Arts: You don't have to be mad to work here . . .: Identifying too closely with a part can damage an actor's sanity: drivel, self-glorification or a real occupational hazard? Kevin Jackson examines the history of thespian breakdowns

Three exhibits, culled from just the past few weeks. On the second night of his one-man show about the boozy, doomed actor John Barrymore, Nicol Williamson dries after a few minutes and abandons the stage. Jeremy Brett's frank discussions in the press of his history of emotional difficulties prompt some critics to speculate about the price he may have had to pay for his brilliant creation of a neurasthenic Sherlock Holmes whose great wits are sure to madness near allied. And in Abel Ferrara's film Dangerous Game, released yesterday, Madonna plays a movie actress suffering bruises and worse thanks to the maniacal realism her on-screen partner (James Russo) brings to his role as a wife-beater.

New examples, but each one in its own way an addendum to an old, not to say ancient, theatrical myth: acting can threaten your stability, particularly if you're unlucky enough to be playing a character who is more than a trifle wobbly in the upper storeys, such as Barrymore, Holmes, or, say, Othello, the role- within-a-role taken by Ronald Colman in that definitive drama about identifying too closely with a part, George Cukor's A Double Life (1947). Even viewers who have never caught the movie on one of its Saturday afternoon repeats may find the plot familiar: the Colman character just can't leave the Moor behind him at the end of the evening, and so he ends up putting out Shelley Winters' light. More fortunate than the poor sap he played, Colman was given an Academy Award for his pains.

Potent and tenacious as this myth may be, however, it requires a certain tentativeness in discussion, partly because it is one of the stock properties of luvvyspeak at its worst; partly because most sensible actors are understandably fed up with the ignorant prejudice that they are a bunch of neurotics, hysterics and flakes; and partly because, like so many other bits of cant or near-cant, it was blasted more than 200 years ago by Samuel Johnson.

One cornerstone of the acting-drives- you-nuts thesis is the view that the actor or actress in some curious way actually becomes the character that he or she is playing. Nowadays, this belief may be associated with such names as Stanislavsky, Lee Strasberg and the Actors Studio, but its origins are of considerably older vintage. For example, among the reasons why puritanical writers (such as William Prynne in Histrio-mastix, the Player's Scourge, 1632) fulminated against the theatre was their belief that actors blasphemed against God's own dramaturgy by creating new beings, and that this process necessarily endangered the integrity of the actor's soul.

As with many other hazardous pursuits, the supposed spiritual danger of acting added an extra layer of glamour to the calling, and by the 18th century that thrill of potential takeover seems to have inspired one line of fashionable chatter - as, in a different form, it does today for people who lap up biographies of such flagrantly neurotic performers as James Dean and Montgomery Clift. The acting and play-going world was roughly divided between the anti-emotionalists, who held that the actor's art is one of icy detachment from the feelings he or she puts on display, and the emotionalists, who believed in the possibility of a complete usurpation of an actor's pysche by a given role. (Diderot's Paradoxe sur le Comedien was one of the texts which helped set the terms of a debate which has muddled on ever since.)

Johnson, suspecting that an actor of his acquaintance - John Philip Kemble - might have subscribed to the emotionalist line, set about refuting it thus: 'Are you, sir, one of those enthusiasts who believe yourself transformed into the very character you represent?' Upon Mr Kemble's answering that he had never felt so strong a persuasion himself: 'To be sure not, sir (said Johnson;) the thing is impossible. And if Garrick really believed himself to be that monster, Richard the Third, he deserved to be hanged every time he performed it.'

Like his crack about no man but a blockhead writing except for money, this is splendidly robust, but also a shade disingenuous. Johnson knew better than anyone about the demonic powers of the imagination: he was so terrified by the mere reading of one Shakespeare play that he had to rush out into the street to regain his composure, and found one scene in Lear so gruelling that he could not bear ever to reread it. One suspects that his vehemence to Mr Kemble - deserved to be hanged - is as much due to his lifelong terror of falling insane as to his pleasure in exploding modish drivel.

Today, even the most loyal Johnsonians ought to think twice before dismissing tales of cracked actors as no more than the profession's foolish self-glorification. That drama involves a strong element of irrationality on the part both of performers and of audience is well attested. Anyone who has experienced nervousness at the prospect of standing up to speak before an audience of wedding guests or jurors will have some inkling of the baseless, but none the less excruciating, miseries of stage fright, and recognise that those performers who have been afflicted with the condition - Olivier, Ian Holm - were not simply being 'temperamental' but suffering from an authentic illness.

Nor does it take a much greater leap of imagination to understand why the sustained pretence of being another person, combined with the stress of performance, might take a grievous toll both on the nerves and on the sense of personal identity. Stories about people who feign madness and end up mad indeed, as in the old Colditz episode, may be apocryphal but they are often plausible. Granted, Ronald Colman's fate in A Double Life now looks far-fetched, yet there is evidence that men who act Othello may actually suffer violent jealousy in their private lives. As one historian wrote of Edwin Forrest, the great American tragedian, 'When Forrest made audiences tremble at Othello's grief, and weep at his betrayal, he was working out, over and over, a private agony.'

Forrest was far from an isolated case. Bela Lugosi came to identify so strongly with his best-known role as Dracula - a performance he flogged around the music halls in his last years - that he took to sleeping in a coffin. (As did Sarah Bernhardt, though for non-vampiric reasons.) Method or post-Method actors are so notorious for living out their parts in real life that the practice has become the regular stuff of jokes, of which a mild example is the actress played by Michelle Pfeiffer in Sweet Liberty who falls passionately in love with Alan Alda, but only for as long as the production lasts.

It is eerily appropriate that the most widely publicised theatrical collapse of recent years should have been that of Daniel Day-Lewis in Hamlet, our greatest play about madness and acting. According to reports at the time, Day- Lewis (well known for his intense preparation for every role) found playing the Dane more than usually disturbing, and, at the height of his distress, experienced hallucinations of his dead father's presence. On the whole, these reports met with a sympathetic response, and there were few latter-day Johnsons to insist that the young enthusiast was simply deceiving himself.

Every profession attracts its malingerers, but the annals of thespian woe are now too long and too thoroughly documented for the occupational hazards of stage and screen to be dismissed - not least because some endlessly haunting performances have been executed by actors who later succumbed to full-blown lunacy: Antonin Artaud is the paradigm. Nowadays, every film and theatre critic sometimes has to echo the same question which literary critics have traditionally asked about the Prince of Denmark: are they just pretending, or are they really mad?

(Photograph omitted)