U.S. nuclear laboratories in a nuclear-zero world

Abstract

In an era of nuclear disarmament, the role of the national labs must be redefined because continued modernization of nuclear activities is a step backward, rather than forward.

OVER A YEAR AGO PRESIDENT BARACK OBAMA DEclared in Prague that the United States was committed to a world without nuclear weapons, but that the path to nuclear zero would involve complex challenges. The impact of nuclear disarmament on the nuclear weapons laboratories is one such challenge. For decades, nuclear weapons have provided the rationale for the national weapons laboratories—Los Alamos National Laboratory, Livermore National Laboratory, and Sandia National Laboratories. In addition to their work to develop successive generations of nuclear weapons, the laboratories have also made major contributions to nonproliferation policy (e.g., through the development of verification technologies) and to the national science and technology base (e.g., through developments in computing). Under the ongoing moratorium on nuclear testing, they are tasked with monitoring the stockpile of nuclear weapons, as well as supervising the disassembly of nuclear weapons that have been removed from the national inventory. Presumably this technical work on verification and disassembly will be even more important during the transition to nuclear zero, even as the support for new nuclear weapons disappears.

The role of the laboratories has not been wholly benign, however. In the past, the labs have undermined support for arms controls whenever their leadership judged that those measures might lessen support for their institutions. For example, during the 1999 Senate debate on the Comprehensive Test Ban Treaty (CTBT), the lab directors testified that they could not guarantee the long-term safety and reliability of the U.S. nuclear stockpile under the treaty, although they had earlier said that, if fully funded for the Stockpile Stewardship Program (SSP), they could. 1 And just last January, they wrote letters to the ranking minority member of the House Armed Services Committee's subcommittee on strategic forces in which they asserted—contrary to a JASON report—that the SSP was not enough to guarantee the reliability of the nuclear weapons in their charge in future years. 2 Given the prestige that Washington traditionally accords the national laboratories, this position poses risks to the goal of nuclear zero.

A history of patronage. The origins of nuclear weapons lay in basic scientific discoveries in the 1930s made by academic scientists in a number of countries. The development in the United States of weapons based on those discoveries was, however, a massive industrial effort, involving thousands of workers in mines, power plants, separation facilities, university physics laboratories, testing facilities, and, of course, Los Alamos, where the first atomic bomb was designed and built. In a few short years, an entire complex of laboratories and production facilities was created, along with a knowledge base that went far beyond the basic physics of nuclear fission.

The U.S. nuclear weapons complex grew until the early 1990s, when the breakup of the Soviet Union called into question the whole raison d'être of the nuclear enterprise. Outlays for research and development for “atomic energy defense” declined in real terms by more than 25 percent through 1995 before recovering gradually to the level of the late 1980s; since 2006, the funds have fluctuated around a slightly lower level. 3

Today the three nuclear weapons laboratories—Los Alamos, Lawrence Livermore, and Sandia—have approximately 22,000 employees and a combined budget of $4.3 billion from the Energy Department. 4 Other funding comes from “Work for Others”—that is, other government agencies, as discussed below. Administratively, the complex is lodged in the National Nuclear Security Administration (NNSA), a “semiautonomous” agency established within the Energy Department in 1999 to insulate the laboratories from Energy's perceived excessive bureaucracy. Whether that ambition has been realized is, however, debatable.

Following the 1992 test moratorium and in anticipation of a permanent CTBT, the laboratories' role was redefined as one of stockpile stewardship. Stewardship is an elastic concept, but it typically includes maintaining the safety and reliability of a smaller stockpile of nuclear weapons into a distant future; dismantling excess warheads and disposing of nuclear materials; and in some versions retaining a “rebuild” capability. 5 In January 1996, Energy formally established the Stockpile Stewardship and Management Program (later the Stockpile Stewardship Program). Under the program, every year the Defense Department and NNSA are required to jointly certify that the nuclear arsenal is safe, secure, and reliable. The 2010 budget for this part of the SSP is $1.5 billion; an additional $1.6 billion is authorized for the science and engineering “campaigns” supporting the stewardship function. 6

The program relies on computer simulations and physical tests of non-nuclear components to support its judgments about the reliability of the weapons in the stockpile, and it maintains a capacity to refurbish and remanufacture components as needed. It is also tasked with maintaining a test-readiness program as a hedge in case nuclear testing is ever again deemed necessary. In addition, in a move that was widely interpreted as a quid pro quo for the support of the laboratories for the Comprehensive Test Ban Treaty, the SSP included increased funds for science at the laboratories and the construction of expensive new facilities, such as the National Ignition Facility at Lawrence Livermore and the Dual-Axis Radiographic Hydrodynamic Test facility.

… there are internal politics in the laboratories that fall along functional lines—weapons designers vs. production engineers, or experimentalists vs. theorists and computational experts—and also along the generational lines separating the older scientists from the younger ones, who have never participated in a nuclear test.

Nuclear weapons have often been touted as a cheap security guarantee (“more bang for the buck”), compared to the cost of force capabilities based solely on conventional forces, but the sums are still staggering. Estimates of the cumulative resources devoted by the United States to the development and production of nuclear weapons between 1940 and 2005 run as high as $7.5 trillion (in 2005 dollars), or around $115 billion a year; analyst Stephen Schwartz estimates that nuclear weapons accounted for 29 percent of total defense spending between 1940 and 1996. 7 Most of this spending was in the Defense budget and went to develop and produce delivery systems, command and control systems, and defensive systems, including air defense and iterations of ballistic missile defense systems. The budget for the nuclear components of the weapons, which falls under Energy, was only 7 percent of the total costs through 1996. 8 An update of these estimates for 2008 suggests a higher share for Energy in recent years: The total appropriations for nuclear forces and operational support were $29 billion, of which $22.5 billion was in the Defense budget and $6.6 billion in the Energy budget. The grand total for 2008, including missile defense and deferred environmental and health costs, was $52.4 billion. 9

The workplace. The weapons laboratories are a special kind of workplace. They are distinguished by their remote locations (no longer so remote for Lawrence Livermore and Sandia); the compartmentalization of the scientists in the name of secrecy; the moral dilemma present in work on weapons of mass destruction; and the cultivated sense, especially in the early years, of being part of an elite group working for national security. 10 The same conditions that foster a strong sense of a special community make it difficult for scientists to leave the laboratories for other jobs; in particular, the secrecy requirements mean that many of their accomplishments are not recognized in the open literature.

A number of ethnographic studies of the national laboratories have delineated laboratory culture at the micro level. 11 These accounts make it clear that there are internal politics in the laboratories that fall along functional lines—weapons designers vs. production engineers, or experimentalists vs. theorists and computational experts—and also along the generational lines separating the older scientists from the younger ones, who have never participated in a nuclear test. The shift from the nuclear testing regime to reliance on computer simulations as the basis for credible deterrence has been accompanied by a rise in status of the computer modelers relative to experimental scientists, something that many scientists and engineers have found difficult to accept. 12

The laboratories' ability to fulfill their mandated security role depends strongly on recruiting and retaining talented scientists and technical workers. Even before the recent interest in nuclear disarmament, the problems of managing a declining enterprise were evident. Downsizing at Los Alamos and Lawrence Livermore has largely been accomplished through attrition, leading to an increase in the average age of the scientific workforce, even as the collective experience of the Cold War scientists disappears. And more than creative scientists are needed: The tasks of dismantling surplus nuclear weapons during downsizing and remanufacturing pits as part of the lifetime extension program require a workforce of trained engineers and technical workers. An additional problem is that these jobs are, in general, open only to U.S. citizens, while the pool of science and engineering graduates in the United States is 40–60 percent foreign born. 13

The role of the labs in national policy debates. The internal politics of the laboratories, as sketched above, have not been closely linked to the “external” politics of formulating national policy with respect to nuclear arms control and disarmament. Laboratory scientists and engineers are naturally interested in changes to their budgets; after 20 years, however, they take SSP for granted and see Washington as remote from their workaday lives. The intra-laboratory struggles have been about competing claims for credibility within the nuclear weapons community and—in another area of contention—about resistance to the NNSA's perceived micromanagement of the labs. Lab employees have also frequently been critical of changes made after for-profit corporations assumed management of Los Alamos and Lawrence Livermore—changes that included shedding hundreds of jobs to offset the costs of the big increase in the annual management fees and the loss of tax-exempt status. 14

… the modernization questions are inseparable from the question of U.S. policy on nuclear weapons; they depend on the decisions about the size of the stockpile, whether new warheads should be developed to replace the current versions being maintained under the SSP, and the projected path to nuclear zero.

As with other institutions that rely on federal government funding, however, the top management of the laboratories engages directly with the Washington policy community. Policy positions are developed in consultation with Energy and Defense, but the laboratories also maintain their own representatives in the capital. Laboratory leaders testify at congressional hearings, meet with agency and congressional staff, maintain contacts with advisory groups such as the Defense Science Board and the JASONs, and sponsor special events, like the fall 2009 Workshop on Nuclear Forces and Nonproliferation organized by Los Alamos at the Woodrow Wilson International Center for Scholars in Washington, D.C. In these various fora they are able to make the case for the role played by the laboratories in guaranteeing the nuclear deterrent and, implicitly and explicitly, to argue for continued reliance on nuclear deterrence as the centerpiece of U.S. security policy. They deploy their prestige and expertise to promote particular policies, such as on the reliable replacement warhead, and have a record of success in maintaining support for the labs' primary mission. 15

A particularly clear example of this behavior occurred in the mid-1990s during the Clinton administration, when lab budgets were being cut and there was much talk of converting them to commercial research. The response from the labs was, on the one hand, to fight for the nuclear weapons mission and, on the other, to secure new support under the rubric of stockpile stewardship. Thus, Livermore director John Nuckolls testified to Congress that nuclear weapon design and testing were the “very building blocks of civilization,” at the same time that an article in the Bulletin of the Atomic Scientists bore the title “Shopping Spree Softens Test-Ban Sorrows.” 16

Transformation plans. Since the end of the Cold War, numerous commissions and task forces have considered the future of the U.S. nuclear weapons complex and issued their recommendations. 17 The issues at stake include the appropriate size for the complex, whether to maintain redundant weapons design capabilities in Los Alamos and Lawrence Livermore, the replacement of aging infrastructure, and security and management issues. At a more fundamental level, however, the modernization questions are inseparable from the question of U.S. policy on nuclear weapons; they depend on the decisions about the size of the stockpile, whether new warheads should be developed to replace the current versions being maintained under the SSP, and the projected path to nuclear zero.

In December 2007, the NNSA administrator, Thomas D'Agostino, announced a plan for “Complex Transformation,” one that would reduce the “footprint” of the nuclear weapons establishment by shedding redundant activities, while simultaneously modernizing other facilities. This is the so-called “modernization in place” option; it avoids the near-term disruption of moving all weapons activities to a central site, at the cost of maintaining all the existing sites for the foreseeable future. Under this plan, employment at the three weapons laboratories is projected to shrink by an additional 20–30 percent. The labs will maintain their core functions, and competition between Los Alamos and Lawrence Livermore will continue. 18 In selecting this plan, the NNSA explicitly rejected a “curatorship” alternative that would have implied an end to U.S. capabilities to design and develop replacement nuclear components and weapons. 19

The Obama administration has signed on to the Complex Transformation plan to the extent that it has increased the NNSA budget for infrastructure in the fiscal year 2011 budget by $2.3 billion. Whether that decision is consistent with the path to nuclear zero depends, of course, on how long the transition is expected to take. The current facilities at the labs pose safety and security risks, but much of the new money is directed toward expanding the production capacity for plutonium pits.

Thus, the U.S. weapons complex in general, and the three national laboratories in particular, are on a path that will maintain an active program of research on nuclear weapons and a downsized, but still considerable, capability to produce new nuclear weapons. The arguments for such a capability are rooted in a doctrine of continued dependence on nuclear weapons as a core element in national security; if that premise is granted, the decision to modernize old, inefficient facilities seems reasonable. The NNSA has acknowledged that a reduction in the size of the nuclear stockpile below 1,000 warheads “could result in a need to reassess the transformation options for the Complex,” but it has chosen a plan that assumes no reductions beyond the figure of 1,700–2,200 deployed warheads which was agreed in the Moscow Treaty of 2002. 20 The New START, signed by Presidents Obama and Dmitry Medvedev in Prague on April 8, 2010, sets the limit for deployed warheads at 1,550 for each side, a figure the United States is already approaching. 21 In effect, the Complex Transformation plan is already out of date. By contrast, a commitment to nuclear zero would require a different set of decisions, both in the near term and, especially, in the longer term. The laboratories might be retained for reasons discussed in the next section, but the planned investment in new weapons production facilities would be dropped.

Maintaining large and modernized weapons laboratories, however, carries a proliferation risk. For example, dispersed sites for special nuclear materials and expertise arguably provide more opportunities for illegal diversion to other countries or non-state actors.

Nuclear zero and the technology base. The weapons laboratories occupy a special position in the U.S. science and technology base: They are the largest recipients of federal funds for research and development, and they perform research in a number of scientific areas in addition to their primary focus on nuclear weapons research. Through its impact on the national laboratories, a move to nuclear zero would affect the national technology base; how much is open to discussion. At least two perspectives on this issue deserve attention:

Nuclear weapons technical expertise will be needed during the transition to zero. Such expertise is necessary for maintaining the functionality of a diminishing stockpile of weapons, dealing with the dismantlement and disposal of the weapons taken out of service, and providing a robust verification regime. Maintaining large and modernized weapons laboratories, however, carries a proliferation risk. For example, dispersed sites for special nuclear materials and expertise arguably provide more opportunities for illegal diversion to other countries or non-state actors.

The physical and intellectual capital accumulated in the national weapons laboratories, which constitutes a national treasure, should be preserved. As noted above, the laboratories have a long history of supporting fundamental scientific research, and the facilities at the laboratories for computing, for example, are perhaps not available anywhere else.

The laboratories' role in the transition to zero. The transition to nuclear zero would require continued technical capabilities similar to those currently provided by the weapons laboratories. The obvious issue is how to guarantee that the necessary competence will remain available, while preparing it for a world in which it will become irrelevant. Some have argued that the tension between the two requirements poses an impossible dilemma; at the least, it is a formidable challenge. 22 Organizations have, however, been known to change with changing missions. The classic example is the March of Dimes: Following the introduction of an effective polio vaccine, the organization transformed itself to one that fights birth defects and premature births, an open-ended goal that virtually guarantees its indefinite survival.

In the years following the end of the Cold War, when funding for the weapons laboratories decreased and their mission shrank, they faced the same concerns for their future which we hear today, including the threat of declining budgets, loss of scientific personnel to retirement and lay-offs, and lowered morale of the remaining workforce. 23 In the 1990s, the labs were “saved” by the Stockpile Stewardship Program. The science component of the program and the promise of new facilities to support it were an important element in gaining approval from the labs for the CTBT because it served as a warrant that the labs would remain interesting places to do science. 24

Significantly, the charge for the SSP included a requirement to “maintain the science and engineering institutions needed to support the nation's nuclear deterrent, now and in the future.” 25 Note the unquestioned presumption of a permanent nuclear capability, a presumption that will not be an appropriate basis for policy making during the transition to zero. While some have argued in favor of maintaining a “virtual” nuclear weapons capability as a hedge, such a policy would undercut the possibility of gaining universal adherence to nuclear zero because other states would claim the same privilege.

Stockpile stewardship provided political cover for a substantial reduction in the number of U.S. nuclear warheads by offering reassurance that the remaining weapons in the stockpile were fully functional. A further decrease in nuclear weapons, however, will not provide a similar boost for stewardship activities. As numbers decline to a few hundred and then a few tens of warheads, the overall stewardship responsibilities will presumably shrink as well, albeit not uniformly. Verification technology will continue to be in demand for the foreseeable future—indeed, its importance will increase during the potentially unstable transition to zero—and the need will persist after zero, however “zero” is defined. A robust nuclear forensic capability would also be useful as a deterrent to clandestine developments.

It is not clear, however, how much is needed in the way of new research and development to provide these interim and long-term capabilities. Strengthened diplomatic and intelligence efforts and new confidence-building measures might well be more effective in managing the transition than investment in new technology. To the extent that current technology is deemed sufficient for these tasks, the argument for continued investment in the research programs and laboratory infrastructure is weakened. 26 If further R&D at the weapons laboratories is deemed essential, then the political problems created by simultaneously promoting nuclear developments while preaching disarmament cannot be avoided.

Ironically, the effort that the national laboratories have made to counter the disappearance of tacit knowledge has served to maintain continuity of knowledge about how to build nuclear weapons.

A related problem also requires attention. So long as the laboratories are a repository of nuclear weapons knowledge, they represent a proliferation risk—the risk that the spread of technical knowledge about how to make nuclear weapons will lead to additional states acquiring them. Strenuous efforts to keep the secret of nuclear weapons during the Manhattan Project were nullified by the work of a few Soviet spies, and subsequent proliferators have all received help, either directly or through espionage, from existing nuclear weapon states. 27 The most recent example is the A. Q. Khan network, which transmitted technology for uranium enrichment to Iran, Libya, and North Korea before it was shut down in 2003. 28

With the publication of the Smyth Report in 1945, which made public the general principles of atomic weapons, it could truly be said that there was no longer an “atomic secret.” Indeed, knowledgeable scientists understood what had been involved in constructing the bomb as soon as they heard of the attack on Hiroshima. How then to understand the high level of security classification maintained on all nuclear weapons knowledge, and the difficulty that would-be proliferators have had in acquiring the ability to build nuclear weapons? The answer to this apparent paradox lies in the considerable success that states have had in controlling access to fissile materials, along with the importance of specific technical knowledge, especially tacit forms of that knowledge relating to any effort to build a workable device. It is one thing to understand the principles of atomic fission and another to master the technically difficult task of constructing a bomb. Current proliferation threats from North Korea and Iran bear out the basic argument that it is difficult to produce a nuclear weapon, even when blueprints or contraband materials and components are available, but they also show that a determined regime can persevere and achieve a nuclear weapons capability.

Ironically, the effort that the national laboratories have made to counter the disappearance of tacit knowledge has served to maintain continuity of knowledge about how to build nuclear weapons. By interviewing retiring scientists to create a more complete record of the historical tests and providing formal training programs to new hires, the archiving programs have implicitly lowered the barriers for proliferators. 29

Whether to maintain a potential nuclear capability is a question faced by any technically advanced country, not just “rogue” states. 30 Countries such as Japan and Sweden, with their histories of early nuclear weapons projects that were terminated before producing actual bombs, might be considered to be “virtual” nuclear weapon states, in the sense that they are usually considered to possess a latent nuclear weapons capability.

These examples—Canada is another case—make it clear that judging potential nuclear weapons capability should rely on political analysis as much or more than on the possession of an advanced science and technology base; technological know-how is not a threat in countries committed to nuclear nonproliferation. Political situations can change, but the majority of the non-nuclear states seem to have reached the point at which nuclear weapons are no longer an imagined option; in this they are in sharp contrast to the nuclear weapon states, whose political elites have difficulty imagining a security scenario without nuclear weapons.

Preservation of the national weapons laboratories for other national goals. The nuclear weapons laboratories do more than oversee the nuclear stockpile. They also engage in efforts to share their research with the civilian sector, a mission that could be enhanced to justify maintaining the laboratories indefinitely, albeit with a revised mandate. These technology transfer programs are based on the proposition that military technologies can be “spun off” to benefit the civilian technology base—as, for example, computer technology did during the 1960s—and they have a particular appeal during periods of declining funding.

Thus, today, the U.S. laboratories and their supporters emphasize the broad capabilities of the laboratories and argue that they deserve to be maintained with new rules that would allow them to interact more easily with other patrons besides Energy. This is the position taken in a 2009 report by the Stimson Center, Leveraging Science for Security. 31 A similar argument was put forward in 1993 by Nobel laureate Hans Bethe, who argued that the nation should “redirect the nation's huge investment in the nuclear weapons laboratories towards research aimed at serious problems that we are sure to face as a society and in the global market place, but that universities and private industry do not have the resources to address.” 32

The laboratories, however, have found it difficult to attract resources for projects that are not connected to their security missions. For example, the program for Cooperative Research and Development Agreements with industry, which in the 1990s was promoted as a way to sustain the laboratories, failed to live up to the early hype. 33 A major problem was the loss of congressional support after the Republicans took control of the House in 1995 and struck down any program that could be tarred with the label of industrial policy; another problem has been the complex set of rules and rights that govern laboratory-industry interactions. 34 The “crown jewels” argument for continued support for the laboratories cannot be sustained if the jewels remain inside their confined walls.

Other critics point to the labs' “cowboy culture” as the reason why Energy has had to add more and more layers of oversight. The underlying tension between the two sides has been exacerbated by a series of safety incidents and security scandals involving lost or misplaced classified data, misuse of government procurement cards, and sloppy accounting practices.

The latest emphasis has been on reinventing the laboratories as general national security laboratories with expertise in all kinds of weapons research. The Stimson Center's aforementioned report makes the case for an expanded mission for the labs, supported through cooperative agreements with other agencies in the security sector—Defense, Homeland Security, and the CIA—and Sandia's success in reducing its dependence on NNSA funding to below 50 percent shows that there is potential to increase sharply the role of “Work for Others” at the laboratories. It remains to be seen, however, whether the “others” will be willing to pay for the long-term costs of renewing laboratory infrastructure. Work for Others typically takes the form of short-term contracts; this short timeframe leads to a fragmentation of effort and difficult working conditions for scientists and engineers, who are forced to allocate their time among numerous contracts. A recently declassified JASON report charges that the costs associated with Work for Others are higher than elsewhere because of the legacy of inefficient facilities from the Cold War. 35

The case for long-term support of the laboratories in a period of declining importance for the nuclear mission has, moreover, a serious weakness: the management problems that have plagued the laboratories for years. The insular laboratory workplace culture described above clashes with the persistent tendency for Energy to manage (or to micromanage, in the view of its critics) the laboratories from Washington. 36 Other critics point to the labs' “cowboy culture” as the reason why Energy has had to add more and more layers of oversight. 37 The underlying tension between the two sides has been exacerbated by a series of safety incidents and security scandals involving lost or misplaced classified data, misuse of government procurement cards, and sloppy accounting practices.

The most notorious case was that of Wen Ho Lee, a physicist at Los Alamos who was charged with transferring nuclear secrets to China; Lee spent nine months in solitary confinement before he was released when federal investigators were unable to prove his involvement in espionage. Other scandals at Los Alamos included the loss of two computer disks holding classified data in 2004 (eventually it was determined that the disks never existed, an outcome that drew further attention to the sloppy practices at the laboratory), and a string of other security breaches. 38 These incidents precipitated the firing of one laboratory director, a stand-down in 2004 of several months at Los Alamos, and the opening of management contracts at Los Alamos and Lawrence Livermore to corporate teams. Given this record, any argument to preserve the laboratories for their value to the national science and technology base would need first to establish that the resources could not be better used elsewhere to accomplish the same mission.

Conclusions. A serious move to nuclear zero will have major implications for the U.S. weapons laboratories. To the extent that the laboratories depend on their role in maintaining nuclear deterrence through stockpile stewardship programs or work on modernization programs (such as the proposed reliable replacement warhead), survival in their present form is unlikely in an era of nuclear disarmament. In the near term, the laboratories will be needed to guarantee the reliability of the stockpile, to oversee the dismantlement of nuclear warheads, and to support verification technologies. The first two functions will, however, decline and eventually disappear during the drawdown to zero. The verification function will last longer; however, in a nuclear-zero world—when the problem will be detection of possible cheating—an international verification effort lodged in the International Atomic Energy Agency or a successor agency would be a preferred solution because it would be more trusted than a national capability.

In order to best define the role of the labs during the next 20 years, a roadmap guiding the transition to zero is needed. How long is the long run? Should the laboratories recruit and train another generation of scientists, or will the current cohort be sufficient? Is new knowledge needed to perform verification, or is current technology adequate? Related to these questions are the challenges of maintaining morale in organizations that are losing their main mission and of sustaining political support for the cost of running the laboratories during the transition period.

A serious commitment to nuclear zero at the labs would require a long-term plan for phasing out nuclear weapons–related activities, with milestones to match political and technological developments, such as the entry into force of the CTBT, or the dismantlement of the last nuclear warhead. These milestones should be viewed as irreversible, and thus there would be no need for further modernization of the weapons complex. To argue that interim nuclear activities require continued modernization is to postpone the time when nuclear disarmament becomes an inevitable outcome, not just an aspiration.