When Everything is Secret, Nothing is Secret

By Benjamin Wittes
Sunday, October 31, 2010, 9:43 AM

For more than a decade now, Steve Aftergood of the Federation of American Scientists has been on a lonely crusade to get the government to declassify the annual top-line budget number for the intelligence community: How much the United States spends on intelligence activities each year. Steve's crusade has involved several litigations (which he, a non-lawyer, has sometimes handled expertly and professionally as a pro se litigant).

His settlements some time back prompted the declassification of the top-line numbers from fiscal years 1997 and 1998. But then the government refused to do more. It resisted Steve's subsequent requests, and litigated successfully to keep the FY 1999 and 2002 budget figures classified. And while the government started declassifying the top-line figures for the intelligence community components in 2007, it did not include military intelligence programs in that figure, making comparison with the 1997 and 1998 figures impossible.

Even where Steve's quixotic campaign has failed, it has forced very senior government officials to defend the frankly ridiculous practice of keeping the top-line number secret, as when then-DCI George Tenet filed an affidavit claiming that supplementing the 1997 and 1998 figures with a comparable one for FY 1999 "could be expected to cause damage to the national security and would tend to reveal intelligence methods." And it has drawn significant attention, and appropriate ridicule, to the practice.

This week, the government released the aggregate number for FY 2010: $80.1 billion. As the Washington Post reported, "The so-called National Intelligence Program, run by the CIA and other agencies that report to the Director of National Intelligence, cost $53.1 billion in fiscal 2010, which ended Sept. 30, while the Military Intelligence Program cost an additional $27 billion." The release, Steve tells me, did not come in response to any direct pressure from him; he had no litigation pending. And yet Steve is one of the main reasons this subject is on the policy table at all, so the release is an occasion for congratulations.The top-line budget number is, in and of itself, not very important. It shows enormous growth in intelligence spending over the past few years--which is exactly what one would expect in an era in which America has been fighting two active wars and using covert intelligence actions instead of getting involved in several others. The number is important for symbolic reasons, however: Its classification is a vivid illustration of the government's inability to distinguish real secrets from fake ones. And irrational secrecy makes people outside of government--and particularly in the press--contemptuous of government secrecy claims in general, many of which deserve in my opinion greater respect than they tend to receive. If everything is secret, then nothing is secret. And if the government can't tell the difference between genuinely sensitive information and the data points of a trend line whose trajectory is, well, intuitively obvious, it can't then expect the press to make such distinctions either.

Secrecy policy is about discipline--the discipline of keeping secret material secret and also the discipline of not trying to keep non-secret material secret. This one is an easy case, and the administration deserves credit for not digging in its heals. As Steve put it in an email this morning, "The government's capacity to 'change its mind' on contentious topics such as this is what makes public policy debates meaningful and interesting. So for that reason it's a really encouraging step, even if it took much too long." Mostly, however, Steve deserves credit for keeping people thinking about this in-and-of-itself unimportant number whose irrational protection stands for something very important.