It is unusual for politicians to face up to the obvious, but the Scottish Executive seems for once to have done so: it has recognised what has long stared it in the face, namely that dishing out methadone to drug addicts is not the answer to their problems or to the problems that they cause society. A different approach is needed.

Perhaps in 100 years historians will wonder why so many of the governing elite, from senior doctors to Cabinet ministers, persisted for so long in the belief that doling out methadone was the answer. The explanation, I think, will be that they wilfully misunderstood the nature of the problem.

Many years ago I used to dole out methadone like the best (or the worst) of them. This was before I thought at all deeply about the question of drug addiction and accepted uncritically all that I had been taught about it by doctors senior to me. I began to change my opinion when I worked in prison where it was the clinical policy to give addicts methadone. I noticed that, far from creating an atmosphere of contentment and satisfaction, it created one of perpetual tension and irritation. Shortly after having been prescribed a dose, the prisoner would return and say, in an intimidating fashion: “It's not holding me, doc, it's just not holding me,” and sometimes announce that, unless he was prescribed more, he would end up attacking other prisoners, and then it would be the doctor's fault.

In Scotland the great majority of addicts prescribed methadone by their doctors never stop taking it, and most of them take other drugs as well. A particularly dangerous combination of drugs is methadone and benzodiazepines (drugs such as Valium), and yet drug clinics and other doctors persist in prescribing this often fatal combination - largely, I suspect, because they are too frightened of their patients to refuse them anything.

The number of people admitted to hospital having taken a dangerous overdose of methadone (556 in 2006-07) is greater, proportionately, than the number of people admitted to hospital having taken a dangerous overdose of heroin (1,530 cases). In Dublin recently, more people have died of methadone poisoning than of heroin overdose. The supposed cure causes as many problems as the supposed disease. If addicts prescribed methadone are given the opportunity to divert it on to the black market, they will: which suggests that they do not really need it in the first place.

In France, addicts are often prescribed a different drug, buprenorphine, which soon became the street drug of preference in Finland, to which it was illegally re-exported by the addicts. More recently, a huge epidemic of buprenorphine addiction has occurred in Georgia (the ex-Soviet republic), numbering scores of thousands of addicts, who take buprenorphine diverted from France. If the addicts really needed the drugs, they would take them rather than divert them on to a black market.

In the prison in which I used to work, a buprenorphine tablet that had been prescribed for an addict to alleviate the symptoms of withdrawal from heroin on arrival in the prison, and which an addict had put in his mouth and spat out for sale to another prisoner, was known as a “furry” because of its rough surface. Again, this suggests that addicts did not really need what they were prescribed, and that the whole basis of prescription was flawed.

The fundamental error that the Scottish Executive has now admitted is in having regarded addiction to heroin as a technical medical problem, to be solved by technical medical means. But that old approach amounts to a surrender to blackmail: give me what I want or I will continue to behave badly and to hold you responsible for the ill-effects of my own behaviour.

Suppose we gave money to burglars to induce them to stop burgling. No doubt most of them would stop for a length of time depending upon how much we gave them. But this does not mean that money is the treatment of the dreadful disease of burglary, or because we prevented certain individuals from continuing to burgle it means that we had reduced the disease of burglary in society as a whole. Rather, we would have encouraged its spread.

This is precisely the logic that has been applied to drug addiction. Just how precisely is evident from the Government's recent declared policy that clinics should now give drug addicts money or other rewards for not taking drugs (as least as proved by drug-free urine samples, something experienced drug addicts have long learnt to provide). This is the first time in the history of medicine, so far as I know, that bribery has been considered a medical treatment.

Contrary to what everyone supposes, withdrawal from heroin is not a serious medical condition - unlike, say, withdrawal from alcohol when it results in delirium tremens (the DTs). The suffering is grossly exaggerated and, in so far as it is genuine, is largely produced by anticipatory anxiety that is itself the consequence of years of mythologising the fearsomeness of withdrawal.

Addiction to heroin is a medical problem only to a minor extent, which is why predominantly medical means will never solve the problem. Most of Britain's 300,000 addicts are drawn from broken families, have a poor education, are without much hope for (or for that matter fear of) the future and have no cultural life, intellectual interests or religious belief. Delusory euphoria - the paradise at three pence a bottle that De Quincey described in his Confessions of an English Opium Eater - is the best that they think that they can hope for in life. This is not a medical problem. Where addiction is concerned, it is time to throw physic to the dogs.

Theodore Dalrymple is a retired prison doctor and author of Junk Medicine: Doctors, Lies and the Addiction Bureaucracy

Times Recommends

• Copenhagen will fail – and quite right too
• They’re still wriggling to avoid the flak on Iraq
• Rule the waves? Not any more we don’t