Again, in the first category, it is generally the case that such regulations derive from statute. Congress makes a law, entrusting its enforcement to a bureaucracy which issues rules and rulings to carry out the law. This is a feature of the administrative state that appeared in the United States in the early 20th century, roughly between the administrations of Theodore Roosevelt and Woodrow Wilson. Thus, the Department of Commerce and Labor was established in 1903; the Federal Reserve Board in 1913; the Federal Trade Commission in 1914. An executive gazette-- the Official Bulletin--was inaugurated in 1917. (The Official Bulletin was published for only two years. It was the precursor to the Federal Register, in which all new regulations are published, which began in 1936.)
Secrecy became a persuasive mode of regulation with the advent of the national security state at mid-century, although its origins also go back to the beginning of the century. The statutory base of secrecy is modest; two or three laws, of which the National Security Act of 1947 is emblematic. Withal, its spare reference to the protection of "sources and methods" led to a vast secrecy system almost wholly hidden from view. There would be no Official Bulletin.
Three general propositions will emerge from this "Brief Account." The first is that from the time of the First World War, the beginning of the great power conflicts that would continue for the better part of the century, the United States recurrently faced espionage attacks by foreign governments, and on occasion, sabotage of notable proportion. A recurrent pattern of these crises is the involvement of ethnic groups, often first-generation immigrants who have retained strong attachments to their ancestral homes and, not infrequently, to political movements that were prominent at the time of immigration.
The ethnic dimension of international conflict has repeatedly created a fear of internal conspiracy in aid of external threat. This was succinctly stated by Theodore Roosevelt in October 1917:
Arguably, one consequence of the "Hun within" syndrome is that the United States developed a pattern of extensive defensive secrecy far greater than would have been required to deal with an essentially external threat. A kind of backward formation took place. Whereas, in the usual situation (if there is such) the existence of secrets required defensive measures, in the American experience of the 20th century, the secrets came about largely because there was a perceived threat. Loyalty would be the arbiter of security. Given that loyalty could not be assumed, a vast secretive security system emerged.
The second proposition is that the statutory basis for secrecy has been, and remains, so elusive that violations of secrecy occur with relative impunity. Edward A. Shils defined secrecy as "the compulsory withholding of knowledge, reinforced by the prospect of sanctions for disclosure."2 This was written in 1956, when the morale of the Cold War system was high, and discipline was readily maintained. In 1946, as will be discussed, the Army Security Agency (formerly the Army Signal Security Agency) decoded the first of several thousand VENONA 3 messages sent by the KGB [Komitet Gosudarstvennoi Bezopasnosti (Committee for State Security)]4 and other Soviet intelligence agents identifying spies working within the American Government. The consequences for American counterespionage were spectacular; the VENONA project continued until 1980. Early on, the Soviets learned of its existence through a spy in the Army Security Agency itself, but as for the American public, not a whisper was heard until the 1980s, and only with the establishment of the Commission on Protecting and Reducing Government Secrecy has this extraordinary archive been made public.
In time, however, the system degraded, largely in consequences of having grown to grotesque proportions. A specific example would be the celebrated "Pentagon Papers," essentially an official history of the war in Vietnam. Most of which were "Top Secret." The New York Times, and later the Washington Post, obtained copies and proceeded to publish selections. The United States Government moved to enjoin publication. The Supreme Court overruled the Executive Branch. Soon after, Harold Edgar and Benno C. Schmidt, Jr. published an article on the case in the Columbia Law Review. Just what was the law here? they asked. They replied, after 158 pages, that they could not possibly tell.5
It has now become routine for information of the highest classification to appear in the press, most commonly as a tactical move in some intra-government policy dispute. There are no sanctions. A fairly routine example of what might be called "deregulation" occurred on October 22, 1996, when the Washington Times published details of a "Top Secret" CIA analysis of the control system of Russian nuclear weapons. The following day, the Washington Post had a "follow-up" story by Reuters:
The Central Intelligence Agency has concluded that Russia’s control over its nuclear arsenal has been weakening, but the chance of unauthorized launch or blackmail remains low, CIA officials said yesterday.
"The Russian nuclear command and control system is being subjected to stress that it was not designed to withstand as a result of wrenching social change, economic hardship and malaise within the armed forces," according to a classified report prepared last month, the officials said.
The CIA report, "Prospects for Unsanctioned Use of Russian Nuclear Weapons" and stamped top secret, was disclosed by the Washington Times in its editions yesterday. CIA officials confirmed the accuracy of the material quoted in the article.
"It reinforces the urgent need for a missile defense to be put in place as soon as possible for the United States as well as for its allies and friends," said James Lilley . . . who served as U.S. envoy to China and South Korea under presidents Ronald Reagan and George Bush.
Shils’s definition to the contrary, however, there are now no sanctions for disclosure. Not, that is, for anyone at the Deputy Assistant Secretary level or above. In the manner of maturing bureaucracies, most agencies involved with security matters have developed a range of publications concerning their activities. The Department of Defense Security Institute publishes Recent Espionage Cases. The May 1996 issue recorded all cases since 1975. It is melancholy reading. Of 89 such cases, 55 involved persons who on their own decided, typically, to try to sell secrets to the Soviets. Only fifteen were "recruited" successfully and there were only nine real-life foreign agents. Hardly a "Hun within" in the batch. But there is one notable case, that of a civilian analyst with the Office of Naval Intelligence who supplied Jane’s Publications with classified photos showing a Soviet nuclear-powered carrier under construction. The photographs were subsequently published in Jane’s Defence Weekly (July 1984). The employee was sentenced to two years’ imprisonment. The Defense Security Institute comments that this was "the first individual convicted under the 1917 Espionage Code for unauthorized disclosure to the press."6
Along with the de facto immunity of senior officials who release classified information, there developed a form of Congressional oversight, beginning with the House Committee on Un-American Activities and the Senate Subcommittee on Internal Security, which could and did protect the intelligence community, as it came to be known, and let out a fair amount of information to the public. But in the process, the public also "learns" a good many things that are not so. As Evan Thomas, the author of a recent book on the early days of the CIA, notes in a recent issue of Studies in Intelligence, a publication of the Central Intelligence Agency: "Polls show that nearly 80 percent of Americans believe JFK died as a result of a conspiracy, and about half believe the CIA was somehow involved."7 Secrecy begets suspicion, which can metastasize into belief in conspiracies of the most awful sort.
Despite the growing frequency of high-level disclosure of classified materials, the public perception is not wrong; the vast proportion of classified material remains classified. This reflects the principled character of the men and women of the Armed Services and the assorted intelligence and related agencies. It also reflects the sheer dimension of the secrecy system. It would be a fair guess that if every page of every newspaper published in the United States on a given week-day were given over solely to reprinting the classified documents created that day, there would not be enough space. This, in turn, reflects the criterion of classification, which is to say, national security.
Harold C. Relyea, of the Congressional Research Service, notes that, "A perusal of the Federal statutes indicates that national security suddenly began to appear with some frequency as the undefined term in laws enacted around the time of U.S. involvement in World War I."8 National defense was not enough; that had been the concern of admirals and generals: dockyards and arsenals and order of battle. This was something more. The world was a far more dangerous place; ideological conflict was as serious as military conflict: indeed, more so, and far more elusive in its details. For the better part of a century the United States would hardly know a moment’s peace of mind. We would gradually see, in Donald L. Robinson’s term, "The Routinization of Crisis Government."9
The decisive moment in this regard was the enactment in 1947 of the National Security Act, which established the unified Department of Defense, the Central Intelligence Agency, and the National Security Council, the latter a standing committee in the White House designed to deal with emergencies of all sorts. In testifying in support of such legislation before the Senate Committee on Military Affairs, James F. Forrestal, then Secretary of the Navy, was explicit in choosing the term "national security" over "national defense." Unifying the Army and Navy was not nearly enough. Forrestal set out a list of "eight requirements against which to measure any plan for national security":
That means in particular that you don’t create military demands beyond your capacity to fill them or that will do injury to other great and urgent demands. And that question of balance, in my view, is one of the most important considerations in war.
(3) A more efficient organization for the translation of strategic requirements into
requirements for materiel and personnel.
(4) Provisions for the coordination of military and other war budgets.
(5) Adequate means for the elimination of waste and duplication in and between the military departments.
(6) An efficient coordinated intelligence organization serving all Government departments and agencies.
(7) An organizational means for fostering scientific research and development within the military departments and among civilian organizations.
(8) Full opportunity of each branch of the military services to develop for its specialized task.10
A succession of post-World War II presidents issued executive orders published in the Federal Register asserting this particular form of regulation, but without defining it. Truman in 1951:
But succinctness is not the same as clarity. Under these executive orders, "national security" is in the eyes of the "appropriate classifying authority." Of which there are at present roughly 5,300 persons within the Federal Government with the authority to classify "originally," but an estimated two million additional persons in the Government who then can classify "derivatively" by citing already-classified documents or by using "classification guides" prepared by their agencies, and another one million in private industry with such ability.17
A third and final proposition is that secrecy, unless carefully attended to, is a source of consider-able sorrow in government. That there can be a need for it, none should dispute. The Framers so provided in Article 1, Section 5 of the Constitution:
But, as Joseph Story wrote in Commentaries on the Constitution of the United States, the object of the clause requiring the keeping of a Journal is "to insure publicity to the proceedings of the legislature, and a correspondent responsibility of the members to their respective constituents."18
And so, at the very outset we encounter the unavoidable tension between the right of the public to know and the need for government, in certain circumstances, to withhold knowledge.19 Relyea has observed: "Ideally, all information held by government belongs to the citizenry."20 And yet, it can be very much in the interests of the same citizenry that some information not be generally available, and within the capacity of a mature democracy to make the distinction. Provided only that the system be kept under review.
However, secrecy can confer a form of power without responsibility, about which democratic societies must be vigilant. A disturbing instance occurred after the discovery, beginning with the Army Security Agency’s code-breaking in 1946, of a most considerable Soviet espionage apparat in the United States, including, by all the evidence, senior officials of the United States Government. The person who most needed to know this was the President of the United States. The issue was national security and he was Commander-in-Chief.
It would appear, however, that President Truman was not told. In their superb account of these events, VENONA: Soviet Espionage and the American Response, 1939-1957, published by the National Security Agency and the Central Intelligence Agency (in connection with a major October 1996 conference on VENONA), Robert Louis Benson and Michael Warner write:
Benson and Warner continue:
The tacit decision to keep the translated messages secret carried a political and social price for the country. Debates over the extent of Soviet espionage in the United States were polarized in the dearth of reliable information then in the public domain. Anti-Communists suspected that some spies--perhaps including a few who were known to the US Government--remained at large. Those who criticized the government’s loyalty campaign as an over-reaction, on the other hand, wondered if some defendants were being scapegoated; they seemed to sense that the public was not being told the whole truth about the investigations of such suspects as Julius Rosenberg and Judith Coplon. Given the dangerous international situation and what was known by the government at that time, however, continued secrecy was not illogical. With the Korean war raging and the prospect of war with the Soviet Union a real possibility, military and intelligence leaders almost certainly believed that any cryptologic edge that America gained over the Soviets was too valuable to concede--even if it was already known to Moscow.22
The decision to share or to withhold information could be--can be--highly personal and political, or purely professional. The Central Intelligence Agency was not informed about VENONA until 1952. The KGB cables indicated that the Office of Strategic Services (OSS) in World War II had been thoroughly infiltrated with Soviet agents. As the CIA was widely regarded as the successor to the OSS, the Army and the FBI were appropriately cautious in sharing their secrets. That is a problem not to be avoided. But when secret information is withheld for personal or political reasons, the democracy can be put at risk.
1 Albert Bushnell Hart, ed., Theodore Roosevelt Cyclopedia (New York: Roosevelt Memorial Association, 1941). "Roosevelt in the Kansas City Star," 1 October 1917, 8.
2 Edward A. Shils, The Torment of Secrecy, with an introduction by Daniel Patrick Moynihan (Glencoe: The Free Press, 1956; reprint, Chicago: Ivan R. Dee, Inc., 1996), 26.
3 The term "VENONA" is an arbitrary codeword which describes more than 2,900 Soviet diplomatic telegrams sent between 1940 and 1948 and the efforts by the United States Government to decode the messages and to identify Soviet agents mentioned therein. Robert Louis Benson and Michael Warner, eds., VENONA: Soviet Espionage and the American Response, 1939-1957 (Washington, D.C.: National Security Agency, Central Intelligence Agency, 1996), vii-viii.
4 Harvey Klehr, John Earl Haynes, and Fridrikh Igorevich Firsov, The Secret World of American Communism (New Haven: Yale University Press, 1995), xxvi.
5 Harold Edgar and Benno C. Schmidt, Jr., "The Espionage Statutes and Publication of Defense Information," Columbia Law Review 73, no. 5 (May 1973), 930.
6 U.S. Department of Defense Security Institute, Recent Espionage Cases: Summaries and Sources (Rich-mond: 1996), 12.
7 Evan Thomas, "A Singular Opportunity: Gaining Access to CIA’s Records," Studies in Intelligence 39, no. 5 (1996): 23.
8 Harold C. Relyea, "National Security and Information," Government Information Quarterly 4, no. 1 (1987): 16.
9 Donald L. Robinson, "The Routinization of Crisis Government," Yale Review 63 (Winter 1974): 161.
10 Relyea, "National Security and Information," 17, quoting Senate Committee on Naval Affairs, Unification of the War and Navy Departments and Postwar Organization for National Security, 79th Congress, 1st Session, Washington, D.C.: Government Printing Office, 1945), 578-79.
11 Relyea, "National Security and Information," 18.
12 Ibid., 17.
13 President, Executive Order 10290, "Regulations Establishing Minimum Standards for the Classification, Transmission, and Handling, by Departments and Agencies of the Executive Branch, of Official Information Which Requires Safeguarding in the Interest of the Security of the United States," Federal Register 16, no. 188 (27 September 1951): 9797.
14 President, Executive Order 10501, "Safeguarding Official Information in the Interests of the Defense of the United States," Federal Register 18, no. 220 (10 November 1953): 7049.
15 President, Executive Order 11652, "Classification and Declassification of National Security Information and Material," Federal Register 37, no. 48 (10 March 1972): 5209.
16 President, Executive Order 12958, "Classified National Security Information," Federal Register 60, no. 76 (20 April 1995): 19825.
17 These estimated figures were supplied by the Information Security Oversight Office, which issues an annual report on classification decisions. See Information Security Oversight Office, 1995 Report to the President (Washington, D.C.: Information Security Oversight Office, 1996). "Derivative" classifiers are responsible for 94 percent of all classification decisions.
18 Joseph Story, Commentaries on the Constitution of the United States (Boston: Little, Brown, 1891; reprint, William S. Hein, 1994), 609-10.
19 Secrecy was present at the creation. The Constitutional Convention of 1787 met in closed session. Before final adjournment, in answer to an inquiry by George Washington, the presiding officer, the Conven-tion resolved "that he retain the Journal and other papers subject to the order of Congress, if ever formed under the Constitution." Max Farrand, ed., The Records of the Federal Convention of 1787, vol. 1 (New Haven: Yale Univ. Press, 1934), xi.
20 Harold C. Relyea, The Evolution of Government Information Security Classification Policy: A Brief Overview (1775-1973) (Washington, D.C.: Congressional Research Service, 11 September 1973), 1.
21 Benson and Warner, VENONA, xxiv.
22 Ibid., xxix.
Proceed to Part 2 of Appendix A
Go Back to the Top Page of the Commission Report