Targeting Nuclear Facilities During War: What Does International Law Have to Say?
Since it was seized by Russia in early 2022, Ukraine’s Zaporizhzhia nuclear power plant has seen several attacks, a fire in one of its cooling towers, and most recently, a drone strike near its perimeter. The Chornobyl Nuclear Power Plant in Ukraine was similarly seized by Russia, though these troops have since withdrawn. The International Atomic Energy (IAEA) has meanwhile expressed concern about the risks to the Kursk Nuclear Power Plant in light of Ukraine’s incursion into Russia. Moscow has also alleged that Kyiv attempted to attack the plant. Sadly the war has gone on long enough for some of these frightening developments to be viewed as par for the course. Mindful of this normalization, the IAEA has repeatedly urged restraint and warned of the catastrophic humanitarian consequences if these nuclear power plants are significantly compromised. Such calls for restraint are anchored in an international legal discourse on preventing military attacks against nuclear facilities that precede this war — and Russian occupation of Zaporizhzhia — by several decades. These legal measures emerged from an interplay of geopolitical consensus, intent, and a specific set of preconditions. They each have unique characteristics that distinguish them, observed most clearly in the scope of their application. Finally, all offer takeaways for the Russia-Ukraine war.
Legal obligations prohibiting attacks on nuclear facilities
Legal obligations prohibiting attacks on nuclear facilities can be divided into three broad categories: international, regional, and bilateral. The chief provisions at the international level, which paved the way for a lot of the subsequent discourse around ‘non-attack’, can be traced back to Additional Protocols (APs) I and II to the Geneva Conventions of 1949. Both APs were adopted in 1977. These discussions over time made their way into IAEA resolutions as well. The 1996 African Nuclear-Weapon-Free-Zone Treaty, or Treaty of Pelindaba, contains clauses specific to non-attack and is the only such instrument in the regional category. Finally, the 1988 India-Pakistan Non-Attack Agreement, known formally as the Agreement between India and Pakistan on the Prohibition of Attacks Against Nuclear Installations and Facilities, inhabits the bilateral category. Again, it is the only agreement of its kind in existence.
Apart from their obvious convergence on purpose, which is to prohibit attacks on nuclear facilities, these instruments have three things in common. They were each motivated by political consensus on the humanitarian consequences of war, an intent to prevent or limit these consequences, and a set of circumstances that demonstrated their legal necessity, thus compelling them into being. Geopolitical changes such as decolonization, the Cold War, fear of nuclear use, and the changing face of modern warfare shed light on the shortcomings of the Geneva Protocols of 1949, and provided the context for the APs to be negotiated. The other measures, too, were borne in similar circumstances. The APs sought to regulate combatant behaviour and safeguard civilians in non-international armed conflict.
The IAEA General Conference adopted GC(XXXI)/RES/475 on the “Protection of nuclear installation against armed attacks” in 1987; it is one of several IAEA resolutions on the subject Preparations for the Pelindaba Treaty began in 1991, one year after South Africa dismantled its nuclear weapons programme. The treaty derives its name from South Africa’s main nuclear research centre, Pelindaba. It signalled a near-continental consensus on securing Africa from the effects of nuclear weapons. The India-Pakistan Non-Attack agreement, too, reflects consensus between adversaries on the humanitarian fallout of nuclear facilities as military targets. One of the agreement’s precipitating factors was Pakistan’s fear of reported Indo-Israeli plans to attack its nuclear facilities at Kahuta. In the agreement’s broader political context, ‘non-attack’ was one of a series of issues under discussion between India-Pakistan in the period 1985-1989, in the interest of resetting ties. That New Delhi and Islamabad signed the agreement thus indicates a cognizance of shared risks, perhaps particularly as both India and Pakistan were, at the time, undeclared nuclear states.
Convergences and divergences in existing legal measures
Despite their commonality of intent, these measures diverge in their scope of application. This is most powerfully illustrated by using the India-Pakistan agreement as a point of reference: it is the narrowest in scope and displays the greatest degree of divergence from the others. The most obvious, of course, is that it is a bilateral agreement, while the others are broadly scoped international guidelines or a continent-wide instrument. The India-Pakistan agreement is a short document with one sole purpose: the two parties’ commitment to refrain from attacking each other’s nuclear installations. The APs and Pelindaba are very broad, with non-attack featuring as one element of a large set of nuclear-relevant stipulations. The APs, as foundational blocks of international humanitarian law, enshrine international norms of behaviour. Pelindaba lays the conditions for Africa to be recognized as a nuclear weapon-free-zone. Another difference—and this leads to analogous readings with Russia-Ukraine—is that the India-Pakistan agreement is between adversaries engaged in territorial conflict, albeit both India and Pakistan are nuclear-armed, whereas in the Russia-Ukraine case, only one is.
Somewhat confusingly, the APs contain both specificity and generality in their application, which could make them weaker than intended. They speak of prohibiting attacks on critical infrastructure “containing dangerous forces,” including nuclear facilities, dams, and dikes during armed conflict. Within nuclear facilities, however, its stipulations extend only to “nuclear electrical generating stations” or civilian nuclear power plants. While the India-Pakistan agreement is focused exclusively on “nuclear installations and facilities,” this emphasis is unqualified — at least on paper — thus technically guaranteeing protection to the adversary’s entire nuclear enterprise. Pelindaba, too, adheres to a much wider understanding of nuclear facilities than the APs, i.e. “any other installation or location in or at which fresh or irradiated nuclear material or significant quantities of radioactive materials are present.”
The APs also set out conditions under which the prohibition on nuclear power plants as military targets would cease to apply. This significant caveat is not found in any of the other measures. Despite these limitations, though, the APs assume chronological significance. Article 5, clause 6 of AP I calls for the conclusion of “further agreements [among themselves] to provide additional protection for objects containing dangerous forces”— an exhortation later mirrored in Pelindaba, the 1987 IAEA resolution, and the India-Pakistan agreement. Notably, the 1987 IAEA resolution also observed the “urgency of concluding an international agreement.” It recommended reviving a draft multilateral treaty prohibiting military attacks on nuclear facilities that the Conference on Disarmament (CD) had reportedly put together in the early 1980s. While the CD did not proceed beyond the initial draft compilation, it created context and familiarity for discussions that led to the 1987 IAEA resolution.
Why political consensus must drive legal necessity
Zaporizhzhia’s occupation is the first time in history that a working nuclear power plant—Europe’s largest—has been subject to military takeover by an adversary. Since then, the India-Pakistan agreement has been frequently invoked for emulation and adaptation. Policy and scholarly debates have recommended strengthening the norm and universality of non-attack. Yet, the political consensus that germinated the APs, IAEA resolutions, Pelindaba, or the India-Pakistan agreement are nowhere to be seen. The greatest contemporary utility therefore of these legal measures — from the broadest to the narrowest — is to be found in the political circumstances that surrounded them, even more perhaps than their text and stipulations. The symbolic value of adversaries having a shared sense of risk is their most consequential outcome, and until this understanding is approximated by the warring parties, it is difficult to foresee norm strengthening or new legal instruments. Ultimately, even the most exacting international laws can be overturned by bad faith.
About the Author
Ruhee Neog is the Director of the Institute of Peace and Conflict Studies (IPCS) in India. Her research looks at the role of norms, language, and institutions in shaping choices and behaviour in international relations. She specialises in Indian foreign policy and nuclear strategy. At IPCS, Ruhee leads the design and execution of projects across the foreign policy spectrum, including Track 1.5 and Track 2 dialogues. She holds additional appointments as Board Member of the International Nuclear Security Forum hosted by Stimson Center and Fellow with Sandia National Laboratories at the US Department of Energy. Past positions include a dual fellowship with the Project on Managing the Atom and International Security Programme at Harvard Kennedy School’s Belfer Center; South Asian Voices Fellowship at Stimson Center; and research fellowship supported by the Nuclear Threat Initiative (NTI). Ruhee has a postgraduate degree in History of International Relations from the London School of Economics and an undergraduate degree in Literature from St. Stephen’s College, Delhi University.
The opinions articulated above represent the views of the author(s) and do not necessarily reflect the position of the Asia-Pacific Leadership Network or any of its members. APLN’s website is a source of authoritative research and analysis and serves as a platform for debate and discussion among our senior network members, experts, and practitioners, as well as the next generation of policymakers, analysts, and advocates. Comments and responses can be emailed to apln@apln.network.
Image credit: A view of the cooling towers at Zaporizhzhia Nuclear Power Station near Enerhodar city, Ukraine, before Russia’s invasion. Photo by Ihor Bondarenko / Shutterstock.
