Not content with feeding the appetite of crime fiction readers worldwide, Val McDermid is now helping the University of Dundee's Centre for Anatomy and Human Identification in a distinctly novel way.

Working at CAHID is clearly a bit of a wheeze. So enjoyable is the work they do, helping police identify corpses found in a range of highly unusual and complicated situations, they now want others to share the fun. Thus, lay people without so much as a GSCE in biology are being invited to take part in a completely free online course. Lasting six weeks, this will use a short story McDermid has written specially for the purpose. Given instruction in various forensic techniques, students can then try to solve the crime. Later, once the identity of the victim and the cause of death have been established, McDermid will publish the story that lies behind the crime, fleshing it out, as you might say.

Reading about this I had a moment of revelation. One of the reasons I rarely read crime fiction, and certainly not the sort where scientific terms are commoner than commas, is that I am never curious enough about why someone has been killed. Probably because I know it will all be explained to me in due course, I can't be bothered trying to work out what's going on and why, ahead of schedule.

Watching TV crime dramas, my mind wanders to the actors' accents, or the set design. My favourite thrillers are where the murderer is known from the start, and the tension comes from the chase. Otherwise, I know that anything I might start to work out for myself must be wrong, because it will be too obvious. Even so, I sympathise with savvier readers who become apoplectic when a piece of unknowable information, which turns out to be the clue on which the crime hinges, has been held up the writer's sleeve until the final chapter.

I can, however, see the appeal of the Dundee course. Although it is not aimed at readers of fiction, quite a few of those will undoubtedly enroll. The idea of solving a murder, however, when there is no manipulative fictional hand tampering with the facts, is tempting even if crime is not your genre. And while this particular venture is something of a gimmick, it taps into an instinct most of us share: the need to know why something has gone badly, tragically wrong.

Only a few crime writers are as close to the scientific cutting-edge as McDermid, hence her choice for setting the hare running for Dundee's make-believe crime scene. She may of course have also unwittingly aided and abetted the demise of her own sort of fiction as the would-be sleuth's first post of call. One can foresee a day when those who at the moment get their deductive thrills from novels discover the richer satisfactions of virtual detecting, armed with the tools of the detective's trade to a degree that few crime novelists could match. Indeed, it would come as no surprise to learn that some of those signing up are writers keen to keep abreast of forensics.

You can't help wondering if the deeper pleasures of crime fiction are in danger of being lost beneath the demands of realism, as science increasingly plays a larger role than psychology in real-life detective work, and events in test-tubes supercede setting and character. It's little wonder that so many writers have gone back in time for their thrillers and crimes, to the days before DNA had even been given a name. After all, there are only so many variations possible on the idea of murder, its justification and cause. As McDermid told me herself last year, detection methods are becoming so sophisticated and swift, certain crimes are now only committed by the desperate or the dim, who can expect the police on the door within hours. For the novelist, this puts an even greater strain on their powers of invention. Could the contemporary crime novel's days be numbered?