BOOK REVIEW / A plain person's guide to catching a serial killer: 'Criminal Shadows: Inside the Mind of the Serial Killer' - David Canter: Harper Collins, 16.99 pounds

GRUESOME] Sensational] Hannibal Lecter called in] This is how the tabloid press reacted last summer to Scotland Yard's announcement that a serial killer had murdered five gay men in London. Front pages were cleared, stills from The Silence of the Lambs reprinted, and columns filled with lunatic speculation about the murderer's personality.

In the event Colin Ireland, who was sentenced to life imprisonment for the murders in December, was trapped not by advice from Denis Nilsen or Michael Lupo, two previous serial killers, but by a security camera which filmed him with his final victim at a London railway station. From statements he gave to the police, Ireland seems to have encapsulated the identity problem frequently found in serial killers in stark form: he had despaired of making his mark in any other way, he said, so took to murder as a way of getting into the history books.

Yet if the police had listened to the crime reporters with whom they routinely mix, there is no reason why they should ever have caught up with Ireland. He certainly isn't the serial killer beloved of popular culture: charismatic, cunning, outwitting the police at every turn. He was reasonably careful - he showed a degree of 'forensic awareness', to borrow a term from David Canter's book - but nevertheless left a fingerprint in the flat of one of his victims.

The significance of the Ireland case, leaving aside the agony of the victims, their families and friends, is as a textbook example of the gulf between the sordid reality of serial killing and its representation in popular culture. Thirteen years since Peter Sutcliffe was convicted of the 'Yorkshire Ripper' murders, the same ludicrous assumptions which obstructed that inquiry are still being made, as though all these criminals are reducible to a single neat formula.

According to this ill-digested mess of Freudian and other theories, serial killers are sinister loners unable to form stable relationships or hold down a job. What never seems to occur to the journalists who write this rubbish is that, if serial killers are so obviously bizarre, it is odd that they are also the most difficult criminals to catch. That is why this book, by the academic psychologist who has pioneered 'offender-profiling' in this country, is so important.

Violent criminals, says David Canter, inhabit the same world as the rest of us - drinking in pubs, making friends, even managing to hold down jobs for long periods. 'Even the most depraved serial murderer is not an alien being driven by processes totally beyond normal experience,' he writes matter-of-factly. 'If he were it is highly probable that the police and others would have become aware of him long before his crimes became a series.'

This fundamental insight has been known since 1888, when a London newspaper remarked that the murderer known as 'Jack the Ripper' was not so obviously deranged as to alarm his victims, but it cannot be repeated too often; the popular press is still, more than a century later, in hot pursuit of a cloaked figure with a bloody knife. A simple psychological mechanism is at work here, distancing 'normal' people (such as journalists) from violent killers and legitimising an enduring fascination with this macabre group of supposed outsiders. What emerges from David Canter's thoughtful, unsensational book is that a great deal is known about serial killers and rapists, but it falls into the realm of broad patterns - the 'shadows' of his title - whose application to a particular police inquiry is not always obvious or unproblematic.

Canter, who is Professor of Psychology at Surrey University, became involved with the subject almost by chance in 1986. On a train journey from London to Guildford, he jotted down an analysis of a series of 24 sexual assaults in London in the previous four years, using material published in the Evening Standard. He was sufficiently struck by the pattern which emerged to send his conclusions to a detective he knew at Scotland Yard.

The police responded by asking him to draw up a 'psychological profile' of the attacker, the so-called 'Railway Rapist'. Canter and the detectives assigned to work with him examined the man's actions year by year, discovering that he was moving outwards 'like a marauding brigand whose confidence grows as he continues to evade capture, spreading his dark influence over an increasing area of the map, the region of each succeeding year encompassing the preceding year'. Making the logical assumption that the rapist had started near home and moved further away as he grew more confident, Canter correctly predicted that he lived in or near Kilburn, the area bounded by the first three attacks.

The profile fitted one suspect in a list of 2,000. John Duffy was placed under surveillance, arrested, and linked by forensic evidence with the rapes and several murders. The case is a neat illustration of what the process can and cannot achieve: Canter was able to pinpoint some of the criminal's likely characteristics without being able, as some detectives jokingly asked him, to supply his name and address.

For the press, offender-profiling quickly became part of the mythology. Reporters have displayed an almost mystical reverence for the process when it is really about common sense - analysing a rapist's behaviour towards his victim, say, on the assumption that it reflects his general attitude to (and previous offences against) women. Yet common sense has been notably lacking in previous serial killer inquiries; Peter Sutcliffe's surviving victims insisted he was a Yorkshireman but detectives were lured on a wild goose chase by a bogus tape recording which claimed he was a Geordie.

In effect, the police and the press have their own overlapping mythologies about criminals and their behaviour. They feed each other's fantasies, especially in the frantic atmosphere of a murder inquiry when everyone fears the criminal will strike again. One of the things which emerges from Canter's book is the value of an academic approach which draws on careful analyses of hundreds of previous cases; with the rise in violent crime, the need to demystify what offenders actually do and how it might be reflected in their everyday lives is more and more urgent.

Canter does not make grandiose claims for offender-profiling but he offers a framework which, as long as it is treated with caution, can be useful in differentiating between two rapists operating in the same area or indicating the kind of criminal record a killer might be expected to have. His conclusion is modest, and a far cry from the Hollywood scenarios in which omniscient experts claim to know everything about a murderer including his favourite colour and shoe size: 'Criminals write their own stories,' he says; 'psychology may help us to read them.' It isn't as sexy as calling in Hannibal Lecter but the low-key approach outlined in this book is more likely to save lives.

(Photograph omitted)