It is now apparent that the American scientific community is approaching a critical point in its relations with the federal government. Until recently, the conduct of most scientific work in this country proceded on a well-founded assumption that it would remain free from official intrusion or state regulation. Since 1979, however, the federal government has frequently acted in the name of national security to impose restraints on important aspects of the scientific endeavor. Most notably, in an effort to curb the export of "militarily useful" technologies, the Administration has applied the existing set of export controls to domestic scientific symposiums, university research programs, and the presentation of scientific papers.

This effort to restrict the dissemination of applied scientific knowledge has sparked heated debate .. . . The debate has thus far addressed the government's effort to control the export of applied scientific knowledge as a broad question of public policy. It seems likely, however, that the major issues in the controversy will soon be tested under narrower, legal principles in a court of law. If so, the government will almost certainly rely on one of two congressional statutes as authority for its restraints on the transmission of technological knowledge.

One statute, the Arms Export Control Act², empowers the State Department to license the export of all military articles listed in the International Traffic in Arms Regulations³. As defined by those regulations, the relevant articles consist not only of warmaking devices such as aircraft and explosives but also for "any information" used in the production of military arms. Equally important, the regulations broadly construe the term "export" to include the noncommercial transmission of information in domestic settings such as scientific symposiums.

The other statute is the Export Administration Act of 1979⁴, which differs from the arms regulations in two respects. First, it authorizes the Commerce Department to license the export of "dual use" technologies that are subject to both military and civilian applications. Second, it deals principally with the export of technologies to "controlled countries" such as the Soviet Union, Poland, and East Germany. Like the arms regulations, however, the Export Administration Act restricts the domestic release of any information used in the production of commodities having a military value⁵. Furthermore-and again like the arms regulations-the Export Administration Act imposes stiff criminal penalties on those who willfully violate its licensing requirements.

In these statutes Congress has provided considerable authority for governmental restraints on the export of "militarily useful" technologies. This fact alone, however, will not end the legal inquiry in cases where the government has invoked the statutes to restrict the open, domestic communication of applied scientific knowledge. On the contrary, in such a case, a major issue will arise concerning the validity of the legislation under the free-speech clause of the First Amendment.

To resolve this type of issue, the Supreme Court has consistently relied on a well-defined analytical framework designed to determine whether the state's interest in regulation is sufficiently important to justify an abridgment of First Amendment freedoms. In the rest of this article, I will examine the ways in which the Court's mode of analysis can accommodate the difficult First Amendment issues arising from the imposition of restraints on the open, domestic communication of technological knowledge ....

Rejecting the notion that all speech is absolutely immune from official regulation, the Court has determined the degree of protection to be accorded to various categories of expression by looking to the major values that underlie the free-speech guarantee. These values, according to the Court, can be summarized in three propositions. First, the right of free speech advances the citizen's interest in self-fulfillment by enabling him to realize his full potential through the free expression of opinions, beliefs, and ideas. Second, the guarantee of free speech serves an important social function by promoting the widest possible circulation of socially useful information. Finally, the right of free speech is essential to a democratic form of government, for it ensures that all information bearing on various policy issues is fully disseminated to the public.^6,7

Though the Court has not yet adjudicated the issue, it seems clear that scientific communications contribute to each of these interests and thus warrant as much protection as political tracts, literary works, or any other variety of speech. Indeed, a system of free scientific expression not only enables scientists to draw on the work of colleagues but also tests the validity of hypotheses against current data and opposing views. In these ways, it promotes the discovery of scientific truth and fosters the intellectual advances that contribute to the collective wisdom.⁸

In the case of technical data, however, more difficult questions arise. For example, does technical information having only military uses warrant the same degree of constitutional protection as political speech or basic scientific knowledge? In all likelihood the Court will answer in the negative, for it has previously held that analogous "lesser" forms of expression do not stand on the same constitutional footing as more traditional varieties of speech . . . . At this point, a crucial issue will arise: given the strong constitutional presumption in favor of free speech, just what burden of proof must the state carry to justify its imposition of restraints on the information? Or, to put it in legalistic terms, what standard of review will the Court apply to the government's stated justification for the challenged restrictions?

To determine the relevant standard of review, the Court will focus on two broad questions. First, does the government have a possessory interest in the underlying information? If so, the Court will apply a mere "reasonableness" standard to any governmental restraints imposed on government employees in an effort to preserve the secrecy of the data. Thus, for example, in Snepp v. United States 444 U.S. 507 (1980), a recent case involving a book published by a former CIA agent, the Court broadly upheld the state's power to impose "reasonable restrictions" on the dissemination of governmental information obtained by government employees. In addition, the Court pointedly noted that this general principle applies "even in the absence of an express agreement" between the government and the employee.

In like manner, the Court will probably sustain any reasonable restraints imposed on the dissemination of information resulting from the government-funded research of private parties ....

On the other hand, if the state attempts to regulate the dissemination of nongovernmental information by private parties, the Court will apply a far more demanding standard of review. In such a case, the weight of the state's burden will be determined by a second line of judicial inquiry focusing on the precise way in which the government has restricted the free-speech right.

On this issue, there are two major possibilities: either the state has imposed a "subsequent punishment"usually in the form of criminal penalties-on individuals who have already published the restricted information, or it has blocked the dissemination of the data by issuing a "prior restraint." In the case of a subsequent punishment, the Court will uphold the action only if the state can demonstrate a "compelling" interest in regulation⁹-a burden of proof that stands as the modern analog of the well-known "clear and present danger" test formulated by Oliver Wendell Holmes.¹⁰ In the case of a prior restraint, the Court will apply an even more demanding standard of review, since the government is seeking to block the timely dissemination of information and ideas. Indeed, on the evidence of the so-called Pentagon Papers decision (New York Times v. United States) 403 U.S. 713 (1971) the Court will uphold the restraint only if the government can show that a "grave" and "irreparable" harm will almost surely result from publication of the data in question.

Clearly, under either standard of review the state is faced with an exceedingly difficult task. Nevertheless, the Court has indicated that in some "exceptional" cases, principally in the area of national security, the government's interest in regulation may be sufficient to warrant a direct infringement on fully protected speech. The remaining question, therefore, is: Just how will the Court assess the importance of the state's concerns to determine whether they are adequate to justify an abridgment of First Amendment freedoms?

The Court has held that the strength of the government's interest in regulation is determined in large part by two independent factors: the nature of the harm that the state is seeking to avert and the likelihood of its occurrence. In particular, the crucial inquiry centers on whether the "gravity of the 'evil,' discounted by its improbability, justifies such invasion of the free speech right as is necessary to avoid the danger."¹¹ With this approach, the seriousness of the threatened danger will affect to some extent the showing required of the government on the "likelihood of occurrence.". . .

Together with the gravity of the threatened harm to national security, the Court's finding on the likelihood of occurrence will generally determine whether the state's interest in regulation is sufficient to warrant the restriction of First Amendment rights. Assume, for instance, that the government can show that the Soviets have sufficient skills to acquire a new military capability by exploiting an American breakthrough in directed energy weaponry. On these facts, the Court will no doubt agree that the government's concerns are sufficiently compelling to warrant an abridgment of First Amendment freedoms. This will probably hold true, moreover, even if the government concedes that the Soviets will eventually acquire the capability anyway, since the maintenance of a military lead time can be highly advantageous. On the other hand, if the threatened harm to the nation's security is less serious, the state's case will be correspondingly weakened, and all the more so if the receiving nation is shown to lack the requisite skills or resources to absorb the technology.

The crux of conventional First Amendment analysis lies in the Court's effort to determine whether the restricted information gives rise to a substantial danger and thus warrants governmental regulation. However, if this issue is resolved in the state's favor, the Court will pursue a further line of inquiry focusing on the government's regulatory technique. In particular, the Court will determine whether the restraints on speech imposed by the state are more extensive than necessary to serve its underlying concerns.¹² Accordingly, even if the government can demonstrate a "compelling" interest in regulation, the Court will invalidate the challenged restraints if it finds that a "less restrictive alternative" could serve the asserted interest equally well ....

What is most striking about the Court's method of First Amendment adjudication is that it takes into account virtually all the commonsense perceptions that have informed the general policy debate on the government's effort to control the export of scientific and technical knowledge. Indeed, if the Court applies its standard analysis to this issue, it will not only give due weight to the value of scientific freedom but will also examine critically the nature and magnitude of the threatened harm to national security. In addition, it will address a variety of other considerations, such as the technical skills of the receiving nation and the reasonableness of the regulatory technique. By incorporating each of these factors into a methods of adjudication that formally allocates the burden of proof, the Court's approach provides a well-defined analytical framework for accommodating the claims of scientific freedom with the legitimate interests of national security.

Footnotes
1. The paper is an edited/cut version of an article that appeared in Science 12 August 1983, vol. 221, pp. 620-624. It is printed here by permission of the American Association for the Advancement of Science and of the Author.
2. Arms Export Control Act, 22 US Code, Sect. 2778 (1976).
3. International Traffic in Arms Regulations, Title 22 Code Fed. Reg. parts 121-128 (Feb. 1976).
4. Export Administration Act of 1979, 50 US Code Appendix, Sect. 2401-20 (1979).
5. Title 15, Code Fed. Reg., part 379 (1982).
6. T. Emerson, A System of Freedom of Expression, (Random House N.Y., 1970), pp. 6-7.
7. Virginia Pharmacy Board v. Virginia Consumer Council, 425 US 748 (1976).
8 National Academy of Sciences, Scientific Communication and National Security, (Washington D.C. 1983) pp. 42-45.
9. L. Tribe, American Constitutional Law, (Foundation Press, Mineola, N.Y., 1978).
10. Schenck v. United States, 249 US 47(1919).
11. Nebraska Press Association v. Stuart, 427 US 539, 562 (1976).
12. In re Primus, 436 US 412 (1978).