Statement of R. James WoolseyMr. Chairman, Members of the Committee. It is an honor to have been asked to testify before you today on S. 1801, "The Public Interest Declassification Act of 2000".
Committee on Governmental Affairs
United States Senate
July 26, 2000
Let me say, first of all, that although the tools proposed by this bill are relatively modest, the bill seems to me to be a positive attempt to begin to come to terms with one of the most vexing problems in the important field of government secrecy - the issue of special searches of classified material for declassification review.
I think that the beginning of wisdom about this aspect of government activity - secrecy and declassification - is to recognize the need both for reform on the one hand and for caution and experimentation on the other. This bill is crafted in that spirit. I will turn in a moment to a few of its specific provisions, but let me first explain why I believe that this overall stance is the right one.
Reform is important because the system is broken and will soon be even more so, as the digital age adds reams of records - e-mails, to mention only one - to the government's vast store of classified material. It is obvious that much of this classified material would be useful to historians and other citizens for a range of important purposes -- and it is equally obvious that some of it was improperly classified in the first place.
I ordered the declassification of a number of files on covert actions during the Cold War some six years ago when I was Director of Central Intelligence. Some of that material has been released after subsequent review. Some has been said, subsequently, not to exist any longer in the government's files. And some (relating to Iran), has been lost inside the government but then, in effect, "released" through a leak to the press. On the Iranian material historians have benefitted, in part by being able to see that the U.S. Government's effect on the 1954 coup in Iran was far from central. After reading the material in the press I can see no defensible reason why, after review, it was not released officially by the government as I had ordered.
(You may have noted, by the way, that I have avoided mentioning the name of the government agency I used to head (let's call it "the C Agency") that was involved in these matters. This is because I returned from travel very recently and was only able to prepare this testimony late yesterday afternoon, when it was already past due to the Committee. Consequently there was no time to submit it for security clearance by that Agency, a procedure I was recently informed by the Agency's Chairman of Publications Review that I should follow if I even "mention" the Agency's name. Hence a slight coyness in prose style to comply with the letter - admittedly not the spirit - of what seems to me to be a rather bizarrely inclusive rule related to classification.)
I have recently represented several Iraqis in an immigration case in which the men were imprisoned based almost entirely on classified evidence introduced by the government that neither the men nor their counsel were permitted to see. After several influential Senators wrote to the Attorney General about the matter two years ago, the government said, in effect, "whoops", and released about 90 per cent of the evidence, saying that it had been improperly classified initially. Yet six men served two years in prison and two men are there still, in no small measure because of this improper classification. And the written version of the immigration judge's recent decision, announced from the bench in non-final form, to rule in favor of the remaining two men and release them has been held up for many weeks in the Justice Department. Need I say that the reason given by the Justice Department is that the judge's written decision is still undergoing classification review?
On the other hand, there is good reason for the government to be cautious with the release of some types of information, even many years after it is acquired.
It is not only the details of code-breaking or espionage tradecraft or agent identity that must be protected -- and frequently for many decades. Often the substance of what is known about a foreign government, or the time when it was known, can indirectly lead to the betrayal of, for example, an agent identity or a broken code. These things must be assessed carefully by experts before intelligence records are released.
Moreover, much of what the U.S. obtains in intelligence is obtained through liaison arrangements - essentially trading intelligence - with foreign countries. Those valuable liaison relationships will dry up if we release material, even after many decades, even if it is material that we ourselves would like to release, without the permission of the foreign intelligence service from which it was obtained. And these relationships save American lives. I daresay that any American who was a tourist in Jordan at the beginning of this year -- and whose life may well have been saved by very professional and cooperative Jordanian intelligence actions that thwarted terrorist attacks on American targets there - would probably not be an advocate of releasing material received from Jordan without Jordanian consent and thereby undermining U.S.-Jordanian intelligence cooperation in the future.
Because of the complexity of making many of these judgments, and the professional experience required, reforms should be carefully considered. They should not, in my view, at least in the intelligence field, rely on broad and automatic rules - such as a certain number of years since the document was created - for declassification criteria. Reforms should be tailored carefully to protect what must be protected for sound reasons, all of it, even if it is many years old, and to declassify the rest as promptly as possible.
As anyone who has worked in this area in government should know, this is not a simple job. There is more than one way to err. Statisticians talk of "type 1" and "type 2" errors - essentially sins of omission and sins of commission. A radar, for example, should ideally detect all enemy aircraft and have no false alarms, e.g. should detect no birds. But in the real world one often has to choose between catching all enemy aircraft as well as a few birds, or picking up no birds but missing a few enemy aircraft. One comes closer and closer to the ideal over time the better, the more experienced, the smarter one's radar designers. There is no substitute for careful experimentation, trial and error.
In this overall context the Public Interest Declassification Board established by the bill seems to me to be a positive step. Experimentation, trial and error are needed. Special searches have been overdone, but they can be valuable tools for the historian and others in some circumstances. One wants only the useful special searches, not the repetitive and trivial. The Board will not be able to achieve this ideal balance on its own, but its recommendations should contribute toward a positive set of rules about which searches are undertaken. Even cutting down on duplicative searches will be a step in the right direction for the hard-pressed professionals in the agencies that are trying to deal with this very difficult problem of making declassification decisions.
I do believe that it would be useful, as Professor Kimball has suggested, for the Board to meet at least two or three times a year and to consist of persons other than current officers or employees of the U.S. Government. It should also, in my judgment, be selected with an eye toward diversity of experience in the world of classified material. There should be both historians and former intelligence and military officers, for example. It is only out of debate over this difficult subject -- between people of good will who both have something to teach and the humility to realize that they also have something to learn - that we are likely to get any useful recommendations for improving the current unsatisfactory system.