President Bush praised the Robb-Silberman commission report for its scathing and perceptive analysis of "intelligence failures" in the "axis of evil" states of Iraq, Iran and North Korea. Indeed, the report contains many useful recommendations for improving intelligence on weapons of mass destruction. But the fallacy in the administration's appointment of a commission to study intelligence failures is that there is almost never such a thing as a pure intelligence failure. Intelligence failure is usually linked to policy failure.
It's easy to see why Bush, or any president, would not want to call attention to that link. But the commission should have.
Let's take the case of North Korea. While the commission's chapters on North Korea's nuclear program are rightly classified, the unclassified summary suggests that spies and satellites have yielded very little information about that country's nuclear weapons efforts. But what does it matter? North Korea has admitted, indeed boasted, of its growing nuclear arsenal, and the United States has done nothing to stop it. How could a few more details provided by the CIA make a difference? If you don't have a policy, intelligence is irrelevant. North Korea's runaway nuclear program is a policy failure, not an intelligence failure.
What's worse, policy failure has actually caused intelligence failure in North Korea. From 1994 to 2003 North Korea's plutonium was at a known location, Yongbyon, where it was measured, handled and surveilled by international (including American) inspectors. We could inspect it -- or bomb it -- at any time. But when North Korea threw the inspectors out and threatened to truck the plutonium away to a hidden location, the United States did nothing. In due course the North Koreans made good on their threat and took the plutonium away. Are we now supposed to believe that it is an "intelligence failure" that we don't know where it is?
A second member of the axis of evil, Iran, demonstrates the same point. Iran, unlike North Korea, denies it has a nuclear weapons program. The Bush administration firmly contends that it does and is almost surely right, even though the intelligence is apparently not a "slam dunk." But since the administration does not plan either to attack Iran's nuclear sites or to try to negotiate them away (the Europeans are supposed to be trying the negotiation route), it hardly matters whether we know all the details.
The "intelligence failure" that prompted the creation of the Robb-Silberman commission was, of course, Saddam Hussein's missing weapons of mass destruction. Here there surely was a policy -- full-scale invasion, no less -- and no one can accuse the Bush administration of inaction. Knowing what we thought we knew, invasion was absolutely the right decision. WMD are too dangerous to take chances. But Bush has since made it clear that even if he knew then what we know now -- that the information on Hussein's weapons was "nearly worthless," in the words of the Robb-Silberman commission -- he would have invaded anyway. There were other reasons for his policy -- Hussein's mistreatment of his population and the wider implications for the Middle East of his continued rule in Iraq. Future historians will decide whether his policy was a failure or a success, but they will know from his own testimony that the CIA's "intelligence failure" was not the determining factor.
It therefore is a fact that in the three most important cases studied by the commission -- Iraq, Iran and North Korea -- the intelligence failures the commission so carefully identifies and makes recommendations to correct made no difference to policy success or failure.
The commission's recommendations focus on improving intelligence on classical proliferation targets -- rogue regimes such as the three axis-of-evil states and Libya. But in the post-Sept. 11 world, we have to fear WMD not just in the hands of national governments but in the hands of terrorists. Here the policy-intelligence mismatch is also evident. Osama bin Laden has declared it a "sacred duty" of jihadists to get nuclear weapons. We hardly need more intelligence on his intentions. But to make a bomb, bin Laden's followers must get either plutonium or highly enriched uranium. U.S. efforts to safeguard these materials worldwide, even after Sept. 11, have been halfhearted. The tremendous success of the Nunn-Lugar program in denuclearizing the former Soviet Union in the 1990s has not been replicated in the post-Cold War era of terrorism. If the United States had such a vigorous set of policies to combat nuclear terrorism, it would need good intelligence to implement those policies. But until we get the policy right, it hardly matters that the intelligence is imperfect.
Without a comprehensive policy to combat WMD, better intelligence will not improve U.S. security. Bush was right to say that keeping the worst weapons out of the hands of the worst people is a U.S. president's highest national security priority. Since Sept. 11, under his leadership, we have scored many successes against the worst people. But U.S. policy toward the worst weapons is still in a pre-Sept. 11 state. Indeed, since Sept. 11 the United States has suffered greater setbacks in counterproliferation than at any time since the 1980s, when Pakistan went nuclear. Until this changes, preventing intelligence failures will still not matter.
The writer is co-director of the Harvard-Stanford Preventive Defense Project and was assistant defense secretary in the Clinton administration. He testified before the Robb-Silberman commission.