NARA and Declassification

Statement of William Burr

National Security Archive before the Public Interest Declassification Board

July 22, 2010

I thank the board for the invitation to appear on this panel to discuss the problem of historical Formerly Restricted Data or FRD. My name is William Burr and I direct the National Security Archive’s nuclear history documentation project.

I’ll begin with my principal recommendation: the U.S. government should rescind the secret status of historical U.S. nuclear weapons deployments in NATO Europe, the Western Pacific, and elsewhere. The objective of a new policy should be the maximum possible disclosure of overseas deployments 25 years old or older. A more open stance would follow directly from the Obama administration’s general approach to nuclear weapons policy and to government secrecy generally. With billions of dollars spent on thousands of nuclear weapons deployed overseas during the Cold War, it is a matter of accountability for the U.S. government to take a more transparent position. Moreover, if the history of nuclear weapons and the Cold War is to be better understood, the closed archival record on the role of nuclear deployments in U.S.-European relations and U.S. diplomacy in East Asia needs to be opened.

As arms control expert Joshua Pollack has put it, a "Big Shhhh" has loomed over past and present deployments of U.S. nuclear weapons in the territory of key allies in NATO Europe. Pollack's observation also applies to historic nuclear deployments in Canada and East Asia. Even though knowledge of the deployments is widespread, the presence overseas of U.S. nuclear weapons during and after the Cold War has been an official secret for decades, classified Formerly Restricted Data (FRD) under the Atomic Energy Act. Secrecy remained tight, at least in terms of declassifying information, even after the 1991 President's Nuclear Initiative when the Bush administration withdrew thousands of weapons deployed abroad leaving a few hundred nuclear bombs at air bases in NATO Europe.

While secrecy has been the U.S. government’s overriding priority, it has ultimately proven unattainable because the identities of the countries and overseas territories where the U.S. has deployed nuclear weapons have been in the public record for years. For example, ten years ago, in 1999 and early 2000, Robert S. Norris, William Arkin, and I published articles in The Bulletin of the Atomic Scientists identifying the U.S. overseas nuclear deployments during the Cold War. What inspired the articles was the partial declassification of a 1978 Defense Department study, History of the Custody and Deployment of Nuclear Weapons July 1945 – September 1977. The study included a heavily excised annex listing the locations of the major overseas deployments with a separate breakdown for NATO countries. Significantly, the reviewers of the History declassified West Germany and the United Kingdom, suggesting some flexibility about which host countries could be officially acknowledged. Determining where the other deployments were was tricky, but through previous government official and unofficial disclosures, the authors filled in the blanks , identifying NATO countries, including Italy, Turkey, Greece, Belgium, and Netherlands, and other Cold War allies such as the Philippines, South Korea, Japan, and Taiwan.

The disclosures in The Bulletin suggested that the policy of secrecy had failed, that it has become more of a formality or a political artifice than a meaningful system of control. The articles attracted much attention but they did not inspire a relaxation of policy. Indeed, the publication coincided with an archival clampdown: the first phases of the scrubbing of inadvertently released Restricted Data (RD) and FRD mandated by the Kyl-Lott amendment. While the FRD category includes significant information that deserves protection, many of the FRD documents were put back in the archival vaults only because they included information about the historical deployments. There the matter has remained, with the Defense Department and the Department of Energy insisting that information about former locations of nuclear weapons is properly classified as Formerly Restricted Data under the Atomic Energy Act

The pointless nature of the secrecy becomes evident when former NATO Secretary General George Robertson recently co-wrote an article that mentioned U.S. nuclear basing in Europe. He could not cite his own knowledge of the facts, but had to use as his source material from the nonofficial Federation of American Scientists. Another example of the sometimes preposterous degree of secrecy can be seen when government agencies excise the word "Turkey" (and sometimes Italy) from references to the Jupiter missile deployments that were at the heart of the Cuban missile crisis settlement (and which has been in the public record for years). This points to a fundamental problem, that much of the historical FRD information about overseas deployments, for example, Greenland, Taiwan, or South Korea, has long been overtaken by events.

A more credible approach would be to apply a policy of transparency to the historical deployments. Correcting this flaw in the classification/declassification system would follow directly from the Obama administration’s general approach to nuclear weapons policy. According to the recent Nuclear Posture Review, transparency is a condition for "strategic stability" and improving “confidence” in the U.S.’s relationship with major nuclear powers China and Russia. Furthermore, the recent Defense Department “Fact Sheet” on the U.S. nuclear weapons stockpile asserts that “Increasing the transparency of global nuclear stockpiles is important to nonproliferation efforts and to pursuing follow-on reductions” to the recent New START Treaty. To make good on initial steps toward a new nuclear posture, greater openness and accountability are essential. With the U.S. government as the only nuclear power that has systematically deployed nuclear weapons in far-off locations, it is appropriate that Washington take the initiative on this matter. By taking the lead in transparency, Washington may encourage other countries, such as Russia, with its large tactical nuclear force, to reciprocate.

Declassifying information on historical overseas deployments of U.S. nuclear weapons is a complex problem, which involves legal and diplomatic issues. The Atomic Energy Act makes it difficult to declassify FRD, but solutions to this problem are available. One way may be to amend the Atomic Energy Act to give the Energy Department and the Defense Department discretion to downgrade FRD such as that concerning the deployments to “defense information.” Alternatives might be the approaches suggested by PIDB itself several years ago in its report, Improving Declassification, which included specific recommendations on “Clarifying the Status and Treatment of Formerly Restricted Data.” One proposal was an amendment to the Executive Order which would treat FRD as “defense information,” which can be routinely declassified. An alternative proposal was to transition FRD to “the normal declassification system” after twenty five years. According to PIDB, either of those approaches “would provide the public with the same rights of access that it has to other information classified pursuant to the order.”

Yet a “right” does not automatically lead to declassification. Even if the status of all or some FRD is changed through legislation or an executive order historical data on the deployments could remain secret. Therefore, the legislation or the executive order would need affirmative language calling upon government agencies to expedite the declassification of historical information (25 years old or older) on overseas nuclear deployments. For example, with the historical numbers of the U.S. nuclear weapons stockpile already in the declassified public record, it should be possible to declassify numbers of weapons specifically allocated to overseas commands and forces (e.g., European Command, Pacific Command, 6th Fleet, Atlantic Fleet, U.S. Army Europe, etc.) without harm to national security. All of these numbers have been overtaken by historical events and have no bearing on current military planning and defense requirements. The proposed amendment could stipulate that declassification would extend to the following topics:

1) Countries and territories where nuclear weapons were deployed
2) U.S. commands and numbered forces that had responsibility for the deployments
3) Numbers of weapons allocated to the commands and numbered forces
4) Numbers and types of delivery systems assigned to the commands and numbered forces.
5) Nuclear cooperation agreements with foreign governments covering the deployments

The United States government, once empowered by legislation or executive order, could declassify the historic deployments, even if sensitive details on storage security arrangements (e.g., permissive action links, specific stockpile sites) remain classified.

Yet, legal authority might not be enough to move forward on declassification because the deployments were often bound up with formal or tacit diplomatic understandings developed between Washington and East Asian and Western European allies during the Cold War. Some governments have never acknowledged that the United States has deployed nuclear weapons on their soil and may remain reluctant to do so.

Given the benefits and importance of transparency, it might not be a formidable problem for Washington to persuade allies to accept a positive approach to declassification. Indeed, a number of key allies are likely to be supportive enough to make significant progress possible. For example, for many years the government of Japan refused to acknowledge the former presence of U.S. nuclear weapons on Japanese soil (e.g., Okinawa, Bonin Islands), but this is no longer true. With last year’s change in government and the declassification of Japanese records on the nuclear relationship, Tokyo is now likely to welcome the declassification of U.S. government information on historical deployments on its territory.

Some NATO allies such as Canada have gone far in declassifying information on nuclear agreements with the United States, including deployments of air defense and anti-submarine weapons, so their cooperation on a more transparent approach could be expected. While the debate over the future role of nuclear weapons in NATO is still unfolding, it is possible that some NATO governments would object to declassification. That others, such as Germany and the United Kingdom, are likely to cooperate might create momentum for others to follow suit.

To ensure that U.S. government agencies take a positive approach to declassification and work with allies in that direction, an executive order should include affirmative language, such as this: In keeping with the greater degree of transparency required by U.S. nuclear arms control, it is the policy of the U.S. government to declassify twenty-five years or older information, previously treated as Formerly Restricted Data, relating to the overseas deployments of U.S. nuclear weapons. The United States will work closely with alliance partners to ensure that declassification is mutually beneficial and that sensitive information on weapons systems and security arrangements is protected.

There is no question that building a consensus in government to take declassification policy in this direction will be difficult; it goes against ingrained habits of thought and conduct built up over the decades. Nevertheless, changing policy on FRD would make government work better by permitting the declassification of historically dated information. Whether greater transparency is possible on the historical nuclear deployments will depend on innovative leadership in government, an area where the Board is in a position to play a constructive role.