Shrouding the Endless Frontier - Scientific Communication and National Security

of military secrecy." Dr. Berkner, and many other scientists, were very distressed about this state of affairs. "Since more and more of our scientific activity is coming within the purview of secrecy," he observed, "the need for appraisal of the effects of secrecy on our scientific stature and progress, and therefore on our national security, becomes of increasing importance."'' After identifying and discussing these effects," he then offered certain considerations for attaining "the best balance of technological secrecy as weighed against free information."'"

At present, the Federal departments and agencies clearly may exert restrictions on the communication or dissemination of scientific research and knowledge produced by, for, or on behalf of the government. Such controls, of course, must be exercised in accordance with prevailing law regarding, for example, security classification, contracting authority, and statutorily mandated or required confidentiality. Generally, the government cannot prevent scientists unaffiliated with it or working without its financial support from communicating their research findings or knowledge unless, for example, they have produced Restricted Data as defined in the Atomic Energy Act," sought a patent for an invention deemed to be subject to a secrecy order," or engaged in pursuits governed by export licensing requirements." Space limitations preclude a review here of current national security authority for controlling scientific communication or the significant policy concepts pertinent to its application. However, these important elements have been considered in the larger overview from which this article derives."

10

BEGINNING THE BALANCE SHEET

Various normal and essential scientific communication activities, including unclassified research dissemination, publication, and exchanges in the open classroom and among scholars, have been limited recently through more vigorous enforcement and more stringent application of existing national security controls. Dr. Berkner's ideas are recounted as points of balance both to justifications for these restrictions and to arguments favoring even broader government authority to constrain scientific communication for reasons of national security. They are discussed here, together with other pertinent considerations, in the contemporary context. This discussion does not presume to be systematic or exhaustive." These ideas are not necessarily organized in any particular priority. Finally, these are not absolute laws or postulates, but general premises and, as such, are mindful of exceptions and allowances.

The communication of basic scientific research findings and knowledge should not be subject to national security restriction, except perhaps in times of declared war. "Basic research," we are reminded, "'is performed without thought of practical ends,...results in general knowledge...," and "provides the means of answering a large number of important practical problems, though it may not give a complete specific answer to any one of them." Moreover, basic research provides scientific capital,...creates the fund from which the practical applications of knowledge must be drawn,” and “'is the pacemaker of technological progress.' Admittedly, the distinction between basic and applied research is difficult to make in some areas of science. Nevertheless, this judgment can be made in many cases, and should be pursued in other more difficult ones if there is to be adherence to the principle enunciated in some pertinent policy instruments that basic

GOVERNMENT INFORMATION QUARTERLY Vol. 1/No 1/1984

research findings and knowledge shall not, for the most part, be subject to national security controls.

In July 1970, a report of the Defense Science Board Task Force on Secrecy gave support to this consideration, saying "the types of scientific and technical information that most deserve [security] classification lie in those phases close to the design and production, having to do with detailed drawings and special techniques of manufacture,'' and added that most of the force of attention in classifying technical information should be directed to these phases rather than to research and exploratory development. '*'* The idea received policy expression in July 1978, in the Carter Administration order on security classification (E.O. 12065), and was continued by a provision of the succeeding Reagan Administration directive (E.O. 12356) which states: "Basic scientific research information not clearly related to the national security may not be classified.''''

1116

In advocating the automatic inclusion of information release terms in all Department of Defense research contracts, the recent report of the Defense Science Board Task Force on University Responsiveness to National Security Requirements offered the following important caveat: **The Department of Defense is assiduously rejecting any control guidelines that would restrain the development and dissemination of the fruits of basic research.""" In March 1982, Acting Deputy Under Secretary of Defense George Millburn repeated this sentence in his prepared statement before two subcommittees of the House Committee on Science and Technology holding a joint hearing on the impact of national security considerations on science and technology." In his testimony before these same two panels, Admiral Bobby Inman, who was then Deputy Director of the C.I.A., said he had never presumed that draconian (control) measures against the basic research side were either warranted or likely to occur. ***0 In general, there appear to be strong indications that very little restriction of the communication of basic scientific information is warranted or planned, whether that information is produced for, by, or on behalf of the Federal Government or independently generated.

21

Government applications of national security restrictions to the communication of scientific research findings and knowledge should be for narrowly defined policy purposes and in accordance with specific criteria. "Freedom of communication among scientists," it has been noted, is essential for scientific progress, and for the critical validation or invalidation of scientific findings."'"' In the United States, scientific communication enjoys protection under the First Amendment." However, neither the functional need nor the civil right of scientists to communicate freely is absolute." Certain particular qualifications and exceptions have been acknowledged in the past. But in this same vein, attempts to limit the dissemination of scientific research findings or knowledge should be for narrowly defined policy purposes and in accordance with specific criteria. Certainly there have been indications from within Congress which have been supportive of this point of view. For example, a 1973 House committee report on security classification policy criticized the Nixon Administration's new order (E.O. 11652) prescribing procedures for creating official secrets because of its use of such overly broad and undefined terms of policy purpose as "national security" and "foreign relations." A more recent House committee report reproved the classification criteria of E.O. 12356, the operative directive. During the past thirty years, succeeding presidential executive orders have narrowed the bases and discretion for assigning official secrecy to agency records. E.O. 12356 reverses this trend in a variety of ways." The House report took

24

Shrouding the Endless Frontier-Scientific Communication and National Security

issue with the elimination of the "identifiable" damage standard for applying security classification and viewed new classification categories as overbroad, of uncertain need, and not qualified or defined.**20

In 1980, a House committee report making recommendations regarding legislative reconsideration of the Invention Secrecy Act urged, among other suggestions, that Congress "Make the necessary findings and declaration of public policy that would justify the exercise of invention secrecy powers in peacetime." In addition, the report sought to "Change the basis for issuance of a secrecy order from the opinion of an agency head that disclosure might or would be 'detrimental to the national security,' to a more demonstrable standard of damage to the national defense."*27 All of these ideas are exemplary of the conditions contemplated by the major premise concerning narrowly defined policy purposes and specific criteria.

Primary policy and procedures concerning government application of national security restrictions to the communication of scientific research findings and knowledge should be established through the legislative process. In his study of national security and individual freedom, Dr. Lasswell makes the important comment that, because all security policies entail risk, "the public interest calls for the calculation of risk by a procedure that balances each policy against every policy and arrives at a judgment to which many minds have contributed." By using ‘`a procedure that takes conflicting views into account and subjects them to the discipline of debate and exposure to available knowledge," the public interest can be protected and public confidence can be gained or vindicated in the wisdom of a national security policy outcome."

30

28

The Constitution vests all legislative power in Congress and prescribes the manner in which statutory laws shall be created. Through this process, Congress has established the arrangements principally with the enactment of the Administrative Procedure Act of 1946, as amended—whereby department and agency regulations are issued and take effect. Only minimal publication requirements have been established for other executive directives, but Congress has been counseled to extend and expand this obligation.” Federal courts, on occasion, have struck down Executive Branch regulations and directives and invalidated actions taken pursuant to them; similarly, Congress, in exercising its legislative powers in an area specifically entrusted to it, has repealed agency regulations and presidential orders. However, the legislature may not divest the President of a constitutional function by legislation. As a result, there are areas of shared power. Exercising general executive authority (Article II, Section 1, Clause 1) and pursuing his responsibilities as Commander in Chief (Article II, Section 2, Clause 1), the President may prescribe policy and procedures for the protection of department and agency records for reasons of national security. Congress, relying upon its mandate to provide for the common defense (Article I, Section 8, Clause 1), to "make rules for the government and regulation of the land and naval forces" (Article I, Section 8, Clause 14), and to "make all laws which shall be necessary and proper for carrying into execution the foregoing powers" (Article I, Section 8, Clause 18) also seemingly may legislate on this subject.” Thus, during the past few years, varying suggestions have been made from within Congress that a statutorily-based security classification system be pursued.'

Recently, a House committee suggested that, even though a presidential executive order on security classification policy and procedure may be developed and issued without being subject to the public notice and comment requirements of the Administrative Pro

GOVERNMENT INFORMATION QUARTERLY Vol. 1/No. 1/1984

cedure Act, both the government and the public would be better served by adherence to the spirit of that provision. The panel recommended that future revisions of classification rules be announced publicly, that they be circulated publicly for sixty days to permit public comment, and that they "be provided to the Congress with sufficient time to permit interested congressional committees to consider the proposals, to hold hearings, and to prepare comments.

35

Since the conclusion of World War II, Congress has enacted various national securitybased restrictions bearing upon scientific communication, and committees in both Houses have developed expertise regarding such law. The congressional forum appears to offer the best opportunity for the realization of Dr. Lasswell's proposed national security policymaking procedure. Also, as a Senate special study committee suggested not long ago in another context, administrative discretion in the execution of statutes regulating scientific communication for reasons of national security should be reduced to a minimum and guided by an instructive legislative history. Termination dates, reporting requirements, and accountability procedures were strongly suggested for legislation bearing upon fundamental civil liberties. Provision also might be made to "give Congress some type of veto over Executive branch rules and regulations judged to be inconsistent with the legislative intent of the authorizing statute.'

1136

All of these ideas are illustrative of the intent underlying the major premise regarding the establishment, through the legislative process, of primary policy and procedures concerning government application of national security restrictions to the communication of scientific research findings and knowledge.

Government application of national security restrictions to the communication of scientific research findings and knowledge, as a matter of policy, should be subject to administrative review and, in the case of privately developed information, court challenge. In the event the communication of scientific research findings or knowledge is restricted for reasons of national security, clear procedures should exist, preferably in statutory law, to permit a reconsideration of the limiting action on a de novo basis by an appropriate government official or, in the case of privately generated information, ultimately by a court to determine if the material in question warrants continued protection in whole or in part.

At the present time, E.O. 12356 provides a mandatory review procedure whereby a contractor or grantee may request that a classified work product be reviewed for possible declassification." This arrangement, however, was designed for the general public to make declassification requests and makes no allowance for a contractor or grantee to know the details of the classification action or to enter into a discussion with government officials regarding it.

The recent report of the Defense Science Board Task Force on University Responsiveness to National Security Requirements does not appear to make any allowance for a hierarchial appeal of initial decisions by Department of Defense officials prohibiting the dissemination of scientific information produced under departmental contract. However, peer review mechanisms discussed elsewhere in the report may include some such appellate procedure when they are fully developed.”

The Invention Secrecy Act allows a private individual to appeal the imposition of a secrecy order to the Secretary of Commerce.” Although no such formal procedure exists for a private person found to be in possession of independently developed scientific

Shrouding the Endless Frontier-Scientific Communication and National Security

7

research findings or knowledge constituting Restricted Data as defined in the Atomic Energy Act, the individual in question, nevertheless, may petition the Department of Energy to remove the information from the protected category. An injunction against the communication or continued possession of independently generated scientific research findings or knowledge alleged to be Restricted Data may be challenged in court.“

Clear administrative and judicial procedures should be available whenever the government applies official secrecy or communication restrictions, for reasons of national security, to scientific information. Current awareness of the importance of such arrangements would seem to be evident from the criticism recently expressed by the Department of Justice concerning the failure of proposed International Traffic in Arms Regulations **to provide prompt judicial review of State Department decisions barring disclosure" of certain technical data."2

42

Government application of national security restrictions to the communication of scientific research and knowledge should not embrace the entirety of large undertakings, only core ideas requiring protection. This consideration is directed at two practical aspects of information protection. Large-scale projects involve many people and, in time, leaks will occur regarding aspects of the undertaking. The Manhattan Project is often cited as an example of a highly successful large-scale secret enterprise, yet there reportedly were "over 1,500 cases in which classified Project information was transmitted to unauthorized persons.

1943

In addition, in a large-scale secret endeavor of a scientific nature, perspectives can become blurred so that support systems and knowledge related to the project, otherwise open within the scientific professions and literature, can become unnecessarily classified or restricted for reasons of national security. Both this point and the previous one were addressed in a 1956 report by a special study committee to the Secretary of Defense. It recommended that the government ''(c)ease attempts to do the impossible and stop classifying information which cannot be held secret." By way of explanation, the report indicated it was referring to “information which cannot be withheld because it inevitably is known to too many people.........the physical appearance and general performance data of new weapons when they have become widely known," and also "compiled data composed of unclassified items and information which is already public, where official confirmation would not be of substantial value to a potential enemy, even though it will require additional machinery to keep track of what information has been publicly revealed."'"

The number of officially secret projects involving scientific research and knowledge should be sharply limited. In more elementary terms, there are important reasons for carefully controlling the amount and duration of national security restrictions on the communication of scientific information. One obvious consideration was offered long ago by the prestigious Commission on Government Security:

that unnecessary restrictions upon the dissemination of scientific and technological information may in the long run actually be detrimental to the national security. Positive contributions to the national security through scientific and technological advancement must not be lost as the result of an overzealous effort to classify."