| This page with graphics | Disarmament Diplomacy | Disarmament Documentation | ACRONYM Reports |

| Acronym Institute Home Page | Calendar | UN/CD | NPT/IAEA | UK | US | Space/BMD |

| CTBT | BWC | CWC | WMD Possessors | About Acronym | Links | Glossary |

Disarmament Diplomacy

Issue No. 69, February - March 2003

Opinion & Analysis

International vs. Domestic Nuclear Safeguards: The Need for Clarity in the Debate over Effectiveness

By Roger G. Johnston and Morten Bremer Maerli

Introduction

According to Article III of the Non-Proliferation Treaty (NPT) each non-nuclear-weapon state party to the treaty undertakes to accept safeguards, as set forth in an agreement to be negotiated and concluded with the International Atomic Energy Agency (IAEA) in accordance with its statute1 and safeguards system. The purpose of such safeguards is to verify the fulfilment of the state party's obligations under the NPT - to prevent diversion of nuclear energy from peaceful uses to nuclear weapons or other nuclear explosive devices.2 As such, IAEA safeguards constitute the most important example of multinational nuclear treaty monitoring.3 By the IAEA's own definition, the safeguards system comprises an extensive set of technical measures by which the IAEA Secretariat independently verifies the correctness and the completeness of the declarations made by states about their nuclear material and activities.4

While this definition goes a long way in describing the safeguards process from the point of view of the IAEA, it fails to describe concisely and substantively the intentions (and limitations) of IAEA safeguards. To add to the confusion over the term "safeguards", the United States government uses the word "safeguards" in a rather imprecise way, often in combination with "security", to cover a wide range of domestic nuclear non-proliferation activities, from physical protection and containment to accounting of nuclear material, grouped under the heading of Material Protection, Control & Accounting (MPC&A). It is not surprising that many observers complain that a clear, concise, and consistent safeguards definition is still missing.5

As a result, there may be a risk not only of mixing the meaning of the different safeguards terms - but also of confusing the distinct goals of each nuclear security measure installed. Through our work, we have observed at first hand a disturbing tendency to uncritically use (or consider) domestic MPC&A hardware and methods for international treaty monitoring applications, including for IAEA safeguards.6 There is a long tradition of the IAEA using domestic (usually US) safeguards technology and approaches with little or no modification for use in IAEA safeguards. Examples include the metal cup (e-cup) seal and the T-1 seal. Each of these were developed for US domestic purposes and then later applied by the IAEA. US national laboratory personnel claim that this only happened after the seals were modified; however, examinations of the "customised" seals shows that the changes are very minor.7

Other examples of nuclear arms control technology diffusion include equipment and schemes for non-destructive enrichment measurements using gamma ray spectroscopy8, neutron coincidence measurements9, and calorimetry.10 Indeed, technical assistance provided by the United States and other countries to the IAEA typically involves personnel who are primarily domestic safeguards experts. Even for planning or discussion purposes, it is common to invoke conventional domestic safeguards measures in any consideration of how a hypothetical on-site inspection regime might be implemented.11 Sometimes this is done as a sincere but misguided attempt to save time or money. In reality, however, this is likely to optimise neither cost-effectiveness nor long-term nuclear security.

To avoid such "one-size-fits-all" approaches to nuclear security, in this paper we hope to show that the two types of nuclear safeguards are vastly different activities, even though both are concerned with responsible nuclear stewardship and with the prevention of diversion. We will do so by comparing a set of "real-life" (operational) differences between "international safeguards" and "domestic safeguards". Hopefully, this will serve as a reminder of how extremely different the two activities really are. Perhaps it will also encourage more effective and creative approaches to treaty monitoring and corroboration.

After briefly discussing and clarifying relevant terminology, the paper compares domestic safeguards with international (NPT) safeguards in a range of different categories. This assessment is followed by some concluding remarks on the dangers of confusing or mixing domestic safeguards with international safeguards.

Terminology and Conceptual Clarifications

Throughout this text, admittedly in a somewhat simplified fashion, "domestic safeguards" will be used interchangeably with "domestic MPC&A", or just "MPC&A", and "international safeguards" will be used as a term equivalent to "IAEA safeguards under the Non-Proliferation Treaty".

According to Merriam-Webster's Collegiate Dictionary, a "safeguard" is defined as "a precautionary measure, stipulation, device, or a technical contrivance to prevent accident".12 In the IAEA context, safeguards can thus be taken to mean a set of preventive, protective, or deterrent measures to be put in place for the purpose of avoiding state diversion of nuclear materials. Domestic US safeguards, on the other side, comprise a set of preventive, protective, or deterrent measures to be put in place for the purpose of primarily avoiding non-state diversion of nuclear materials.

It is obvious that a state operates domestic safeguards inside its own borders and facilities, and that it does so largely for its own benefit and purposes. Treaty monitoring, in contrast, works in an international arena and multinational setting. IAEA safeguards involves trained Agency personnel attempting to ascertain if a state that signed the NPT is in compliance with its obligations, declarations, and promises. The framework for the monitoring and on-site inspections is outlined in agreements with the respective signatories to the NPT.13,14 When a signatory has its nuclear facilities inspected, it is called the "host" (inspected) state.

As mentioned, "domestic safeguards" is a term commonly considered as identical to domestic MPC&A. One problem with this definition, however, is that the US Government often talks about nuclear "security and safeguards", where "security" is meant to be the P portion (protection) of MPC&A, and "safeguards" is the C&A portion (control and accountability).15 It is, furthermore, important to recognise that treaty monitoring (including IAEA safeguards) is not international MPC&A, nor is it domestic MPC&A demonstrated to foreigners.16 While "cooperative MPC&A" has evolved since the end of the Cold War, there currently is no "international MPC&A". Cooperative MPC&A involves one country assisting another with its domestic MPC&A, such as the programmes undertaken by the United States Department of Energy (DOE) to help the Russians strengthen their own domestic nuclear MPC&A. There are currently no internationally-designated "nuclear islands", however, where it may be envisioned17 that various countries could send their nuclear weapons, materials, or waste for joint protection, control and accounting.

The IAEA claims to be "safeguarding the atom".18 While popular descriptions like this may be useful for raising support and recognition for IAEA's core activity, such vague security concepts may tend to blur the intention and distinct function of safeguards activities. Contrary to what might be common popular belief, for example, IAEA "safeguards" do not comprise any provisions for physical protection of nuclear materials. There are, moreover, no treaties covering the domestic physical protection of nuclear materials, though there have been proposals.19 The IAEA's Convention on Physical Protection applies only for nuclear material in transit, and efforts to broaden the scope of the treaty are underway.20 Recommendations for physical protection have been developed by experts from member states under the auspices of the IAEA.21 These recommendations, however, do not have the formal status of an international "standard" and are completely voluntary. The IAEA, furthermore, does not guard, protect, control, or own any substantial amount of nuclear materials, nor is it responsible "either for the provision of a state's physical protection system or for the supervision, control or implementation of such a system."22 The IAEA also carries out no primary accounting of nuclear materials, but rather relies on the member states' system of accounting and control.23

Two other terms that are useful in comparing, understanding, and discussing domestic MPC&A and treaty monitoring, are "Protagonist" and "Antagonist". The Protagonist is the party that implements the domestic MPC&A measures or the NPT safeguards inspections. The Antagonist (adversary) is the party that attempts, or may potentially attempt, to defeat the domestic MPC&A measures or to violate the treaty for nefarious purposes. In simplistic terms, "good guys" are the Protagonists and "bad guys" are the Antagonists. The IAEA (and its inspectors), for example, are the Protagonist for IAEA safeguards, while the host (inspected) state is the Antagonist. Note that Antagonists for domestic MPC&A can be either insiders or outsiders. "Insiders" work inside a nuclear facility, while "outsiders" are not authorised to gain unsupervised access to a given facility.

Note further that it is possible - though we know of no examples - that safeguards inspectors could, in the course of their inspections, suggest to the host state ways to improve its own domestic MPC&A practices. This is not part of the inspectors' regularly assigned duties. It does, however, represent a potential situation where domestic MPC&A and treaty monitoring could interact, and where both the inspectors and the host state could, at least briefly, simultaneously play the role of Protagonists. Another possible overlap between domestic MPC&A and treaty monitoring could occur in the future should states someday sign international agreements promising to meet certain minimum standards in regards to domestic nuclear MPC&A. Unfortunately, this situation does not currently exist.24

Attributes

In the following we consider how different characteristics of safeguards differ for domestic MPC&A and international safeguards under the Non-Proliferation Treaty.

Goals and Objectives

Probably the most important difference between domestic MPC&A and international safeguards concerns their goals and core objectives. The purpose of domestic MPC&A is to protect, control, and to account for nuclear materials. This includes protecting nuclear weapons or materials from sabotage, vandalism, terrorism, espionage, theft, diversion, or loss. International NPT safeguards, on the other hand, are concerned with obtaining evidence that each state that signed an agreement or treaty is indeed complying with its obligations, declarations, and promises.

Observe that denying terrorist and other sub-nationals access to fissile material through stringent standards of physical protection is a principal role of domestic safeguards. The robustness of international safeguards for such terrorist scenarios is much more questionable. This is further described below.

The Protagonist

For domestic MPC&A, the Protagonist is the state that owns the nuclear facility placed under MPC&A, or (depending on the context) the facility personnel who implement the MPC&A as directed by their government. The Protagonist for NPT safeguards, in contrast, will be an international agency - IAEA or the European Atomic Energy Community (EURATOM) - that conducts on-site inspections. The host (inspected) state does not play the role of Protagonist, at least when its own nuclear facilities are under inspection. Instead, the inspected state plays the role of Antagonist, i.e., the potential adversary that might try to hide evidence that it is violating the treaty. In the case of bilateral treaties, however, each state will usually play both roles: Antagonist (when the host) and Protagonist (when inspecting the other nation's facilities).

Note that it is possible that unauthorised insiders or outsiders might take actions that undermine treaty compliance, such as diverting nuclear materials, without the knowledge or consent of the host state. This situation is more properly handled by domestic safeguards, and serves as a reminder that the need for domestic MPC&A does not go away when procedures for treaty monitoring (such as IAEA safeguards) are implemented. Indeed, separate domestic MPC&A and international treaty monitoring measures must operate simultaneously and without mutual interference inside a given nuclear facility - a reality that is often overlooked in planning on-site verification regimes.25

The Antagonist

The Antagonist for domestic MPC&A can be outsiders, such as thieves, spies, saboteurs, terrorists, or blackmailers wishing to penetrate a nuclear facility for nefarious purposes. The Antagonist can also utilise disloyal insiders working inside the facility. The most likely Antagonist is a single or a group of rogue individual(s), whether insiders or outsiders, working at cross-purposes to the state that owns the facility. The exact identity of all potential Antagonists for domestic safeguards is usually not known. In the case of NPT safeguards, the Antagonist is not an individual or small group. It is a state - the state that signed the NPT. Treaty monitoring is implemented in order to detect cheating or treaty breakout by the Antagonist.26 Note that a disloyal (e.g., bribed) safeguards inspector can also attempt to hide evidence of violations, but in doing so, he or she is presumably in the employ of the host state. It is thus still the host state that is the underlying Antagonist.

Loyalty

The Antagonist who tries to defeat domestic MPC&A is operating at cross-purposes to the nuclear facility and the state that owns it. If an insider, the Antagonist is probably being disloyal to his or her country. Individuals trying to hide safeguards violations, however, will often be operating under secret encouragement or orders from their own government (the host state). They are then typically being loyal to their country. This is a different situation than the one involving corrupt inspectors (above).

Attack Resources

The Antagonist for domestic MPC&A - being an individual or group - will typically possess fairly modest resources for trying to defeat domestic MPC&A measures.27 The Antagonist for safeguards, on the other hand, is the host state, and will typically have available the resources of an entire nation - most importantly, a scientific apparatus already boasting sufficient technical sophistication to possess and handle nuclear materials, and perhaps clandestinely pursue the development of nuclear weapons. Based on group membership alone, this type of Antagonist can marshal and control a million to a billion times more resources (people and funds) than an individual or small group.

Indeed, the host state is an Antagonist who typically possesses (or can hire) world-class experts in espionage, surreptitious entry, electronics, computers, encryption, nuclear instrumentation, social engineering, etc. Thus, the hardware and approaches used for treaty monitoring will typically deal with a much more sophisticated adversary than is the case for domestic MPC&A. For this reason alone, international safeguards are a much more challenging task. We should not, therefore, automatically expect methods and hardware that are adequate for domestic MPC&A to be up to the more difficult (and varied) challenges of treaty monitoring.

The Facilities

In the case of domestic safeguards, the nuclear facilities placed under MPC&A measures are owned and controlled by the Protagonist. This is completely different from the situation obtaining in international safeguards where the nuclear facilities are owned and controlled by the Antagonist. Note that conventional security applications (such as home or industrial security monitoring) more closely resemble domestic MPC&A in this sense than international treaty monitoring.28 There are few security applications where the "good guys" openly install security and monitoring hardware onto the walls of facilities owned by the "bad guys". Thus, conventional (domestic) security methods and hardware are not automatically germane to IAEA safeguards.

The full realisation that the Adversary owns the facility being monitored would seem, for example, to call into question the current IAEA practice of storing verification hardware and calibration standards unattended inside the facility to be inspected. Such equipment is often protected by tamper-indicating seals, but current seals do not generally provide the high levels of security that would seem warranted.29,30

Ownership of Items Being Monitored

With domestic MPC&A, the nuclear items being monitored are owned and controlled by the Protagonist. In the case of IAEA safeguards, they are owned and controlled by the Antagonist. This is one of the most profound differences between domestic and international safeguards. It means that the operation of NPT safeguards is not simply a routine extension of conventional security models where the "good guys" own and control the valuables, and the "bad guys" try to steal them.

The Role of Facility Personnel & Infrastructure

The personnel and infrastructure (including security hardware and structures like fences) inside a nuclear facility actively assist the Protagonist in implementing domestic MPC&A measures. With treaty monitoring, however, the facility personnel and infrastructure can potentially be exploited by the host state to assist in cheating or breakout. Indeed, there are current examples worth considering of confusion or potential confusion in distinguishing between the domestic and international nuclear husbandry functions. We are assured, for example, that existing security fences at Russian nuclear facilities can be leveraged to assist with international treaty monitoring31 - even though the owners of that fence (the Russians) are the ones being monitored! Thus, the all too common idea that fences and guards inside a nuclear facility can be helpful for treaty monitoring is fallacious, and a dreary testimony to confusion over the true nature of domestic MPC&A vs. treaty monitoring.

Atmosphere & Implementation

Domestic MPC&A is typically implemented with at least some degree of secrecy. Information about the system of physical protection, the quantities and qualities of fissile material at a specific location, and the date and time of transport are usually classified. Details about security procedures, alarm response times, and the number of security guards are similarly not openly discussed. This secrecy assists the Protagonist in implementing and maintaining effective domestic MPC&A, and complicates the task for any potential Antagonist. Sharing US domestic safeguards technology and approaches with the international community for use in other settings could thus potentially compromise US national security and needs to be carefully analysed.32

In addition to operating in a secret environment, domestic MPC&A is inherently adversarial towards the Antagonist. For example, security guards are often under orders to use deadly force if necessary to prevent the diversion of nuclear materials.

All of this is in stark contrast to international safeguards, where cooperation, corroboration, and a certain level of mutual trust are critical issues. Total secrecy is not possible because the host (inspected) state could insist on understanding and approving any monitoring hardware brought into its nuclear facilities. This is due to its legitimate concerns about safety, security, and espionage. Moreover, an excessively adversarial stance on the part of any participant is not conducive to effective treaties because it will tend to doom any international agreement.33

The fact that international safeguards system is not subject to the same level of secrecy does not by any means render it useless. To maximize its effectiveness, the system uses tamper-indicating devices and features. The principle is that the international authority is able to detect any tampering with the system. This will, however, require, optimal and correct uses of tags and seals (see below).

Another difference between domestic MPC&A and treaty monitoring involves the degree to which a "compliance mode" may predominate. In such a mode, security personnel and others may be lulled into a false sense of security, due to a lack of incidents and tedious routine work. In discussions with US domestic MPC&A personnel, it is common to hear complaints that sometimes the emphasis is on blindly following formal regulations, and/or on satisfying the whims of domestic auditors and bureaucrats, rather than focusing on true effectiveness.

This appears to be much less of a concern for IAEA safeguards, though some have questioned the IAEA's ability to holistically and aggressively detect and report problems.34,35,36 The reason for this may be that domestic MPC&A is much older and has had a chance to establish a more complex bureaucracy, with less room for flexibility, robustness, and individual initiative to improve security. Alternatively, it may be that IAEA safeguards, unlike domestic MPC&A, automatically receive critical scrutiny from an international cast of participants so that slipping into a complaisant mode of compliance is less likely, in particular as inspectors intrinsically look for evidence of wrongdoing.

Espionage, Safety, and Security Concerns

The Protagonist who implements domestic MPC&A will be greatly concerned about preventing espionage, assuring nuclear safety and security, and protecting national interests. These concerns are the main reason that domestic MPC&A is implemented in the first place. The Antagonist for domestic MPC&A does not share these concerns, and will work actively against them.

In the case of international safeguards, in contrast, it is the Antagonist (host state) that will customarily have the most immediate and practical concerns about these issues. This is because the Antagonist owns the nuclear facility and its contents, and has the ultimate responsibility for what happens to them. This is why the IAEA approval process includes full scale testing of new equipment to ensure that the equipment meets the safety and security standards agreed upon in the international safeguards community. If it can be shown that IAEA equipment caused a relevant incident in a facility under safeguards, the operator will request that the IAEA remove the equipment, investigate the cause, and correct the problem.

Operational Environment

The success of both domestic and international safeguards can be influenced by local external factors that are difficult to predict reliably such as earthquakes, extreme weather conditions, industrial accidents, electrical power fluctuations and failures, labour-management conflicts, political uprisings, financial problems, bureaucratic turf battles, political interference, and the need for site-specific training and knowledge. These factors, however, may generally be easier to predict for domestic MPC&A practitioners than for international safeguards inspectors because domestic MPC&A will tend to operate in a better known and more time-tested environment. For example, the local politics, weather, labour situation, safety and security cultures, and bureaucratic procedures will usually be well understood by the facility operator, and are likely to be taken into consideration when domestic MPC&A equipment is fielded.

International safeguards, however, may have to contend with a lot more unknowns. On-site verification equipment and monitoring procedures will often be developed and approved by a number of different states, including those lacking a good understanding of the local conditions that must be endured.

Confidence in the Chain of Custody

The Protagonist who implements domestic MPC&A will generally have confidence in the reliability of preceding and subsequent handling steps. If, for example, a shipment of nuclear materials arrives at a given nuclear facility with proper paperwork and approvals, there will usually be few questions about its legitimacy and true origin. Similarly, when a shipment leaves a nuclear facility for a domestic transfer, the personnel in that facility are usually not interested in, or responsible for, its ultimate (national) disposition.

In the case of international safeguards, however, the Protagonist (inspectors) will tend to be much more suspicious of earlier and subsequent handling of nuclear materials. Who dispatched what, and to whom, are issues of critical interest.

Consequences to the Antagonist of Getting Caught

An Antagonist who is caught attempting to defeat or circumvent domestic MPC&A measures faces the very real possibility of being injured, incarcerated, or killed. For international safeguards, however, individuals who are secretly authorised by the host state to violate treaty obligations are unlikely to face any substantial physical danger or legal liability. The cheating state itself may be able to avoid international wrath and even sanctions through political "horse-trading", stalling tactics, or promises of better behaviour in the future. Currently, there are no a priori sanctions for international safeguards violations; how the international community has dealt with years of non-compliance both in Iraq and North Korea are indicative of the troublesome consequences that can rush to fill this vacuum.

Reporting of Anomalies

There is a substantial psychological difference between domestic MPC&A and international safeguards when problems are detected. Domestic MPC&A personnel who report anomalies, evidence of security deficiencies, losses, or diversion will usually present these concerns to supervisory security personnel who will tend to be less than pleased to learn there are problems. Such failures will inevitably reflect negatively on the past and current performance of domestic MPC&A personnel, especially supervisors and managers. In a healthy domestic security/safeguards programme, the problems will be recognised and addressed anyway. In less professional situations, however, security supervisors and managers may be unwilling to admit to themselves or others that problems exist.

The situation may differ considerably with respect to international safeguards. There, the inspectors are specifically hired to find problems. If anomalies are found, their supervisors, at least in theory, are eager to hear about evidence of treaty violations, and inspectors will be praised for detecting them. On the other hand, the potential political impact on the international scene, as well as the likely accusations and international squabbling that may result, are likely to make the need for substantial and clear evidence of violations stronger than for domestic MPC&A breaches.

Response Time

Domestic MPC&A measures are usually designed for an immediate response (seconds to minutes) to any Antagonist attack. With international safeguards, however, evidence of cheating or breakout may not be detected until months after the incident. Moreover, the evidence will typically be analysed at great length before a final conclusion is reached. A response, if any, will thus be even more delayed (see the section on "Consequences to the Antagonist of Getting Caught").

Presence of the Adversary

With domestic MPC&A, the Adversary, if an outsider, should never be present inside the nuclear facility. If the Adversary is an insider, his or her access should at least be limited and closely monitored. In the case of international treaty monitoring, in contrast, the Adversary is always present (as owner and operator of the facility), and will have extensive unsupervised access to most or all of the facility infrastructure, contents, and international monitoring equipment present (see, e.g., the section on "Video" below).

Tags and Seals

Tags and seals are frequently used for both domestic and international nuclear applications.37,38 Tags uniquely identify or fingerprint an item or container, while seals detect unauthorised access or tampering. Tags and seals will usually be installed and inspected by the Protagonist in the case of domestic MPC&A, though it is possible that the person responsible for tag/seal installation or inspection could be an Antagonist, i.e., a disloyal insider. For international safeguards, however, it may sometimes be necessary for the Adversary, using host facility personnel, to install the tags or seals, perhaps under the watchful eye of the inspectors. This is because espionage concerns may prohibit foreigners from handling nuclear materials and containers owned by the host. This could make the tags and seals more exposed to unauthorised handling and tampering. It is important to note for both domestic and international applications that there are currently no tags or seals designed with the idea of revealing any hidden agendas the installer or inspector may have. Such tags/seals, however, are possible.

Video Monitoring

Video cameras are a useful tool for both domestic MPC&A and for some international safeguard activities. Video monitoring for domestic MPC&A is very similar to video monitoring for conventional non-nuclear security applications: the Protagonist owns and controls the video cameras and the walls they are mounted on, and does the interpretation of the video images. The Adversary should not have access to the cameras, video images, or surveillance systems. The video images are used to watch for outsiders trying to penetrate the facility, or to double-check the activities of insiders.

Video monitoring is quite different for international safeguards. The Antagonist owns the facility and the walls on which the video cameras are mounted. The Antagonist has continuous, often unsupervised, access to the equipment, and the problem of protecting the cameras and their enclosures from tampering in this kind of environment is a largely unsolved problem.39,40 Moreover, the video cameras are meant to monitor the activities inside the nuclear facility, and are not focused on any outside adversaries.

Transport

For domestic MPC&A, the Protagonist knows and controls the dates, times, logistics, and itinerary of the transport of nuclear weapons or materials. Ideally (as discussed above in the section on "Atmosphere & Implementation"), the Antagonist is ignorant of the details, which are usually classified.

For international safeguards inspections of transfers, the Antagonist is in charge of nuclear transport. The Protagonist (inspectors) may need to be given transport information, but exactly what information can be shared with foreigners is a largely unresolved issue.41

Cultural & Language Differences

With domestic MPC&A, the Protagonist and the most likely Antagonists belong to the same nation and culture, and probably speak the same language. This is not the case with international treaty monitoring. Any practical implementation must take into account cultural and language differences, including what can be vastly different values and attitudes about security.42

Politics

Political leaders and the general public will usually be completely in favor of domestic MPC&A. Unless specific problems become apparent, MPC&A efficacy will rarely be questioned.

International safeguards may be a different story. There will tend to be a great deal of concern and suspicion from political leaders, the media and the public about its effectiveness, especially when new arms control regimes are being considered. As painfully evidenced by the Iraq crisis, the reliability of treaty verification will be a critical issue for both national security priorities and for political support for new treaties or disarmament programmes.43

Conclusion

Domestic and international safeguards - despite both being called "safeguards" - are profoundly dissimilar. Domestic safeguards are primarily concerned with nuclear materials protection, control, and accounting (MPC&A). Most of the safeguards currently undertaken by the IAEA, in contrast, involve monitoring under the NPT.

The fact that there may be some overlap in the usefulness of certain technological equipment used to support these distinct purposes should not be unexpected. This paper, however, has considered a wide range of areas in which domestic MPC&A and international safeguards have clear and inherent differences. For many of these areas, the two tasks have almost diametrically opposed characteristics in terms of adversaries to defeat and context in which to operate. Their highly disparate nature makes it thus unlikely that domestic MPC&A and IAEA safeguards can be well served by uncritically using the same (or similar) philosophies, tools, and methods. The common idea that monitoring equipment can be "shared" between domestic MPC&A functions and international safeguards functions in order to save time and money is equally dubious.

International safeguards are not, as some seem tempted to assume, merely a routine extension of conventional nuclear security measures within states. We should thus be suspicious of any quick attempts to apply domestic MPC&A practices and hardware to international safeguard monitoring applications (and vice versa), especially when this is done without major modifications or careful analysis of the task at hand.

Notes and References

1. The charter of the International Atomic Energy Agency prescribes the organisation's twin objectives: to promote the peaceful (and safe) use of nuclear energy and to safeguard civil nuclear materials against diversion to military use. Since its founding in 1957, the IAEA has played a pivotal role in the development of the "peaceful atom" and, increasingly over the years, in the detection of clandestine nuclear weapon programs.

2. Article III of the Treaty on the Non-Proliferation of Nuclear Weapons. The treaty entered into force on March 5, 1970. According to Article III, procedures for the required safeguards "shall be followed with respect to source or special fissionable material whether it is being produced, processed or used in any principal nuclear facility or is outside any such facility. The safeguards required shall be applied to all source or special fissionable material in all peaceful nuclear activities within the territory of such state, under its jurisdiction, or carried out under its control anywhere". For the full text of the treaty, see the United Nations, http://disarmament.un.org/wmd/npt/npttext.html.

3. In accordance with voluntary safeguards agreements, the IAEA performs carefully negotiated and limited inspections in nuclear weapon states outside and inside the NPT. While important, such inspections will be kept outside the scope of this discussion, as will safeguards performed by the European Atomic Energy Community (EURATOM). Material being placed under EURATOM safeguards is normally been made liable to inspection by the IAEA.

4. Goldschmidt, P. 1999. "The IAEA Safeguards System Moves into the 21st Century". Supplement to the IAEA Bulletin 41, S-2, http://www.iaea.org/worldatom/Periodicals/Bulletin/Bull414/article8-suppl.pdf.

5. See e.g. Hackel, E. 2000. "Implementing Safeguards in Weak and Failed States", in E. Hackel and G. Stein (Editors), Tightening the Reins. New York: Springer, p. 142.

6. Morten Bremer Maerli and Roger Johnston. Spring 2002. "Safeguarding This and Verifying That: fuzzy concepts, confusing terminology, and their detrimental effects on nuclear husbandry". Nonproliferation Review 8, http://www.cns.miis.edu/pubs/npr/vol09/91/abs91.htm.

7. Examinations performed by Roger G. Johnston and the Vulnerability Assessment Team in the Applied Monitoring and Transparency Laboratory (AMTL) at Los Alamos National Laboratory.

8. Reilly, D., Ensslin, N., Smith, H. and Kreiner, S. (Editors). 1991. Passive Nondestructive Assay of Nuclear Materials. Washington, D.C.: Nuclear Regulatory Commission, NUREG/CR-5550, http://www.nis5.lanl.gov/Publications.

9. Ibid.

10. ASTM. March 2000. Standard Test Method for Nondestructive Assay of Plutonium, Tritium and 241Am by Calorimetric Assay, ASTM C 1458-00, New York: American National Standards Institute.

11. Johnston, R.G. Spring 2001. "Tamper Detection for Safeguards and Treaty Monitoring: Fantasies, Realities, and Potentials". The Nonproliferation Review 8, pp. 102-115.

12. Brittanica. 2001. Merriam-Webster's Collegiate Dictionary, http://www.britannica.com/dictionary?book=Dictionary&va=safeguards.

13. IAEA. 1972. INFCIRC/153, "The Structure and Content of Agreements Between the Agency and States Required in Connection with the Treaty on the Non-Proliferation of Nuclear Weapons", http://www.iaea.org/worldatom/Documents/Infcircs/Others/inf153.shtml.

14. IAEA. 1998. INFCIRC/540(corrected), Model Protocol Additional to the Agreement(s) between State(s) and the International Atomic Energy Agency for the Application of Safeguards, www.iaea.org/worldatom/Documents/Infcircs/1998/infcirc540corrected.pdf.

15. For the purposes of this article and to avoid confusion, we will instead use the term "domestic MPC&A" throughout the rest of this paper, though we will continue to think of the term as referring to "domestic safeguards".

16. Maerli and Johnston, op.cit, endnote 6.

17. For such an envisioning, see Rinne, R. May 1999. An Alternative Framework for the Control of Nuclear Materials. Stanford University: Center for International Security and Cooperation.

18. ElBaradei, M. April 1999. Safeguarding the Atom, IAEA Bulletin 41, http://www.iaea.org/worldatom/Periodicals/Bulletin/Bull414/article1.pdf.

19. Bunn, G. Summer 2000. "Raising international Standards for Protecting Nuclear Materials from Theft and Sabotage". The Nonproliferation Review 7, pp. 146-157.

20. Even with the increased focus on physical protection in the aftermath of the September 2001 terrorist attacks, the revision of the convention seems set to become quite limited. As it now stands, the "well defined amendment" to the convention would exclude: any requirements to submit reports to the international community on the implementation of physical protection, a peer review mechanism, a mandatory application of INFCIRC/225 (see endnote 22), a mandatory international oversight of physical protection measures, and any nuclear material and facilities for military uses. See Stig Isaksson, "Improving the International Physical Protection Regime", presentation given at the Nordic Society for Non-Proliferation Issues Seminar, Bergen October 2002.

21. IAEA. 1999. INFCIRC/225/Rev.4 (Corrected), The Physical Protection of Nuclear Materials and Nuclear Facilities, http://www.iaea.org/worldatom/program/protection/inf225rev4/rev4_content.html.

22. Ibid, Section 3.2.b.

23. Under comprehensive safeguards agreements concluded with the IAEA, a state must establish a State System of Accounting and Control (SSAC) of nuclear material on a national or regional basis. The SSAC co-operates closely with the IAEA in the implementation of safeguards, regularly providing the Agency with information on matters related to the nuclear material accountancy system in force and the state's adherence to the reporting requirements. See Svein Thorstensen, "Safeguards & illicit nuclear trafficking: Towards more effective control", IAEA Bulletin, special issue on Safeguards and Industry, Volume 38, Number 4, http://www.iaea.or.at/worldatom/Periodicals/Bulletin/Bull384/thorsten.html.

24. Bunn, op.cit, endnote 19.

25. Gerdes, E.R., Johnston, R.G., and Doyle, J.D. 2002. "A Proposed Approach for Monitoring Nuclear Warhead Dismantlement." Science & Global Security (in press).

26. "Breakout" here means unilaterally and suddenly abrogating the treaty or agreement.

27. Particularly powerful criminal adversaries and a threat from another state may be exceptions to this rule. However, terrorist groups generally have limited resources available (at least as compared with state resources), and any military intervention of state in a neighbouring country would likely be a declaration of war.

28. Maerli and Johnston, op.cit, endnote 6.

29. Johnston, R.G. and Garcia, A.R.E. Spring 2000. "An Annotated Taxonomy of Tag and Seal Vulnerabilities." Journal of Nuclear Materials Management 28, 23-30.

30. Johnston, R.G., Garcia, A.R.E., and Grace, W.K. Fall 1995. "Vulnerability Assessment of Passive Tamper-Indicating Seals". Journal of Nuclear Materials Management 24, 24-30.

31. As suggested by Ivan C. Oelrich, "Production Monitoring for Arms Control", pp. 109-123, in Michael Krepon and Mary Umberger, eds., Verification and Compliance: a Problem-Solving Approach, (Cambridge, Massachusetts: Ballinger, 1988), p. 116.

32. Johnston, op.cit, endnote 11. Moreover, private companies may have invested vast resources in the development of specialised equipment and systems for domestic nuclear applications. Protection of proprietary rights and trade secrets is thus likely to play a role in any consideration of disseminating information about domestic MPC&A hardware or software. There will often also be complex trade agreements, export control issues, and customs barriers that may limit the wide distribution of MPC&A products.

33. There is also often a basic symmetry and equality (lacking in domestic MPC&A) between the Protagonist and Antagonist, especially for bilateral treaties where the Protagonist and Antagonist are partners in the treaty. In bilateral treaties, the Protagonist and Antagonist will usually have to swap roles to complete bilateral inspections.

34. Kay, D. 1994. "The IAEA: How can it be Strengthened?" in M. Reiss and R.S. Litwak (Editors), Nuclear Proliferation After the Cold War. Washington D.C.: Woodrow Wilson Press, pp. 319-332.

35. Anderson, D. 1997. Nuclear Safeguards. Australian Foreign Affairs, Parliamentary Research Service, http://www.aph.gov.au/senate/committee/uranium_ctte/report97/ch12_0.htm.

36. Leventhal, P. 1997. Safeguards Shortcoming - A Critique, http://www.nci.org/plsgrds.htm.

37. Johnston, op.cit, endnote 11.

38. Gerdes, Johnston & Doyle, op.cit, endnote 25.

39. Johnston, op.cit, endnote 11.

40. Gerdes, Johnston & Doyle, op.cit, endnote 25.

41. Johnston, op.cit, endnote 11.

42. Krause, K.R. (Editor). 1999. Culture and Security: Multilateralism, Arms Control and Security Building. London: Frank Cass.

43. On the reliability issue, see Gallagher, N. 1999. The Politics of Verification. Baltimore: John Hopkins University Press.

Roger G. Johnston, Ph.D., CPP is Team Leader for the Vulnerability Assessment Team in the Applied Monitoring and Transparency Laboratory (AMTL) at Los Alamos National Laboratory (LANL). His research interests include specialty tools and instrumentation, non-proliferation, and tamper detection. He received his CPP certification from the American Society of Industrial Security in 1997, and served as a Science Fellow at the Center for International Security and Cooperation (CISAC) at Stanford University for 2000-2001. Morten Bremer Maerli is a Researcher at the Norwegian Institute of International Affairs (NUPI). He has been working on issues relating to nuclear non-proliferation assistance to Russia since 1995. For the 1999-2001 academic years, he was a Visiting Research Scientist at Sandia National Laboratories, California, and a Science Fellow at the Center for International Security and Cooperation (CISAC) at Stanford University. Harry Dewey, Jim Doyle, and Eric Gerdes contributed to some of the ideas explored in this piece. The views expressed in the paper are those of the authors and should not necessarily be ascribed to the AMTL, LANL, the United States Department of Energy, or NUPI.

Return to top of page

© 2003 The Acronym Institute.