FAS | Government Secrecy |||| e-Prints | Index | Search |

Declassification: Where Are We Today?

by Jim David

February 1997

When President Clinton assumed office, his administration faced the decades-old problem of a growing mountain of classified Executive Branch records and the ineffectiveness of existing laws (including the Freedom of Information Act) to reverse this trend. The need to drastically reduce the literally billions of pages of these records, some of which date back to the 1930s, is clear. Whether it is the fate of our POWs and MIAs from the Korean and Vietnam Wars, Swiss complicity in protecting Nazi Germany's wealth, dangers faced by those living near our nuclear weapons facilities or any number of other important issues, the public has a right to know the considerable relevant information contained in these secret records that no longer merits protection. After an inter-agency task force examined the problem for more than one year, President Clinton signed Executive Order 12958 (E.O. 12958) in April 1995. The new order, designed in part to speed up the declassification of these records, went into effect in October 1995.

As described in detail below, the progress of most agencies in implementing the new order is slow and uneven at best, and in many cases it appears that large numbers of records, especially the most important high-level ones, will still not be reviewed for declassification by the established target dates and thus will remain totally inaccessible to the public. Moreover, there is a related and rapidly growing problem in that large numbers of declassified records are still completely inaccessible to the public and might remain so for many years. The administration and Congress must now seriously consider further changes in the law to ensure that the greatest number of important records are released and that the public has timely access to them. Along these lines, several recommended changes are set forth below.

E.O. 12958 mandates in part that all permanent pre-1975 classified records be automatically declassified by 2000, regardless of whether they have been reviewed or not, unless the agency has applied to the President for an exemption. (Federal law requires that all records be appraised as permanent or temporary. Permanent records are never to be destroyed, while temporary records can be destroyed after varying lengths of time.) Exemptions must be based on a determination that the records contain one or more of eight categories of information (e.g., information that would reveal the identity of a confidential human source or application of intelligence sources or methods; assist in the development or use of nuclear, chemical or biological weapons; or reveal current military war plans). All agencies must establish a systematic declassification review program to review records, including those exempt from automatic declassification, and agencies are required to declassify at least 15% of their non-exempt records annually. Significantly, the new order excludes information concerning nuclear weapons ("Restricted Data" under the Atomic Energy Act of 1954) from its operation, and thus many hundreds of millions of pages of classified records (primarily in Department of Energy and Department of Defense holdings) do not even come within its purview.

How many permanent pre-1975 classified records are there subject to E.O. 12958? As required by the new order, all agencies have submitted proposed implementation plans to the Information Security Oversight Office (ISOO) in the National Archives, and most of these contain estimates on the number of subject records. Although some of the proposed implementation plans are themselves classified and a few agencies inexplicably will not give permission to the public to review even their unclassified plans, the majority are available for examination and they disclose a staggering number. (As a yardstick, there are about 13.2 million pages in one mile of records.) Some holdings are relatively small, such as those of the U.S. Information Agency (13 million pages), State Department (35 million pages), National Aeronautics and Space Administration (12 million pages), and Arms Control and Disarmament Agency (3.2 million pages). Others are huge. The Department of Defense (Office of the Secretary of Defense, Air Force, Army, Marine Corps, Navy, Joint Chiefs of Staff, National Security Agency, Defense Intelligence Agency, etc.) estimates it may have as many as 1.5 billion pages, and the Central Intelligence Agency (CIA) 166 million pages. The Federal Bureau of Investigation (FBI) apparently has not estimated how many records it has subject to the new order, but the figure must total at least several hundred million pages because a 1981 joint National Archives/FBI report stated the FBI held about 750 million pages of both unclassified and classified records. (When considering that government reports disclose that all Executive Branch agencies together reviewed for declassification less than 500 million pages from FY 1980 through FY 1995, one wonders whether the drafters of E.O. 12958 ever realistically thought these billions of pages could be reviewed by 2000.)

As difficult as it may be to comprehend, some of these estimates are low due to two factors. First, a number of agencies have only incomplete information on how many records they hold and the types thereof because of poor records management practices through the years. Second, some agencies have huge numbers of unappraised records and no estimate prepared by any agency includes any pre-1975 classified records in this category that should be appraised as permanent.

What has been the progress of the agencies in complying with E.O. 12958? To determine the progress two questions must be answered: How many pages have been reviewed and exactly what records have been reviewed? With respect to the former, all agencies have submitted their figures on the numbers of pages reviewed in FY 1996 (the period roughly corresponding to the first year that the new order was in effect) to ISOO for inclusion in its FY 1996 report to the President. However, this office will not release these figures to the public until the report is submitted - and this very well may be a long wait since the FY 1995 report was not submitted until September 1996, almost one year after the end of the fiscal year. Moreover, in most cases the individual agencies are refusing to release their figures pending the submission of the report. The task of learning exactly what records have been reviewed is even more difficult. Unfortunately, there is no one office in the Executive Branch that has this information. The public must first find out who the personnel are in each agency in charge of the declassification review effort and contact them, and this alone can take many hours of effort. After contacting these personnel, the public in most instances will get at least some minimum information on the records the agency has reviewed. However, it might take the agency several weeks to provide this information and the information could very well be in a format which is difficult to understand or use.

Notwithstanding these major obstacles, information derived from various sources gives a partial picture of the progress (or lack thereof) thus far. All the agencies cite a shortage of resources as a key impediment, and there is no question this is a real problem. Yet, some have managed under these constraints better than others. As always, every agency is extremely reluctant to declassify with little or no review more than a handful of records. Consequently, agencies in most cases are still conducting varying levels of detailed and time-consuming examination to determine whether the records contain information exempt under the new order, not subject to the new order at all (e.g., "Restricted Data") or which cannot be declassified by the agency holding the records and thus needs to be referred to another agency for a declassification decision.

The individual components of the Department of Defense (DOD), for example, have each proposed varying amounts of records for exemption from automatic declassification. Although reporting that all of its components together reviewed 68 million pages in FY 1996, the department will not release the figures for each individual component. Within DOD, the Air Force appears to have made the most progress. It is reviewing very important collections of high-level records from the 1940s into the early 1970s, including the files of the Secretary of the Air Force and the Chief of Staff. Even with this outstanding effort, though, it is difficult to see the Air Force reviewing all of its estimated 177 million pages of exempt and non-exempt records by 2000 with the present resources. Other DOD components, some of which have even larger numbers of records, have smaller programs or in some instances are just starting their programs. The CIA has proposed that 106 million pages (roughly 2/3 of its subject holdings) be exempt. It will not release the number or exact nature of the records reviewed to date, but based on past experience it is likely that the number totals at most only a few hundred thousand pages. The FBI has proposed almost all of its unknown number of records subject to the new order for exemption. It established a special team to review the roughly 236,000 pages of records on violent domestic protest groups in the late 1960s and early 1970s. However, the review is still not complete. One of the few agencies that may actually review all of its subject holdings by 2000 is the State Department, due chiefly to the fact that a 1991 statute required it to begin reviewing all of its records 30 years old and older. For example, over 10 million pages were reviewed by the department in FY 1996 alone.

The most important element in this entire process is to declassify to the maximum extent possible the records of greatest interest to the greatest number of people. These records are, in short, those of the high-level offices and the advisory groups thereto. To give but one example, the records of the Secretary of the Army, Army Chief of Staff, and their key assistants are far more critical than the much more voluminous records of the Army's Ordnance Corps, Quartermaster Corps, Transportation Corps, and their successors. This must be the principle that guides the administration and Congress in this area, even if it results in considerable delay in reviewing the much larger numbers of lower-level records.

How can the objectives of E.O. 12958 be achieved? One obvious solution would be to provide more resources, but this is unlikely in this period of tight budgets. Another possible solution would be to modify in several important respects the present mandate that all permanent pre-1975 classified records be reviewed by 2000. First, there would be the new requirement that the records of the agencies' high-level offices and the advisory groups thereto be reviewed first. These offices and groups should be specifically denominated so there is no confusion or ambiguity about what offices and groups are included in this category. If, and only if, an agency has demonstrated good faith progress and adequate cause would an exemption be granted from automatic declassification for the remaining lower-level records that the agency has been unable to review. Second, there would be new public reporting requirements, requiring at a minimum that agencies report several times a year exactly what records have been reviewed, the dates thereof, the quantity thereof, the approximate numbers declassified in whole or part, and the repository in which they are located. Moreover, agencies would be required to report publicly on the numbers of unappraised pre-1975 classified records and their plans for appraising them. There is no reason why potentially hundreds of millions of pages of records are not subject to the new order simply because the agencies have failed to appraise them as required under federal law.

An emerging problem relating to the entire declassification effort is the considerable difficulties faced by the public in getting access to many declassified records. This problem has several aspects to it. First, when a particular collection is reviewed only rarely are all the records therein declassified. Therefore, before the public can examine the collection officials must perform the time-consuming task of going through it and removing all the material that is still classified (i.e., "processing" the collection). "Processing" is only performed at the National Archives, the governmental repository to which all permanent records are required to be sent. Second, although many such collections have been transferred to the National Archives because of resource limitations they are not being "processed" and may not be for some time. One example of this is a collection of almost 8 million pages of pre-1964 Office of the Secretary of Defense (OSD) records which was reviewed for declassification between 1981 and 1996 and transferred to the National Archives in early 1996. (OSD records, among other things, contain the files of the Secretary of Defense and all the Assistant Secretaries of Defense. They are the single most valuable group of military-related records that exist.) No portion of it has been "processed" yet and the National Archives cannot say when it will do so - and until this is done the public will continue to be unable to see even one page in the collection. Third, in some instances collections have been reviewed by an agency but they are not even being transferred to the National Archives. An example of this is another collection of over 6 million pages of pre-1964 OSD records that was also reviewed between 1981 and 1996, but for which there are no plans for transfer to the National Archives. Until this transfer is made and the records "processed", the public will be unable to see any of it. Again, increased resources would help but these are unlikely to be forthcoming. The National Archives, in conjunction with interested members of the public, must develop a priority list for both "processing" collections it already has and arranging the transfer to it of further important collections.

These are some proposed solutions to the current problems, and there have and will be others made. For example, the Commission on Protecting and Reducing Government Secrecy (chaired by Senator Moynihan) will issue its report in the near future. Undoubtedly, it will contain many good recommendations on these and related problems. All proposals must be seriously assessed and the best of these adopted so that the critical and long-delayed goal of increased public access to our government's records is met.

FAS | Government Secrecy |||| e-Prints | Index | Search |