There's something about the debate over animal experiments that turns perfectly nice people into rabid dogs. Take the exchange on the Today Programme between a leading cancer scientist and the head of an anti-vivisection campaign. It had been prompted by the publication of a letter signed by 500 scientists declaring their support for the use of animals in research. Within 60 seconds, the two speakers were at each other like a couple of lab rats on steroids.
But then, this is one scientific debate notable for its lack of evidence or rationality. On the one side, you have screaming zealots who refuse to face facts. On the other, you have the anti-vivisection lobby. For years, scientists have been insisting that the benefits of carrying out experiments on animals is beyond doubt. And to back their case, they wheel out random examples of breakthroughs said to be impossible without the use of animals.
As an example of unscientific argument, this takes some beating. For a start, the plural of "anecdote" is not "data" - and for every case of animal research proving helpful, there's another where it proved hopelessly misleading. Indeed, there's one going through the courts right now. According to tests on mice, the now-notorious painkilling drug Vioxx should do the human heart a power of good. It doesn't: just ask the pharmaceutical company Merck and the thousands of patients affected by the drug.
Despite all the posturing from defenders of animal testing, the fact is that no systematic study of the predictive value of animal tests has ever been carried out. What little evidence there is suggests we may as well train Roland the lab rat to toss a coin to decide if a drug is safe. The problem is not just with bad drugs slipping through, either. Who knows how many drugs for curing humans have gone begging just because Roland went paws-up?
Contrary to what many scientists would have us believe, there is another way.
This week, hundreds of researchers are in Berlin for the fifth world congress on alternatives to animal experiments. Considering the dismal level of funding for such research, impressive progress is being made. There is currently much excitement over so-called "human microdosing", in which new drugs are given to patients in doses too low to have health effects, but high enough to reveal their effect on cells.
Such developments are being keenly watched by pharmaceutical companies under intense pressure to find new drugs - and preferably ones that don't land them in court. Privately, they admit that research using animals - even those genetically modified to mimic certain human traits - has failed to produce insights of sufficient reliability. The Vioxx case, meanwhile, has highlighted the dangers of extrapolating from mice to men.
If you want to know the future of animal experiments, forget the 500 scientists. Their arguments are as scientific as organ-grinders insisting they must keep their monkeys. Watch instead the pharmaceutical companies, whose patience with the farce of animal experiments can't last much longer.