Phillip Knightley: A wilderness of mirrors

The failure of British intelligence over Iraq was catastrophic but predictable

Poor old James Bond has had a terrible thrashing this week. First the former British ambassador Sir Peter Heap accused Bond and his colleagues in the British Secret Intelligence Service (SIS) of being useless spies who frequently made things up. Then the Iraq Survey Group (ISG) reported that it could find no weapons of mass destruction in Iraq, thus underlining a fundamental failure of intelligence and removing at a stroke Britain's justification for going to war.

What's going on and who's to blame? First of all, Bond himself - or rather his creator, Ian Fleming. Fleming had been in naval intelligence during the Second World War and then ran a private worldwide intelligence network based on the Kemsley Mercury news service. He knew how real intelligence worked but preferred to write about the glamorous fantasy world of Bond. As a result, most of us have no idea what spies do, how dodgy a lot of their reports are and how often their services get it wrong.

Iraq, far from being unusual, is merely another item in a long list of intelligence failures, not just British ones but American and Soviet ones as well. According to the Royal Institute of International Affairs, Western intelligence's success rate in predicting Soviet moves during the Cold War was no better than many a think-tank. SIS and the CIA failed to predict the first Soviet atom bomb, the Chinese invasion of Korea, the Hungarian revolt, the siting of Russian missiles in Cuba, the Soviet invasions of Czechoslovakia and Afghanistan. Above all, both services failed even to imagine the collapse of Soviet communism and the end of the Cold War, and had no idea that Saddam Hussein would invade Kuwait.

The KGB's overall performance was no better. The Russian historian Vladislav M Zubok said the KGB had "delusions of grandeur" about its work. One of its generals, Oleg Kalugin, admitted: "When people in the West say that Soviet intelligence penetrated the higher echelons of Western government, I know that this is not true."

The problem is a flaw at the heart of intelligence gathering which is difficult to eliminate. It was behind Sir Peter Heap's moan about spies - one shared by many an ambassador - that he was required to allow SIS officers pretending to be diplomats to work out of his embassy.

As an ambassador in Ankara once said to me, "I'm here trying to foster good relations between Britain and Turkey and I have to share my embassy with British spies who spend their time trying to persuade Turkish citizens to be traitors. Is it any wonder I'd like to see the back of them."

This situation arises because no matter how well-trained an SIS officer is, no matter how good his languages, he can hardly head off to Baghdad and go around asking Iraqis if they've seen any weapons of mass destruction recently.

So from his base in the embassy (and with diplomatic immunity if he is caught) he tries to recruit as agents Iraqis who might have access to the knowledge he seeks. Ideally, he tries to find someone who will do this (and risk his life) for ideological reasons. If he cannot, then he offers large sums of money.

But agents selling information are not the most reliable sources. Money tempts them to exaggerate, inflate the importance of their contacts, copy information from obscure journals that their controller is unlikely to have seen, and even make things up. This gives the SIS officer running them many a headache. He may claim in his reports to London that his agent has been reliable in the past but he may not really know whether he is on this occasion.

He may not know how his agent is getting his information. As Sir Peter Heap put it: "It would make a huge difference in assessing the value of a report from, say, 'a source close to the president' to know whether that source is the vice-president, or a household servant, or someone with whom the president lunches occasionally."

British spymasters are well aware that an agent's information could be dodgy and there are procedures to catch it before it makes its way too far up the consumer chain - to the Joint Intelligence Committee (JIC) and then on to various government departments. The spymaster may call on his professional instincts: does this report "feel" right?

But then spymasters are human and attuned to the mood of the politicians who pay their salaries. If SIS and the JIC were aware that the Government was determined to invade Iraq but needed to make a case for war to get the British public on side, then wishful thinking may have led them to believe the 45-minute claim and the existence of weapons of mass destruction.

So if it were a case of: decision to go to war and then intelligence to support that decision - and it certainly looks that way - the British intelligence community is dangerously compromised, in the pocket of the Government, politicised and no longer objective.

What SIS appears to have lacked in its intelligence assault on Iraq was an agent in Iraq's own intelligence service. This is the five-star coup that all spymasters dream about. George Blake (alive and well and living in Moscow) was an SIS officer who was recruited by the KGB. He became one of the most successful spies in history, negating several major Western intelligence operations by revealing them to the Russians in advance. Oleg Penkovsky was a Soviet military intelligence officer recruited by SIS. He brought to London Soviet missile manuals which enabled us to identify what Moscow was up to in Cuba during the missile crisis that nearly led to a third world war. Kim Philby was a British SIS officer recruited by the Russians who was able to keep them informed during the Second World War of German efforts to persuade Britain to negotiate a separate peace with Germany and unite in a war against the Soviet Union.

But again there is a flaw in such intelligence operations that tends to blow them apart. One of the major emotions in all intelligence services is paranoia. In a business riddled with double-cross and betrayal there is always the fear that you are being played for a sucker.

So it would have crossed the mind of some SIS officers that the 45-minute claim might have come from an Iraqi intelligence plant with the aim of lulling the West into a false sense of security and then pouncing - that actually the WMD could be ready in five minutes.

When spies begin to think like this, the flaw triumphs and intelligence becomes useless. For the better your spy in the enemy camp is, the less likely you are to trust and believe him. Thus Moscow ignored some of the Cambridge spies' best information because it could not believe they could have gathered it unless they were part of a devilish SIS plot to mislead the KGB.

What's true and important? What's true and not important? What is disinformation designed to mislead? What is a mixture of both? Which agent is trustworthy? Which is an opposition plant? Are we conning them, or are they conning us? Have we been penetrated? If so, who is the traitor and what damage has he done?

And so the intelligence game goes on, a wilderness of mirrors. In most cases it produces no winners, no losers - except the taxpayer - and no information of lasting value. It's not the world of James Bond, but that's how it really is.

Phillip Knightley is the author of 'The Second Oldest Profession' (Pimlico), a history of spies