International Law for a Fragile World: Global Risk and the Limits of Classical International Law

[Dr Sergey Sayapin is Professor of Law at KIMEP University (Almaty, Kazakhstan) and Distinguished Visiting Global Scholar at the NUS Centre for International Law (2025)]

International law was largely designed for a world in which harm could be identified, responsibility attributed, and violations remedied. Its core concepts – breach, obligation, responsibility, reparation – presuppose a legal universe structured around discrete acts, identifiable actors, and reasonably linear chains of causation. Today’s global order no longer fits that model. Climate collapse, pandemics, technological disruption, financial instability, and democratic erosion are not isolated events caused by individual wrongdoing. They are systemic risks produced by deeply interconnected social, ecological, and technological systems. Yet international law continues to operate as if uncertainty were an exception rather than a defining condition of global life.

This post argues that the difficulty international law faces in responding to contemporary crises is not primarily a problem of non-compliance or weak enforcement. Rather, it reflects a deeper structural mismatch between classical legality and the nature of global risk in the twenty-first century. International law was built to regulate conduct – today it is increasingly asked to govern conditions. It was designed to respond to harm after it occurs – it is now confronted with threats that must be anticipated, managed, and mitigated long before responsibility can be meaningfully assigned. If international law is to remain relevant in a fragile world, it must be reconceptualised not merely as a system of rules and prohibitions but as a framework for global risk governance.

From Responsibility to Risk

At the heart of classical international law lies a model of legality centred on responsibility: states assume obligations, breaches occur, responsibility follows. This architecture is visible across doctrinal fields, from the law of state responsibility to international criminal law, human rights law, and even environmental law. While these regimes differ in scope and technique, they share a common assumption: that law’s primary task is to identify unlawful conduct and attach legal consequences to it.

Global risk unsettles this assumption. Risks are not events but probabilities. They do not arise from a single act but from the cumulative operation of complex systems – economic, technological, ecological, and political. Climate change is not caused by one state’s emission decision; pandemics are not the result of a single regulatory failure; financial crises are not traceable to one unlawful transaction. In such contexts, responsibility becomes diffuse, causation opaque, and temporality non-linear. Harm unfolds gradually, often irreversibly, and responsibility – when it can be established at all – comes too late to be meaningful.

The COVID-19 pandemic offers a stark illustration. Long before the virus spread globally, epidemiologists had warned of pandemic risk arising from dense urbanisation, global travel, industrial farming, and weak health systems. Yet international law had little to say about these structural vulnerabilities. When the crisis hit, legal debates rapidly turned to questions of border closures, emergency powers, vaccine access, and potential responsibility for delayed reporting. These questions mattered – but they addressed the consequences of failure rather than its systemic causes. The legal architecture proved far more comfortable assigning blame than organising preparedness.

This is not to suggest that responsibility has become irrelevant. Rather, it has become insufficient. A legal order oriented primarily toward ex post accountability struggles to address ex ante vulnerability. As a result, international law often finds itself reacting belatedly to crises whose basic contours were foreseeable, even well-documented, long before they materialised.

Risk as a Structural Condition of Global Order

The problem is compounded by the tendency to treat global crises as exceptional disruptions rather than as manifestations of a deeper condition of fragility. Legal discourse frequently frames climate change, pandemics, or technological threats as “challenges” to an otherwise stable international order. Yet what defines the contemporary global system is not stability punctuated by crisis but permanent exposure to systemic uncertainty.

Modern global risks share several defining features. They are transboundary, operating across jurisdictions and generations. They are cumulative, produced by countless individually lawful actions whose combined effects are catastrophic. They are non-linear, marked by tipping points and feedback loops rather than gradual escalation. And they are largely irreversible once certain thresholds are crossed. These characteristics do not sit easily with legal doctrines built around territorial jurisdiction, individual consent, and retrospective adjudication.

Climate change again provides the clearest example. For decades, scientific consensus warned of rising temperatures, extreme weather, and ecological tipping points. Yet international law struggled to translate this knowledge into binding obligations commensurate with the risk. The result has been a proliferation of targets, pledges, and reporting mechanisms that acknowledge danger without decisively constraining its sources. The international legal system did not fail because states secretly violated climate law – it failed because lawful behaviour, aggregated across time and space, pushed planetary systems toward collapse.

Seen in this light, fragility is not an aberration but a structural condition of global governance. International law is not failing because states occasionally violate its rules – it is struggling because the world it seeks to regulate no longer conforms to the assumptions embedded in its legal form. The persistence of this mismatch risks rendering international law reactive, marginal, and normatively hollow.

The Limits of Prohibition and Sanction

One response to this diagnosis has been to call for stronger rules and tougher enforcement. From climate law to cyber governance, reform proposals often emphasise clearer obligations, more robust compliance mechanisms, and enhanced sanctions. While such measures may be necessary in specific contexts, they do not address the core problem. Prohibition and sanction presuppose that undesirable outcomes can be prevented by forbidding particular acts. Global risks, however, often arise not from prohibited behaviour but from socially embedded practices that are economically rational, politically entrenched, and legally permitted.

Financial crises illustrate this dynamic particularly well. The 2008 global financial crisis was not triggered by a single illegal act but by the lawful accumulation of leverage, risk-taking, and regulatory arbitrage across interconnected markets. International economic law proved far better at disciplining state intervention after the fact than at governing systemic vulnerability beforehand. By the time responsibility could be debated, the damage had already cascaded globally. Similar patterns are emerging in the governance of artificial intelligence. The most significant risks posed by AI – opacity, bias, autonomy, and concentration of power – do not arise from clearly unlawful conduct. They emerge from rapid innovation, competitive pressures, and the scaling of technologies whose societal effects are only partially understood. A legal order focused on sanctioning misuse after harm occurs risks perpetually lagging behind technological reality.

From Attribution to Anticipation

What global risk demands is not the abandonment of legality but its reorientation. Instead of asking primarily who is responsible after harm occurs, international law must increasingly ask how harm can be anticipated, mitigated, and distributed fairly before it materialises. This requires a shift from attribution to anticipation.

Anticipatory governance is not foreign to international law. Elements of it already exist in precautionary principles, environmental impact assessments, early-warning mechanisms, stress testing, and reporting obligations. Yet these tools remain peripheral, often treated as technical supplements rather than as core elements of legal design. Moreover, they are frequently constrained by sovereignty and consent, limiting their capacity to address genuinely global risks.

A risk-oriented international law would place foresight and resilience at its centre. It would treat uncertainty not as a defect to be eliminated but as a condition to be managed. It would accept that legality in a fragile world must be adaptive, iterative, and capable of learning from near-misses as well as disasters. Crucially, it would recognise that waiting for breach is often a recipe for irreversible harm.

Fragility as a Legal Concept

To make this shift intelligible, international law requires a more explicit engagement with the concept of fragility. Fragility is not merely vulnerability to harm: it is the condition of systems whose stability depends on constant maintenance and whose failure produces cascading effects. Ecological systems, global supply chains, financial markets, health systems, and information ecosystems all exhibit this quality.

Treating fragility as a legal concept has several implications. First, it challenges the assumption that stability is the natural baseline of international order. Second, it reframes prevention not as exceptional intervention but as a routine obligation. Third, it foregrounds distributive questions: who bears the costs of risk-taking, who benefits from it, and who is protected when systems fail. Fragility also has a temporal dimension. Many global risks unfold across generations, raising profound questions of intergenerational justice that classical doctrines of reciprocity and consent struggle to accommodate. A legal order that takes fragility seriously must grapple with obligations owed not only to other states but to future populations whose interests cannot be represented through traditional diplomatic mechanisms.

From Crisis Management to Risk Governance

None of this suggests that international law must abandon its foundational principles. Sovereignty, consent, responsibility, and legality remain indispensable to global order. But they can no longer operate as the sole organising concepts of international law in a world defined by systemic risk. A legal order designed primarily to respond to breach will always arrive too late when harm is cumulative, diffuse, and irreversible. In such circumstances, effectiveness lies not in stronger condemnation after failure, but in earlier intervention before catastrophe becomes inevitable.

The central challenge, therefore, is not enforcement but orientation. International law must learn to govern conditions rather than merely sanction conduct, to manage vulnerability rather than simply assign blame, and to anticipate disruption rather than wait for its legal crystallisation. This requires a shift in emphasis from prohibition to precaution, from reaction to resilience, and from retrospective responsibility to prospective governance. It also requires acknowledging that uncertainty is no longer a marginal problem to be solved, but a permanent feature of global life to be organised.

Crucially, this is not a call for technocratic de-legalisation or for replacing international law with technical expertise. On the contrary, governing risk is an inherently normative task. Decisions about acceptable levels of risk, distribution of vulnerability, and protection of future generations are deeply political and moral choices. International law cannot avoid these questions by retreating into formalism. If it does, those choices will be made elsewhere – by markets, technologies, or geopolitical power – without transparency or accountability.

Reimagined in this way, international law becomes neither a utopian project nor a mere crisis manager. It becomes a framework for collective foresight: a means of structuring cooperation under conditions of uncertainty, of embedding precaution into legal design, and of building institutions capable of learning before disaster strikes. Its success should be measured not only by compliance or adjudication, but by whether foreseeable harms are reduced, resilience strengthened, and catastrophic thresholds avoided.

The risk, if international law fails to evolve, is not simply ineffectiveness. It is irrelevance. A legal system that speaks most clearly only after systems collapse will gradually lose its authority in shaping the future. Conversely, a legal order that takes fragility seriously – one that treats anticipation as a core legal function rather than an auxiliary concern – can remain central to global governance even in an unstable world.