exception is James Hopkins, philosophy lecturer at King's College, London, who challenges Crews's account of when the father seducer first came into the picture. Crews quotes Freud as initially saying that governesses and servants were the seducers and only later declaring that father was at the bottom of it all. According to Hopkins, Freud regarded daddy as a prime suspect all along.

Most, however, ignored the substance of the charges and paid tribute to Freud's brave, pioneering work and his legacy of a new understanding of the mind. They claim that repression, transference, the role of early development and the importance of dreams and symbolism are all valuable discoveries. Others admitted that there had, quite understandably, been errors in the early days - what science did not make mistakes? - but that the psychoanalytic profession is far better regulated now.

Much of this might be dismissed as a squabble over the origins of a declining sect. In the case of child abuse, however, Freud's legacy extends beyond the consulting rooms of Hampstead and mid-town Manhattan. Crews believes that not only did Freud's ideas about infantile sexuality both provide a sophisticated rationale for incest - 'It's what children want anyway' - and make genuine cases harder to identify - 'It's a common fantasy' - but that once the issue came out of the closet they made it even harder to uncover the truth.

There is now a fierce debate both in Britain and the US on the issue of 'false memory' - whether the memories of early sexual abuse that thousands of (mainly) women claim to have uncovered during therapy are real or the result of suggestion. 'A vulgarised but recognisably Freudian idea of repression lies at the heart of our 'false memory prosecutions',' says Crews. 'And making matters worse is the way that Freud utterly rejected the possibility that 'recollections' could be implanted by the therapist.'

(Photograph omitted)