The Lincoln Plawg - the blog with footnotes

Politics and law from a British perspective (hence Politics LAW BloG): ''People who like this sort of thing...'' as the Great Man said

This page is powered by Blogger. Isn't yours?
Wednesday, September 10, 2003

Iraq intel: can systemic failures justify a bad faith charge?

The whole sorry saga under investigation by Lord Hutton started with the allegation in the infamous Today broadcast by Andrew Gilligan at 0607 on May 29 that HMG knew that 45 minute claim was
even when they put it in the dossier. By his 0732 broadcast, wrong had changed to questionable.

And, though the Inquiry has dealt at some length with the minutiae of the process by which the dossier was drawn up, the only finding worth having from Hutton on the dossier question will be one of bad faith imputable to Tony Blair.

Looking at the claim in isolation, it may well be that Hutton will conclude the charge against Blair unproven: that his delegating the drafting of the dossier to JIC Chairman John Scarlett is sufficient to pass the buck, whatever the volume of suggestions from Alastair Campbell and others from the No 10 machine.

But that does not exhaust the possibilities. Testimony of Anthony Cordesman to the House Intelligence Committee [1] suggests an alternative approach.

Cordesman is scathing about the quality of the intelligence about Iraq available to USG and HMG before and during the war. But, for our purposes, hindsight needs to be excluded. On page 20, he identifies a number of features that seriously compromised the pre-war intelligence on Iraqi WMD - and, so far as I can see, those features would all have been apparent to Blair and his men at the time the dossier was created in September 2002. They relate to organisation of intelligence evaluation, as, for example:
Security compartmentation within each major aspect of collection and analysis severely limits the flow of data to working analysts. The expansion of analytic staffs has sharply increased the barriers to the flow of data, and has brought large number of junior analysts into the process that can do little more than update past analyses and judgments. Far too little analysis is subjected to technical review by those who have actually worked on weapons development....
or artefacts of the process:
Analysis tends to focus on technical capability and not on the problems in management and systems integration that often are the real world limiting factors in proliferation. This tends to push analysis towards exaggerating the probable level of proliferation, particularly because technical capability is often assumed if collection cannot provide all the necessary information.

Cordesman is talking primarily about the way the US agencies operate: perhaps their UK counterparts don't suffer from the same failings.

But he also points out that
...the users of intelligence are at best intolerant of analysis that consists of a wide of qualifications and uncertainties even at the best of times, and the best of times do not exist when urgent policy and warfighting decisions need to be made. Users inevitably either force the intelligence process to reach something approach a definitive set of conclusions, or make such estimates themselves.

Intelligence analysts and managers are all too well aware of this fact. Experience has taught them that complex intelligence analysis --filled with alternative cases, probability estimates, and qualifications about uncertainty --generally go unused or make policy makers and commanders impatient with the entire intelligence process. In the real world, hard choices have to be made to provide an estimate that actually be used and acted upon, and these choices must either by the intelligence community or the user.

He concludes (p22) that
In practical terms, any political effort to try to communicate the true level of uncertainty and probable outcomes inherent in most estimates of proliferation seems almost certain to make it difficult or impossible to gain a political consensus for timely and effective domestic or international action. Communicating uncertainty may be a good way of arguing against action but only because its impact is to create nearly endless discussion and debate and block agreement on any policy that requires broad political agreement on a single course of action or the use of military force. In practical terms, the US and its allies may again have to act on the basis of something approaching "worst case" assumptions, and this is a risk that proliferating nations and extremist movements may have to learn they take when they proliferate.

Cordesman, who, from memory, was a supporter of the invasion, doesn't exactly seem outraged by the prospect that future preemptive wars will be sold on the basis of deliberately skewed intelligence! I suspect that the straight Scots-Irishman with Blair's fate in his hands may not take quite so sophisticated a view.

The idea of the deception of the public for its own good is reminiscent of the appalling frankness speech of Prime Minister Stanley Baldwin on November 12 1936, in which (very roughly) he said he not made a call for rearmament before the then recent general election (despite intending to rearm) because he knew he would never have been re-elected if he had done so!

Cordesman offers a further nugget on the dossier:
Moreover, a detailed comparisons (sic) of the British and CIA reports shows that the British document often implied that intelligence had more certainty than the US document, although both governments shared virtually the same intelligence. It is clear from the investigation by the British parliament that this was partly because the British report had a much heavier degree of editing by the Prime Minister's office.

Clearly, he fails to appreciate the subtleties of the HMG organisation chart. Or perhaps he's unconvinced by the repeated protestations of John Scarlett's utter and total independence...

  1. On July 24 (PDF) - scarcely breaking news!

free website counter Weblog Commenting and Trackback by