We must guard intelligence from corruption

By Lawrence Freedman

Financial Times

Published: July 14 2004 19:40 | Last Updated: July 14 2004 19:40

This has been a bruising time for intelligence agencies. On Wednesday, Lord Butler noted "serious flaws" in the British intelligence used to support war against Iraq. The American agencies have fared worse than their UK cousins. Last week, the US Senate intelligence committee lambasted the Central Intelligence Agency for judgments that were "either overstated, or were not supported by, the underlying intelligence reporting". Still to come is the report of the independent commission investigating the terrorist attacks of September 11, 2001. The CIA is now demoralised and George Tenet, its director, has departed. Lord Butler has determined that British intelligence should not suffer the same fate. Whatever the mistakes made by the Joint Intelligence Committee (JIC), he supports its former chairman, John Scarlett, as the next director of MI6.

In a formidable narrative history of the development of both policy and assessments, Lord Butler's report makes clear that the intelligence case was sincere and not contrived. It draws attention to the power of inherited assumptions, confirmed in one assessment after another and, in this case, reinforced by "groupthink" on an international scale. His critique i s far less damning than the US Senate's of the CIA: on other pressing issues the JIC got a lot right; with Iraq it got the intention right but the capabilities wrong and here the problem was the thinness of the intelligence - not its misrepresentation. There was not the same reliance as in the US on émigrés.

It was prudent and safe, given Saddam Hussein's record, to assume he was up to his old tricks. Optimistic assumptions would have been irresponsible so long as the objective was to keep the pressure on Iraq to comply with Security Council resolutions and get United Nations inspectors back into the country. But the presumption was not at all safe as a rationalisation for a controversial war. It would have taken a brave intelligence chief to decide at this stage that all previous assessments were mistaken. However, it would have been wise to acknowledge the extent of the uncertainty, especially when it was being publicly presented, and it was strange, after UN inspectors did get back in and f ound little, not to revisit the assessments.

The report makes clear one reason why the inherited assumptions had not been tested with sufficient rigour - the JIC's agenda had become extremely crowded. For those bothered about weapons of mass destruction, there was Iran and North Korea, the efforts of A.Q. Khan, the Pakistani scientist, to spread nuclear technology, dangerous tensions between India and Pakistan and the constant fear of the next big al-Qaeda attack. This all reinforced the lesson of the 1990s: every time the intelligence community had a chance to look deeper into an issue, starting with the Iraqi nuclear position after the 1991 Gulf war, things appeared worse than previously assumed.

This crowded agenda has not evaporated and the JIC will be called upon to comment in the future. The institutional checks suggested by the Butler report are sensible enough, but the main check will probably be a small voice saying "remember Iraq" at every temptation to offer bold assertions about a deve loping threat. The risk now is of excessive caution so real dangers will be missed, or warnings discounted.

This saga warns of how intelligence, when used to serve a wider political purpose, can be corrupted. By allowing the Blair government to make its case under the JIC's imprimatur, Lord Butler notes, "more weight was placed on the intelligence than it could bear". The dossier format did not lend itself to the nuance and subtlety that the seasoned intelligence officer can insert into even an apparently straightforward assessment. Yet strict separation can also be dysfunctional, because the agencies need to know the questions that bother the policy-makers. Nor is it possible to return to the traditional relationship that depended on levels of secrecy that can no longer be maintained. In this age of insecurity, much activity appears to be intelligence-driven. Aircraft flights are cancelled, individuals are arrested, and military operations are authorised on the basis of the "latest intelligence". It wa s the public's reluctance to accept assertions of danger on trust that led to the government's dossier of September 2002.

Lord Butler offers two possibilities for the future, in addition to a general injunction to acknowledge limitations. Either the government can draft its own dossiers, with the JIC checking any intelligence-based assertions, or the JIC "could prepare and publish itself a self-standing assessment, incorporating all of its normal caveats and warnings, leaving it to others to place that document within a broader policy context". Lord Butler strongly prefers the first course but that risks a repetition of this sad story. Intelligence should not be pushed back into its shell but drawn out even more.

Not long ago, no British government would even admit to the existence of intelligence agencies. Now the latter have their own websites. It should not be a great leap forward for their assessments, inevitably sanitised to protect sources and with full respect for inherent uncertainties, to be made known as a matter of course. It was better for the government to attempt to share intelligence on Iraq than to keep it hidden. But to avoid future accusations of mendacity, the agencies must learn to speak for themselves.

Sir Lawrence Freedman is professor of war studies at King's College London