NOTE: This article originally appeared in the January-March 2011 issue of Trust & Verify, a publication of Verification Research, Training and Information Centre (VERTIC).
The ‘Additional Protocol’ is so-called because it supplements a state’s safeguards agreement with the International Atomic Energy Agency(IAEA). The Additional Protocol (AP) substantially strengthens levels of assurance on the peaceful nature of nuclear activities in countries that have ‘comprehensive’ safeguards agreements, by broadening the information to be reported to the IAEA and the access given to inspectors. Without these extra measures, the IAEA’s ability to detect undeclared nuclear activities is substantially reduced. The AP is, consequently, an essential component of the nuclear non-proliferation regime.
All non-nuclear-weapon states (NNWS) that are party to the 1968 Nuclear Non-Proliferation Treaty (NPT) are required to have what are known as comprehensive safeguards agreements in place covering all their nuclear material. When the need to strengthen IAEA safeguards with respect
to the detection of undeclared nuclear activities became obvious, the ‘Model Additional Protocol’ was developed. The Model AP (INFCIRC/540), which provides a template for what an AP supplementary to a safeguards agreement should look like, was agreed by the IAEA’s Board of Governors in 1997.
During negotiations on the Model AP by the IAEA Board (which ran from 1996 to 1997), some participants said they regarded the adoption of an AP by a state as a ‘voluntary’ undertaking. The two IAEA Directors General involved at the time, Hans Blix and his successor Mohamed ElBaradei, agreed with this proposition. More recently, the final document of the 2010 NPT Review Conference noted ‘it is the sovereign decision of any state to conclude an additional protocol’.
As of 4 March 2011, 135 states—two-thirds of all NPT parties—had signed an AP. Of these, 106 have their APs in force. Although the AP is not yet universal, it is important to note the very high uptake by those NNWS party to the NPT that have ‘significant nuclear activities’. Such activities encompass any amount of nuclear material in a facility or ‘location outside facilities’, or nuclear material in excess of the exemption limits specified in paragraph 37 of the standard comprehensive safeguards agreement (IAEA document INFCIRC/153). There are 62 NNWS parties to the NPT with significant nuclear activities: 55 of these—almost 90 per cent—have signed an AP, and 47 (or more than 70 per cent) have their AP in force. These figures show that the combination of a comprehensive safeguards agreement and an AP is now clearly established by international practice as the NPT safeguards standard.
There are, however, seven NNWS parties to the NPT with significant nuclear activities that have not yet signed an AP. One of these—Algeria—had a draft AP approved by the Board in 2004 but has not signed it. Six—Argentina, Brazil, DPRK, Egypt, Syria and Venezuela—have not commenced negotiation of an AP with the IAEA, and at least three—Brazil, Egypt and Syria—have said that they have no intention of doing so (DPRK is included here because, although it gave notice of withdrawal from the NPT in 2003, the validity of this withdrawal has not been determined). Also of concern is Iran, which signed an AP in 2003, and began applying it on a provisional basis, but then ‘suspended’ it in 2005.
This paper discusses whether the AP is, in fact, ‘voluntary’, whether the remaining AP holdouts are right to regard it as optional, and what is to be done about the situation.
Background—why the Additional Protocol
When a NNWS joins the NPT it undertakes to accept safeguards, as set out in an agreement to be concluded with the IAEA in accordance with the IAEA’s Statute and ‘the Agency’s safeguards system’, for ‘verification of the fulfilment of its obligations assumed under this Treaty with a view to preventing diversion of nuclear energy from peaceful uses to nuclear weapons’ (NPT Article III.1). Safeguards are to be applied on all ‘source or special fissionable material’ (a phrase encompassing depleted, natural and enriched uranium as well as plutonium and thorium), in all peaceful nuclear activities within the state’s territory, under its jurisdiction or carried out under its control anywhere. A standard safeguards agreement for this purpose—i.e. INFCIRC/153—was approved by the Board in 1971, and has been in use ever since.
The Agency’s ‘safeguards system’ that has evolved since 1971 is based on the standard agreement set out in INFCIRC/153, but it is now much more than this. To have a full appreciation of the safeguards system it is necessary to look beyond INFCIRC/153, both upwards, to the IAEA’s Statute and decisions of the Board, and downwards, to: (a) instruments supplementary to safeguards agreements, such as subsidiary
arrangements and facility attachments; (b) IAEA Secretariat documents such as Safeguards Manuals, the Safeguards Criteria, the integrated safeguards conceptual framework, etc.; and (c) safeguards implementation procedures and techniques.
These various components provide a degree of flexibility in ‘the Agency’s safeguards system’, which has enabled the Board and the Secretariat to adapt the safeguards system to meet changing circumstances over some four decades. This adaptability is a key strength, essential to maintaining and improving the effectiveness of the system.
It is clear from the terms of the NPT that NNWS parties are obliged to declare all their nuclear material for safeguards, and the IAEA has both the right and the obligation (as set out in INFCIRC/153 paragraph 2) to verify that states do so—now generally termed the ‘completeness’ of states’ declarations. For the first two decades after the NPT, however, safeguards practice focused on verifying declared nuclear material (which, in essence, involved only confirming the correctness of declarations). It was thought that any undeclared nuclear material or activities would be revealed through diversion of declared nuclear material or misuse of declared facilities—and in any case neither the procedures nor the technologies had been developed for detecting undeclared nuclear programmes.
The discovery of Iraq’s clandestine nuclear weapon programme in 1991 showed the weaknesses in safeguards practice at that time. It became evident that if a state has undeclared nuclear material or activities it is quite likely there will be no obvious links between these and the declared nuclear programme. International attention turned to how to strengthen the safeguards system, with particular emphasis on the need to develop capabilities to detect undeclared nuclear material or activities. These efforts became known as ‘Programme 93+2’ (so called because it scheduled to start in 1993 and was expected to take two years), undertaken by the IAEA with the assistance of supporting states.
Programme 93+2 addressed two broad areas: technical (the procedures and technologies needed to detect undeclared nuclear material or activities) and legal (whether any additional formal arrangements were required to cover new safeguards activities). Amongst experts analysing the legal aspects, there was debate about whether the IAEA really needed additional legal authority. A number of experts considered that the provisions already contained in INFCIRC/153 for special inspections were sufficient to support the new safeguards procedures under discussion, and simply needed to be put to use. Special inspections, which may involve access to locations or to information, may be carried out in cases where information made available by the state is not considered adequate for the IAEA to fulfil its responsibilities. To date, for various reasons, special inspections have been rarely used, but they are an important part of the safeguards toolbox nonetheless.
Other experts maintained that an ad hoc approach based on special inspections would not be the best way for the IAEA to obtain the information and access necessary to effectively fulfil its responsibilities. By the end of Programme 93+2 this view prevailed, and it was decided to develop a standard set of reporting requirements and procedures, applicable to all states. Hence the idea of the AP came into being.
Is the AP really voluntary?
In a formal sense all treaties are ‘voluntary’—it is a sovereign decision for a state whether it accedes to a particular agreement or not. However, the circumstances of the AP are different to those of a new agreement.The AP does notpresent significant new commitments, but rather is an elaboration of existing commitments, already undertaken in the NPT and INFCIRC/153 safeguards agreements:
(a) under the NPT, NNWS agree to accept IAEA safeguards to verify non-diversion of nuclear energy, and to accept IAEA safeguards on all their nuclear material. An important aspect of this is that there should be no nuclear material ‘undeclared’ to safeguards. The IAEA has both the right and the obligation to verify the fulfilment of these commitments. As Director General Yukiya Amano emphasised at the 2010 NPT Review Conference, ‘the AP is of vital importance for the Agency to be able to provide credible assurance not only that declared nuclear material is not being diverted from peaceful uses, but also that there are no undeclared nuclear material and activities in a state’;
(b) under the NPT, NNWS agree to accept ‘the Agency’s safeguards system’. This system is not static—indeed INFCIRC/153 did not exist in 1970 when the NPT came into force. The Agency’s safeguards system comprises many elements, and has undergone substantial evolution over the years. Almost 90 per cent of NNWS parties to the NPT with significant nuclear activities have signed an AP. International practice therefore demonstrates that the contemporary form of the Agency’s safeguards system is the combination of a comprehensive safeguards agreement and an AP.
The AP was developed specifically to address weaknesses in safeguards implementation, in order that the IAEA can more effectively fulfil its mandate under the NPT and safeguards agreements. The AP should not be regarded as ‘optional’; the safeguards system, in which the AP plays an integral role, is not a smorgasbord, where states are free to choose the bits they like and leave out those they don’t. The IAEA Board, representing the organization’s membership, has determined that the procedures under the AP are essential for the IAEA to meet its responsibilities under the NPT to ensure that all nuclear material in NNWS remains in nonexplosive use.
The high uptake of the AP indicates that the view it is optional is not widely shared—but some significant states are amongst the holdouts, including states that hope to expand their nuclear programmes or to establish new programmes. It seems a number of factors are in play, including:
(a) a politicization of attitudes towards safeguards that has occurred in recent years—particularly worrying as it could imply a weakening of consensus support for non-proliferation. This is reflected by the Non-Aligned Movement’s (NAM) opposition at the 2010 NPT Review Conference to a consensus statement that the AP is the NPT safeguards standard. It seems some members of the NAM have lost sight of the importance of safeguards as a technical verification mechanism that benefits, not the West or the global ‘North’, but every state. Safeguards should be recognized as being an essential tool to dispel suspicions, and to help states demonstrate to neighbours and the international community that they are meeting their treaty commitments. To argue for less effective safeguards—safeguards without the AP—is contrary to their own national interest, which is to have a safeguards system that is more, not less, effective;
(b) a lack of understanding as to what the AP involves. For example, at the end of January 2011, Reuters reported statements by the Syrian President Bashar al-Assad that ‘nobody will accept to sign it’ (in fact 135 states have done so) and inspectors can ‘come any time to check anything under the title of checking nuclear activities, you can check anything’. These comments suggest that he has not been well briefed on the AP. In addition, remarks from some Brazilian officials reinforce the impression that the AP provisions on access, including managed access, are not well understood;
(c) the unwillingness to date of major nuclear suppliers to insist on the AP as a condition of supply. The Nuclear Suppliers Group has struggled unsuccessfully for some years to reach agreement on the AP as a condition for new nuclear supply. It has not helped in this endeavour that two NSG members (Brazil and Argentina) are amongst the AP holdouts. The major nuclear suppliers are members of the G8, and, for several years, G8 summits have endorsed the AP as a condition of supply. Yet a number of G8 members are considering nuclear supply contracts with states that not only have no AP, but even make a point of refusing it (e.g. Egypt).
What can be done?
Refusal to accept the most effective form of safeguards, of which the AP is an essential component, erodes confidence in the peaceful intent of the states involved. There is no reason to refuse the AP to protect legitimate national security and commercial interests. When INFCIRC/540 was negotiated, these interests were raised by a number of participants, and the AP text is careful to specify the extent of the IAEA’s access rights and the state’s right to establish managed access arrangements. The international community will draw the conclusion that, like with Iran’s ‘suspension’ its AP, those states that continue to refuse an AP are worried about what the IAEA will find.
Supporters of the non-proliferation regime need to do all they can to persuade those who have not concluded APs do so without further delay. This is particularly the case for nuclear suppliers, which have considerable ‘leverage’ at their disposal. Actions that could be taken to persuade the holdouts to reconsider include:
(a) outreach: to the extent that refusal of the AP may reflect inadequate understanding of its provisions, and inadequate understanding of NPT commitments, further diplomatic and technical outreach efforts may help to dispel misplaced concerns. Further outreach should also help to persuade those states without significant nuclear activities that have yet to sign an AP to do so, and to assist those that have signed but not yet ratified to move forward. A number of states and the European Commission are assisting the IAEA with AP outreach. Other states should consider what they can do to contribute to these efforts in their own region;
(b) establishing the AP as a condition of nuclear supply: the G8 has adopted the AP as a principle but its members have yet to follow through. The NSG is close to adopting the AP as a condition but is not quite there yet. Effective safeguards are essential to ensure that nuclear cooperation and trade are for peaceful purposes only; suppliers must stop prevaricating on the requirement for an AP;
(c) if necessary, the IAEA must consider other mechanisms for obtaining the information and access required to meet its safeguards responsibilities. The AP is an elaboration, a standardization, of information and access the IAEA could require under the special inspection provisions of INFCIRC/153. Taking an ad hoc approach through these provisions is not ideal, but is preferable to the continued application of less effective safeguards in the holdout states.
The NPT requires NNWS to accept the ‘Agency’s safeguards system’ to verify that all their nuclear material is in nonproscribed use. An essential part of this is to be able to derive assurance that there no nuclear material and activities exist outside safeguards. It cannot be considered satisfactory that the IAEA has to issue qualified safeguards conclusions—for those states with an AP it can provide credible assurance of the absence of undeclared nuclear material and activities, but for those states without an AP it cannot. This situation does not meet the terms of the NPT. It would be an affront to the great majority of states who have accepted the AP for this situation to continue indefinitely.
High-level representations are warranted with every state that claims less effective safeguards are good enough. It is to be hoped that changes in leadership in Brazil and Egypt, two key states for the AP, will lead to a reconsideration of attitudes to date. Since refusal of the AP has been made a political issue by the main holdouts, it may be necessary to consider a campaign at heads of government level, similar to the Nuclear Security Summit process that was so successful last year.
The author wishes to acknowledge the support of NTI (Nuclear Threat Initiative) for the preparation of this paper. The views expressed however are his own.