The pressure to overhaul the nation's intelligence agencies has become so intense that there is probably a greater risk that the job will be done poorly than not done at all. Nevertheless, the pressure is being applied in the right area.
With three catastrophic failures in less than two years, the American intelligence system is a demonstrable shambles, unable to deliver the goods. It was unable to turn up the Sept. 11 plot, despite some telling clues; it was flat wrong on the presence of weapons of mass destruction in Iraq, helping to propel the United States into a war whose primary justification was incorrect; and it failed to predict the fierce insurgency that arose after Saddam Hussein was toppled. It is impossible to plausibly conclude that the federal intelligence function requires anything less than a massive overhaul.
Predictably, the intelligence agencies, themselves, are resisting the notion that their failures are evidence of a need for fundamental reform. The acting CIA director, John McLaughlin, for example, is already rejecting the expected call to create a new cabinet-level intelligence chief.
But of all people, McLaughlin should understand the need for change. Although the head of the CIA is responsible for the nation's intelligence function, he has little authority over its many component parts, which include the Pentagon, the FBI, the State Department and even the Department of Energy. Indeed, the Department of Defense claims 80 percent of the federal intelligence spending, said Peter Brookes, senior fellow for National Security Affairs at the conservative Heritage Foundation.
It should be obvious to everyone at this point that the system is too fractured for the nation's well-being. A cabinet-level officer, with the authority to compel the various agencies to share their information sounds like nothing less than common sense. But more than that is required.
The nation continues to need to bolster its "human intelligence" capabilities, which were gutted following episodes of domestic abuses in the 1960s and 70s. And intelligence processes and conclusions need to be subject to confidential outside auditing as a counterbalance to the inevitable inbreeding that occurs in this kind of closed system.
That is the problem that allowed most of the intelligence services to see what they wanted in Iraq rather than what was really there. In fairness, that is also a political problem; the Bush administration was primed to hear only what it wanted about Iraq. But Americans have direct recourse over their political representation. They have no such influence over intelligence gathering.
The need for change is urgent, but not so urgent that it should be done hastily. Immediate needs are best addressed by the internal improvements that have already occurred, and by the hard-won knowledge that we have enemies who mean us harm.
But nothing is more important over the long term than to remake the nation's intelligence function in a way that is smart, useful and durable. It is probably not the kind of change that should be undertaken in the pressure-cooker atmosphere of a presidential campaign, but the subject is appropriate to the debate and the work should begin as soon as possible after the election. We don't have time to waste, but we have the time to do it right.