The United States’ “Withdrawal” from the Iran Nuclear Deal

I wrote this piece for the site E-International Relations. It appeared there today.

The Joint Comprehensive Plan of Action (JCPOA) is a diplomatic accord negotiated among seven states, purposed in addressing the concerns of six of those states (the United States, the United Kingdom, France, Russia, China, and Germany – a.k.a. the P5+1), with the nuclear energy program of the seventh (Iran).  On July 14, 2015, the agreement of the JCPOA was announced, bringing to a successful conclusion almost two years of negotiations, and seeming to bring to an end a period of international tension concerning Iran’s nuclear program that had been ongoing since 2002, and that had at times appeared likely to result in war.

The agreed JCPOA documents consist of 159 total pages of text, including 18 pages of the JCPOA itself, with a further 141 pages divided among five annexes, including a dispute settlement mechanism.  That such a diverse set of state parties, with decades of varying levels of antipathetic relations among them, could come to such a detailed and comprehensive settlement concerning an issue of such high international security sensitivity, was seen by many as nothing short of miraculous, and as a great triumph for diplomatic approaches to the settlement of nuclear proliferation concerns.

On July 20, 2015, the U.N. Security Council adopted Resolution 2231, in which it unanimously endorsed the JCPOA, and brought into force the measures relating to Security Council sanctions agreed therein.  Independently, yet in coordinated parallel with the JCPOA, on July 14, 2015 Iran and the International Atomic Energy Agency (IAEA) signed a “Roadmap Agreement” for the “clarification of past and present outstanding issues regarding Iran’s nuclear programme.”  This agreement, and the report submitted pursuant to it by the IAE Director General to the IAEA Board of Governors on December 15, 2015, brought to a close the IAEA’s multi-year investigation into the possible past military dimensions of Iran’s nuclear program.

By January 16, 2016, the date by which all sides had successfully implemented their initial commitments pursuant to the JCPOA, there seemed to be ample cause for optimism that the JCPOA would in fact fulfill its purpose of reducing international tension over Iran’s nuclear program, and through a carefully negotiated and reciprocal exchange of commitments allow all the parties thereto to achieve their core aims.  For Iran, these were to maintain it’s essential capacities to domestically produce uranium based fuel for its civilian nuclear reactors, including the sensitive step of uranium enrichment, while also having the crippling international financial sanctions that had been levied on the country since 2006 removed.  For the P5+1 the core aims of the JCPOA were to significantly limit Iran’s nuclear fuel production program, and submit it to an unprecedented level of ongoing international monitoring and verification, in order to push Iran much further away from the capacity to produce a nuclear weapon than it was in 2015.  As with any successful diplomatic accord, all sides were able to declare victory to both international and domestic constituencies through the terms of the JCPOA.

Within the context of international nuclear nonproliferation agreements, the JCPOA was a fascinatingly novel and exciting development.  Agreements among states on nuclear arms and nuclear energy have for decades typically taken the form of formally adopted treaties, constituting obligations of international law for the parties thereto.  This was certainly the case with regard to the nuclear arms control agreements negotiated between the United States and the Soviet Union/Russia both during and after the Cold War, and it was true of international agreements concerning nuclear weapons testing and regional nuclear weapon free zones.  The cornerstone of this web of treaties concerning nuclear energy and nuclear weapons proliferation is the Nuclear Non-Proliferation Treaty, which came into force in 1970, and which mandates that all non-nuclear weapon states conclude separate and additional bilateral treaties with the IAEA.  In short, the nuclear nonproliferation normative regime in international relations was, and continues to be, a highly treatified area of formal international legal obligations among states.

But the JCPOA broke that mold.  It was very deliberately negotiated not as a formal legally binding treaty among its parties, but as a legally non-binding diplomatic accord based in political, instead of legal, commitments.  States, of course, frequently enter into agreements with each other and make commitments that are not intended to be legally binding, yet do reflect an agreement among the parties with regard to some issue of their international relations. Sometimes these agreements are termed “memoranda of understanding” or “executive agreements.” In this case the agreement is entitled a Joint Comprehensive Plan of Action, which conveys the legally nonbinding nature of the text.

And yet, as noted above, the JCPOA consists of 159 pages of excrutiatingly detailed recitations of agreed terms and a schedule for their implementation, including specific commitments concerning Iran’s nuclear facilities and technical capabilities, specifically enumerated sanctions laws and regulations which were to be withdrawn, a highly proceduralized dispute settlement mechanism, and a pre-agreed process whereby any one of the permanent five members of the U.N. Security Council – all of whom were parties to the JCPOA – could unilaterally bring the Security Council’s prior economic sanctions on Iran back into application in the event it felt Iran was not in compliance with the agreement (i.e. the so-called “snapback” procedure).

The JCPOA seemed to represent a new wave of international agreements addressing highly sensitive and controversial issues of international relations through non-legally-binding, yet highly detailed and comprehensive recitations of political commitment.  This use of what some have referred to as “soft law” approaches to high-politics issues of international affairs was of course not new overall – non-legally-binding recitations of political commitments had of course been employed before by states on a wide range of issues of international relations (see, e.g., the 1975 Helsinki Final Act, and elements of the 2015 Paris Climate Accord).  But in the context of the nuclear nonproliferation issue area, the JCPOA was novel in its detailed and comprehensive handling of a complex and multi-faceted dispute through the form of a mutually agreed document reciting political and not legal commitments.

From January 2016 through May 2018, the IAEA, as well as the U.S. government, repeatedly confirmed that Iran was in compliance with its commitments under the JCPOA, which entailed among other things the shipping of most of Iran’s stockpile of low-enriched uranium out of the country, reducing enrichment capacity by placing two-thirds of its operating centrifuges into storage, permanently disabling its heavy water reactor at Arak, and bringing into effect an Additional Protocol inspection treaty with the IAEA under which it allowed an unprecedented level of access to the agency’s inspectors.  In 2017 alone, the IAEA conducted 419 safeguards inspections in Iran – far and away more than in any other safeguarded state.  Added to this, in 2017 the IAEA conducted a total of 35 complementary access inspections in Iran. These are sometimes referred to as “snap” inspections because of the shorter notice period given to the state prior to the inspection.  In May 2018, Iran’s nuclear program was the most closely monitored in the world, and it was vastly reduced in scope and activity as compared to 2015.

Nevertheless, this is precisely when U.S. President Donald Trump decided to “withdraw” the U.S. from the JCPOA, and to re-impose the economic sanctions on Iran that had been lifted under it.  Trump had long criticized the JCPOA as “the worst deal ever.”  He cited as his reasons for leaving the agreement the fact that it did not address Iran’s ballistic missile program or its regional behavior, and that most of Iran’s commitments under the JCPOA are limited to a 10-15 year period, after which Iran could resume its enrichment and other covered activities at pre-JCPOA levels.

Much has been written about the reasons for the U.S. withdrawal, and about its prudential character.  I would rather like to briefly discuss the issue of how best to characterize this action taken by the U.S.  Due to the legally nonbinding character of the JCPOA, one cannot use legal language to describe states’ interactions around the agreement. One cannot, for example, look to the rules of treaty law found in the 1969 Vienna Convention on the Law of Treaties to assist with things like interpretation of the agreement, implementation, breach, or termination.  These rules apply only to treaties, i.e. legally binding international agreements. This makes an analysis of the meaning and proper characterization of its purported withdrawal tricky.

But even with regard to legally nonbinding agreements among states, there is at least a principled difference between withdrawing from an agreement, on the one hand, and breaking it on the other. Perhaps it would be best, for analogical purposes, to use the language of contracting which Trump, the self-professed master of dealmaking, at least ought to understand from a principled perspective.  The essential function of a contract in a domestic legal system is to protect the parties’ reasonable dependence on the promises made by the other parties, when the actions recited by the promises do not occur simultaneously.  For example, the cable company will make me sign a contract to pay them over a set period of time in the future before they will undertake the up-front cost of giving me the necessary equipment and turning on the cable signal. If there were no contract in this situation, it wouldn’t make sense for the cable company to simply trust that I will make the agreed payments in exchange for their up-front costs.

The JCPOA, while not a legally binding treaty, was an international agreement comprised of reciprocal promises exchanged by the various parties.  It was a quid pro quo, i.e. Iran’s promises were made in exchange for the promised actions of the P5+1, and vice versa.  And as with the cable company hypothetical above, the actions recited by the parties’ promises were not to happen simultaneously.  In the case of the JCPOA, the bulk of Iran’s commitments – including the shipping out of the country of most of its stockpile of low-enriched uranium, the disassembling and shelving of two-thirds of its centrifuges, the permanent disabling of the Arak reactor, and the bringing into force of an Additional Protocol inspection agreement with the IAEA – were frontloaded to the beginning of the JCPOA’s term, and were in fact essentially accomplished by Iran by Implementation Day – January 16, 2016.  The commitments of the P5+1, on the other hand, were primarily to consist of the lifting of economic sanctions, and of keeping them lifted, for the entirety of the term of the JCPOA.  The essential agreement enshrined in the JCPOA was that Iran would implement the bulk of its promises first, in reliance on the promises made by the P5+1, including the United States, that they would in return implement the bulk of their promises over the succeeding 15 years.  And the U.S. was, of course, integral to the JCPOA coming into being. Because of its economic power, and the effect of U.S. economic sanctions, there would have been no agreement reached without the participation and commitment of the U.S.

Some observers have commented that lawful withdrawals from international agreements do occur, even when there has been no violation of their terms.  And that this is true even in the arms control area. They frequently cite to the U.S. withdrawal from the Anti-Ballistic Missile Treaty (ABM) treaty with Russia in 2002.  The difference in that case, however, was that there was a negotiated withdrawal clause in the ABM treaty, agreed by the parties.  The JCPOA, by contrast, does not contain a withdrawal clause. The asynchronous nature of the agreed timing of the commitments under the JCPOA made such a clause undesirable.

Because of the frontloaded nature of Iran’s commitments under the JCPOA, and its fulfilment of those commitments in reliance on the future fulfilment of the commitments of the P5+1, along with the fact that the JCPOA does not contain a withdrawal clause agreed among the parties, it would be more accurate in my view to characterize the U.S. action in May 2018 not as a withdrawal from the JCPOA, but as a simple breaking of its commitments made therein.  To be clear, I am not arguing that the U.S. violated international law through breach of a legally binding obligation. What I am saying is that the term “withdrawal” conveys a sense of justifiability which simply does not exist here.  There was no basis in the agreement or in its context of implementation on which to justify the U.S. decision to cease the fulfilment of its promises made under the JCPOA.  The U.S. simply decided to break those promises.

In my view, the JCPOA was in fact a tremendous victory for diplomacy. It was a novel, creative, intelligent solution to a problem in international relations that had – and still has – the potential to lead to war.  The U.S. had no principled basis for its decision to abandon the JCPOA. This action was not a violation of law but it was a, frankly, shameful refusal to abide by promises which the United States had made, and on which another state had reasonably relied. To term this action a withdrawal is to sugarcoat a simple breaking of promises.

Nothing in international law prevented President Trump from deciding to not honor the promises made by the United States in the JCPOA.  But let’s call it what it was. And it wasn’t a withdrawal.

Friend or Foe: The Treaty on the Prohibition of Nuclear Weapons and the NPT

I’m very pleased to host a guest post by Dr. Stuart Casey-Maslen. Stuart is Research and Policy Coordinator at the International Campaign to Abolish Nuclear Weapons (ICAN) and an Honorary Professor at the University of Pretoria’s Human Rights Centre. He holds a doctorate in international humanitarian law and master’s degrees in international human rights law and forensic ballistics. I’ve known Stuart for quite a while as one of the high quality people writing in the area of arms control law. He co-edited Nuclear Weapons under International Law, published by Cambridge University Press in 2014 (which included a chapter from me), and is the author of a legal commentary on the Treaty on the Prohibition of Nuclear Weapons, to be published by Oxford University Press in January 2019.

Friend or Foe? : The Treaty on the Prohibition of Nuclear Weapons and the NPT

By: Stuart Casey-Maslen, Research and Policy Coordinator, ICAN

Of all the attacks aimed at the TPNW by nuclear-weapon states (as well as some of the more militant umbrella states), one of the most persistent has been that the Treaty undermines the existing non-proliferation and arms control architecture, especially the 1968 Treaty on the Non-Proliferation of Nuclear Weapons (NPT). For instance, on the day the TPNW was adopted by 122 states in New York, France, the United Kingdom, and the United States issued a joint statement in which they argued that the new treaty risked “undermining the existing international security architecture which contributes to the maintenance of international peace and security” and reiterated their “continued commitment” to the NPT and to further promoting “its authority, universality and effectiveness”.

First and foremost, it is important to recall that the TPNW explicitly acknowledges and supports the NPT. A preambular paragraph dedicated to that treaty declares that the NPT is “the cornerstone of the nuclear disarmament and non-proliferation regime” and reaffirms that its “full and effective implementation” has “a vital role to play in promoting international peace and security”. That is an unequivocal endorsement. It also contains treaty language taken almost directly from Article 11 of the NPT; specifically the obligation in Article 1, paragraph 1 (c) never under any circumstances to “Receive the transfer of or control over nuclear weapons or other nuclear explosive devices directly or indirectly”.

For sure, there is no “get-out-of-jail-free” card in the TPNW as is claimed, seemingly for perpetuity, by the P5 “Nuclear Weapons States” named as such under the NPT (Article VI notwithstanding). All nuclear weapons by any state party must be withdrawn from operational status and destroyed. In addition, the assistance provision in the TPNW (Article 1, paragraph 1(e)) prohibits the provision of source or fissionable material to the P5 for the purpose of producing or maintaining nuclear weapons, which is not the case under the NPT. So the TPNW goes beyond the NPT in restricting vertical proliferation as well as horizontal proliferation. That additional layer of protection is a good thing (unless one happens to believe that the more nuclear weapons the P5 possess, the better for humanity; more on that issue below).

But in April 2018, in a written “Outline of Legal Risks for States Contemplating Joining” (the TPNW), officials from the United States doubled down on the rhetoric, claiming that the new Treaty was “a step backwards on nonproliferation verification by ignoring the Additional Protocol” and that it was “inconsistent with the practice under the NPT of achieving progress on disarmament pursuant to negotiated measures containing rigorous verification and reflecting the realities of the international security environment”.

Let’s look at these issues in turn and see if they have merit.

First, is the TPNW a step backwards in verification and does it ignore the Additional Protocol?

The TPNW obligates every state party to either maintain or — if it does not yet have one in place — to negotiate and bring into force with the International Atomic Energy Agency (IAEA) its Comprehensive Safeguards Agreement (INFCIRC/153 (Corrected)). This is at least as strong as, and certainly more specific than, the obligation in Article III of the NPT to accept safeguards on source or special fissionable material “with a view to preventing diversion of nuclear energy from peaceful uses to nuclear weapons or other nuclear explosive devices”. Moreover, around a dozen states parties to the NPT do not yet have a Comprehensive Safeguards Agreement in force. Surely, any other international instrument that serves to pressure those states into concluding such an Agreement is a good thing?

It is true that the TPNW does not obligate every state party to negotiate and bring into force the Additional Protocol on safeguards with the IAEA. But nor does the NPT. By the end of 2017, though, an impressive 132 states had an Additional Protocol in force with the IAEA.

For those not familiar with it, the Additional Protocol is a binding agreement with the IAEA granting the Agency additional inspection authority to that provided in the Comprehensive Safeguards Agreement. A principal aim is to enable the IAEA inspectorate to provide assurance about the accuracy and completeness of declared activities and the absence of undeclared activities. Under the Protocol, the IAEA is granted expanded rights of access to both information and sites. Activities carried out during such complementary access can include examination of records, visual observation, environmental sampling, use of radiation detection and measurement devices, and the application of seals and other identifying and tamper-indicating devices.

But the TPNW does go further than the NPT in safeguarding. Any state that owned, possessed or controlled nuclear weapons on 8 July 2017 and subsequently eliminated their programme prior to becoming a party to the TPNW must agree upon safeguards that are “sufficient to provide credible assurance of the non-diversion of declared nuclear material from peaceful nuclear activities and of the absence of undeclared nuclear material or activities in that State Party as a whole” (Article 4, paragraph 1, TPNW). In practice, this means the Additional Protocol, even if the instrument isn’t cited explicitly in the text. In this regard, therefore, the TPNW goes significantly beyond the requirements laid down in the NPT.

Now let’s consider whether the TPNW is inconsistent with the so-called “practice” under the NPT. Massive reductions in their nuclear arsenals have been achieved by Russia and the United States since the apogee in the 1980s. Bold agreements (the INF Treaty, START, and New Start) were concluded bilaterally by successive presidents to reduce their stockpiles and make the world a safer place. In recent years, though, this trend has stalled and “modernisation” has become the new buzz word. More “usable” nuclear weapons are being developed. These new trends are provoking a new nuclear arms race at a time when societies can ill afford it.

So what is actually inconsistent is the practice of the P5 with respect to their legal obligations under the NPT. Article VI is explicit: each state party to the NPT “undertakes to pursue negotiations in good faith on effective measures relating to cessation of the nuclear arms race at an early date and to nuclear disarmament”. The International Court of Justice went further in its 1996 Advisory Opinion on the Threat or Use of Nuclear Weapons, holding unanimously that “There exists an obligation to pursue in good faith and bring to a conclusion negotiations leading to nuclear disarmament in all its aspects under strict and effective international control.” (Dispositif F, added emphasis)

The United States has expressed support for a nuclear-free world, but has also committed to maintaining an arsenal for deterrence and defence as long as nuclear weapons exist. So where are the negotiations to make these weapons a thing of the past? A new arms race is a clear violation of Article VI. The 122 states that adopted the TPNW were living up to their NPT obligations. The P5 are not.