The capacity of the United States to protect its scientific and technological (S&T) advantage is vital not only as a matter of maintaining U.S. economic competitiveness, but also as a matter of ensuring U.S. national security. From an economic standpoint, experts estimate that theft by foreign adversaries of research, intellectual property, and trade secrets annually costs the U.S. economy up to $600 billion. With respect to national security, the U.S. military’s ability to successfully achieve technical superiority tomorrow depends on innovation today, including in areas like artificial intelligence, synthetic biology, microelectronics, quantum computing, and robotics.
Yet, while there is widespread agreement among U.S. experts that the United States must protect its S&T advantage, a spirited debate is currently underway between intelligence professionals and members of the scientific community about the measures necessary to do so. At the extremes, on one side, members of the scientific community (rightly) highlight that U.S. scientific innovation depends on openness and transparency, as well as the ability of the United States to attract and retain highly qualified foreign scholars. On the other side, intelligence professionals (rightly) emphasize that some foreign governments—most notably the People’s Republic of China (PRC)—do not abide by widely accepted norms related to scientific research, and instead exploit for their own advantage the very features of openness that make the U.S. research enterprise successful.
Although lawyers have engaged with this debate in academic scholarship and in strategic ways such as influencing national-level law and policy, legal practitioners have often been absent at the operational level. Consequently, legal practitioners appear to be missing opportunities to aid members of the intelligence and scientific communities as they navigate an ever-increasing body of law and policy regulating federally-funded S&T research, instead engaging only when their clients affirmatively bring issues to them. As a result, national security and research and development (R&D) professionals often are left wondering what to make of what they perceive as a morass of conflicting law and policy.
Lawyers know well the value they offer to their clients—they explain laws and policies governing their clients’ equities and correct related misunderstandings; they resolve apparent tension in, and between, various legal and regulatory provisions; they identify risks and ensure compliance; and they propose reforms aimed at advancing their clients’ interests. Each of these competencies holds substantial promise for addressing the challenges surrounding S&T protection, thereby suggesting a need for greater engagement by legal practitioners.
By way of facilitating such engagement, this Article provides a primer on a narrow, but foundational, aspect of the technology protection debate. Recently, experts in the scientific and intelligence communities have been at odds over their respective perspectives about the continuing viability of National Security Decision Directive (NSDD) 189, a 1985 directive issued by President Ronald Reagan that established a national policy favoring openness with respect to federally-funded fundamental research. The wisdom of that position is currently being challenged for numerous reasons, including the revolutionary pace at which innovation now occurs; the increasing conflation between civil and military uses of technology; and the rising threat to the United States posed by the PRC Government—a threat that has prompted a flurry of recent executive branch and legislative action aimed at protecting federally-funded research from malign foreign influence. Given these developments, and particularly in light of arguably conflicting executive orders and laws that have arisen since President Reagan issued NSDD-189, some experts in the scientific community have suggested that the thrust of NSDD-189 has been eviscerated and, therefore, requires forceful reaffirmation. On the other hand, some intelligence experts assert that the policy NSDD-189 embodies is too permissive given current threats and, thus, should be tempered in the interest of protecting U.S. technology. But the discussion between these two communities about NSDD-189’s continuing viability has neglected a profoundly important question: What does the law say?
This Article addresses that gap, with the aim of providing legal practitioners with knowledge they can leverage to contribute to the technology protection debate. To do so, the Article proceeds as follows: In Part I, the Article explores the historical and legal context that frames these issues. More particularly, it surveys the milieu from which NSDD-189 emerged; discusses what the directive says and, as importantly, what it does not say; and surveys various subsequent presidential and congressional actions that, at least to some degree, occupy the same space as NSDD-189. In Part II, the Article considers what force NSDD-189 continues to hold, a central subject at the essence of today’s technology debate. In doing so, Part II examines the questions of what legal force executive actions generally hold; what interpretive theory to apply to them, particularly when they appear to conflict with each other and with related legislative enactments; and what the answers to those questions yield in terms of ascertaining the continuing viability of NSDD-189. Finally, some concluding remarks are provided.
I. Historical and Legal Context
Given the recent flood of attention on the subject, one could be forgiven for mistakenly believing the issue of technology protection is novel. In fact, U.S. law and policymakers have long recognized the need, from a national security perspective, to ensure U.S. S&T superiority. As described below, over the years this has led to a surge of federal legislation, executive actions, and academic studies related to technology protection.
A. Historical Background
Efforts by the U.S. government to protect S&T date back over a century. As early as 1917, the United States enacted the Espionage Act, which prohibited obtaining information, recording pictures, or copying descriptions of national defense information with intent to injure the United States or advantage a foreign nation. Subsequent efforts to prevent the transfer of U.S. technology followed, including passage of the Export Control Act in 1940; creation of a new security classification system by the Office of War Information in 1942; and enactment of the Atomic Energy Act, which established categories of “restricted data,” in 1946.
Following World War II, the looming threat the Cold War would pose began to crystalize and evidence emerged that the Soviet Union was exploiting the United States “to strengthen [its] military industrial base through the legal and illegal acquisition of Western technology.” The U.S. government’s response yielded an array of executive and legislative actions. For instance, Congress passed the National Security Act in 1947, which, among other things, established the National Security Council and restructured the U.S. military and intelligence apparatus. In 1951, President Truman issued Executive Order (EO) 10290, thereby extending the U.S. security classification system beyond military agencies. Three years later, Congress amended the Atomic Energy Act to allow greater civil participation in the U.S. nuclear research enterprise.
This trend was part of a larger effort after World War II to provide federal funding to U.S. universities conducting basic research. The move was deliberate and viewed as a response to success universities had achieved in advancing national security during World War II. However, as civil participation in the U.S. defense enterprise increased, foreign adversaries likewise “shifted from an exclusively military focus to an all-inclusive one, targeting the civilian sector and universities as well.” Against this backdrop, in 1976, the U.S. Department of Defense’s (DoD) Director of Defense Research and Engineering requested that the Defense Science Board (DSB) prepare a report on the implications of transferred technology for U.S. defense. The DSB’s subsequently issued report—known as the “Bucy report” after Fred Bucy, the chairperson of the related task force—“represented a radical departure from contemporary export control concepts which focused primarily on material control.” That is, instead of exclusively concentrating on the harms caused by transfer of the output of U.S. technological innovation, the Bucy report “placed the focus on the exporting of know-how and certain keystone technologies.” The report therefore called, among other things, for a comprehensive DoD study of the vectors by which technology was transferred abroad, including through training that foreign scholars were obtaining from “U.S. technical institutes and universities.”
Around the same time, geopolitical events such as the Soviet Union’s invasion of Afghanistan were causing U.S.-Soviet relations to further deteriorate. New trends also were emerging in the scientific enterprise that exacerbated challenges surrounding technology protection. Namely, because innovation in many fields was occurring at a revolutionary pace, “the distinction between basic and applied research was becoming less relevant. . . . [And] an increasing number of technologies were dual-use in character . . . [making it] difficult, if not impossible, to separate military applications from civilian ones.” The DoD thus began to look for new ways to control U.S. technology, including by categorizing certain areas of information as “sensitive.”
B. The Corson Report
Owing to these and other similar developments, conflict between the scientific and national security communities deepened, such that efforts to impose greater restrictions on technology transfer eventually were met with a “hostile response from the academic community.” In an effort to ameliorate the tension, in 1982, the DoD, in partnership with other organizations such as the National Science Foundation (NSF), asked the National Academy of Science “to examine the relation between scientific communication and national security in light of the growing concern that foreign nations are gaining military advantage from such research.” The resulting panel—chaired by Dale Corson, a physicist and President Emeritus of Cornell University—began its work with a recognition that remains salient today; that is, that the U.S. government and research community both would “lose much” if the nation were unable to “find a policy course that reflect[ed] legitimate concerns” of both the national security and scientific communities. What emerged from the panel’s work was a report titled Scientific Communication and National Security, informally known as the “Corson report.” To this day, the legal regime surrounding U.S. technology protection is grounded in recommendations made in the report.
To begin, the Corson report framed the issue by describing what had, by then, become two entrenched and seemingly irreconcilable positions. On one hand, the report explained that “openness has contributed to American military and economic strength.” On the other, it acknowledged that “recent trends, including apparent increases in acquisition efforts by our adversaries, have raised serious concerns that openness may harm U.S. security by providing adversaries with militarily relevant technologies that can be directed against us.” The report then explored the increasing alarm national leaders were expressing about Soviet acquisition of U.S. S&T and sought to explain its causes.
Mirroring remarkably similar comments made today about China, the Corson report cited one U.S. official who expressed that “the Soviets exploit[ed] scientific exchanges . . . in a highly orchestrated, centrally directed effort aimed at gathering the technical information required to enhance their military posture.” Another U.S. official stated that a network of Soviet intelligence operatives were “exploit[ing] the ‘soft underbelly’” of U.S. society, including “our traditions of an open press and unrestricted access to knowledge,” and “the desire of academia to jealously preserve its prerogatives as a community of scholars unencumbered by government regulations.” The report attributed rising U.S. concern to a perception that the U.S. technological advantage over the Soviets was diminishing. It also observed that “the separation between military operations and scientific research [had] quickly narrowed,” and “a steadily increasing share of these technologies [were] dual-use in nature,” meaning that they had “both military and nonmilitary applications.”
Nevertheless, while acknowledging that the Soviets had acquired a substantial amount of U.S. military technology, the Corson report concluded “that universities and open scientific communication have been the source of very little of this technology transfer.” This view was reflected in the report’s three principal recommendations, which would have the effect of framing the future debate. First, the report recommended “that the vast majority of university research, whether basic or applied, should be subject to no limitations on access or communications.” Second, it urged that “[w]here specific information must perforce be kept secret, it should be classified strictly and guarded carefully.” Third, it suggested that additional restrictions might be appropriate in certain, limited “gray areas,” such as barring researchers from designated countries from participating in federally-funded R&D, or requiring prepublication review of certain research manuscripts. However, to fall within the gray area, the panel concluded that a technology would need to meet each of the following criteria:
· The technology is developing rapidly, and the time from basic science to application is short;
· The technology has identifiable direct military applications; or it is dual-use and involves process or production-related techniques;
· Transfer of the technology would give the U.S.S.R. a significant near-term military advantage; and
· The United States is the only source of information about the technology, or other friendly nations that could also be the source have control systems as secure as ours.
While it represented an understandable effort to find compromise between the scientific and national security communities, this third category of R&D, which would come to be called “gray-area research,” grew into “a major sticking point” in the DoD’s efforts to establish related policy. Not only was gray-area research causing friction between the DoD and university officials, it also became clear that there were divides within the Pentagon. Work to try to define and identify gray-area research was proving too complicated, particularly when the benefits were unclear. As a result, in May 1984, a senior DoD official testified before Congress that the DoD was abandoning “the gray-area concept” in favor of a binary “classification-nonclassification approach”—a move that stunned the scientific community.
C. National Security Decision Directive 189
This approach evolved into the policy ultimately embodied in NSDD-189, issued by President Reagan on September 21, 1985. Just over a page in length, the significance of the directive is belied by its brevity. NSDD-189 begins by announcing that its purpose is to “establish[] national policy for controlling the flow of science, technology, and engineering information produced in federally-funded fundamental research at colleges, universities, and laboratories.” The directive then defines fundamental research by stating that
“[f]undamental research” means basic and applied research in science and engineering, the results of which ordinarily are published and shared broadly within the scientific community, as distinguished from proprietary research and from industrial development, design, production, and product utilization, the results of which ordinarily are restricted for proprietary or national security reasons.
In a background section, NSDD-189 underscores the significance of the national security threat posed by foreign acquisition of U.S. technology, while also acknowledging that U.S. leadership in S&T, which “is an essential element in our economic and physical security,” depends on the open exchange of ideas. Finally, NSDD-189 announces:
It is the policy of this Administration that, to the maximum extent possible, the products of fundamental research remain unrestricted. It is also the policy of this Administration that, where the national security requires control, the mechanism for control of information generated during federally-funded fundamental research in science, technology and engineering at colleges, universities and laboratories is classification. . . . No restrictions may be placed upon the conduct or reporting of federally-funded fundamental research that has not received national security classification, except as provided in applicable U.S. Statutes.
The enduring value experts see in NSDD-189 is reflected by the fact that, in the decades since its promulgation, studies examining fundamental research policy have repeatedly called on the U.S. government to formally and forcefully reaffirm the principles the directive embodies. As one study put it, “NSDD-189’s application to fundamental research conducted in U.S. universities cannot be overemphasized.” Stated simply, by creating a binary regime under which federally-funded fundamental research must be either unrestricted or—in rare cases required for national security—classified, the directive established as national policy a default favoring openness.
In the years following issuance of NSDD-189, the U.S. government’s commitment to the directive’s principles manifested itself in numerous ways. For example, the Department of Commerce’s Export Administration Regulations (EAR) and the Department of State’s International Traffic in Arms Regulations (ITAR) each were revised to acknowledge that, under most circumstances, NSDD-189 “provides an exclusion for certain research activities at colleges and universities in the United States from the application of the export regulations.” Similarly, policies issued by federal agencies demonstrated a responsibility to openness by calling for prompt and wide dissemination of agency-funded research.
Nevertheless, foreshadowing later trouble, the text of NSDD-189 also made certain concessions about the unknowable state of the future. It noted, for example, that the “government-university-industry partnership in research activities” on which it was focused was only then “emerging.” As such, it left open the possibility that “a more significant problem may well develop.” It also acknowledged that, even beyond classification, additional restrictions on federally-funded fundamental research might be “provided in applicable U.S. Statutes.” These caveats to the basic thrust of NSDD-189 went largely unrecognized and, as discussed below, have been a source of subsequent friction.
D. The Aftermath of National Security Decision Directive 189
While NSDD-189 was greeted with a warm reception, gradually, the research community began to voice concerns that the U.S. government was eroding the directive’s permissive approach to unrestricted, federally-funded fundamental research. Among other things, researchers undertook studies that eventually revealed that many awards for federally-funded research were only issued after a research institution agreed to provisions restricting the use of foreign nationals or publication of research results. Complaints also arose about the proliferation of export control measures and the increasing use of controls on federally-funded R&D like the “sensitive but unclassified” (SBU) designation. This led to calls for the U.S. government to reaffirm and abide by NSDD-189.
For its part, the executive branch at least voiced an enduring commitment to the core aspects of NSDD-189. For example, while expressly highlighting that she was communicating “[o]n behalf of the President,” in 2001, Assistant to the President for National Security Affairs Condoleezza Rice stated that “the policy on the transfer of scientific, technical, and engineering information set forth in NSDD-189 shall remain in effect, and we will ensure that this policy is followed.” Likewise, in 2010, then Undersecretary of Defense for Acquisition, Technology, and Logistics Ashton Carter issued a memorandum reaffirming that “the DoD will not restrict disclosure of the results of fundamental research . . . unless such research efforts are classified for reasons of national security or as otherwise required by applicable federal statutes, regulations, or executive orders.” Again, the memorandum explicitly referenced NSDD-189, stating that DoD grants, contracts, and negotiations must be “fully compliant with [NSDD-189].”
Yet, at the same time, concern was rising about foreign adversaries exploiting the openness of the federally-funded R&D enterprise in ways that were harmful to the United States. The attacks of September 11, 2001, and the anthrax incidents that followed shortly thereafter created a “more tangible possibility of bioterrorism” and fueled renewed fears “regarding the performance and publication of dual-use research.” In an effort to quell those fears, in March 2002, Assistant to the President and Chief of Staff Andrew Card, Jr., released a memorandum for heads of executive branch departments and agencies reinforcing the obligation they had to “safeguard Government records regarding weapons of mass destruction . . . includ[ing] biological, radiological, and nuclear weapons.” The memorandum included directives widely viewed as opening the aperture of controls federal agencies could place on information, including more expansive use of the SBU designation.
Congress likewise was legislating in the space. Of most relevance here was the passage of the Homeland Security Act in 2002, which created the Department of Homeland Security (DHS). Likely in response to the contemporaneous debate that was raging about how to manage distribution of fundamental research, the Act stipulated that “[t]o the greatest extent practicable, research conducted or supported by [DHS] shall be unclassified.” However, the Act also not only supported use of the SBU designation, but in fact required the President to “prescribe and implement procedures under which relevant Federal agencies . . . identify and safeguard homeland security information that is sensitive but unclassified.”
Although this suggested possible uniformity, at least, in the executive branch’s treatment of unclassified information that was nevertheless sensitive, in fact, numerous controls emerged to protect such information. As a Congressional Research Service report stated, “Included among these [were] widely used markings such as ‘Sensitive But Unclassified,’ ‘Limited Official Use,’ ‘Official Use Only,’ and ‘For Official Use Only.’” This led to significant confusion and, beyond the inherent policy objections researchers had, concern arose over possible inadvertent noncompliance. A study by the Center for Strategic and International Studies (CSIS) noted, for example, “[s]cientists feel vulnerable to violating rules on categories that are ill defined.” While ostensibly limited in scope to a review of the Department of Energy (DOE), CSIS’s conclusions that the DOE’s SBU “definition is so broad as to be unusable,” and that “[s]ensitive unclassified information is causing acute problems at DOE,” might as well have been lodged against the executive branch, writ large.
E. Executive Adoption of the Controlled Unclassified Information Designation
As a result, in 2008, President George W. Bush issued to the heads of executive branch departments and agencies a memorandum identifying the “Controlled Unclassified Information” (CUI) designation as “the single, categorical designation” for “unclassified information that does not meet the standards for” classification “but is (i) pertinent to the national interests of the United States or to the important interests outside the Federal Government, and (ii) under law or policy requires protection from unauthorized disclosure, special handling safeguards, or prescribed limits on exchange or dissemination.” The memorandum designated the National Archives and Records Administration (NARA) as the executive agent responsible for managing the “CUI Framework,” and charged NARA with creating and maintaining a “CUI Registry.”
In essence, President Bush’s 2008 memorandum codified establishment of the Corson report’s gray area by acknowledging that some information required dissemination control even though it was unclassified. Perhaps owing to the fact that the memorandum had been issued only to heads of executive branch departments and agencies, in 2010, President Barack Obama rescinded the memorandum and issued EO 13556, which again aimed to address the “confusing patchwork” of policies, procedures, and markings that had arisen to manage information not subject to classification under EO 13526 or the Atomic Energy Act but nevertheless still “require[] safeguarding or dissemination controls pursuant to and consistent with laws, regulations, and Government-wide policies.” EO 13556 reaffirmed the CUI program and NARA’s designation as the executive agent responsible for managing it.
EO 13556 also reinforced that “[t]he CUI categories and subcategories shall serve as exclusive designations for identifying unclassified information throughout the executive branch that requires safeguarding or dissemination controls, pursuant to and consistent with applicable law, regulations, and Government-wide policies.” To effect the Order’s mandate, among other things, the EO assigned NARA responsibility for “approv[ing] categories and subcategories of CUI and associated markings to be applied uniformly throughout the executive branch.” NARA subsequently did so, and the organization continues to maintain a robust CUI Registry.
Significantly, in a 2019 report commissioned by the NSF to study fundamental research security, an elite, scientific advisory board supporting the DoD named JASON recognized that “there is no division or category within the CUI Registry directly concerned with the conduct of academic research, and this appears to be broadly consistent with the principles laid out in NSDD-189.” That said, there is no question that certain of the categories of information ultimately included in the CUI Registry clearly have application to fundamental research. For example, JASON stated that “two categories of export controls, namely, ‘Export Controls’ and ‘Export Controlled Research,’ come into play for novel technologies and software that could be considered dual use, or which might adversely affect U.S. national security or nonproliferation objectives.” Another category, titled “Controlled Technical Information” (CTI), has implications for DoD and National Aeronautics and Space Administration (NASA) research, insofar as CTI is defined as “technical information with military or space application that is subject to controls on the access, use, reproduction, modification, performance, display, release, disclosure, or dissemination.”
Notwithstanding JASON’s acknowledgment that EO 13556 and its CUI regime were generally consistent with NSDD-189, the panel also observed that CUI categories “remain to be fully reconciled with NSDD-189.” Even more, JASON reported to the NSF that the CUI designation “is ill-suited to the protection of fundamental research areas.” JASON therefore urged the NSF to “support reaffirmation of the principles of NSDD-189,” and “discourage the use of new CUI definitions.” Others in the R&D community have taken critique of the CUI designation a step further, seemingly operating, in effect, as if it does not exist. Citing the JASON study, for instance, one Princeton University professor recently stated that “the creation of intermediate categories” like CUI “would cause confusion and be counterproductive” because they “do[] not map well to the standards of NSDD-189.” Given the legal force associated with EO 13556, such comments are curious and raise the question of how the Order should be reconciled with NSDD-189.
F. The Rise of China and Issuance of National Security Presidential Memorandum 33
Other, more recent, presidential and legislative activity has served to further cloud the question of what viability NSDD-189 has today. In large part, this activity has been directed at threats posed by the PRC government and reflects concerns about the rapidity with which China has become a near-peer competitor of the United States. As a National Academies report observed, “Over the past two decades, China has systematically pursued strategies for dominating technology development in key areas.” It has done so not only through major investments in its own S&T ecosystem, but also through acquisition of U.S. S&T, “including through illicit means.” That is, among other things, “China’s intelligence services . . . exploit the openness of American society, especially academia and the scientific community” to advance PRC interests and harm the United States. This exploitation occurs through many vectors, including undisclosed conflicts of interest or commitments, intellectual property theft, cyberattacks, and economic espionage.
To address these threats, in 2021, President Donald Trump issued National Security Presidential Memorandum (NSPM) 33. NSPM-33 stated that the “memorandum directs action to strengthen protections of United States Government-supported Research and Development (R&D) against foreign government interference and exploitation.” Unlike EO 13556—which aims primarily to control the dissemination of information like that produced via federally-funded grants—NSPM-33 imposes certain barriers to entry, designed to prevent malign actors from participating in the R&D enterprise. In doing so, NSPM-33 contains several significant features warranting attention.
First, NSPM-33 acknowledges that “[t]he open and collaborative nature of the United States R&D enterprise underpins America’s innovation, S&T leadership, economic competitiveness, and national security.” Nevertheless, the memorandum underscores that “some foreign governments, including the People’s Republic of China, have not demonstrated a reciprocal dedication to open scientific exchange, and seek to exploit open United States and international research environments to circumvent the costs and risks of conducting research.” As a result, NSPM-33 stresses that adversaries have been able to “increas[e] their economic and military competitiveness at the expense of the United States, its allies, and its partners.”
Second, even while attempting to address this challenge, NSMP-33 mentions NSDD-189, stating that “[m]uch of United States Government-supported R&D is broadly shared and includes fundamental research as defined in” NSDD-189. Thus, while imposing measures that are designed “to protect intellectual capital, discourage research misappropriation, and ensure responsible management of United States taxpayer dollars,” in harmony with NSDD-189, NSPM-33 expresses a commitment to “maintaining an open environment to foster research discoveries and innovation that benefit our Nation and the world.”
Third, to harden the federally-funded R&D enterprise against malign foreign influence, among other things, NSPM-33 obligates federal agencies that fund R&D activities to “require that participants in the . . . enterprise who significantly influence the design, conduct, reporting, reviewing, or funding of Federally-funded research disclose appropriate information . . . [to] enable reliable determinations of whether and where conflicts of interest and commitment exist.” The term conflict of interest was given a familiar meaning focused on divided loyalties that might exist due to financial interests or relationships in conflict with the research. The phrase conflict of commitment was more nuanced, centering on “a situation in which an individual accepts or incurs conflicting obligations between or among multiple employers or other entities.” One specific conflict of commitment that garnered attention in NSPM-33 was participation in foreign-government sponsored talent recruitment programs, which the memorandum defined as “effort[s] directly or indirectly organized, managed, or funded by a foreign government or institution to recruit S&T professionals or students.” As explained in the memorandum, “Some foreign government-sponsored talent recruitment programs operate with the intent to import or otherwise acquire from abroad, sometimes through illicit means, proprietary technology or software, unpublished data and methods, and intellectual property to further the military modernization goals and/or economic goals of a foreign government.”
Importantly, NSPM-33 called on the Assistant to the President for National Security Affairs, in collaboration with the Office of Management and Budget (OMB) and the Office of Science and Technology Policy (OSTP), to coordinate implementation of the memorandum’s directives.As part of that effort, the OSTP solicited input from the scientific community. Unsurprisingly, this yielded renewed calls for the Biden Administration to reaffirm NSDD-189 and its commitment to unrestricted fundamental research.
G. Recent Legislative Enactments to Protect Federally-Funded R&D
Around the same time that President Trump issued NSPM-33, Congress enacted a flurry of legislation similarly designed to protect the U.S. S&T advantage. For instance, in the fiscal year (FY) 2021 National Defense Authorization Act (NDAA), Congress mandated that federal research agencies “shall require, as part of any application for a research and development award,” that certain covered individuals “disclose the amount, type, and source of all current and pending research support received by, or expected to be received by, the individual as of the time of the disclosure.” Covered individuals also must certify that the disclosure is “current, accurate, and complete,” and must “agree to update such disclosure at the request of the agency.” The statute grants various enforcement powers to federal research agencies, including, among other things, rejection of noncompliant applications, suspension or termination of pending R&D awards, and temporary or permanent debarment.
In 2022, Congress also passed the CHIPS and Science Act. As relevant here, the Act requires OSTP to “publish and widely distribute a uniform set of guidelines for Federal research agencies regarding foreign talent recruitment programs.” In addition, federal research agencies are mandated by the Act to establish a policy requiring covered individuals to certify they are not a part of a malign foreign talent recruitment program.
Additional recent legislation reflects similar congressional effort to address the challenge of foreign malign influence in the U.S. S&T enterprise. For example, in the FY 2019 NDAA, Congress included a provision requiring the Secretary of Defense to “establish an initiative to work with academic institutions who perform defense research and engineering activities . . . to support protection of intellectual property, controlled information, key personnel, and information about critical technology,” and “to limit undue influence, including through foreign talent programs.” This provision was modified in each of the two subsequent NDAAs—first to require the Secretary to also develop and update a list of academic institutions of the PRC, Russia, and other countries that engage in certain malign behaviors, and later to require the Secretary to also publish a list “of foreign talent programs that pose a threat to the national security interests of the United States.” In addition to imposing the disclosure requirements mentioned above, the FY 2021 NDAA also restricts DoD funding to higher education institutions that host Confucius Institutes, which are assessed by the intelligence community to serve as fronts for PRC intelligence service activities.
H. Recent Activity Related to National Security Decision Directive 189
This bevy of activity has once again led to national-level discussions focused on the continuing viability, or lack thereof, of NSDD-189. Most recently, the Defense Advanced Research Projects Agency (DARPA) and the NSF “asked the National Academies of Sciences, Engineering, and Medicine (the National Academies) to convene an ad hoc committee to consider policies and practices related to the production and commercialization of research in domains critical to national security.” The committee’s work yielded a lengthy study issued in September 2022 titled Protecting U.S. Technological Advantage, which included four recommendations aimed at protecting the U.S. S&T advantage. As applicable here, the first recommendation was that “[t]he President, through an executive order, should clearly reaffirm that it is the policy of the United States that fundamental research, to the maximum extent possible, should remain unrestricted.” The National Academies also recommended that “the executive order should direct the Office of Science and Technology Policy, in coordination with federal agencies, to define criteria for open and restricted research environments within 120 days of issuance of the executive order.”
Subsequent efforts by the National Academies have focused even more attention specifically on NSDD-189. For example, in meetings in November 2022 and January 2023, the National Academies convened experts to consider, among other things, “whether now is the time to revisit NSDD-189 and the practical effects of doing so, based on views of the foreign policy, security, scientific, and legislative communities.” To date, the meetings do not appear to have yielded consensus on NSDD-189’s current status or on the appropriate path forward.
II. National Security Decision Directive 189’s Continuing Force
As noted at the outset of this Article, there has been very little, if any, focus on what the law tells us about the force NSDD-189 continues to hold. That said, it is beyond question that the ongoing debate about the issue is taking place in an unclear legal regime. Even considering only the executive actions discussed above (i.e., NSDD-189, EO 13556, and NSMP-33), there is confusion, for instance, about the legal force of executive orders; what legal significance to give each order relative to the others; and how these specific orders are (or are not) meant to be read in conjunction with one another. The landscape is further muddied by the fact that, as discussed above, both presidential administrations and Congress have continued to issue executive and legislative mandates, respectively, that regulate the conduct and reporting of fundamental research. On their face, certain actions arguably appear to be in tension, if not outright conflict, with one another.
This patchwork of executive action and congressional legislation has created significant confusion and raises several important questions about the current legal framework surrounding fundamental research. Determining the force NSDD-189 has today requires focus on several central questions, three of which are addressed below. Those are: (1) what legal force do executive orders generally hold; (2) what legal distinction, if any, exists between the various forms of executive instruments relevant to this issue; and (3) how is the meaning and effect of various executive orders ascertained, particularly when they appear to be in tension with one another and with surrounding legislation.
A. Lawfully Issued Executive Orders Generally Hold the Force and Effect of Law Until Rescinded or Stricken
As alluded to above, there have been strains of thought in some scientific literature that implicitly reflect a view that NSDD-189 lacks any inherent legal force. For example, in the JASON study discussed above, the authors stated that “[t]he fundamental principles embraced by NSDD-189, along with much of its original wording, were subsequently incorporated into the Federal Acquisition Regulations (FAR) and are therefore the law of the land.” A National Academies study similarly expressed that “[c]ontracting officers and universities sometimes do not recognize that the fundamental principles as well as much of the wording of NSDD-189 are incorporated into the Federal Acquisition Regulations.” Such statements strike legal practitioners as peculiar, given the inherent legal force and effect presidential promulgations like NSDD-189 hold.
To be sure, it is clear that “[t]he Constitution does not mention the president’s authority to issue orders.” Nonetheless, it is also acknowledged that “the president’s power to do so is by now beyond dispute.” The question of what force such orders hold fundamentally distills down to the question of what authority a president may exercise, in the first instance. Put simply, executive authority is limited and must be found in either Article II of the Constitution or in a delegation of Congress’ Article I powers to the president. Thus, by extension, it follows that “[t]he President’s power, if any, to issue [an] order must stem either from an act of Congress or from the Constitution itself.”
This bedrock principle in turn explains why, from a legal perspective, executive orders hold the force of law. That is, properly issued executive orders or directives have as their legal underpinning constitutional power or, in the case of congressional delegations, a mix of constitutional and statutory power. In either case, the upshot is that if executive orders “are based on appropriate authority,” meaning if either Article II or a congressional delegation truly vest authority for the related issue in the president, such orders “have the force and effect of law.” Consequently, presidential orders have the impact of placing “institutions, such as Congress, administrative agencies and the courts, as well as the public in the position of responding to or implementing the policy and law they embody.”
As importantly, as legal directives of the president, executive orders generally retain their force as law until modified, rescinded, or judicially stricken. Thus, orders that are issued “remain effective [even] upon a change in administration . . . and . . . continue to be effective until subsequent presidential action is taken.” That said, a president has nearly unlimited authority to revoke or modify previously issued orders, whether issued by his own administration or a prior one. As Jack M. Beerman states, “All [a] President needs to do to revoke or revise an executive order . . . is issue a new executive order,” which is a power that presidents have exercised numerous times throughout history. Nevertheless, to preserve executive intent, and in the interest of clarity, the law disfavors repeals by implication. Therefore, when later-in-time executive orders are intended to operate as revocations of previously issued orders, it is customary that the former will expressly indicate as much.
As applied here, this framework leads to the conclusion that, as a legal matter, NSDD-189, EO 13556, and NSPM-33 all have the force and effect of law and remain legally viable, so long as they (1) were lawfully issued, and (2) have never been rescinded. With regard to the former, although questions have occasionally arisen over the years about the wisdom of the policy embodied in these executive orders, there appears to be no argument about whether they were lawfully issued. To varying degrees, NSDD-189, EO 13556, and NSPM-33, all concern access to R&D information produced through executive branch efforts and all touch upon national security. Further, in EO 13556 and NSPM-33, President Obama and President Trump, respectively, cited as his authority to issue the order “the authority vested in [him] as President by the Constitution and the laws of the United States of America.” Although NSDD-189 does not expressly indicate the authority on which it is based, on its face it aims to address “a significant threat to our national security.”
From a constitutional perspective, the scope of a president’s Article II powers has been a matter of heated debate that remains somewhat unsettled. Discussions on the topic often begin with Justice Robert Jackson’s concurring opinion in Youngstown Sheet & Tube Co. v. Sawyer, which offered a three-part framework for assessing a president’s powers. Under Justice Jackson’s framing, a president’s authority is (1) at its zenith when he acts pursuant to congressional authorization, (2) at its lowest ebb when he acts in contravention of Congress, and (3) in a “zone of twilight” when he “acts in absence of either a congressional grant or denial of authority.” According to Justice Jackson, when operating within the twilight zone, the “test of power is likely to depend on the imperatives of events and contemporary imponderables rather than on abstract theories of law.”
While Justice Jackson’s concurring opinion in Youngstown may provide a useful conceptual construct, as a practical matter, it has been criticized as “provid[ing] almost no guidance as to how . . . cases should be decided” and, thus, being “of no help.” Regardless, although the exact reach of the executive’s national security powers remains unclear, it is beyond debate that courts have afforded presidents great deference when considering their scope. For example, perhaps considering national security matters to be what Justice Jackson referred to as “imperatives of events” in Haig v. Agee, the Supreme Court affirmed the president’s authority to “withhold passports on the basis of substantial reasons of national security and foreign policy.” More significantly, as in Department of Navy v. Egan, the Court has extended this authority even to instances in which the president relies on inherent authority. There, the Court concluded that the president’s “authority to classify and control access to information bearing on national security . . . flows primarily from this constitutional investment of power in the President and exists quite apart from any explicit congressional grant.”
Thus, in short, the president’s inherent Article II powers are at their zenith in matters of national security, as is true here with respect to the issuance of NSDD-189, EO 13556, and NSPM-33. Moreover, as noted above, at least with respect to EO 13556 and NSPM-33, Presidents Obama and Trump also cited to “the laws of the United States of America” as their authority for issuing the orders, and it is easy to imagine the executive branch making a similar contention concerning NSDD-189. Assessing the efficacy of such claims is complicated by their lack of specificity—a condition suffered by numerous other executive orders. One can surmise, though, that the executive might point to the National Security Act of 1947; Espionage Act of 1917; the National Science and Technology Policy, Organization, and Priorities Act of 1976; FY 2021 NDAA; or any number of other statutes, including congressional authorizations and appropriations for specific federally-funded R&D programs, as the statutory bases for these promulgations.
Precisely which statutory regime the executive branch might point to hardly seems to matter; the upshot is likely the same. As one author concluded following a robust study of the issue, “courts lack a theory of the executive order’s role in our separation of powers system and . . . in the absence of such a theory, doctrine has developed along lines that augment executive branch power at Congress’s expense.” This is true whether a president issued an executive order under constitutional or statutory authority. In the latter case, it can be argued that “while courts often seek to effectuate (some version of) congressional intent when interpreting statutes, their guiding principle when interpreting executive orders—including Article I executive orders—has generally been to give effect to presidential intent.”
Even more, in affirming actions taken under executive orders, “courts sometimes draw on Article II powers to imply statutory authority” supporting them. Exemplary of this approach is Dames & Moore v. Regan. There, the Supreme Court affirmed use of an executive order that, in the context of the Iran Hostage Crisis, suspended pending lawsuits of American citizens against Iran. Although the government had argued that the order was authorized by the International Emergency Economic Powers Act (IEEPA) and Hostage Act, the Court “declined to conclude that the [statues] directly authorize[d] the President’s suspension of claims.” Nevertheless, the Court found the statutes “highly relevant in the looser sense of indicating congressional acceptance of a broad scope for executive action in circumstances such as those presented in this case.” That is, the Court relied on the “general tenor of Congress’ legislation,” and “the enactment of legislation closely related to the question,” in upholding the president’s action under the executive order. Thus, at bottom, whether viewed through the lens of constitutional authority, statutory authority, or some combination of the two, there appears to be no credible argument that NSDD-189, EO 13556, and NSPM-33 were not lawfully issued.
There also is no evidence that any of these orders have been rescinded. As previously explained, it is a customary practice that when a president intends to revoke a previously issued order, he expressly says as much given that the law disfavors repeals by implication. NSDD-189, EO 13556, and NSPM-33 contain no such indication. In fact, on the contrary, NSPM-33 expressly mentions NSDD-189 by way of acknowledging that “[m]uch of United States Government-supported R&D is broadly shared and includes fundamental research as defined in [NSDD-189].” If the intent had been for NSPM-33 to wholly disrupt that regime, one would not expect to see such favorable language. Additionally, although some have suggested that the orders are in conflict with one another, as explained below, the better view is that they are complementary and can be harmonized. Put simply, there is no obvious reason to believe that NSDD-189, EO 13556, or NSPM-33 has been rescinded, either explicitly or implicitly.
B. No Legal Distinction Exists Between Executive Promulgations Based on the Form the Instruments Take
Another question that has arisen in this setting is whether there is any legal significance to the fact that NSDD-189 is a presidential directive; EO 13556 is an executive order; and NSPM-33 is a presidential memorandum. The more general question of how to conceive of different presidential pronouncements has vexed observers for some time. One source suggests that “[a] widely accepted description of executive orders and proclamations comes from a report issued in 1957 by the House Government Operations Committee,” which stated the following:
Executive orders are generally directed to, and govern actions by, Government officials and agencies. . . . [while] [p]roclamations in most instances affect primarily the activities of private individuals. Since the President has no power or authority over individual citizens and their rights . . . the President’s proclamations are not legally binding.
This framework, however, has been criticized for neglecting “the fact that even orders formally directed at other government officials can have practical effects on private parties.”
Ultimately, as one author has noted, “Attempts by legal scholars and political scientists to classify these different forms of orders have been unavailing.” This is because, in general, there is no difference between these instruments as concerning their legal force and effect. As explained by the Department of Justice’s Office of Legal Counsel, “it is the substance of a presidential determination or directive that is controlling and not whether the document is styled in a particular manner.” The only apparent difference between various types of presidential pronouncements is procedural, rather than substantive. That is, under the Federal Register Act, executive orders and proclamations are required to be published in the Federal Register, whereas other instruments only require publication “when the President determines they have ‘general applicability and legal effect.’” That said, in part because there is no legally binding definition of the term executive order, and because “there are almost no legally enforceable procedural requirements that the president must satisfy before issuing (or repealing) an Executive Order or other presidential directive,” the form a presidential promulgation ultimately takes “may reflect nothing more than a bureaucratic choice.” Consequently, as relevant here, there is no legal distinction to be drawn from the fact that NSDD-189, EO 13556, and NSPM-33 each take a unique form.
C. Executive Orders and Statutes Should be Harmonized With One Another
Having demonstrated that NSDD-189, EO 13556, and NSPM-33 hold the force and effect of law, have not been rescinded, and are of equal force despite their different forms, the question remains of how they should be interpreted, both in conjunction with one another and with congressional legislation at least arguably occupying the same space. This is a particularly significant question given concerns that have been voiced by some experts that the executive instruments are in tension, if not outright conflict, with one another and laws passed by Congress.
As Tara Leigh Grove observes, “The federal judiciary . . . has regularly grappled with the meaning of presidential directives for well over a century.” Although limited authority suggests that there is an open question about the interpretive theory applicable to such directives, over time, a substantial body of caselaw has emerged in which “federal courts have repeatedly asserted that presidential directives should be treated just like statutes.” Indeed, in considering a case requiring it to interpret the meaning of an executive order, the Supreme Court has stated that it would “approach the construction of [an] Executive Order . . . as [it] would approach the construction of legislation.” This somewhat uncomplicates the analysis related to NSDD-189, given that the applicable legal landscape is formed by both executive orders and legislation.
To determine the meaning of legislation, courts often employ various canons of construction, which the Supreme Court has referred to simply as “rules of thumb.” Several canons could conceivably lend aid to the question of what to make of NSDD-189’s current force, but two are of particular utility: first, “[a]s is true of interpretation of statutes, the interpretation of an Executive Order begins with its text,” which “must be construed consistently with the Order’s ‘object and policy,’” and second—and perhaps more importantly—“when two statutes are capable of coexistence, it is the duty of the courts, absent a clearly expressed congressional intention to the contrary, to regard each as effective.”
Of course, the first rule of construction is to “[r]ead the statute,” but as explained above, merely reading the plain text of NSDD-189, EO 13556, NSPM-33, and contemporaneous, related legislation like the FY 2021 NDAA does not yield an obvious answer about whether or how the instruments work together. However, considering their text in the context of their “object and policy” is insightful. NSDD-189, NSPM-33, the FY 2019, FY 2020, and FY 2021 NDAA, and the CHIPS and Science Act all arose in the midst of grave national concerns about the threat posed by malign acquisition of U.S. R&D by foreign adversaries. Thus, while NSDD-189 is often overread with an eye entirely focused on the directive’s unrestricted fundamental research provision, as discussed above, NSDD-189 also explicitly recognized the “significant threat to our national security” posed by procurement of U.S. technology by foreign actors. Indeed, the text of NSDD-189 also expressly reflected that further refinement of the directive’s thrust should be expected, given its acknowledgment that “the government-university-industry partnership activities” on which it was focused was only then “emerging.” As such, the directive left open that “a more significant problem may well develop,” and fully conceded that additional restrictions on federally-funded fundamental research might be “provided in applicable U.S. Statutes.” As explained, following issuance of NSDD-189, such legislation has materialized, which has had the effect of placing additional restrictions on fundamental research, despite the effect those measures have had on restricting, to some degree, conduct and reporting of fundamental research.
In fact, the second canon introduced above, sometimes referred to as the “harmonization canon,” also leads to the conclusion that NSDD-189, EO 13556, NSPM-33, and the aforementioned, related legislative efforts must be read “to give effect to each if we can do so while preserving their sense and purpose.” U.S. caselaw is replete with examples of harmonization efforts of this type arising when courts were faced with arguably conflicting laws. Significantly, this is true not only in the context of statutes, but also, as here, with respect to executive orders. At work in this approach are two fundamental principles that resonate here: (1) courts, and by extension executive departments and agencies, “are not at liberty to pick and choose” which laws to enforce, and (2) “[i]n the absence of some affirmative showing of an intention to repeal, the only permissible justification for a repeal by implication is when the earlier and later statutes are irreconcilable.”
When taken together, the aforementioned authority leads to the conclusion that, to the extent NSDD-189, EO 13556, NSPM-33, and related legislative enactments are in any tension, they must be reconciled to give each the fullest effect possible.
D. NSDD-189’s Core Principle Remains Lex Terrae
At the heart of NSDD-189 rests the foundational tenet that, to the maximum extent possible, the conduct and reporting of federally-funded R&D should be unrestricted. That principle remains as potent today as it was when President Reagan issued the directive in 1985. What has changed is not the government’s commitment to that precept, but rather the nature of the threat that the U.S. R&D enterprise faces from malign foreign actors. NSDD-189 itself recognized that possibility and, in various ways described above, the law accounts for it. At bottom, as a matter of law, while subsequent presidential and congressional actions have surely cabined, to some degree, the “maximum extent” to which fundamental research can remain unrestricted, the fact remains that NSDD-189’s core feature remains legally viable.
Conclusion
The debate about what controls to put on the performance and distribution of government-funded research in the name of national security is enduring and unlikely to abate. Largely missing from that conversation, though, is the perspective of legal practitioners, who can bring clarity about what the law says. The value of doing so is perhaps no more evident than in connection with questions currently being considered at a national level of whether now is the time to revisit NSDD-189. That question is inherently a matter of future policy, but to answer it, one must understand what the law tells us about it today.
