PERSONAL PERSPECTIVE : In the CIA: Fighting to Keep Information Unskewed

Whether one favors Robert M. Gates to become Director of Central Intelligence or not, the Senate Intelligence Committee’s confirmation hearings have shown the American people that the world of intelligence is a more complex world than that of James Bond. It is certainly as full of passion--albeit of a different kind.

The Senate hearings pointed up the centrality of intelligence analysis. That is what intelligence is really about: pulling together and sharing all available information on a subject; coming up with the “so what” of that information; submitting these findings to fierce scrutiny and debate, and then instructing senior policy-makers about the true state of the world. This process and this service--not covert-action shenanigans--are the principal purposes of U.S. intelligence.

Good analysis is always beset by numerous hazards. Chief among these: Can analysis be wholly objective? How do we know that it isn’t being skewed? And how do we know that U.S. foreign policy will be based on the intelligence facts and analyses senior decision-makers have received?

Can analysis be fully objective? Not really. Try as we may, every analyst--whether intelligence officer, professor, journalist or whatever--is formed by his or her life experiences, and all new information, every new stimulus, has to pass through those existent prisms. Yet, over the years, the intelligence community--officials of the Central Intelligence Agency, the State Department, the Defense Department and certain other offices--has built up interlocking systems of analysis that, if untrammeled by skewing, create the best opportunity to produce the most objective judgments.

This structure was born in 1947, when President Harry S. Truman and the Congress established the CIA. Its explicit purpose was to end the turf-protecting and squirreling away of intelligence that had contributed to the Pearl Harbor disaster, and so avoid a far more horrible future Pearl Harbor.

The systems of intelligence analysis and national estimates that CIA Directors Walter Bedell Smith and Allen Dulles set up demanded of intelligence officers that their highest calling be integrity of judgment. Their priority was to go where the evidence took them, and then candidly tell it like it is to their bosses and their policy-making customers. The intelligence officers’ first criterion was to play it straight--not to please superiors by catering to their preconceptions or known commitments. It is significant that every President since Truman has continued this system--even though some, on occasion, ignored or scorned the intelligence judgments.

How do we know that intelligence analysis isn’t being skewed or cooked? We don’t--if by “we” is meant the general public. The public is well-served in this respect, however, by the Senate and House intelligence committees.

A chief purpose of such oversight is to help discern when, if at all, senior intelligence officers are so certain of their own omniscient judgments that competing analyses do not get a fair hearing--the word goes down the intelligence chains of command that certain judgments will not be appreciated, and the danger arises that only one analytic line will be served up.

The most difficult skewing to guard against, however, is down in the analytic trenches, when an officer self-censors his or her own judgments--gutless in Gaza--out of fear of displeasing the boss. There is no protection against this except supervisors bred in the tradition of telling it straight, who insist their analysts do the same--even if the latter come up with judgments challenging their own.

At the same time, all the red-penciled editing of supervisors is not necessarily skewing. Ideally, supervisors are more experienced--they write better, they know better what the customer wants in terms of format, length and presentation. For example, my first proud analytic piece, years ago, came back from my supervisor with just one comment: “This stinks.” Years later, I came across that crumpled document in the recesses of my safe. Sure enough--it did.

Other differences of analysis result when two or more analysts interpret available information differently. In such cases, the give-and-take of debate refines the matter, sharpens the analytic message and clarifies where and to what degree dissent exists. The system provides well for such procedures--and for the clear registering of dissent, whether within or between agencies of the intelligence community.

Where these procedures prevail, our tax moneys are well-spent. Where these procedures are skewed, then our taxes have ill-served policy-makers and heightened the possibility that some foreign-policy gaffe or disaster will occur.

Intelligence is not perfect or all-knowing. It can be wrong, and the instincts of policy-makers sometimes better. But the record indicates the contrary, that policy has more likely gone astray--as in the case of the Vietnam War--when professional intelligence judgment was ignored.

This leads to the third hazard: How do we know policy-makers will pay attention to the intelligence community’s judgments? We don’t--and there is no law that says they have to. Only the anomaly of scorning the information and the judgments that millions of dollars of intelligence tax moneys have produced. But countless considerations other than intelligence judgments go into policy--prior commitments, congressional restraints, available moneys, available force, the steadfastness of foreign allies. Public opinion is important, too.

More than ever, now that we face a radically changing world, we citizens need the best, most imaginative and most objective intelligence analysis possible--and the closest marriage of it with policy-making officialdom. The intelligence community is filled with skilled, dedicated analysts. The trick will be to use their talents, free from pressures to toe any line, and to augment their ranks with new blood, including the best brains in the country.

This may or may not require major organizational change. But more important than reorganizations, or flow charts, or fancy new collection systems, will be the quality and integrity of the analyst trying to dope out some foreign political or social situation where data are incomplete and the answer must come mainly from experience and feel. Hence the greater investment must be the developing of human intelligence capital.

The CIA’s analytic tradition of playing it straight, of trying to avoid skewing, is a good model for use elsewhere. One such place is Capitol Hill, where Gates deserves to be confirmed or not on the merits of the case--not on the skewing to which some of his senatorial supporters have resorted.