The Uncertainty Dilemma
CIA can't know everything. So what?


In Voltaire’s Micromegas, a gigantic visitor from another planet asks a sailor why humans bicker so much. “Because we agree on the two or three things we understand,” explains the sailor, “and we disagree on the two or three thousand things we don’t understand.”  The bickering over Saddam’s weapons of mass destruction has become deafening — and we still don’t understand what was at stake.

In his new and much-discussed book, At the Center of the Storm, former CIA director George Tenet writes, “One of the greatest mysteries to me is exactly when the war in Iraq became inevitable.” Indeed, that is a mystery to many people, but the reason is that the underlying policy issue was never properly understood — even in the administration. Resolving the problems that arise at the interface of intel and policy in a security environment dominated by uncertainty may be the greatest intellectual challenge that has ever faced the government of the United States — and we have only begun to understand it.

Tenet writes that no serious debate was given to the imminence of the Iraqi threat. But he also makes it clear that this was beside the point:

The absence of evidence and linear thinking, and Iraq’s extensive efforts to conceal illicit procurement of proscribed components, told us that a deceptive regime could and would easily surprise us. It was never a question of a known, imminent threat; it was about an unwillingness to risk surprise.

In a later passage, he gets to the critical question:  

We should have said, in effect, the intelligence was not sufficient to prove beyond a reasonable doubt that Saddam had WMD. The evidence was good enough to win a conviction in a civil suit but not a criminal case. Would we have gone to war with such conclusions?

But intel is not evidence. Intel is by its nature fragmentary and inferential. The evidence we needed in Iraq was on the ground in Iraq, and Saddam controlled all of it. When Tenet said the case for war was a “slam dunk” he was jumping the barrier that should separate intel and policy. The intel was ambiguous. Intel is always ambiguous, especially when the question is as vast as that posed by Iraq. Because the presumption was already against Saddam — and rightly so — the intel community’s ambiguous answer left Cheney and many other people with no doubt that Saddam had WMDs — as a policy judgment based on history and the totality of the circumstances, of which intel was only a part. 

Policymakers do not have the luxury of coming to no firm conclusion simply because the intel comes to no firm conclusion. What has been consistently missing from this whole debate is an appreciation of the fact that, given Saddam’s history, we had to presume the worst. By 2002, the only thing that could assuage our fears about Saddam was transparency in his regime. Without transparency, we were facing a potential threat of unknown scope that — should it ever materialize — could cause much more damage than a punitive action would be able to remediate.

And that meant that it was Saddam — not the United States — who had the burden of proof as to the WMDs. It was the administration’s failure to understand this and make this clear which has led to the widespread — and absolutely false — perception that because the pre-war intelligence was mistaken, it was a mistake to invade Iraq. If policy should never “cook” the intel, neither should intel “cook” the policy. Even if the administration had known how unreliable the CIA’s intel really was, the problem facing it would have been exactly the same.

Think of it this way: After the stock-market crash of 1929, policymakers in the United States realized that when publicly traded companies commit fraud, a punitive reaction does not suffice to make the victims whole. So they responded by drafting the brilliant securities laws of 1933 and 1934, which placed a heavy burden of financial transparency on corporate America. Henceforth, we would not wait for evidence of fraud before taking punitive action. Today, publicly traded companies are penalized immediately just for failing to file independently audited financial statements according to a strict schedule. 

It has now become clear that without similarly draconian penalties for failing to keep proper records, the attempt to control WMD proliferation is going to fail. 


Sign up for free NRO e-mails today:

NRO Polls on LockerDome

Subscribe to National Review