Further to our item here yesterday on Trump’s apparent inclination to reform (or punish) the intelligence community, it’s worth looking back to another of Herbert Meyer’s reflections on the problems of intelligence that he wrote ten years ago when controversy over a leaked National Intelligence Estimate (NIE) of the Bush Administration was much in the news. Here Herb reflects on the inherent problem of the institutional biases of intelligence agencies, which can’t be fixed by re-organizations or “making a deal.”
One problem inherent to NIEs is that they sometimes reflect nothing more than the institutional biases of each of the 16 participating agencies. A second inherent problem is that sometimes these agencies are so determined to not be proven wrong about what the future holds that they try to have it both ways, for instance by obscuring their projections beneath an avalanche of “on the one hand, on the other hand” sentences.
The best and most concise description of NIEs that suffer from these problems comes from President Reagan’s great Director of Central Intelligence, William J. Casey: “total crap”.
That’s why Casey’s orders to me were to make certain that the NIE’s we produced for President Reagan overcame these problems.
First, I was to sort though the differing judgments of the 16 agencies to understand if they were basing their conclusions on the facts contained in the text of the NIE itself – or merely on long-standing institutional biases. If the latter, my job was to confront the agency representatives and then work with them to align their judgments with the facts.
Second, when an agency wanted to dissent from the consensus, it was my job to assure that this dissent was written as clearly as possible so the President could understand not only what this agency was saying, but why it had chosen to dissent from the majority view.
Finally, when all the bureaucratic fighting had ended and we had hosed the blood off my office walls, it was my job to run the crucial “Key Judgments” of the NIE through my word processor one last time, to assure that the finished product was intelligible to an intelligent but busy policymaker. That meant knocking out all the “on the one hand, on the other hand” sentences and replacing them with sentences that made a point. It meant eliminating the gobbledygook sentences that invariably had crept in, such as: “We judge that Soviet leaders will be neither too hasty nor too reluctant to either over-react or under-react to the developing circumstances flowing from the new initiative.” It meant weeding out Key Judgments that were accurate but worthless – such as the old standby: “We judge that the future of US-Soviet relations will be volatile and subject to change.”
As usual, worth reading the whole thing.