The Next NIE
There is Value in Recognizing Ignorance
By Michael Tanji | December 10, 2007
Ample ink has been spilled by both ends of the political spectrum on what the latest NIE on Iran’s nuclear capabilities means. Partisans in both camps have reason to love and hate the thing, or more precisely what they think is in the thing, given that we are dealing with just four pages of unclassified and high-level conclusions from 150 pages of narrative and supporting material. The folly of judging important books by their covers notwithstanding, one question remains to be asked: with all that may be wrong with an NIE, what should be done to make them more right?
By “right” I do not mean what political slant they should take (ideally, none). To a certain extent I do not even mean how correct they should be. By “right” I mean how accurately they explain what is known and more importantly what is unknown.
Allow me to explain.
As a former intelligence analyst I recall a time when an enlightened boss with sufficient bureaucratic muscle would allow analysts to include a section in their assessments labeled “intelligence gaps.” “Key judgments” are the intelligence community’s version of the outside world’s “executive summary.” I’m not aware of a parallel to “intelligence gaps,” since like most authors trying to convey expertise, intelligence officers do not like to suggest that there is something they don’t know.
The importance of the gaps section in an intelligence assessment cannot be understated. It was not simply a laundry list of factors that you were ignorant about; it served as a management tool that would help craft future collection requirements. Satellites not taking pictures of the right sites? Not recruiting agents in the right locations or asking them the right questions? By pointing out where collection was coming up short and how you were under-serving policymakers, you justified your request for a change in tactics and demonstrated your willingness to respond to consumer needs.
This wasn’t a one-way street. Collectors get more credit when the gather information of high value. One report with a breakthrough piece of information is worth infinitely more than 20 reports telling you some variation on what you already knew. By providing what analyst’s needed, collectors ended up helping themselves. Likewise, if the list of unknowns grew smaller, consumers could better determine if intelligence was serving their needs and bolster their confidence in our assessments.
Of course like any exercise conducted in a bureaucracy, this collector-analyst feedback process was spotty at best. You had to make a serious effort to carve out the time the fill out evaluations and in a job that is often nothing but juggling one “must do now” task after another, collector feedback was often viewed as a housekeeping task that could be put off, usually indefinitely. As much as anything, this internal intelligence failure contributes to the community’s sorry record of gathering meaningful intelligence on a wide variety of targets, particularly hard targets like Iran.
There are many ways to tackle an intelligence problem analytically, but what makes or breaks an intelligence assessment is the information available to be analyzed. One unimpeachable source on one discrete topic can make for a very confident assessment indeed, but few intelligence problems are so narrow and simple. Read the key judgments of this latest NIE backwards to see what I mean. Viewed in this fashion it says, “We are not terribly confident in these discrete elements, but when we view them as a whole our confidence is boosted.” That’s akin to an airline mechanic signing off on the airworthiness of a jetliner knowing that almost everything he just inspected is dodgy. Who wants to board the plane first?
As mentioned previously, the full NIE is 150 pages long, so enough collection may not be a problem. However, the fact that the authors make such an effort to couch and caveat what they think they know indicates that they are making up for a great deal of ignorance. Should we be satisfied with such ambiguous and circumspect descriptions of the information our nation’s leaders will be using to make the critical decisions of our time?
Your author has commented elsewhere on what he feels is really at work with this NIE. While it may very well be a partisan political agenda made manifest, experience suggests that the intelligence community is doing its utmost to avoid another public failure on par with Iraq’s WMD programs (or India’s nuclear weapons test, or the collapse of the Soviet Union, etc.). This includes but is not limited to using every linguistic trick in the book to ensure that anything written about any intelligence problem could be looked at in hindsight as “right” if interpreted in a particular way.
The community owes the leaders of our country, regardless of their political affiliation, better.
People laughed at former Secretary of Defense Donald Rumsfeld when he spoke of the problem of “unknown unknowns.” Odd as his verbiage was, he was driving home the point that you cannot make sound decisions about national security if you are in any way ignorant of your adversary’s intentions and activities. Does anyone really think that one of our toughest intelligence targets suddenly became transparent or is it more likely that there is more going on here than meets the eye?
Anyone in the current administration – and any aspiring presidential candidate – that is interested in leaving a noteworthy imprint on the intelligence community should make mandatory the inclusion of intelligence gaps in every NIE and major intelligence product. Our adversaries do not need to know what we don’t know, but the community would be doing itself and its customers a great service by remembering the significance of ignorance.