Our intelligence agencies need more radical reforms than have so far been proposed

Britain's foreign intelligence agency, better known as MI6, is in the throes of reform - we learn from smoke signals emanating from its Vauxhall headquarters and interpreted on our behalf by the Foreign Office. The changes include a brand new and far more senior head of reporting - a boss spy whose job will be to check the analysis of others - the recruitment of a senior business figure whose task is to improve the running of the agency, and the introduction of more people with operational experience into analytical roles.

Britain's foreign intelligence agency, better known as MI6, is in the throes of reform - we learn from smoke signals emanating from its Vauxhall headquarters and interpreted on our behalf by the Foreign Office. The changes include a brand new and far more senior head of reporting - a boss spy whose job will be to check the analysis of others - the recruitment of a senior business figure whose task is to improve the running of the agency, and the introduction of more people with operational experience into analytical roles.

As with anything relating to intelligence and national security, there are limits on our ability to judge the extent of these reforms from the details that have been confirmed. As the current head of the Pentagon so memorably said, there are known unknowns - the things we know we don't know - but there are also unknown unknowns - the things we don't know we don't know. And so it is with MI6. The place is a thoroughly closed book, and the Freedom of Information Act, which exempts the intelligence services, offers no remedy.

Given the extent of the intelligence failure before the war in Iraq and the sharp criticism of the intelligence services in the Butler report last summer, however, the Government and MI6 surely have a common interest in convincing us that they are engaged in a major overhaul. But if the reported reforms are the best they can offer, we can only conclude that they fall very far short of what is needed.

The truth is that if MI6 and the other intelligence services had not been so grievously mistaken in their assessment of Iraq's weapons capabilities, British forces would almost certainly not be engaged in Iraq today. Saddam Hussein's supposed possession of weapons of mass destruction in defiance of a string of UN resolutions was the official reason why Britain went to war. The ultimate responsibility, of course, rests with the Prime Minister and the Cabinet. It was Mr Blair's judgement of the advice and interpretation of intelligence that he was given that took us to war. And that judgement was wrong.

But, as the Butler report showed, the gloss that MI6 put on the intelligence it had gathered - and the comparative weight it gave to the multifarious strands of information - was also wrong. It is all very well for defenders of the intelligence services to say that they were essentially right in their judgement of Saddam Hussein - that he had never abandoned his intention of acquiring lethal weapons. But there is a great difference between this and the conclusion that MI6 and the Government drew: that Saddam had large quantities of prohibited weapons which constituted an imminent threat to world security.

The enormity of this misjudgement was illustrated again yesterday when the International Survey Group's last inspectors returned from Iraq empty-handed and the weapons search was formally abandoned. All in all, the tale of Iraq's non-existent WMD and the 45-minute detonation time must surely be accounted one of the most colossal and costly failures of intelligence ever. That MI6 actually "withdrew" some of the crucial details and retrospectively discredited the single source that had provided them matters little. The information had already been fed up the chain, "over-egged" or "sexed up" (take your pick), and used to clinch the argument for war.

Given the scale of this débâcle, it is hard to see how a weightier head of reporting, more experienced analysts or business management techniques can prevent a similar failure in future. What is needed is a comprehensive review of how intelligence is gathered and assessed; a thorough examination of the intelligence-sharing agreement with the United States and how it works, a new look at the way in which the intelligence agencies interact with the Government of the day, and much, much more transparency and accountability. Iraq's non-existent weapons showed that secrecy was no guarantee of accuracy; it meant only that the errors went unchallenged until it was too late.