Retooling Efforts to Stop the Proliferation of WMD Expertise

Sharon K. Weiner

As part of its nonproliferation agenda, the United States has given an increasing role to programs focused on the scientists and engineers who make weapons of mass destruction (WMD), with a particular emphasis on reducing opportunities and incentives for scientists to disseminate their skills and knowledge. Terrorist access to nuclear weapons knowledge became the focal point of U.S. nonproliferation policy after the September 11 attacks.

The United States also has worried about the fate of WMD experts in war-torn states such as Iraq and Libya, as well as in Pakistan, where there are concerns about anti-U.S. sentiment and the activities of retired weapons scientists. More broadly, working with scientists has been proposed as a way to create dialogue with North Korea, eventually on nuclear weapons issues. Some U.S. policymakers argue that collaborative research with scientists can help create transparency and raise confidence that biotechnology and nuclear energy programs are not also engaged in weapons work.

Such scientist engagement efforts are not new. U.S. and Soviet scientists worked together on arms control verification measures during the Cold War. However, the focus on WMD expertise as a central component of U.S. nonproliferation policy dates from the early 1990s. The disintegration of the Soviet Union and the subsequent dismal economic conditions in its successor states led to a variety of proliferation fears. In response, Russia and the United States worked together on several efforts that collectively became known as Cooperative Threat Reduction (CTR), or Nunn-Lugar, after Senators Sam Nunn (D-Ga.) and Richard Lugar (R-Ind.), who championed this agenda. Parts of the CTR work focused on helping Russia dismantle and destroy unconventional weapons; other programs sought to improve the security of WMD-relevant materials.

Yet another set of efforts focused on the scientists themselves. Because of concerns that poverty-stricken WMD experts would be tempted to sell their skills to rogue states, especially Iran and North Korea, this part of the CTR agenda focused on creating jobs for former Soviet WMD scientists and engineers. Programs with this mandate were created in the U.S. departments of Defense, Energy, and State.

The job creation programs started in response to the Soviet collapse have provided the basis for current knowledge-nonproliferation efforts. Their activities, procedures, and assumptions will continue to influence this policy agenda as it expands beyond the former Soviet Union. Therefore, it makes sense to understand the successes and failures of these programs and the problems they encountered as these efforts were being put into place. Certainly, past antagonisms from the Cold War created problems for the U.S.-Russian cooperative nonproliferation agenda, as did diverging policy priorities and Russian attempts to reassert a more powerful role in international relations. Yet, the history of this knowledge-nonproliferation agenda also points to another source of problems: bureaucratic politics.

U.S.-Russian nonproliferation efforts were significantly affected by relations between different programs and the assumptions and routines the programs developed for implementing their efforts. These bureaucratic issues often created or reinforced problems that could have been mitigated or even overcome. Because current policy efforts have evolved from these earlier U.S.-Russian programs, they have inherited some of the same bureaucratic problems and tendencies. Therefore, successfully curbing the proliferation of WMD expertise beyond the former Soviet Union depends on heeding the lessons of this past experience.

Shifting Program Goals

The Soviet Union had a massive WMD complex that included multiple redundant facilities. After the country’s collapse, parts of this complex were inherited by each of the successor states. This dispersal raised fears about the proliferation of WMD expertise, as did the weakened economy and internal security in many of the newly independent states, but especially Russia. In response, the United States created five programs, each of which shared the goal of retraining WMD experts and helping to create permanent nonweapons work for them. The U.S. Defense Department was home to two such programs: the Defense Enterprise Fund and an unnamed set of efforts to convert former Soviet weapons facilities. The State Department managed U.S. participation in two multinational Science Centers, one in Moscow and one in Kiev. The Energy Department also had two programs: Initiatives for Proliferation Prevention (IPP), created in 1994, and the Nuclear Cities Initiative (NCI), which dates from 1998.

Although some of these programs had additional goals, each had as a central mission the creation of nonweapons jobs as a means of combating the potential proliferation of WMD expertise. Over time, however, each program came to adopt different goals. Some focused on goals that were easier to measure; others shifted to outcomes that were more likely to demonstrate success.

For example, the Science Centers and the IPP shifted back and forth between “engagement,” which seeks to use cooperative research projects to provide short-term income to scientists, and the more difficult task of “redirection,” which aims to create jobs for WMD workers outside the weapons complex. The IPP also came to argue that engagement was useful because it helped the United States better understand the state of the Russian nuclear complex and activities there. Engagement made sense as an initial strategy because establishing the business connections and training necessary for job creation took time and former Soviet WMD experts already were in desperate economic straits.

Although both programs did develop job creation strategies, they persistently focused on measuring progress toward engagement. The IPP, for example, measured the number of WMD experts involved in any type of project, counted the dollars invested by business partners, and recorded the number of licenses and patents that resulted from its efforts. The Science Centers also counted the number of scientists involved in research projects, but neither program kept an accurate record of the number of jobs created nor took steps to discover how many of those jobs remained in existence after several years.

Focusing on engagement rather than job creation solved certain problems faced by the Science Centers and the IPP. First, engagement was simply easier. Converting facilities and scientists from defense to civilian functions is inherently difficult, much more so in a country with little experience with market forces. Second, measuring the number of scientists engaged allowed these programs to demonstrate success to a U.S. Congress that at times was quite skeptical of this nonproliferation agenda. These programs could report that they had “engaged” tens of thousands of former Soviet experts. The number of jobs created, however, never exceeded a few thousand, if that.

Both the NCI and the Defense Department’s conversion efforts did count the number of jobs they created, but their goals shifted in other ways. The NCI was faced with limited funding from Congress and increasing constraints imposed by internal security forces in Russia. The U.S. national laboratories, which provided the scientific expertise and collaborators for NCI and IPP projects, also were disgruntled because they had less clout than NCI headquarters in decisions on project choices. In response to the increasingly hostile environment, NCI managers chose to focus on a few key projects, rather than broader job creation efforts. When those projects failed, the NCI strategy was called into question, and the program shut down.

Like the NCI, Defense Department conversion programs faced externally imposed problems. The most serious was that most of the Russian institutes interested in conversion made conventional weapons. Rather than search for more appropriate institutes, the Defense Department settled for converting facilities that had little or no relevance to unconventional weapons. This, in turn, made Congress skeptical of the national security benefits of continuing these programs; one was ended in the mid-1990s, and the other saw significant funding reductions before it also was terminated.

Although each of these programs shifted goals for different reasons, the consequences were similar. Some Russian political leaders were committed to the goal of defense conversion, but others were skeptical of U.S. motives. The lack of new jobs reinforced domestic political concerns, especially among the internal security forces, that U.S. programs really were more interested in spying than in nonproliferation. As a result, Russia became less and less cooperative. This, in turn, seemed to confirm the suspicions of program skeptics in the United States who worried that U.S. funding was allowing Russia to spend its own resources on rebuilding its military rather than on nonproliferation.

The shift from redirection to engagement also had another important consequence: U.S. programs came to include WMD experts who were not the primary proliferation concern. As part of redirection, the Science Centers and the IPP were supposed to focus on older WMD experts who had been part of the Soviet complex. These experts were seen as a proliferation risk because they experienced a sharp decline in salary and living standards when the Soviet Union collapsed, Russia was downsizing its complex, and these scientists lacked the skills to get market-oriented jobs in the new Russian economy. In contrast, the goal of engagement meant that U.S. programs could bypass these older scientists. Instead, they engaged younger experts who often had skills that were more relevant to the new economy and were interested in temporary research collaborations, but who had every expectation of remaining in the Russian WMD complex. In other words, engagement meant U.S. programs were more inclined to fund experts who would remain gainfully employed in Russia’s nuclear complex and thus were not a proliferation risk.

Bureaucracies often stray from the policy objectives they have been given and come to adopt goals that are easier to measure or more likely to demonstrate their success. Such changes can result in new or better ways of achieving desired policy outcomes or lead to new policy priorities. This was not the outcome for the programs described above. Their shifting program goals exacerbated underlying tensions between Russia and the United States and within the United States and caused resources to be spent on scientists who were not a priority. As a result, although fighting the proliferation of WMD expertise through job creation remained a top U.S. policy objective, programs found it increasingly difficult to achieve this goal.

Standard Operating Procedures

U.S. nonproliferation efforts traditionally have focused on international legal regimes, such as the one built around the nuclear Nonproliferation Treaty. Fighting proliferation with job creation was a novel approach that arose from a unique event: the sudden collapse of the Soviet state. As a policy goal, job creation also was at odds with the normal tasks given to the Defense, Energy, and State departments, none of which had significant experience with converting military facilities, creating commercial jobs, or working with Russian WMD experts. As a result, they did what all bureaucracies do: they implemented the programs they were given using their own standard operating procedures. Unfortunately, some of these routines actually made nonproliferation goals more difficult to achieve.

The Defense Department, for example, needed U.S. commercial partners to help convert Russian defense facilities. Businesses were invited to bid for such opportunities, and the Pentagon chose the winner by using standard rules that give contracts to the lowest bidder when there are no established suppliers with which the department has a relationship. The lowest bidder, however, did not necessarily have any experience in defense conversion or working in Russia. As a result, U.S. commercial partners often were ill-prepared for the difficulties they would face, and many conversion projects failed. Consequently, the managers and employees of some Russian facilities became less interested. More problems arose because Russia had been promised high-tech conversion projects that built on the skills found in employees at key defense facilities. Instead, many of the commercial partners selected made low-tech consumer goods. Some Russian facilities were disappointed; others felt they had been misled. As one scientist explained, it was as if the United States had asked nuclear weapons scientists from Los Alamos National Laboratory to make baby diapers.[1]

Standard operating procedures had a different effect at the State Department, which was in charge of U.S. participation in the Science Centers. The department’s normal task of diplomatic relations led to an emphasis on maintaining harmonious relations among countries that contributed to the centers. In other words, the department focused on creating a project selection process that would ensure consensus among Science Center funders rather than developing rigorous procedures for identifying proliferation-prone scientists or conversion projects that were likely to lead to commercial success. As a result, few projects led to job creation.

Similarly, when the Science Centers sought more funding for job creation efforts in the late 1990s, the State Department led the way in creating a process that made it easier for other U.S. government agencies to fund collaborative research with former Soviet WMD experts through the centers. The centers were very successful in this respect. At the Moscow science center, for example, more than 90 percent of funding for its new Partner Project job creation effort came from U.S. government agencies,[2] but the vast majority of these agencies were interested in short-term research contracts, not job creation for the WMD experts they engaged.

At the Energy Department, the standard operating procedure that proved most problematic was the decentralized power structure.[3] Congress tried to help remedy problems with insufficient coordination and information sharing by creating the National Nuclear Security Administration (NNSA), a semi-autonomous agency within the Energy Department that is responsible for military-related nuclear functions. Yet even under the NNSA, which was established in 2000, the U.S. national laboratories retained significant decision-making authority. To help Russian WMD experts design, implement, and carry out job creation projects, both the IPP and NCI used experts from the U.S. national labs. Thus, both programs had to contend with these problems.

The independent power of the labs had some advantages. Because of their mutual scientific backgrounds and interests, U.S. and Russian nuclear weapons experts proved eager to work with each other. In turn, they pressured their respective governments to remove some of the barriers that initially made U.S.-Russian cooperation difficult or slow.

Unfortunately for the IPP, however, the decentralized power structure in the NNSA also created problems. The labs bickered with each other over projects and funding. The seven smaller labs often accused the “three gorillas” of Los Alamos, Lawrence Livermore National Laboratory, and Sandia National Laboratories of skewing the IPP toward their own preference for big projects that were unlikely to lead to job creation. The big labs, in turn, argued that the “seven dwarfs” lacked the WMD-relevant skills necessary to engage the Russians and were involved in the IPP only to boost their budgets.[4]

Further, the allocation of project funding often was determined more by the preferences of the individual labs than by the overall strategy developed by IPP headquarters. The labs also refused to heed the pleas from headquarters for more precise record keeping. As a result, the program could not definitively prove its impact. Moreover, periodic external reviews criticized the IPP for spending more money on U.S. lab experts than on Russian ones and for giving money to Russian institutes that were also involved in work with Iran.[5] Some of these criticisms were justified, but others were exaggerated in order to make broader political points.

The NNSA’s decentralized structure had a different impact on the NCI. The decision-making authority for that initiative was concentrated in a central headquarters, giving the labs much less say over program activities. Further, in contrast to the IPP, the NCI’s early projects involved community development and the creation of low-tech jobs for anyone living in Russia’s nuclear cities. The labs did not find much of a role for their expertise in such projects, often arguing that NCI efforts were misguided or counterproductive. Deprived of strong support from the U.S. labs and at times actively criticized by them, the NCI found it difficult to marshal the political and financial resources it needed.

It is only natural for bureaucracies to implement new programs according to their standard operating procedures. As the previous examples show, however, U.S. nonproliferation policy toward Russia suffered because some of these routines made it more difficult to create jobs. Other routines provided ammunition for domestic political forces in the United States and Russia that questioned the value of the threat reduction agenda.


Despite their common goal, there was very little coordination among the five programs that the United States created during the 1990s to convert former Soviet WMD experts to nonweapons work. They did not share lessons learned from past success and failure, nor did they engage in collective public outreach to build support for their policy agenda. With the exception of some program personnel who focused on working with former biological weapons experts and facilities, these programs never developed a joint list of Soviet WMD experts or institutes to help them target or prioritize their efforts. The U.S. programs also did not develop sufficient means to share information about project funding; thus there was no guarantee they did not fund the same scientists for the same work or fund the same project twice.

Coordination also was a problem within agencies. The Energy Department’s IPP and NCI, both of which involved U.S. national labs, did little to coordinate their activities. This remained true even after the two programs were merged in 2002 to form Russian Transition Initiatives. Within the NNSA, some labs also persistently refused to share information with the others. As for the Moscow and Kiev science centers, there was little coordination or communication between them, and at times, they followed different policies for job creation. The Defense Department was an exception in this case; its conversion effort did share some project ideas and personnel with the Defense Enterprise Fund.

In addition to causing waste and inefficiency, the lack of coordination created problems with Russia. U.S. programs tended not to coordinate their visits to Russia. As would be expected, Russian officials were sensitive about giving U.S. government personnel access to the country’s nuclear weapons complex. When IPP, NCI, and Science Center project managers each requested access to the same facilities and experts, this confirmed for some in Russia that the United States was more interested in “nuclear tourism” and spying than in nonproliferation. Those fears were reinforced when U.S. programs repeatedly failed to create many jobs.

Fixing Bureaucratic Problems

Each of the problems described above is an example of standard bureaucratic behavior. Unfortunately, these bureaucratic issues created significant problems for U.S. nonproliferation efforts. Fewer jobs were created for former Soviet WMD workers because the programs in charge of this policy agenda came to pursue slightly different goals, used implementation routines that created inefficiencies or other problems, and refused to coordinate with each other. These bureaucratic tendencies, in turn, exacerbated existing fears in Russia and domestic political debates in the United States. As a result, programs faced not only additional hurdles to job creation, but also uncertainties about their funding and future. They became less inclined to take risks, be innovative, or push for the re-examination of job creation as a nonproliferation strategy.

Such problems need not be a fact of life. Some can be anticipated and mitigated. In the case of U.S. knowledge-nonproliferation efforts, three relatively simple fixes would have helped, either by increasing program efficiency or by limiting the degree to which bureaucratic problems were allowed to reinforce political ones.

The first such fix is to have external stakeholders periodically review the assumptions, goals, processes, and outcomes of nonproliferation programs. Doing so can help identify places where program activities have strayed from policy goals for bureaucratic and other reasons.

Although Congress at times provides such oversight, in general it lacks the expertise and electoral incentives to engage in program review on a regular basis. Moreover, congressional oversight can be problematic when there are partisan or other domestic differences over a policy agenda.

Individual members of Congress, for example, can request investigations by the U.S. Government Accountability Office (GAO) and use the scope of inquiry and timing to try to influence the outcome. The NCI was the subject of two GAO investigations during its first few years, before its projects could reasonably have been expected to produce significant results. Similarly, in its 2008 review of the IPP, the GAO criticized the program for overstating its accomplishments, keeping records sloppily, and funding institutes that also were engaged in work on Iran’s Bushehr reactor. Critics seized on the latter point and used it, inaccurately, to argue that the United States was funding Iran’s weapons program and to bring up old concerns that threat reduction was really a subsidy to the Russian defense complex and being used for weapons that would eventually be aimed at the United States.

For these reasons, external review is more likely to be helpful if it is performed by external stakeholders, such as nongovernmental organizations (NGOs). NGOs either have or have access to the expertise to undertake even technical reviews. Because they are outside the bureaucracy, they are more likely to be able to see the influence of bureaucratic preferences on program activities and thus help analyze whether these help or hinder the pursuit of the original policy goals. Moreover, NGOs can be deterred from skewing their reviews to support their own partisan positions because Congress also can ask for competing reviews from other NGOs or threaten to leave partisan analysts out of such activities in the future.

External reviews could be further improved if the programs published and periodically updated not only their results, but also the metrics and decision-making criteria used in program implementation. Among the U.S. efforts aimed at former WMD experts, the Science Centers had the best track record of making their results publicly available. Eventually the IPP followed suit, but it reported and measured its metrics inconsistently and in some cases inaccurately. None of the programs published its decision-making criteria for external review. Sometimes this information was not even shared with the people in charge of engaging WMD scientists or carrying out job creation projects.

Publishing program metrics and rules would help clarify the assumptions behind program choices and make external review easier. Such documentation also would increase uniformity of the effort within programs. The U.S. Army, for example, publishes field manuals and makes many easily available to the public. In addition to providing detailed information for service members to use in implementation, the manuals give external stakeholders a basis for analyzing how the Army deals with uncertainties and how it can be expected to respond to external constraints. Moreover, the manuals help provide a basis for judging whether the policy the Army implements is overly influenced by the Army’s own bureaucratic interests.

Another area for improvement is program coordination. Indeed, one of the most persistent suggestions for improving U.S.-Russian threat reduction activities was that a cabinet-level “czar” be appointed to oversee this policy agenda. After almost two decades, the first such official was appointed by the Obama administration although that official is responsible for a much broader set of policies.

A high-profile presidential appointee could have helped troubleshoot problems with Russia and raise political awareness in the United States, but many significant coordination problems arose at a much lower level. It probably would have escaped the czar’s notice, for example, that the U.S. labs did not share lessons learned in implementing IPP projects or that the Moscow and Kiev science centers had different criteria for the number of WMD experts required in a project. At such levels, better coordination is a function of the power of program managers, who need the authority to create interagency connections and to overcome barriers to cooperation and information sharing within and among programs.

Beyond the Former Soviet Union

The scientist redirection efforts that formed part of the threat reduction program with the former Soviet Union are winding down. As some programs seek to close down, however, others are looking to expand to new areas. For example, five years ago the IPP changed its name to Global Initiatives for Proliferation Prevention (GIPP) and now has programs in Iraq, Libya, and other areas. Older State and Defense department efforts that focused on biological weapons expertise have been rejuvenated. Engagement of scientists from North Korea is seen by many as a step to creating a broader dialogue with Pyongyang on nuclear issues.

As a means of discouraging proliferation, these new efforts are embracing engagement as a way to encourage transparency or establish the basis for a dialogue that eventually will include more-sensitive national security issues. Some policy experts even have suggested funding long-term research collaborations as an inducement to scientists to refrain from working on unconventional weapons.

Whatever strategy is chosen for future efforts to prevent the spread of WMD expertise, it is foolish to allow programs to perpetuate bureaucratic tendencies that created significant hurdles for past nonproliferation efforts. Providing program managers with more authority, requiring programs to report not only outcomes but also program metrics and decision-making rules, and making the managers subject to periodic review by external stakeholders can help ensure that future nonproliferation efforts are less likely to be hindered by bureaucratic politics.

Working with foreign scientists has the potential to help the United States discourage proliferation in a variety of countries besides the former Soviet Union. Moreover, cooperation between scientists can be productive even when higher-level political negotiations are stalled. To realize the benefits of such efforts more fully, however, it is necessary to improve on past experience. Even if bureaucratic politics cannot be eliminated, a few simple steps would help reduce its more damaging consequences.  ACT

Sharon K. Weiner is an associate professor in the School of International Service at American University. She holds a Ph.D. in political science from the Massachusetts Institute of Technology and is author of Our Own Worst Enemy? Institutional Interests and the Proliferation of Nuclear Weapons Expertise (2011), from which this article is adapted.



1. Adi Ignatius, “U.S. Stirs Russian Resentment With Plans for Defense Conversion,” The Wall Street Journal, September 19, 1994.

2. Sharon K. Weiner, Our Own Worst Enemy? Institutional Interests and the Proliferation of Nuclear Weapons Expertise (Cambridge, MA: MIT Press, 2011), p. 180.

3. For a history of these problems, see President’s Foreign Intelligence Advisory Board, “Science at Its Best, Security at Its Worst: A Report on Security Problems at the U.S. Department of Energy,” June 1999.

4. See Weiner, Our Own Worst Enemy? pp. 232-234.

5. A 1999 investigation by the U.S. General Accounting Office found that less than 40 percent of project money was actually going to institutes in the former Soviet Union; most was going to the U.S. labs. See U.S. General Accounting Office, “Nuclear Nonproliferation: Concerns With DOE’s Efforts to Reduce the Risks Posed by Russia’s Unemployed Weapons Scientists,” GAO-RCED-99-54, February 1999, p. 4. In 2008 the IPP was criticized for funding scientists at a Russian institute that also did work on the Bushehr reactor in Iran. See House Committee on Energy and Commerce, “Department of Energy Program Funds Russian Nuclear Work in Iran,” February 6, 2008.