How the war in Ukraine impacts the law and politics of security guarantees

The paradoxical revival of security guarantees

Russia’s full-scale invasion of Ukraine in 2022 has revived the concept of security guarantees. Their content and scope have been central to scholarly and political debates about Kyiv’s resistance to the Russian aggression (see here and here).

Yet, this revival highlights a paradox: if existing arrangements had been effective, there would be no need for new ones. The 1994 Budapest Memorandum—signed by Ukraine, Russia, the UK, and the USA when Kyiv joined the Non-Proliferation Treaty, and registered as an international instrument with the UN in 2014—has proven toothless. Russia has violated its commitment to refrain from threatening or using force against Ukraine’s territorial integrity or political independence, and attempts by the other signatories to activate the UN Security Council to assist Ukraine have come to nothing—not least due to Russia’s veto.

As States attempt to design new security guarantees for Ukraine and adapt Europe’s security architecture to new geopolitical realities, this paradox persists, raising fundamental questions about the nature and function of security guarantees.

What’s in a name?—A tentative typology

The concept of ‘security guarantees’ is notoriously elusive. It is not a legal term with an authoritative—let alone statutory—definition. Rather, it is used as a generic label by practitioners and scholars to refer to certain types of international peace and security arrangements. This blog post treats security guarantees as an umbrella term for formal and informal, bilateral or multilateral arrangements aimed at maintaining or restoring peace and security under specific conditions.

Regardless of their form, this post categorises security guarantees in two ways: by nature, and by function. In terms of their nature, security guarantees can consist of either positive or negative commitments. In the latter case, actors commit to refrain from doing something. For instance, States that possess nuclear weapons may commit not to use them against non-nuclear weapon States. Certain provisions of the Budapest Memorandum would fall into this category. These types of ‘negative’ commitments are often referred to as ‘security assurances’ (see here and here). In the former scenario, actors commit to taking action under certain circumstances, including by military means. A classic example of a ‘positive’ security guarantee is the commitment to collective self-defence in the event of an armed attack, as set out in Article 5 of the North Atlantic Treaty and Article 42(7) of the Treaty on European Union (TEU).

This post further ascribes at least three functions to security guarantees. Firstly, they act as a deterrent: their mere existence is intended to prevent conflict. This category includes both multilateral and bilateral mutual defence arrangements, such as the 2019 Treaty of Aachen between France and Germany and the 2021 Franco-Greek strategic partnership accord. It also encompasses unilateral pledges, including those made by various NATO allies to Finland and Sweden between their application for NATO membership in 2022 and their subsequent accessions in 2023 and 2024. Secondly, security guarantees can be supportive. They can offer aid and assistance in military, financial, economic, humanitarian, or technical areas. Notably, military support does not necessarily involve the deployment of troops for combat purposes, but can consist of delivering weapons, sharing intelligence, or providing training. Thirdly, security guarantees can contribute to post-conflict stabilisation and reconstruction, thereby having a restorative dimension. Peace operations, for instance, can fulfil this function. Their mandates may entail stabilisation tasks (e.g. KFOR in Kosovo), disarmament and arms control (e.g. MINUSCA in the Central African Republic or MONUSCO in the Democratic Republic of the Congo), and state-building activities (e.g. ONUSAL in El Salvador).

A very brief history of security guarantees

Examining the history of security guarantees reveals that their design and implementation depend on each era’s political, legal, and cultural context. For centuries, they were primarily centred on providing tangible collateral. From ancient times until the late 18th century, individuals served as securities in the form of hostages. Dynastic marriages were also a widespread form of guarantee—think of Marie-Antoinette and Louis XVI. Another common practice was depositing land (including its inhabitants) as collateral—take the stipulations in the 1919 Versailles Treaty (Article 428 et seq.) on the conditional presence or withdrawal of occupation forces on German territory situated to the west of the Rhine.

Especially since World War II, the practice of security guarantees has evolved from physical collateral and bilateral or trilateral security arrangements to more abstract commitments unfolding within a multilateral framework. Examples thereof are multilateral third-party commitments to provide aid and assistance in the event of armed aggression, as set out in Article 5 of the NATO Treaty. Nevertheless, the physical dimension of security guarantees has not entirely disappeared. Often, a third-party guarantee is reinforced by a physical element, such as the deployment of troops or the long-term lease of land for foreign military bases. This is intended to reduce the inherent ambiguity of security commitments and make collective self-defence more likely in the event of an attack. Following this line of reasoning, former German Chancellor Konrad Adenauer insisted that US troops be stationed in Germany to give the transatlantic security guarantee a physical and human dimension—not least because NATO’s Article 5 leaves alliance members considerable leeway in how they fulfil their mutual defence obligation. Having ‘boots on the ground’ after a ceasefire is also an essential element of future security guarantees for Ukraine, as the Franco-British commitment of January 2026 shows.

The end of the Cold War marked another turning point with the emergence of a new form of credible third-party commitment—multidimensional peace operations—which reflected the belief in multilateralism espoused by liberal institutionalists. Peace operations conducted under the auspices of the UN gained particular traction, with mandates extending beyond ‘traditional’ observational peacekeeping tasks to encompass disarmament, arms control, and state-building activities.

While this brief overview suggests that the abstract dimension of security guarantees became incrementally more important than their physical backstop, the war in Ukraine seems to reverse this trend. Although abstract commitments (e.g. EU financial aid) remain widespread, physical involvement has regained prominence. This is not a return to the past, but rather a hybrid model blending the old with the new. However, before examining past and potential future security guarantees for Ukraine, it is important to consider their legal and political particularities.

The law and politics of security guarantees

Although the most frequently referenced type of security guarantee is bilateral or multilateral conventional arrangements on mutual defence, such as NATO and EU mutual defence commitments, the law of security guarantees is far more varied. Some security guarantees are based on unilateral acts, such as the aforementioned security pledges made by various NATO members (e.g. the Nordic States or the UK) to Finland and Sweden on their path to NATO membership. Others are rooted in the mandates of peace operations. Still others are based on administrative agreements or contracts concerning, for instance, the lease of land for a third country’s military base. In short, the law of security guarantees is multifaceted.

However, despite the pacta sunt servanda principle and the standard of good faith applying, most formal security guarantees are not directly enforceable or justiciable in the absence of a global enforcement body. Nevertheless, States receiving security guarantees generally favour legally binding arrangements over political declarations, as demonstrated by Ukraine both in the past and present. Yet, higher degrees of formality do not necessarily lead to higher rates of compliance. Therefore, the quest for formal arrangements is one of the paradoxes associated with the law of security guarantees.

Furthermore, legally speaking, all codified security guarantees are considered equal and equally binding. However, this supposed legal equality does not hold from a political perspective, presenting another paradox. Indeed, politically, security guarantees can be ranked according to their potential for deterrence and credibility. Those involving military support, ideally in the form of a collective defence obligation, are considered stronger, particularly when provided by a nuclear (super-) power, as the nuclear deterrent is believed to increase the potential costs incurred by an aggressor. Finland and Sweden’s application to join NATO exemplify this quest for a multilateral, nuclear-backed security guarantee. However, even a collective defence arrangement involving one or several nuclear powers only serves its deterrent function if it is credible.

Different schools of thought offer different answers to the question of what constitutes a credible security guarantee (see here). Realists argue that it is material (i.e. military) strength and demonstrable interest. Liberal institutionalists, on the other hand, claim that credibility is based on institutions, law, interdependence, and reputational costs. Constructivists, meanwhile, emphasise the importance of shared beliefs, norms and trust. Despite these differences, international relations scholars agree that the reliability of a security guarantee is relational, not legal. This brings us back to the first paradox: while the real-world reliability of security guarantees is primarily determined by factors beyond the law, those receiving them usually prefer legally binding commitments.

Security guarantees for Kyiv: continuities and change

When it comes to Ukraine, we find a variety of security guarantees, most of which involve formal or informal positive commitments that are both supportive and restorative. This is because many were introduced after Russia’s aggression began—and the conflict is ongoing. Nevertheless, the existing security guarantee scheme for Ukraine is remarkably multidimensional, covering military, financial, humanitarian, technical, and economic aspects. As the EU and its Member States, together with the UK, are currently the biggest supporters of Kyiv, the ensuing analysis will focus on these entities acting as guarantors of Ukraine’s present and potential future security guarantees.

In the ‘supportive’ dimension, two types of guarantees stand out. Firstly, direct bilateral and multilateral aid and assistance arrangements are being implemented in the military realm (e.g. the delivery of lethal weapons, military training, and intelligence sharing), the macro-financial area (e.g. loans and grants), and the humanitarian field (e.g. healthcare and evacuation). Notably, as US support waned, the EU and its Member States significantly scaled up their support, with EU institutions taking the lead on financial and humanitarian aid. In 2025, 90 % of financial aid to Ukraine came via EU loans and grants—and this will most likely remain unchanged in the foreseeable future. Military aid, by contrast, is becoming increasingly concentrated, with Northern and Western Europe contributing around 95 %—particularly from Scandinavia, Germany, and the United Kingdom—while allocations from Southern and Eastern Europe continue to fall. Secondly, the EU runs two peace operations: one military (EUMAM) and one civilian (EUAM). These operations support both military actors and civilian structures in withstanding the Russian attack. EUMAM, for example, has to date trained over 90,000 Ukrainian soldiers. In addition, the EU has adopted 20 sanctions packages in response to the Russian aggression, which are unprecedented in scope (see here for an in-depth analysis). Indeed, countermeasures can also constitute security guarantees as they are intended to weaken the aggressor.

In terms of ‘restorative’ security guarantees, there are bilateral and multilateral financial, material, and technical arrangements to help Ukraine repair its energy infrastructure and physically protect nuclear power plants (see here). Additionally, the EU has pledged substantial support for Ukraine’s reconstruction (see here for an analysis). As the conflict is ongoing, disarmament and arms control are not yet being discussed, but are likely to become important agenda item in the future. Regarding future ‘restorative’ security guarantees, France and the UK signed a declaration of intent in January 2026 to deploy troops to Ukraine once a peace deal has been reached with Russia. A US commitment to a military backstop is still missing. 

This brings us to the ‘deterrence’ function of security guarantees, as the deployment of a stabilisation force can have both a restorative and deterrent function. Currently, third parties are reluctant to offer Ukraine either ‘boots on the ground’ or a formal mutual defence arrangement. Indeed, Kyiv’s accession to NATO is regularly questioned (see, for instance, here), and discussions on how to deal with Article 42(7) TEU in the event of accession are ongoing. Instead, following a G7-declaration on supporting Ukraine, almost a dozen European States have signed non-binding ‘security agreements’ with Ukraine containing stipulations on coordination and assistance in case of a future attack. Only time will tell whether such agreements will have sufficient deterrent potential.

Importantly, a new type of security guarantee is emerging in relation to Ukraine that cuts across these three functional categories: defence industrial integration. Since 2022, a series of legislative measures—with the acronyms ASAP, EDIRPA, EDIP, and SAFE—have been adopted at the EU level to strengthen the joint production and procurement of defence materials (for an analysis, see here), with Ukraine enjoying a special status in these programmes. This directly links Ukraine’s defence industrial capacity and, ultimately, defence capabilities, with those of EU States, creating a two-way security guarantee that serves supportive, restorative, and deterrent purposes simultaneously. Therefore, this analysis posits that this emerging defence industrial intertwining represents a novel form of physical interconnection, based on economic and industrial interdependence rather than troop deployments or base leases.

Interestingly, the Ukrainian case provides empirical insights for all three theoretical perspectives introduced earlier. The preference for nuclear-backed guarantees, as evidenced in the push for NATO membership, aligns with the realist emphasis on material power. Meanwhile, the EU’s multidimensional support—financial, humanitarian, and industrial—reflects the liberal institutionalist faith in interdependence and reputational costs. And while both the multidimensional support and the industrial imbrication illustrate constructivism by underlying trust and shared norms, the reluctance of many EU Member States to offer Ukraine a fast-track EU accession suggests insufficient confidence.

What Ukraine teaches us for designing security guarantees in a multipolar world

The war in Ukraine revitalised the concept of security guarantees. In the absence of a formal collective defence commitment, we have witnessed the proliferation of bespoke arrangements offering various security guarantees, some formal and some informal. The EU’s defence industrial integration with Ukraine is a prime example of this, adding a new integrative dimension to the traditional purposes of security guarantees.

This development reflects two broader shifts. Firstly, security guarantees are becoming more tailored and pragmatic, as evidenced by the modular nature of the arrangements for Ukraine, which are designed to address specific threats and needs. Secondly, they are becoming regionalised in a multipolar world where global consensus is elusive, but regional solidarity remains achievable. France’s recent offer to extend its nuclear umbrella to interested European partners is indicative hereof. For legal scholars and policymakers, the lesson is clear: the future of security guarantees will not be shaped by grand treaties (alone), but by the cumulative effect of pragmatic, modular—and probably regional—solutions.

Yet, the central paradox endures. States continue to pursue new guarantees despite their historical limitations, often preferring formal commitments even when it is factors beyond the law that determine reliability. Ukraine’s experience suggests that the most effective guarantees may blend legal form with pragmatic feasibility and reliability, whether through physical backstops, regional frameworks, or industrial interdependence. In this sense, the war has not made security guarantees obsolete but has redefined how they operate.

This blogpost forms part of the Mini-symposium on international legal issues concerning a peace treaty for Ukraine. The introduction to the symposium by Anne Peters, Gregory Fox, and Florian Kriener can be found here. The symposium builds on presentations and discussions during the Conference on International Law and Peace in Ukraine organized at the Max Planck Institute for Comparative Public Law and International from April 17-18, 2026 and supported by the German Foreign Office.