On May 22, 1957, E.E. “Pete” Gardner was piloting his tiny Cessna aircraft into Kirtland Air Force Base on the high, dusty deserts of Albuquerque, New Mexico. Gardner noticed that about a mile away a massive B-36 Air Force bomber also was preparing to land. The largest aircraft ever built by the United States, the B-36 had a wingspan two-thirds the length of a football field. As Gardner watched the lumbering bomber, he saw what he assumed was a paratrooper fall from the B-36, which was then at about 1,700 feet elevation.
What Gardner didn’t know and wouldn’t know for years, was that the B-36 accidentally had dropped a Mark-17 hydrogen bomb, with parachutes attached to slow its fall. Stretching 24 1/2 feet in length and five feet in diameter, the Mark-17 was an awe-inspiring weapon, weighing 42,000 pounds. The bomb’s explosive yield, officially still classified information, is known to have been at least 10 megatons, or about 625 times the explosive force of the bomb dropped on Hiroshima. Recalls Gardner:
“I told my passenger that if he could sit up, he would see a man parachuting from a bomber off our left wing tip. Almost immediately thereafter, the parachute impacted the ground (and) there was one hell of an explosion. I distinctly recall my passenger saying, ‘Gee, that’s pretty. I wonder if they are going to do it again?’”
The non-nuclear explosive in the weapon detonated, blasting a crater 25 feet wide, hurling metal parts a mile distant. Unsure what he had witnessed, Gardner mentioned the incident to a reporter, who contacted the air base. A bomb indeed had fallen, said an Air Force spokesman. It was ‘just an ordinary bomb’ producing no damage when it hit, the Air Force official said.
The bombing of Albuquerque was a secret for decades, as the federal government classified documents describing the accident. It simply was too embarrassing to admit the United States had built the world’s most powerful weapon and then accidentally dropped it outside one of its own cities.
In 1981, the Defense Department finally admitted some of what happened. Five years later, the Albuquerque Journal used the Freedom of Information Act to obtain additional government documents and details. What was learned is that the accident occurred because of human error, when a crew member steadying himself during air turbulence grabbed a release mechanism, dropping the bomb on Albuquerque’s outskirts.
At the same time the government was classifying information about the embarrassing New Mexico incident, a commission appointed by President Dwight Eisenhower issued an 807-page report on government secrecy and classification abuse. The Wright Commission, in its June 23, 1957 report, recommended abolishing the “confidential” classification. This was the lowest level marking and one which would end up on some documents detailing the Mark-17 accident. There was no danger in releasing most information stamped “confidential,” the commission said.
There is no indication that the Wright Commission report had any impact on New Mexico’s federal officials, or that they even read it. What is clear is that there would have been no danger to admit to the accident, as the government did years later.
Classifying the documents didn’t make the United States a safer nation. It didn’t strengthen the armed forces or make American borders more secure. It did, however, prevent Congress from asking annoying questions about the training of bomber crews and the design of bombers. It kept New Mexicans from asking if the Air Force had disaster plans, and whether it made sense to store and transport nuclear weapons near a population center.
Mountains of Secrets
In 1985, the last year for which somewhat reliable figures are available, the federal government classified an estimated 22,322,895 documents, a 14 percent increase over the previous year and a dramatic 82 percent increase since 1973. The volume of material classified just in 1985 equals 418 stacks of documents, each as tall as the Washington Monument. It isn’t surprising that a study by two congressional subcommittees concluded in 1985 that “the classification process is out of control.” The study found that:
“Observers from all points on the political spectrum are in basic agreement on several key points: too much information is classified…Only a fraction of the information classified is of a military nature and of value to the Soviets. Much material is classified to protect diplomatic relationships, hide bargaining positions, or prevent premature disclosure. All too often, documents are classified to protect politically embarrassing information or to hide government misconduct.”
Over-classification and the secrecy it engenders are harmful in a number of ways. There are finite resources to protect classified documents, which require special storage, treatment and protection. By producing an unmanageable mountain of documents, over-classification stretches these resources, making it more difficult to protect truly sensitive information.
Over-classification also hides error and misjudgment and restricts congressional access to vital information. It hampers the open exchange of data scientists insist is fundamental to survival in a technologically competitive world. It threatens the health and safety of Americans. Such as when residents of southern Utah were prevented from learning of health dangers to which they and their children were exposed each time open-air nuclear tests in Nevada showered them with fallout.
At the heart of the dispute over secrecy and classification is a fundamental concept: that secrecy robs a democracy of accountability. In recent years, America has witnessed crime, scandal and questionable policies concealed by secrecy, most notably in the Vietnam War, Watergate and the Iran-Contra affair. Presidents of both parties acted in secret, arbitrarily and imperially, avoiding the inconvenient checks and balances of Congress and the clumsy restraints of democracy. They ordered actions in the dark that they knew could not stand the light of day.
In the last three decades, nearly 40 studies, hearings or reports have examined over-classification and secrecy. All have agreed that too much information is classified. To understand its growth is to understand how this situation continues.
The Beginning
The modern classification system, today massive and cumbersome, was born in modest circumstances on March 22, 1940 when President Franklin Delano Roosevelt issued Executive Order 8381. Documents had been classified before the Roosevelt order, of course. The armed forces had on occasion used markings such as “confidential” and “secret” at least as early as the War of 1812. A formal classification system began to take shape during World War I. But the first direct presidential involvement in what virtually would become a presidential classification system didn’t occur until Roosevelt’s order.
It was a humble beginning. Roosevelt’s decree was drawn narrowly, citing as legal authority a 1938 statute on protection of “vital military and naval installations.” The order emphasized protecting information concerning weapons and military or naval plans. Equally important, Roosevelt gave authority to classify just to the Secretary of War, Secretary of Navy, and their staffs.
The Truman Order
These limitations disappeared on September 25,1951 when President Harry S. Truman issued a new executive order on classification. Truman’s order boldly took the federal government where no classifier had gone before. Citing no specific law to justify his decree, Truman instead relied on “the authority vested in me by the Constitution and statutes, and as President of the United States.” As scholars Richard C. Ehlke and Harold Relyea note:
“Politically such reliance upon implied constitutional powers strengthened the President’s discretion to make official secrecy policy: it intertwined his responsibility as Commander-in Chief with the obligation to ‘take care that the laws be faithfully executed.’”
In a breathtaking expansion, Truman also gave the authority to classify to “all Departments and agencies in the Executive Branch of the Government,” and enlarged the classification criteria from “national defense” to “national security.” As government classifiers soon demonstrated, the Truman terminology provided a standard of much elasticity and breadth. The same day Truman issued his order, the Office of Price Stabilization issued an inter-office memo ordering employees not to release information that might embarrass OPS.
The Truman plan was much criticized. Forty-four GOP senators signed a manifesto, pledging to “resist any attempt to conceal facts from the American people.” On November 9, 1953, President Dwight Eisenhower replaced the Truman order with one of his own. Unfortunately, the Eisenhower classification plan didn’t change matters much. Historian, Arthur M. Schlesinger, Jr., concluded that the Eisenhower order “provided no effective control over the stampers and no workable method for the declassification of documents…” For nearly 20 years, until it was superseded by President Richard Nixon’s order in 1972, the Eisenhower plan was the foundation of the nation’s classification system. It left a lasting imprint.
Abuses and Reforms
Over the years, classification has concealed some humorous items, such as the time the Army classified a report on the bow and arrow, which it called a “silent, flashless weapon.” A memo between the Joint Chiefs of Staff said too many documents were classified improperly. The memo was then classified. A recorded government weather report concluded with a warning that the information broadcast was classified. And an American military officer classified the menu of an American government-sponsored banquet for former Queen Fredricka of Greece.
Other concealed items haven’t been as amusing. Improper classifications hid problems at nuclear power plants and denied citizens information about food, air and water purity. Classification covered up intelligence agency misdeeds and permitted illegal activities in the executive branch.
Obviously, some classification is needed. Reforms for salvaging the system range from its complete restructuring, to strengthening of oversight agencies, to better education, accountability and record-keeping.
In 1973, the House Committee on Government Operations placed much of the blame for Watergate secrecy excesses on the presidential classification system, saying:
“The committee therefore strongly recommends that legislation providing for a statutory security classification system should be considered and enacted by Congress. It should apply to all executive departments and agencies…”
A statutory classification system raised separation-of-powers questions. How far can Congress go in controlling executive branch classifiers? Supreme Court decisions suggest the two branches of government share powers in this area, but Congress hesitated in launching such a frontal attack on the presidency.
Relyea, a Congressional Research Service authority on secrecy and classification, is among those who would like Congress to authorize the president to develop a classification system, yet reserve for itself the right to approve or disapprove it.
Congress also could demand a greater role in the drafting of executive orders, through hearings or by requiring draft orders to be opened to public comment. This proposal has support from Steven Garfinkel, director of the Information Security Oversight Office that monitors executive branch classified information programs. “We’re asking for after-the-fact criticism that can be avoided if Congress views itself as a player,” he said.
Some authorities have called for a review office or an ombudsman to resolve classification disputes. Review power also could be given to the General Accounting Office, a congressional watchdog with a history of dealing responsibly with classified documents, or to a restructured version of the Information Security Oversight Office that places the agency under legislative control.
Scrap The System
Other improvements were suggested in the 1985 “Study on the Protection of National Secrets” prepared by two House subcommittees. It suggested scrapping the categories or “confidential,” “secret” and “top secret” for a two-tier system of “high-level security data” and “confidential data,” the latter covering all other information the government feels should not be made public, but which is not militarily sensitive.
This two-tiered system distinguishes between significant secrets and less important “administratively controlled information.” This latter category, while classified, often is distributed widely within government. Proponents say the new system would make it more difficult to perpetually classify less important data, while emphasizing protection of vital secrets.
Two agencies now exercise some authority over classification. The Information Security Oversight Office was created by President Jimmy Carter and charged with administering the executive branch’s information security program. The director was promised independent authority and given some power to achieve this task, such as the ability to order documents declassified.
But the office today has a staff of 13, which is not enough to oversee a system that creates millions of classified documents each year. The unit also reports to the National Security Council, making the office less an independent overseer than a creature of the executive branch whose information it is supposed to supervise.
Some would like to see the unit converted into an independent oversight agency. The Senate Select Committee on Intelligence recommended in 1986 that the office be given greater authority to oversee special access programs and be given a larger staff, including a “permanent inspection element.” In a recent Congressional Research Service report to Congress, Frederick Kaiser suggests making the unit a creature of Congress rather than one owing its existence to a presidential executive order. The office “would then have to report regularly to the authorizing panels about its activities and findings, problems in monitoring agency compliance with existing requirements, and difficulties in convincing agencies to implement its proposals.”
The other, often-overlooked agency with a traditional role in declassification is the National Archives and Records Administration. In 1982, President Ronald Reagan killed a requirement in the Carter executive order that most documents be reviewed for declassification after six years, and again at 20 years. However, some limited systematic review continues under the guidance of the Archives, which reviews permanently valuable classified records for declassification after the items become 30 years old.
The threshold is 50 years old for cryptology and intelligence files.
There are limits to what the Archives can achieve, noted Edwin Thompson, director of the Archives declassification division. Transfer of documents to the Archives from some agencies is years behind schedule. Review also occurs under guidance of the original classifier, who can veto release of any material.
The declassification staff also was cut about 40 percent in 1982, Thompson said. Today, the staff numbers about 45. The backlog of documents needing examination is so massive that many will not be reviewed until more money is provided, Thompson said.
Sanctions
In 1962, the House special Government Information Subcommittee examined the nation’s classification system and concluded that a “security system which carries no penalties for using secrecy stamps to hide errors in judgment, waste, inefficiency, or worse, is a perversion of true security.”
Ten years later, the same subcommittee re-examined the system, finding 2,505 cases where executive branch employees were punished for improperly releasing information. It found no case of punishment for improper or over-classification.
The House subcommittee wasn’t the only group to urge sanctions for abusers of government secrecy. In 1977, a report for the Commission on Federal Paperwork suggested willful, improper secrecy be punished by a fine of up to $10,000 and imprisonment for up to five years.
Despite the appealing simplicity of sanctions, the most important step toward accountability in classification is to make declassification easier. The temptation to conceal decreases as the possibility of declassification increases.
The first declassification schedule was put into place by President John F. Kennedy. President Carter speeded the process by requiring a justification for retaining classification markings past six and twenty years. President Reagan scrapped that timetable, saying in his executive order that documents could remain classified “as long as required by national security considerations.” In effect, this ended most automatic review.
The Reagan rejection of the Carter plan did more than just affect when documents will be released. It emphasized that what one president has authority to grant, another can take away. It underscored the recommendation in a 1959 congressional report that a declassification system with independent authority was needed. The President cannot cancel what Congress creates.
Was six years too short a period for review? Opinions vary. Historian Schlesinger argues for 10 years, or two and a half presidential administrations. Secretary of State Dean Acheson urged a 25-year review. Undersecretary of State George Ball said five years of secrecy was enough for most documents.
The 1985 Stillwell Commission also recommended the Defense Department require classifiers to challenge classifications believed to be improper. Advocates of openness want classifiers to be evaluated regularly, with their reviewing skills tied to their promotions and their authority to classify.
The Whistleblower Protection Act, recently vetoed by former President Ronald Reagan, also would help classifiers trying to do their job fairly.
Other accountability measures could include Congressional earmarking of agency monies for declassification and requiring minimum standards for an agency’s internal oversight of classification, a step recommended by the Information Security Oversight Office.
Complicating matters is the federal government’s lack of a central log of classification actions. Investigators studying classification abuse rely on sampling techniques. Past researchers have concluded that abuses are widespread. “Reached serious proportions,” said the 1956 Coolidge Committee. “Out of control,” found the two 1985 House subcommittees. Politics also clouds the picture. In 1985, the Information Security Oversight Office changed its sampling techniques. Although previous samplings revealed the number of classified documents grew with each passing year, the new sampling suggested the Carter administration classified more documents in 1980 than did the Reagan administration in 1987–a conclusion many greeted skeptically. The change has prompted some to suggest the General Accounting Office should examine the oversight office’s sampling methods. Others have suggested that Congress require a yearly classification assessment, based on widely accepted sampling techniques, or require classifiers to keep a log of their actions.
Special Access Means Special Secrets
Reformers also want a list and an evaluation of the laws and special access programs now in use by the executive branch to withhold information from Congress. Special access programs are highly classified programs to which access is extremely restricted. Created by the president or executive branch agencies, their very existence often is secret. They carry a bewildering array of code names–names that are often classified. Congressional investigators admit they don’t know how many special access programs exist and estimates range as high as 10,000. As such, congressional oversight of the secret programs is exceedingly difficult. As the Iran-Contra hearings made clear, illegalities can be hidden by such programs.
Review of the programs could yield a list of the laws now cited as authority to withhold information, a list that has never been compiled. In 1958, Rep. John Moss (D-CA.), the “father” of the Freedom of Information Act, said his subcommittee found 75 laws used by the executive branch to withhold information. In 1984, the American Society of Access Professionals found 135 statutory provisions used by federal agencies to deny information requested under the FOIA, according to Lois Alexander, the Society’s president.
The Derivative Blues
Record keeping reforms also could focus on those who create the records. Garfinkel, of the Information Security Oversight Office, argues that reducing the number of “original” classifiers (those with authority to initially classify documents) limits improper security. If restrictions on original classifiers were extended to “derivative” classifiers, a larger but less regulated group of government employees and private contractors, openness would be better served, Garfinkel said. Derivative classifiers have no authority to create a classified document, but can restate information in such a way–such as inserting a classified document in an unclassified report–that the new package requires classification. In 1987, the majority of documents that were classified–83 percent–were stamped by derivative classifiers. Garfinkel suggests that these classifiers, like the “originals,” be identified by name or position and be held accountable for their actions.
Under guidance of the National Security Council, the Information Security Oversight Office attempts to educate classifiers in the 65 executive branch agencies it oversees. But the focus is on the need for secrecy, rather than the importance of openness. In one recent 47-minute training film, called “Information Security Briefing,” it warned that, “leaks may be just as damaging to our national security as outright espionage, and leakers should expect to be treated accordingly.” The film largely is silent on the dangers of excessive secrecy. Such questions as whether national security would have been helped or harmed if the Iran-Contra affair had been exposed earlier are left unasked. Some have suggested that congressional review or production of educational materials used to instruct classifiers could result in a more balanced presentation of the dangers of excessive secrecy.
Another target group for openness education in the information process is the judiciary. In 1974, Congress gave federal judges the authority to review classified documents that are requested under the Freedom of Information Act. Judges were charged with determining if documents were properly classified according to the relevant executive order. If not, the judge could release them. The executive branch could not merely assert that a document was classified properly; it had to supply proof. However, the judicial oversight of stamp-happy bureaucrats that Congress had in mind is not working.
A recent study concludes that, “courts seldom second-guess the classification decisions of agency officials.” The solution, suggests FOIA expert Allan Adler, is education through tough questioning of judicial nominees when they come before the Senate Judiciary Committee for confirmation. Adler urges senators to closely question judicial nominees about whether they would use this review power that has been given to them by Congress.
Educating Consumers
Another target, as always, is the public at large, as the ultimate consumers and inheritors of what the secrecy system produces.
One method that reformers suggest is a yearly congressional status report that would examine secrecy, classification and the relationship of politics to the classification system. Such a report could have analyzed Daniel Ellsberg’s unauthorized release of the Pentagon Papers in 1971 and his resultant indictment for espionage.
It might also have discussed how former President Lyndon Johnson, who in 1971 was a private citizen with no special right to classified documents, also disclosed segments of the Pentagon Papers. Johnson’s unauthorized publication came in his autobiography “The Vantage Point.” Johnson was not indicted, but paid for his disclosures.
Students of the classification system believe it can be reformed. Until it is, however, the secrecy it creates strips Americans of a fundamental right: the opportunity to lend meaningful consent to their government’s actions, or to exercise their constitutional right to dissent.
©1989 David Morrissey
David Morrissey, on leave from the Albuquerque Journal, is investigating government secrecy during recent presidential administrations.
