FAS | Government Secrecy | E-Prints ||| Index | Search | Join FAS

Paper I

Appropriate Criteria for Classification of Historical Records:
Balancing the Public's Right to Know with the Protection of Sensitive Information

With the end of the Cold War there has been widespread recognition both inside and outside the federal government that the national security information system needs major revision. One of the thorniest aspects of the reform debate has been resolving differences over the appropriate criteria for continued classification. From the perspective of the historical community, the key issues are: recognizing that the public's right to know should be balanced with the need to protect sensitive information, accepting the age of a document as one of the most important indicators of declining sensitivity, and developing precise and narrowly defined criteria for those few sensitive records that must remain classified beyond the targeted date for declassification. The following paper provides an overview of the evolution of current policies, the problems with the current system and the calls for reform with special attention to the issue of establishing the appropriate criteria for continued classification of historical records.


On April 26, 1993 President Bill Clinton issued Presidential Review Directive 29 noting the end of the Cold War and calling for a re-evaluation of our security classification and safeguarding systems, as articulated in Ronald Reagan's executive order, to ensure that they are in line with the reality of the current, rather than the past, threat potential.1 He ordered those agencies most involved with the system not only to come up with steps that could be taken to "increase individual accountability for the operation of the classification system" but, more importantly, to find ways "to declassify information as quickly as possible."2 The following month, in May, 1993, the Secretary of Defense and the Director of Central Intelligence established a Joint Security Commission with a mandate to examine all aspects of existing security practices and procedures within the intelligence and defense communities to identify improvements to the security system. In charging the commission members with their task, Deputy Secretary of Defense William Perry and Director of Central Intelligence James Woolsey noted that the current national security system is fragmented, complex and costly.3

Public interest groups and scholars have also urged reform of the national security information system. The American Historical Association and the Organization of American Historians have in recent years passed resolutions urging that declassification policies for historical records be reshaped "in line with the post-Cold War environment and the cost-containment that would follow."4

In a 1989 article in Government Information Quarterly, Frederick M. Kaiser of the Congressional Research Service concluded that the "enormous and escalating amount" of classified information "raises a number of serious concerns for a democratic society and its governance."5 Kaiser argues that the proper functioning of democratic government rests upon the informed consent of the citizens. This principle, he maintained, should be built into all executive orders on national security information. However, he noted that the current system, which has been in place since President Reagan issued E. O. 12356 in 1982, fails when measured by this standard. Federal government reports themselves have regularly documented yearly increases in the number of classifying actions, while the number of declassifications has remained the same or decreased.6

Indeed the mountain of classified information grows at an astounding pace. Over seven million documents are classified every year, creating economic ripples throughout the government. Although the full scope and cost of the classification system is itself classified, a recent study by the Office of Management and Budget estimated that over 32,400 people work full-time to maintain a system that costs an estimated $16 billion per year.7 To protect its secrets the government pays for everything from locks, vaults, and personal background checks to expensive page-by-page review of documents that are often several decades old. Given its vast size, the Department of Defense spends more than any other agency on information security, but even such agencies as the Department of the Interior and the General Services Administration each spend over one million dollars per year for such purposes. The National Archives currently holds more than 325 million pages of old--some from the 1920s--classified documents and spends $9.5 million annually on its classification/declassification responsibilities. With a large percentage of National Archives' space and resources being spent managing classified documents, there is a legitimate fear among historians that the National Archives is in danger of being a storage vault of largely unavailable information. The amount of classified material held by the Archives has increased 2.5 times since 1984.8

Identifying the problem has proved much easier than finding solutions. On March 1, 1994, the Department of Defense and the Central Intelligence Agency Joint Security Commission issued a 157-page report calling for streamlined procedures for classifying and declassifying information.9 The report had very specific recommendations for handling the classification of future documents so that they can be declassified more expeditiously. Yet the commission had little to say about the declassification of the enormous amounts of still-secret material over thirty years old. While the National Archives holds a significant amount of classified information, the large majority of secret material, which some contend is as much as a billion pages, is held in-house by individual agencies. "Detailed review of these documents is not feasible," the report stated, "and arbitrary bulk or automatic declassification schemes are perceived as risking the loss of information that still requires protection. Thus the report concluded both that line-by-line declassification is too expensive and time consuming and that agencies are generally not willing to declassify information without line-by-line review.10 In effect the report determined the issue too difficult to solve and moved on to other issues.

The task force set up by the White House to revise the executive order proved even less productive. After sixteen months, the task force has produced four internal drafts, but, as of December 1, 1994, there is still no official draft released for comment by the public or congress. Despite the critical need for action, the revision of the executive order has stalled on the thorny problem of redefining in the post-Cold War era the criteria appropriate for determining what national security concerns require the continued classification of historical records.

More is at stake here than just the enormous economic costs of the present information security system; perhaps even more critical are the threats the system poses to an informed citizenry. The debate over balancing legitimate security review and the needs of a democratic society began in earnest in the years immediately following the use of the atomic bomb. In 1950 Life magazine devoted several issues to exploring various aspects of atomic weaponry and highlighted the issue of "necessary security and unnecessary secrecy." The article quoted various leaders of the day including Joseph and Stewart Alsop and J.R Oppenheimer to stress the point that "there was a growing disparity between required security and officially imposed secrecy" and that such a policy prevented the kind of informed debate essential in a democracy. Describing the existing policy as one of "specious security," the article notes that the government withheld from citizens "facts that are useless to an enemy or known to him." The article concluded that some information about atomic weapons must be kept secret, but "public information can and should be substantially increased."11

As a consequence of the current classification system, American foreign and military policy during the Cold War remains poorly understood, not only by the general public but also among scholars and policymakers. Knowledge of the forty years of the Cold War is essential for dealing with the complex relations that are evolving; yet, large gaps remain. The public inaccessibility of historic records from the State Department, the Department of Defense, and other agencies relating to foreign policy has been largely responsible for this situation. Until the last few decades historians were generally able to gain access to historical records over twenty-five years old, but the current system has made it difficult for historians to obtain the primary documents on which to base an evaluation of the policies of the 1950s, 60s, and 70s. For example, 1,092 cubic feet of records of the Office of the Secretary of Defense and 805 cubic feet of records of the Joint Chiefs of Staff remain classified after thirty years. Lack of access to these documents prevents historians from constructing a balanced interpretation of the past and from pointing out the importance of previous experience for understanding contemporary problems.

The current security classifications policies also separate policy-makers from the lessons of the present. As Representative Lee H. Hamilton, the Democratic Congressman from Indiana who currently chairs the House Foreign Affairs Committee, has written: "When important decisions are made in secret, or when information relevant to policy decisions remains unnecessarily secret, this [public] scrutiny is not possible, and policy failures are more likely."12

Although current national security rests on the assumption that much information must remain secret because its release would be embarrassing, there is strong evidence that keeping documents secret can also be an embarrassment. In an October 30, 1994 New York Times article, Tim Weiner explores in depth an episode in which efforts to keep secret CIA activity in British Guyana in the 1960s proved an embarrassment to the Clinton administration. Although the efforts to destabilize the government of Dr. Cheddi Jagan succeeded in the 1960s, Jagan returned to power as president in 1992 in the country's first democratic election in thirty years. In June, the Clinton administration considered as nominee for Ambassador to Guyana a person who had been involved in the 1960s destabilization efforts. According to Weiner, the administration was "apparently unaware that the prospective nominee had helped to undermine the restored leader." In an interview with Weiner about the nomination, President Jagan said he was flabbergasted and conveyed his unhappiness to the Clinton administration. Furthermore, Jagan noted that "Everybody in Guyana knows what happened, I don't understand why they should be left secret."13 The insistence by the State Department and the CIA that these documents on Guyana remain classified resulted from a dispute between agency declassifiers and the State Department's Advisory Committee on Historical Diplomatic Documentation. The committee recommended that the Foreign Relations of the United States (FRUS) volume dealing with Guyana in the Kennedy Administration not be published because the omission resulted in a distorted account.

The balance to be struck between the basic requirements of national security and the imperatives of democratic government and official accountability is a difficult one, and historians recognize that there are legitimate national security needs that must receive serious consideration in decisions regarding public access to classified federal records. Some types of information obviously need continued protection, such as the identity of confidential, living human intelligence sources whose lives would be endangered if records pertaining to military plans and crypotologic systems currently in use were declassified. Foreign policy records can be especially sensitive, since they often involve negotiations that occur behind closed doors with expressed understandings of confidentiality. Policy-making officials clearly need a guaranteed period of time during which they can entertain full, free, and uninhibited debate with the assurance of confidentiality. Under our system of government, military officers respond to but do not make national defense policy, yet they are responsible for making contingency plans for potential conflicts across the globe which could be harmful or compromising to foreign policy efforts if disclosed. These legitimate concerns must be weighed against the need of the public to investigate the character of the nation's foreign relations. Twenty-five years of classification, and in most cases a much shorter period, should provide the necessary measure of confidentiality to present negotiators.

The small amount of information which many agree should remain classified for longer periods must be designated in clear and narrow terms and well protected. The current classification system inverts this relationship. Government resources are expended to protect a mass of information that is not truly sensitive. Almost a decade ago, the Stilwell Commission, established by the Department of Defense, concluded in its 1986 report Keeping the Nation's Secrets that "Too much information appears to be classified and much at higher levels than is warranted."14 A similar point was made by the Permanent Select Committee on Negligence of the House of Representatives in a February 4, 1987 report, which called for the reduction of classified information, noting that overclassification "damages the credibility of appropriately classified information."15 Several striking examples of such problems have been revealed, including the CIA classification as secret the measures it took to comply with the Americans with Disabilities Act or the continued classification by the Department of Defense of American troop movements during World War I.16 The current system fosters such situations by mandating strict sanctions for the release of information that could cause damage to national security. The presumption of current policy dictates that "when in doubt, classify." This endangers the security of genuine secrets and also contributes to a "cavalier" attitude towards classified material among some government employees.17 The latter have their own vested interest in the system. Steven Aftergood, a senior research analyst with the Federation of American Scientists, has noted after his conversations with government insiders, "no one is going to streamline himself out of a job."18

As the Aldrich Ames espionage case illustrates, a system that tries to protect everything is often unable to protect that which is most sensitive. The end of the Cold War offers an appropriate opportunity for recasting the standards of the classification/declassification system so that there is a more rational relationship between national security concerns and the need for openness, accountability, and informed policy-making. A new approach to security classification policy offers the possibility of devising a system that identifies a relatively small amount of genuinely sensitive information and protects it well. At the same time, such a system would release useful information that does not involve security risks, For example, new issues, such as environmental protection and international economic competitiveness, have come to the fore, and the information created and collected over the years by federal agencies has the potential to be a tremendous asset in addressing these problems.19 Gregory A. Gwash, Deputy Director of the National Industrial Security Program, in an interview in June, 1993 with Steven Aftergood of the Federation of American Scientists, noted that ". . . we waste a lot of time trying to safeguard information after the horse is out of the barn, so to speak. Basic scientific research is discovered, developed, publicized, broadcast to the world, and then we try to apply classification to its military application. It's a nightmare. It creates the problem of classifying everything that pertains to the application."20 The critics' view of the negative implications of this approach to information policy is summarized by Ralph Vartabedian, who wrote in a 1993 article for the Los Angeles Times: "If the military services continue to classify industry’s most promising technology, they risk posing a formidable obstacle to President Clinton's proposed $20-billion effort to help the defense industry convert to commercial enterprises . . . ."21 The flip side of this dilemma is that if high-tech, military technologies are opened up for commercial utilization, they are also made available to arms manufacturers and sellers to the third world. Yet many argue that a greater balance needs to be achieved between the need to protect national security interests and to make information more accessible. Currently researchers outside the federal government have very limited access to considerable amounts of government information, access to which many inside and outside government believe would not endanger the U.S. national security and would increase U.S. economic competitiveness abroad. Yet striking this balance has not proved easy.

Evolution of Classification/Declassification Policies

To place into context this consideration of the appropriate criteria for withholding information it is important to review the evolution of current classification/declassification policy. Since World War II, both the Executive Branch and the Congress have contributed to the establishment of national security classification policy. Under the Constitutional power of the Commander in Chief of the armed forces, the President has claimed authority for issuing executive orders on classification and declassification that establish policies for handling national security related material. In 1951 President Harry Truman's E.O. 10290 responded to Cold War concerns and expanded the protection of traditional military and diplomatic matters to include the concept of national security.22 Two years later, in 1953, President Dwight D. Eisenhower issued E.O. 10501, which eliminated one level of classification, limited who could classify information, and introduced new declassification polices. But during this period Congress also passed security classification legislation. The Atomic Energy Act of 1946 included provisions stipulating that all information relating to nuclear weapons be classified automatically. Likewise, the National Security Act passed in 1947 and the 1949 Central Intelligence Agency-Act provided for the classification of intelligence sources and methods.

The first comprehensive executive order on classification/declassification was President Richard M. Nixon's Executive Order 11652 issued in 1972. Although in 1971 President Nixon had directed a review of security classification procedures, the publication of the classified Pentagon Papers in 1972 in the New York Times and the Washington Post served as the catalyst for renewed discussion of declassification policy. An interagency committee established by Nixon studied overclassification problems and the need for putting in place a process for reviewing and releasing previously classified information. This study formed the basis of Executive Order 11652.23 A key component of the Nixon order was the specific provision that documents were to be automatically declassified at the end of the "thirtieth full calendar year" following their creation. Only agency heads were given the discretion to keep information classified for longer periods. All thirty-year-old or older information should be automatically declassified "except for such specifically identified information or material which the head of the originating Department personally determines in writing at that time to require continued protection because such continued protection is essential to the national security or disclosure would place a person in immediate jeopardy."24

From the perspective of this paper, the section listing the specific criteria for exempting classified information from scheduled declassification and establishing procedures to accomplish this are most pertinent. The Nixon order specified that "the use of the exemption authority shall be kept at the absolute minimum consistent with national security requirements."25 If only a few people had the authority to exempt material from automatic declassification, the rationale was that most material over thirty years old would be declassified. The Nixon order established only four categories of information that could be classified beyond thirty years. The first pertained to information furnished by foreign governments or international organizations and held by the United States on the understanding that it be kept secret, and the second exempted information protected by statute that deals with cryptology and intelligence sources and methods. The third category of information was that related to specific projects or foreign relations matters, while the fourth included information that might place individuals in jeopardy if disclosed. Through a mechanism called mandatory declassification review, a citizen could request the review of such material after ten years. However, such requested information could be kept classified if the review indicated it continued to meet any of the criteria listed above.

Although the Nixon order narrowed the criteria and established the policy of systematic declassification, it was never fully implemented and therefore failed to adequately deal with the growing mountain of declassified material. In a 1979 report, the General Accounting Office (GAO) noted basic problems with the implementation of the order and concluded that the order did not provide for in-depth reviews of the classification process and failed to provide needed authority and sufficient staff to ensure agency cooperation.26

Jimmy Carter made "open government" one of the themes of his Presidential campaign. Soon after becoming President, he directed his staff to study the security classification policy. Carter had three basic concerns--that over-classification worked against the principle that the people have a right to know what their government is doing, that classifying information for too long hampered historical research, and that routine over-classification made it difficult to protect items that needed to be kept secret. In 1982 former White House official Richard M. Neustadt responded to a request for information from the House Subcommittee on Government Information and Individual Rights and described the Carter administration's process of redrafting the classification security executive order. Neustadt wrote: "Everyone who participated in the process of writing E.O. 12065 agreed that vast amounts of routine paperwork containing no real national security information had been classified at the 'Confidential' level." A problem that he called "grade escalation" had reached the point "that in the White House it was widely assumed that any document classified below the 'Top Secret' level was not worth reading."27

President Carter’s Executive Order 12065, signed in June of 1978, changed not only the criteria but also the time period for classification. It called for the declassification review of twenty-year-old records. Of special significance to historians, the new order devoted a section to the establishment of a "public interest" test that was very much in keeping with his campaign promises. The provision required the inclusion of the public's need to know in decisions to keep material classified. It stated that even if material scheduled for declassification still met the stated criteria for continued classification "in some cases . . . the need to protect such information may be outweighed by the public interest in disclosure of the information, and in these cases the information should be declassified."28

While the Carter order did refer to "automatic" declassification of older records, in actuality it required that all records, no matter how old, should go through a systematic review process based on agency guidelines.29 The Carter order stated that "information may not be considered for classification unless it concerns" one of seven specified categories. These were: military plans, weapons, or operations; foreign government information; intelligence activities, sources, or methods; foreign relations or foreign activities of the United States; scientific, technological, or economic matters relating to the national security; United States government programs for safeguarding nuclear materials or facilities; or other categories of information which are related to national security and which require protection against unauthorized disclosure. In addressing the declassification side of the equation the order stated: "Even though information is determined to concern one or more of the criteria in Section 1-301, it may not be classified unless an original classification authority also determines that its unauthorized disclosure reasonably could be expected to cause at least identifiable damage to the national security."

Each agency was to prepare "guidelines for systematic review covering twenty-year-old classified information under their jurisdiction."30 These guidelines sought to identify types of information that could be declassified "automatically," that is in bulk, as opposed to types of records requiring page-by-page review. The only instruction for the design of these guidelines was that they "state specific, limited categories of information which, because of their national security sensitivity, should not be declassified automatically but should be reviewed item-by-item to determine whether continued protection beyond twenty years is needed."31 All information twenty years old or older that did not fall into one of the "sensitive" categories of the guidelines was to be declassified. On the subject of foreign government information, the Carter order stated that unauthorized disclosure of foreign government information or of the identity of a confidential foreign source was presumed to cause at least identifiable damage to the national security.32

While the "public interest" section of the Carter order indicated a general commitment to openness, other aspects of the order limited the declassification of material during the twenty-year review process. The order gave the Secretary of Defense and the Director of Central Intelligence the authority to establish "special procedures" for the review and declassification of cryptologic information and "information concerning the identities of clandestine human agents."33 This effectively exempted these categories of information from the standard systematic review process slated to begin after twenty years. Similarly, the broad category of "foreign government information" received special protection in a section that expressly extended the date for systematic declassification review of foreign government information until thirty years from its date of origin.34

Additionally, the Carter order retained the somewhat vague standard of "national security," but nonetheless went some distance to decrease the ambiguity inherent in that formulation by stipulating that "damage" to national security must be "identifiable" in order to retain classification beyond the regularly allotted time.35 This criterion sought to prevent the invocation of a nebulous national security interest as a way to elude declassification. Blanche Wiesen Cook, a former Vice President for Research of the American Historical Association, noted at a 1993 conference that this important provision of the Carter order advanced a proper balance between protection of genuine national security interests and the public's right to know about the actions of its government.36

Shortly after Carter issued his order, the General Accounting Office released a report which basically concluded that the new order may assist in some needed improvements but that the basic deficiencies in the national security classification system would not be corrected because of the lack of a strong enforcement mechanism.37 While the Carter order was in effect for only about three years, the GAO assessment proved accurate.

The 1979 acceleration of the Cold War with the Soviet invasion of Afghanistan and the seizure of American hostages in Iran provided the context for President Ronald Reagan's efforts, beginning in the early 1980s, to put in place policies that would provide stricter controls over national security information. In April, 1982, President Reagan issued Executive Order 12356, which, in a number of ways, made major departures from the Carter policies. Most importantly, the order contained no targeted time period, such as twenty or thirty years for the declassification of historical records. Instead, it stated only that "information shall be declassified or downgraded as soon as national security considerations permit."38

As scholars have noted, the Reagan order establishes as policy the principle that information can remain classified for an indefinite period, based upon the "intrinsically subjective character" of determinations about "national security" and "damage."39 Furthermore, the Reagan order criteria for continued classification was broader than any previous executive order. It increased the range of information that may be classified and created greater hurdles for declassification. Executive Order 12356 stated that information shall be considered for classification if it concerns: military plans, weapons, or operations; the vulnerabilities or capabilities of systems, installations, projects, or plans relating to the national security; foreign government information; intelligence activities, or intelligence sources or methods; foreign relations or foreign activities of the United States; scientific, technological, or economic matters relating to the national security; United States government programs for safeguarding nuclear materials or facilities; cryptology; a confidential source; or other categories of information that are related to the national security and that require protection against unauthorized disclosure as determined by the President or by agency heads or other officials who have been delegated original classification authority by the President.40 The order specifically stated that the "Unauthorized disclosure of foreign government information, the identity of a confidential foreign source, or intelligence sources or methods is presumed to cause damage to the national security."41

Not only did the Reagan order add to the Carter order three categories of information -- specifically the vulnerabilities or capabilities of systems, installations, projects or plans relating to the national security; cryptology; and a confidential source -- but it eliminated the word "identifiable" when treating the possible damaging effect of disclosure and widened the context within which information could be considered relevant by adding the phrase "by itself or in the context of other information."42

There was considerable concern expressed at the time the Reagan order was issued about the implications of the phrase "by itself in the context of other information." Dr. Anna Nelson, representing the American Historical Association and the Organization of American Historians, stated before a House Committee on Government Operations that "the very purpose of historical research is to gather together information from a variety of sources which will have meaning in the context of other information. Perhaps those who wrote that line did so over concern that an intelligence source might emerge from a 'mosaic' of information. If so, the section is a clear case of overkill." Dr. Nelson concluded that such ambiguous phrases in Executive Order 12356 were "open invitations to close any information so desired by the declassifier."43

The language in the Reagan order frequently made a presumption of classification, while the Carter order worked from a presumption of non-classification. Doubt over whether to classify at all was to be resolved in favor of non-classification in the Carter order. This presumption was reversed under the Reagan policy. As Steven Aftergood has observed, "the Reagan order makes it plain that, when in doubt, information is to be classified."44

The results of Executive Order 12356 were predictable. Historians and public interest groups uniformly recognize that Reagan-era information policy has resulted in huge increases in the amount of government information being classified and in a growing backlog of information awaiting declassification review.45 A maze of guidelines, rules, and regulations have evolved to implement the order, further faxing the resources of the Executive Branch and preventing researchers from gaining access to historical records.

Frustration with the Reagan executive order and particularly with the large amounts of historical material that remained classified motivated Congress in October 1991 to pass its first law mandating systematic declassification of State Department records. Although Congress had passed in 1966 and substantially strengthened in 1977 the landmark Freedom of Information Act, that law specified that requests for access to national security information would be subject to the classification standards established in the current executive order. Thus an executive order with a presumption of "when in doubt keep classified" seriously undercut the effectiveness of the Freedom of Information Act. Public Law 102-138, the "Foreign Relations Authorization Act" for fiscal years 1992 and 1993, signed by President Bush on October 28, 1991 included a small section calling for the declassification of almost all the older, historically valuable records of the State Department and established criteria for continued classification that were not as broad as the Reagan order. For historians, the enactment of PL 102-138 was a major step forward in a decades-long effort to improve access to the historical records of the State Department and to ensure the accurate and timely publication of the State Department's Foreign Relations of the United States (FRUS) documentary series.

The legislation was prompted in large part by the outcry over distortions in the 1989 FRUS volume on relations with Iran from 1952 to 1954. The volume omitted any mention of widely-recognized covert C.I.A. activities that had helped overthrow Mohammed Mossadeq, Iran's prime minister.46 The Advisory Committee on Historical Diplomatic Documentation, the advisory group of outside scholars charged with overseeing the publication of FRUS, had for over a decade been complaining about State Department declassification policies. In protest over the publication of the Iran volume, Dr. Warrant Cohen, professor of history at Michigan State University, resigned his chairmanship of the Committee. In addition, several national associations of professional historians sent resolutions to the Secretary of State asking that the integrity of the FRUS series be restored.47

The controversy over publication of the Iran volume underscored the inadequacy of declassification policy. Committee Chairman Cohen pinpointed the problem when he wrote that "Delays [with the FRUS series] have been caused by an overly elaborate, costly declassification process that encourages distortion and coverup."48 Cohen emphasized the harmful effect of the malfunctioning declassification process in his resignation letter to Secretary of State James Baker.

Public Law 102-138 did not depend on the executive order as the Freedom of Information Act does, but instead stipulated that, with the exception of those records that remain sensitive for national security reasons, State Department records "shall be declassified not later than 30 years after the record was prepared." In addition to provisions for continued classification of a record wholly prepared by a foreign government, the law listed four criteria that would be grounds for exemption from automatic declassification. These were records that would compromise weapons technology important to the national defense of the United States or reveal sensitive information relating to the design of United States or foreign military equipment or relating to United States cryptologic systems or activities; would disclose the names or identities of living persons who provided confidential information to the United States and would pose a substantial risk of harm to such persons; would demonstrably impede current diplomatic negotiations or other ongoing official activities of the United States government or would demonstrably impair the national security of the United States; or would disclose matters that are related solely to the internal personnel rules and practices of the Department of State or are contained ID personnel, medical, or similar files the disclosure of which would constitute a clearly unwarranted invasion of personal privacy.49

The declassification standards put forth in Public Law 102-138 represented a significant widening of access to historical government documents over that provided by the Reagan order on classification/declassification. Indeed, the State Department has shown increasing awareness that the post-Cold War environment requires a new policy on declassification. However, the implementation of PL 102-138 has not been easy or fully satisfactory. After warning in 1993 that the Department ran the risk of being in violation of the law, the State Department's Advisory Committee on Historical Diplomatic Documentation in a July, 1994 report to the Secretary of State, again expressed concern "about the progress of systematic declassification review and transfer to the National Archives and opening of 30-year old State Department documents."50

Against this backdrop of past executive orders and legislation, it is appropriate to consider the current efforts of the Clinton administration to revise Executive Order 12356. Following Clinton's April, 1993 directive, the task force undertaking the revision held public hearings in June, 1993 and completed a first draft by the end of August, 1993. Comments on drafts one and two provided both by the agencies and concerned citizens, including historians, outside of the federal government led to the development of a March 17, 1994 draft. This draft established a target period of twenty-five years after which records should be declassified and specified that exemption decisions for continued classification beyond twenty-five years are to be made by the agency head. Documents could remain classified if they met any of the following criteria: identify a human intelligence source or reveal information about an intelligence source or method; reveal information that would assist in the development or use of weapons of mass destruction; reveal information that would damage U.S. cryptologic systems; reveal information that would impair the application of state of the art technology within the U.S. weapon system, intelligence gathering system, or counterintelligence system; reveal current U.S. military war planes; reveal information harmful to U.S. foreign relations; violate a statute, treaty, or international agreement.51

The criteria in the March, 1994 Clinton draft were narrower than those of the Reagan executive order. Two of the broadest categories in the Reagan order "the vulnerabilities or capabilities of systems, installations, projects, or plans relating to the national security" and "other categories of information . . . as determined by the President"--no longer appear. Several other categories in the Reagan order were combined or narrowed. Clinton's March, 1994 draft narrowed the intelligence criteria by adding the qualifier "human" to "source" and the qualifier "application" to "sources and methods." The qualifier "exceptionally grave damage" was also added to the very broad category of "foreign relations" to raise the threshold for continued classification.52

The March, 1994 Clinton draft also narrowed the "cryptology" criteria by providing more specific language, stating "information that would impair U.S. cryptologic systems or activities." The Clinton draft replaced the Reagan order’s general "military plans, weapons, or operations" with a reference to only current war plans. Additionally, the March, 1994 drafters of the Clinton order substituted the Reagan order's all-encompassing clause on "scientific, technological, or economic matters" with a significantly reduced scope that refers only to "state of the art" weapons and intelligence systems.53

The research and public interest community has had some concerns about the formulation of the criteria but have basically supported the March 17 draft. In an April 22 letter to Steven Garfinkel, the chair of the task force with responsibility for drafting a new executive order, Professor Gary Nash, the President of the Organization of American Historians (OAH), indicated approval for the new automatic declassification provisions and provided one specific suggestion for improving the exemption criteria. The OAH suggested that the provision regarding information that would cause exceptionally grave damage to the relations between the United States and a foreign government be amended to state: "reveal information that would demonstrably cause exceptionally grave damage."54 This change would guard against exempting information simply because it might reveal past actions that were or might continue to be embarrassing to the United States, but would not in themselves cause significant, demonstrable harm to current foreign relations.

Sheryl Walter, General Counsel of the National Security Archive, offered another specific suggestion for refining the March 17 draft. She urged the task force to include a clause specifying that records above the classification level of "Confidential" would be the only "old" records subject to exemption from automatic declassification. She noted that such a provision would facilitate the release of much information that is no longer sensitive, not only because of its age but also because of the likelihood that large amounts of information originally classified at the "Confidential" level were never sensitive enough to be hidden from public view in the first place.55

The March, 1994 Clinton draft order on national security information pointed toward significant reform of the classification/declassification system. It promised to commit the federal government to greater openness by establishing benchmark dates for the release of all classified information, by reinstituting an automatic declassification date of twenty-five years, and by narrowing the criteria for exemptions. The draft implicitly adopted the age of a document as the primary criteria for its declassification. It also employed a "balancing test" in declassification decisions similar to that used in President Carter’s Executive Order 12065. The Clinton plan would eliminate the current emphasis on the risk of disclosure and replace it with a policy that balances the need for protection of sensitive information with a commitment to greater accountability and openness in government.

In testimony in June, 1993 before the Clinton task force charged with responsibility for redrafting E.O. 12356, historians recommended a return to the "balancing test," the establishment of a benchmark for the release of information with all but the most sensitive material released within twenty years of its creation, and specific language for narrowing the exemptions. Many historians believe that the Reagan order’s broad language that "information shall be declassified as soon as national security considerations permit"56 has served as an enormous loophole to keep huge amounts of records closed. Many historians have argued that for records over twenty-five years old there should be no consideration of potential damage to relations between the U.S. and another country. To tighten the language for other exemptions, historians have recommended language such as "information relating to the design, construction, or operating characteristics of components and systems currently used in electronic warfare, the release of which would pose a hazard to U.S. forces." Likewise, instead of the broad categories on intelligence information, historians have recommended inserting a provision stating that "information relating to the sources and methods of gathering intelligence in which a clear case can be established that the release of such information would place a living intelligence source at probable risk of suffering physical harm."57 In the area of foreign government information, historians have recommended a return to the standard of the 1972 Nixon order, under which only foreign government documents were restricted from disclosure. Under this standard, as a courtesy to another government, any document that government wholly created and provided to the U.S. government should continue to be withheld until that government agrees to its disclosure.58

The reforms to the current national security classification executive order suggested by historians and others are geared toward replacing a presumption of secrecy with a presumption of openness. Fragile democratic ideals have endured for so long in this country largely because Americans have been able to participate in informed debate and have been able to hold their leaders accountable to the public record. Furthermore, historians cannot glimpse the lessons of the past without access to federal records, and this seriously affects the wisdom of current policies. With the end of the Cold War, it is time for the Executive Branch, the Congress, and citizens to reexamine the appropriate criteria for withholding sensitive national security information.

At the heart of the reform efforts is the need to resolve difference over the appropriate criteria for continued classification. The debate centers on arriving at a target date after which most records can be opened and establishing the criteria for continued classification beyond the targeted date. From the point of view of the historical community the age of a document is one of the most important indicators of declining sensitivity. Yet two troubling aspects of declassification policy that make bulk declassification of older material at a targeted date difficult are the handling of "foreign government information" and problems created by an agency’s authority over its records that are in the physical custody of another agency. These issues as well as the need for increased interagency coordination are addressed in depth in the following three papers.

Paper One - Endnotes

  1. President, Presidential Review Directive 29, "National Security Information," April 26, 1993, p. 1.
  2. Ibid., p. 1.
  3. Washington Post, May 27, 1993.
  4. Council of the American Historical Association, "Resolution on Declassification Policy," December 27, 1992; Organization of American Historians, "Resolution on Revision of Executive Order 12356," April 16, 1994.
  5. Frederick M. Kaiser, The Amount of classified Information: Causes, Consequences, and Correctives of a Growing Concern," Government Information Quarterly 6 (No.3), p.247; Washington Post, May 27,1993.
  6. Kaiser, Government Information Quarterly, pp.250-251.
  7. R. Jeffrey Smith, "32,400 Workers Stockpiling U.S. Secrets," Washington Post, March 15,1994; Steven Aftergood, "The Perils of Government Secrecy," Issues In Science And Technology 8 (Summer 1992), p. 81.
  8. Smith, Washington Post.
  9. Joint Security Commission, "Redefining Security," February 28, 1994.
  10. Ibid., pp.32-33.
  11. "The Atomic Bomb," Life, February 27, 1950, p.91.
  12. Lee H. Hamilton, "The Costs Of Too Much Secrecy," Washington Post, April 13, 1992.
  13. Tim Weiner, "A Kennedy-CIA Plot Returns to Haunt Clinton," New York Times, October 30, 1994.
  14. Commission to Review Department of Defense Security Policies and Practices, Keeping the Nation's Secrets, February 11, 1986, p. 31.
  15. U.S. House of Representatives, Permanent Select Committee on Intelligence, "United States Counter-intelligence Security Concerns – 1986" (GOP: February 4, 1987), pp. 3, 13.
  16. Tim Weiner, "U.S. Keeps a Seal on Even Faded Secrets," Philadelphia Inquirer, December 19, 1991; Smith, Washington Post, March 15, 1994.
  17. Kaiser, p. 253.
  18. Aftergood, pp.84, 85.
  19. Anna K. Nelson and Richard H. Kohn, "The U.S. Must Declassify Its Cold-War Documents," Chronicle of Higher Education, September 16, 1992.
  20. Steven Aftergood, "Industrial Security Review," Secrecy and Government Bulletin, Issue No. 23, (June 1993), p. 3.
  21. Ralph Vartabedian, "Old Rules Keep Promising U.S. Technology A Secret," Los Angeles Times, July 14, 1993.
  22. Harold Relyea, "The of Government Information Security classification Policy: A Brief Overview," (1972-1992), Congressional Research Service, 1992. p. 3, 4.
  23. Ibid., p. 2.
  24. President, Executive Order 11652, "Classification and Declassification of National Security Information and Material," Section 5 (E). Hereafter cited as Executive Order 11652; See appendix for full text of pertinent sections of the executive orders on classification/declassification discussed in this report.
  25. Executive Order 11652, Section 5 (B).
  26. U.S. General Accounting Office, Improved Executive Branch Oversight Needed For the Government's National Security Information classification Program. LCD-78-125. March 9, 1979. Washington D.C., p. 8.
  27. U.S. Congress House. Committee on Government Operations. Executive Order on Security classification. Hearings, 97th Congress, 2nd session. March 10 and May 5, 1982. Washington D.C., U.S. Govt. Print. Office., 1982, pp. 332-3.
  28. President, Executive Order 12065, Section 3-303
  29. Ibid., Section 3-402.
  30. Ibid.
  31. Ibid
  32. Ibid., 1-302.
  33. Ibid., 3-403.
  34. Ibid., 3-404.
  35. Ibid., 1-301.
  36. Blanche Wiesen Cook, "Presidential Papers in Crisis: Some Thoughts on Lies, Secrets, and Silence," unpublished paper prepared in 1993, p.6.
  37. U.S. General Accounting Office. Improved Executive Branch Oversight Needed for the Government's National Security Information Classification Program, pp. 3,4.
  38. President, Executive Order 12356, Section 3.1. Hereafter cited as Executive Order 12356.
  39. Aftergood, "The Perils of Government Secrecy," p.84.
  40. Executive Order 12356, Section 1.3.
  41. Ibid.,1.3 (c).
  42. Ibid.
  43. Anne Nelson, Hearings Before A Subcommittee On Government Operations House Of Representatives, March 10 and May 5, 1982, pp. 109, 110.
  44. Aftergood, "The Perils of Government Secrecy," p.84.
  45. Ibid., p. 84, Kaiser, pp. 250-1; Relyea, p. 95.
  46. Karen J. Winkler, "Historians Criticize State Department for ‘Distortions’ and ‘Deletions’ in Its Record of U.S. Foreign Policy," The Chronicle of Higher Education, April 4, 1990, p. A-12.
  47. Ibid.
  48. New York Times, May 8, 1990.
  49. 22 U.S.C., Sec. 4354, Oct. 28, 1991, pp. 687-88.
  50. "Report of the [State Department] Advisory Committee on Historical Diplomatic Documentation for the Period January 1993-June 1994," pp. 4-5.
  51. March 17, 1994 draft of Executive Order on "Classified National Security Information," Section 3.4 (b).
  52. Ibid.
  53. Ibid.
  54. Letter from Professor Gary Nash to Steven Garfinkel, April 22,1994.
  55. Memorandum from Sheryl Walter to Mark Srere, Kate Martin and Gary Stern, April 16, 1994, re: 4/15 draft EO comments. In the files of the National Coordinating Committee for the Promotion of History.
  56. E.O. 12356, Sec. 3.1 (a).
  57. Memorandum to Steve Garfinkel and Members of the Task Force charged by the April 26 Presidential Review Directive with drafting a new national security directive, from Page Putnam Miller, Director of the National Coordinating Committee for the Promotion of History, June 14, 1993.
  58. Ibid.; The complexities surrounding the treatment of "foreign government information" are addressed in the second paper of this series.

Back to...

On to...

FAS | Government Secrecy | E-Prints ||| Index | Search | Join FAS