Mission Critical

Paul R. Pillar (“Think Again: Intelligence,” January/February 2012) is right to rate intelligence analysis as merely one factor among many in presidential decision-making. In correcting an inflated view of intelligence, though, Pillar may have understated the role intelligence plays in shaping the quality of policy execution, if not its fundamental direction. Accurate intelligence and, every bit as important, timely intelligence can be critical.

Pillar points out, for example, that the intelligence community predicted some of the instability that would arise in Iraq after the fall of Saddam Hussein. This assessment, however, came in January 2003, three months after Congress had voted to authorize the use of military force. Had those same alarming assessments been available to lawmakers in September and October of 2002, the outcome of the debate would likely not have changed. But the airing of serious concerns about the aftermath of an invasion, backed by solid intelligence reporting, might have sharpened the thinking of policymakers responsible for planning for that aftermath.

A decade earlier, U.S. intelligence overstated the difficulty that coalition forces would face dislodging Iraqi forces from Kuwait. But the availability of those warnings early in the process — months before Congress voted to authorize force — helped the first Bush administration in its planning and may well have contributed to the buildup of an overwhelming air and ground force that guaranteed swift victory.

Pillar, an outstanding career intelligence officer, touches on a point worthy of deeper inquiry: the special difficulty that confronts the intelligence community when the problem at hand is to predict the results of various U.S. policy options. When the policy in question is a preemptive war, this kind of impartial analysis is critically important. As Pillar suggests, though, it may be information that is particularly unwelcome to the very decision-makers who need it most.

JOHN DIAMOND
Author, The CIA and the Culture of Failure
Washington, D.C.

Paul R. Pillar replies:

John Diamond makes several valid and useful observations. Intelligence is indeed used routinely to support the execution of decisions, and it tends to be employed more for that than for the making of major decisions. Additionally, the assessment of policy outcomes does place intelligence officers in a difficult situation, made all the more difficult when what is being assessed is not merely an option but rather the likely consequences of a course already decided upon. All this raises the question of how much we should expect from intelligence officers in saving policymakers from their own folly. Diamond is also correct that completing the assessment of post-Saddam challenges in Iraq earlier probably would not have changed the outcome of what passed for debate on the war.

But no policymaker in either the executive or the legislative branch asked for that assessment. I initiated it, and it was completed only after a thorough process that contrasted with the three-week rush job that produced the infamous October 2002 weapons estimate. It is unlikely that earlier completion would have affected planning for the invasion’s aftermath since the war planning was not following a schedule based on a congressional vote. In any event, the war planners were overpowered by a civilian Pentagon leadership determined to try to fight the war on the cheap — quite unlike the 1991 war in Kuwait. Those leaders ruthlessly rejected Army Chief of Staff Eric Shinseki’s estimate that “several hundred thousand soldiers” would be required for success in Iraq.