John Bolton has an excellent editorial in the Wall Street Journal discussing intelligence community (IC) organization. He ultimately recommends doing away with the director of National Intelligence and resubordinating the IC to the National Security Council (NSC), a proposal with some merit. He also makes the exceptional point that one of our biggest issues with intelligence is not so much collecting or analyzing it as assessing its implications.
I confess to being something of a hobby-horsewoman on the latter topic, and I am energized to see Bolton bring it up. He correctly implies that it is squarely within the purview of the president and his advisers to assess the implications of intelligence. Everyone on earth may agree that Iran has a nuclear program that could be readily adapted to the production of nuclear weapons, but only the president of the United States has been elected to decide the implications of that for the policy of his nation. Assessing these implications is inherently a political process, and that means the IC should not be performing it at all, except under policy guidance from the executive.
Intelligence cannot answer the question “Should we strike Iran now?” That is a policy question, but too often a great confusion arises on that head, abetted by the handling of intelligence itself. The “Iraqi WMD” controversy is a superb example of such confusion. No form of intelligence could have ensured that we struck Iraq at precisely the right moment to guarantee a smoking gun. The president made a policy decision; his critics, however, have argued ever since that he made an intelligence decision. Their unspoken premise is that intelligence can naturally appear in a form that obviates policy assessments and eliminates risk, and that Bush truncated some accepted process by not waiting for it to do so.
Intelligence doesn’t do that, however. Bolton is right to raise the issue of assessing policy implications, because it’s in this area that the controversy and vulnerability of the process are concentrated. Every celebrated instance of public dissatisfaction with intelligence is predicated on assumptions about the policy implications of that intelligence. Those assumptions are rarely examined in any systematic way, and presidents almost never address them or seek to shape them in the public mind.
Keeping open the national-security option of preemption, which is peculiarly reliant on intelligence, means that the question of policy implications from intelligence will recur for us in the future. The organizational correctives we apply should honor the executive’s constitutional responsibility for assessing policy implications. Bolton may be right about having the IC report to the NSC; that arrangement would in some ways mimic how the military plugs intelligence into planning and execution. At some point, however, we will have to come to grips with the persistent and legitimate possibility that, in any given situation, what we are seeing among our politicians is a disagreement on policy implications rather than on the intelligence itself. The IC is not responsible for the management or the outcomes of such disputes, and it should not be organized in a way that encourages it to participate in them.