International Law for a Fragile World: Political Risk – Authoritarianism, Information Warfare, and Democratic Fragility (Part I)

International Law for a Fragile World: Political Risk – Authoritarianism, Information Warfare, and Democratic Fragility (Part I)

[Dr Sergey Sayapin is Visiting Principal Research Fellow at the NUS Centre for International Law and Professor of Law at KIMEP University (Almaty, Kazakhstan; on leave).]

The preceding posts in this series have examined fragility in ecological, technological, human, and economic systems. All of these forms of risk, however, converge upon a further system whose resilience international law has largely taken for granted: the political institutions through which societies decide how every other risk is to be governed. Climate adaptation requires governments capable of planning beyond the electoral cycle; pandemic preparedness depends upon public trust in credible institutions; technological governance presupposes authorities able to regulate powerful private actors; and economic resilience requires governments capable of making contested distributive choices under uncertainty. When political institutions themselves become fragile, the capacity to govern every other category of risk deteriorates with them. Democratic erosion should therefore be understood not merely as a problem of constitutional design or domestic political organisation but as a form of systemic risk – and, in a sense, as the meta-risk of a fragile world.

The empirical evidence is difficult to dismiss. Freedom House reported in March 2026 that global freedom declined in 2025 for the twentieth consecutive year, with 54 countries deteriorating and only 35 improving; the share of the world’s population living in countries rated “Free” has fallen from 46 to 21 per cent in two decades. International IDEA’s Global State of Democracy 2025 found that 94 of the 173 countries it assesses – 54 per cent – had declined in at least one key dimension of democratic performance compared with five years earlier, that its indicator of credible elections had fallen to its lowest level in thirty years, and that press freedom had suffered its most far-reaching decline since the dataset began in 1975. The V-Dem Institute’s Democracy Report 2026 counts 92 autocracies against 87 democracies and estimates that almost three-quarters of humanity now lives under autocratic rule. These assessments differ in method, but they converge on a single conclusion: decline is no longer confined to fragile or transitional states. It is a structural feature of the contemporary international order.

This two-part post argues that international law is insufficiently equipped to understand this phenomenon because it protects democracy primarily through individual rights, electoral standards, and the prohibition of unlawful external interference. These protections remain indispensable. Yet democratic fragility increasingly develops through processes that are cumulative rather than episodic, informational rather than territorial, and frequently lawful when each measure is considered in isolation. Part I examines how democratic erosion operates and why the architecture of international law struggles to capture it. Part II turns to information warfare, the paradox of democratic self-defence, and the contours of a legal approach to democratic risk governance.

From Breakdown to Erosion

Classical legal analysis encourages us to look for identifiable acts. Was an election fraudulent? Was an opposition leader arbitrarily detained? Was a newspaper censored? Did a foreign state coerce another in matters of its political choice? These questions matter enormously, but they increasingly obscure the dominant pattern of democratic decline. For much of the twentieth century, democracies typically died abruptly – through military coups, the suspension of constitutions, or the cancellation of elections. As Nancy Bermeo observed a decade ago, these “flagrant” forms of backsliding have given way to subtler ones: executive aggrandisement, in which elected leaders incrementally dismantle checks on their power through formally legal means, and strategic electoral manipulation, in which elections continue to be held but are tilted in advance. Steven Levitsky and Lucan Way captured the resulting regime type as competitive authoritarianism – systems in which democratic institutions are real enough to be the principal means of acquiring power, yet so abused by incumbents that competition is genuine but unfair. Anna Lührmann and Staffan Lindberg have shown that this gradual, legally camouflaged mode of autocratisation is what distinguishes the “third wave of autocratisation” from its predecessors.

Constitutional scholars have identified the legal techniques involved. David Landau’s notion of abusive constitutionalism describes the use of constitutional amendment and replacement to make incumbents difficult to dislodge. Kim Lane Scheppele’s autocratic legalism captures the use of electoral mandates to enact a legal programme that systematically disables the institutions capable of restraining power. Her earlier image of the “Frankenstate” is particularly instructive for international lawyers: a monstrous whole assembled from parts each of which, considered individually, can be found in some respectable democracy elsewhere. A judicial appointments procedure, a media regulator, an electoral law, a reform of the prosecution service – each can be defended by reference to comparative practice. It is their combination that produces a system no longer capable of correcting itself.

Hungary over the past sixteen years illustrates this dynamic with unusual clarity. Following its 2010 two-thirds parliamentary majority, the governing coalition adopted a new Fundamental Law, reshaped the Constitutional Court, entrenched policy choices in “cardinal” laws requiring supermajorities to amend, and redrew the electoral system. In 2018, close to five hundred pro-government media outlets were donated by their owners to a single foundation, KESMA, creating a conglomerate whose formation the government exempted from competition review as a matter of “national strategic importance”. OSCE/ODIHR concluded that the 2022 elections were well administered but “marred by the absence of a level playing field”, and the European Parliament declared that Hungary had become “a hybrid regime of electoral autocracy”. None of this required a single act that international law would readily classify as a flagrant violation. Democracy was not abolished; the conditions of competition were rewritten.

The sequel is equally instructive. In April 2026 the opposition Tisza party won a two-thirds majority and Viktor Orbán conceded – even though international observers found once again that there had been “no level playing field”, that the misuse of public office and resources had blurred the line between state and party, and that monitored media had displayed a clear bias in favour of the ruling party. The election demonstrated that a competitive authoritarian system can still be voted out while a residual capacity for self-correction survives. It also exposed what may be called the restoration dilemma. The new government faces a president, a chief prosecutor, and heads of independent bodies appointed for long, entrenched terms; the President has refused to resign and has turned to the Venice Commission. Poland faces the same predicament: since 2023 its government has struggled to resolve the status of judges appointed through the politically captured National Council of the Judiciary, a situation the Venice Commission in March 2026 described as “unprecedented” in scale. Erosion achieved through law cannot simply be reversed by law without risking the very legality that restoration seeks to protect. Here political fragility displays the defining feature of other systemic risks examined in this series – path dependence approaching irreversibility.

Democratic Decline as Systemic Risk

These experiences suggest that democratic erosion is better understood as risk than as a sequence of violations. A fragile democracy is not merely one in which rights have already been breached. It is one in which the institutional mechanisms capable of detecting, resisting, and correcting abuses are progressively losing their resilience. Independent courts, professional civil services, electoral management bodies, universities, civil society organisations, investigative journalism, parliamentary oppositions, local governments, and politically neutral security institutions perform functions analogous to redundancy in complex engineered systems. They prevent the concentration of failure in a single point and create alternative sites of scrutiny, correction, and resistance. Their apparent inefficiency is precisely what makes democratic systems resilient – a point that echoes the argument in the economic posts of this series that systems optimised for efficiency tend to lose their capacity to absorb shocks. From a risk perspective, checks and balances are not constitutional luxuries but safeguards against systemic failure.

This understanding also changes the temporality of democratic collapse. Political systems rarely move instantaneously from democracy to authoritarianism. Vulnerability accumulates, informal norms of restraint disappear, polarisation intensifies, and citizens lose confidence that institutions operate impartially. Eventually a shock – a contested election, an economic crisis, a security emergency, a mass protest – tests a system whose capacity to absorb disruption has already been diminished. The decisive moment may then take a form that looks entirely lawful. In July 2021 Tunisia’s President Kaïs Saïed invoked the emergency clause in Article 80 of the Constitution to suspend parliament – the very parliament that the constitution required to remain in session during an emergency – and a temporary measure became the foundation of a new constitutional order. In El Salvador, the 2021 dismissal of the Constitutional Chamber of the Supreme Court enabled a reinterpretation permitting immediate presidential re-election, and in July 2025 the legislature approved constitutional amendments allowing indefinite re-election. In Myanmar, the military justified its February 2021 coup by alleging electoral fraud that the electoral commission had rejected – a reminder that manufactured distrust in elections can itself serve as the trigger for breakdown. In each case, the decisive event was less important than the vulnerabilities accumulated before it.

The same logic applies to the threshold of peaceful transfer of power, which is perhaps the most important single indicator of democratic resilience. The storming of the United States Capitol on 6 January 2021 and the assault on Brazil’s three branches of government in Brasília on 8 January 2023 both followed sustained campaigns delegitimising the electoral process. In Brazil, the Supreme Federal Tribunal in September 2025 convicted former President Jair Bolsonaro and senior military officers of plotting to overturn the 2022 election, sentencing him to more than 27 years’ imprisonment. Whatever one’s view of these episodes, they show that accountability – where it occurs – comes after the threshold has been tested, and depends on institutions that have survived the test intact.

International Law Protects Democratic Acts More Easily Than Democratic Systems

International law is by no means indifferent to democracy. Article 25 of the International Covenant on Civil and Political Rights guarantees the right to take part in public affairs, to vote and to be elected at genuine periodic elections, and to have access to public service. The Human Rights Committee’s General Comment No. 25 describes Article 25 as lying at the core of democratic government based on the consent of the people, and its General Comment No. 34 emphasises that the free communication of information and ideas about public and political issues between citizens, candidates, and elected representatives is essential to the effective exercise of political rights. Thomas Franck’s celebrated claim in 1992 that a “right to democratic governance” was emerging in international law found partial expression in regional instruments. The Inter-American Democratic Charter of 2001 proclaims that “the peoples of the Americas have a right to democracy” and envisages suspension of a member state following an unconstitutional interruption of the democratic order. The African Charter on Democracy, Elections and Governance prohibits unconstitutional changes of government, a category that expressly includes the amendment of constitutions to infringe the principles of democratic change of government. Within the European Union, the Court of Justice has held that the values in Article 2 TEU define the very identity of the Union as a common legal order, upholding a regulation that conditions access to EU funds on respect for the rule of law, and has condemned disciplinary regimes that expose judges to political control.

Yet much of this architecture remains organised around discrete rights and identifiable interferences with them – or, at the other extreme, around flagrant ruptures such as coups. What lies between them is precisely the terrain on which contemporary erosion occurs. International law can ask whether a journalist’s freedom of expression has been unlawfully restricted more easily than whether an information ecosystem has ceased to sustain democratic deliberation. It can assess whether a particular election satisfied legal requirements more easily than whether years of institutional transformation have made competition structurally unequal. It can scrutinise an individual interference with judicial independence more readily than the cumulative conversion of a judiciary into an institution unwilling to constrain executive authority. The Hungarian case is emblematic: observers could repeatedly certify that elections were technically well run while concluding that they were not fair – a conclusion for which international law offers few consequences.

Even where regional regimes do address systemic decline, their tools are blunt and late. Sanctions against unconstitutional changes of government presuppose a rupture visible enough to be named; they are poorly suited to aggrandisement that never crosses a single bright line. They also depend on membership: when Mali, Burkina Faso, and Niger left the Economic Community of West African States in January 2025 rather than submit to its democratic protocol, they demonstrated how readily conditionality can be escaped by exit. The European Union’s financial conditionality, the most developed instrument of its kind, required years of procedure before funds were frozen – by which time the institutional transformation it sought to prevent was largely complete.

This produces a disjunction between formal legality and systemic resilience that mirrors the patterns identified throughout this series. Just as climate harm emerges from the aggregation of individually lawful emissions, and financial crises from the accumulation of individually lawful exposures, democratic erosion emerges from the aggregation of individually defensible measures. The interactions matter as much as the components: weakened media pluralism degrades the information environment in which elections take place; distrust in elections weakens institutional legitimacy; attacks on courts weaken the mechanisms through which electoral disputes are resolved; and polarisation makes institutional compromise more costly. Democracy, in this sense, is an ecosystem – but international law still largely regulates its components separately.

The difficulty is compounded by a development that has transformed the very medium of democratic politics: the digital information environment. If institutional erosion attacks the structures of democratic self-correction, information warfare attacks the cognitive conditions on which self-correction depends. Part II of this post turns to that dimension, to the paradox it creates for democracies seeking to defend themselves, and to what a legal approach to democratic risk governance might look like.

Topics
Featured, General, Public International Law
Tags:
Subscribe
Notify of
guest

0 Comments
Oldest
Newest Most Voted