Death Penalty under International Law: Amnesty’s 2025 Report

Introduction

The death penalty under international law became harder to defend after Amnesty International’s Death Sentences and Executions 2025 report recorded at least 2,707 executions in 2025, the highest figure Amnesty has reported since 1981, excluding China (Amnesty International, 2026). The number is not just large. It is legally disturbing because it reflects a 78% rise compared with 2024 and comes at a time when international human rights law has moved steadily toward tighter restrictions on capital punishment.

The report’s central finding is blunt: executions increased sharply, but the increase was not spread evenly across the world. Iran alone accounted for at least 2,159 executions, around 80% of all known executions recorded by Amnesty in 2025. Saudi Arabia also remained a major executing state, while China was excluded from the global total because official figures remain secret, despite Amnesty’s assessment that thousands of executions likely took place there (Amnesty International, 2026). The result is a report that shows both what is known and what remains deliberately hidden.

That secrecy matters. A death penalty system cannot be properly assessed when states conceal execution numbers, charges, trial records, appeal outcomes, or clemency decisions. Lack of transparency weakens public scrutiny, prevents independent verification, and makes it harder to detect wrongful convictions, discrimination, torture-tainted evidence, or political misuse of criminal courts. In capital cases, opacity is not a minor administrative problem. It affects the protection of life itself.

Amnesty’s 2025 report should not be read as a general campaign document detached from law. It is a factual record that raises direct questions under international legal standards. Which offences may still attract the death penalty? What kind of trial is required before a state may take life through criminal punishment? Can drug trafficking, espionage, corruption, or broadly defined terrorism offences justify execution? What happens when confessions are allegedly extracted under torture? These questions place the report inside the legal framework of the right to life, fair trial guarantees, human dignity, and the global movement toward abolition.

The strongest legal issue concerns the gap between domestic penal power and international limits. Many states that retain the death penalty present it as a necessary response to public insecurity, drugs, terrorism, murder, or threats to the state. International law does not prevent governments from punishing serious crime. It does, however, restrict how far punishment may go. Article 6 of the International Covenant on Civil and Political Rights protects the inherent right to life and allows capital punishment only under narrow conditions for states that have not abolished it (United Nations, 1966). The Human Rights Committee has made clear that the “most serious crimes” threshold must be read restrictively and is limited to crimes of extreme gravity involving intentional killing (Human Rights Committee, 2018).

That point is decisive for Amnesty’s findings on drug offences. The report records 1,257 executions for drug-related crimes in 2025, representing 46% of all known executions worldwide (Amnesty International, 2026). However grave drug trafficking may be as a social problem, it does not usually involve intentional killing by the accused person. On that basis, executions for drug offences expose one of the clearest conflicts between retentionist criminal policy and contemporary human rights law.

The procedural concerns are just as serious. Amnesty reports death sentences linked to unfair trials, alleged torture or other ill-treatment, in absentia proceedings, mandatory death penalties, and special or military courts in several jurisdictions (Amnesty International, 2026). These defects are not technical flaws that can be ignored once a domestic court has imposed sentence. Where the punishment is irreversible, the quality of the proceedings becomes central to legality. An execution following an unfair trial may amount to arbitrary deprivation of life.

The report also shows that the death penalty is increasingly isolated, even while some states use it more aggressively. Amnesty records that by the end of 2025, 113 countries were abolitionist for all crimes (Amnesty International, 2026). That figure does not erase the reality of executions in Iran, Saudi Arabia, the United States, Singapore, Egypt, Yemen, Somalia, and other retentionist systems. It does show that the legal and moral direction of the international system is moving away from capital punishment, not toward its normalization.

This article uses Amnesty’s 2025 report as the factual foundation for a legal analysis of the death penalty under international law. It does not treat every state practice as automatically unlawful merely because it involves execution. The better question is more precise: when do the executions, death sentences, offences, procedures, and secrecy documented in the report breach international standards? Answering that question requires close attention to the right to life, the “most serious crimes” rule, fair trial guarantees, torture-based evidence, protected categories of persons, methods of execution, extradition risks, and the duties of abolitionist states.

The central argument is that Amnesty’s 2025 report does not show a global return to lawful acceptance of capital punishment. It shows a sharper confrontation between a small group of highly active executing states and an international legal order that has made the death penalty increasingly exceptional, restricted, and difficult to reconcile with human dignity.

1. Amnesty’s 2025 Report and Its Legal Relevance

Amnesty International’s Death Sentences and Executions 2025 is the factual anchor for this article. It does not create law. It does not bind courts, states, or treaty bodies. Its value is evidentiary: it records how capital punishment was used across the world during 2025 and allows those facts to be measured against international standards.

That distinction matters for any serious analysis of the death penalty under international law. The legal rules come mainly from treaties, customary standards, treaty body interpretation, and human rights case law. Amnesty’s contribution is different. It gathers data on executions, new death sentences, secrecy, offences punished by death, methods of execution, and procedural concerns. The report supplies the factual record; international law supplies the test.

1.1 The Report as a Factual Trigger

Amnesty explains that its 2025 figures cover the judicial use of the death penalty between January and December 2025. The organization relied on official figures, judgments, information from people sentenced to death, lawyers, families, media sources, and civil society organizations. It included only cases for which it found reasonable confirmation (Amnesty International, 2026, p. 5).

That methodology gives the report practical weight. It helps reveal patterns that domestic systems often present as isolated criminal cases. A single execution may appear to be a national matter. Hundreds or thousands of executions, especially for non-lethal offences or after unfair proceedings, create a different picture. They suggest structural problems in the way some states use criminal punishment.

The report should also be read with its institutional position in mind. Amnesty opposes the death penalty in all cases. That does not make its factual findings irrelevant. It simply means the legal reader should separate two things: Amnesty’s abolitionist position and the verifiable information it provides. The correct method is to use the report as evidence, then test that evidence against international obligations.

This approach avoids two weak extremes. One would be to treat Amnesty’s conclusions as if they were binding law. The other would be to reject the report because it comes from an organization with a clear human rights position. Neither response is analytically useful. The better question is whether the practices documented in 2025 comply with the right to life, fair trial guarantees, the prohibition of torture, and the narrow conditions that international law places on executions.

1.2 Minimum Figures and State Secrecy

The report’s global figure is deliberately cautious. Amnesty recorded at least 2,707 executions in 2025, but it also states that the real number is likely higher. The reason is simple: several governments do not publish reliable information on capital punishment. China and Viet Nam treat death penalty data as state secrets, while little or no information was available for countries such as Belarus, Laos, and North Korea because of restrictive state practices (Amnesty International, 2026, p. 5).

This is not just a statistical limitation. It is a rule-of-law problem.

A state that hides execution figures makes it harder to verify whether trials were fair, whether defendants had effective lawyers, whether confessions were obtained through torture, whether appeals were meaningful, or whether clemency procedures existed in practice. Secrecy also prevents families and the public from knowing how the most severe form of punishment is being used.

Capital punishment is irreversible. That makes transparency more than an administrative preference. It is part of accountability. A secretive system may still have courts, judgments, and formal procedures, but outsiders cannot assess whether those procedures genuinely protect the accused.

China is the most important example. Amnesty excludes Chinese executions from its global total because official data remain secret. At the same time, the report states that available information indicates that thousands of people are executed and sentenced to death in China each year (Amnesty International, 2026, p. 5). The result is an incomplete global picture in which the largest executing state may be absent from the numerical total.

This affects the legal analysis directly. The absence of public data does not prove compliance. It may point in the opposite direction. Where a state executes at scale but withholds figures, charges, court records, and clemency information, it becomes almost impossible to test the system against the “most serious crimes” rule, fair trial guarantees, or non-discrimination standards.

1.3 The 2025 Execution Surge

The central finding of the report is severe: Amnesty recorded at least 2,707 executions in 2025, a 78% increase on the 1,518 executions recorded in 2024. This was the highest annual figure Amnesty had recorded since 1981, excluding China. The increase was driven mainly by Iran, where at least 2,159 executions were recorded, more than double the 2024 figure and around 80% of all known executions in 2025 (Amnesty International, 2026, pp. 6–8).

The numbers are alarming, but their distribution is just as important. The surge did not reflect a broad global return to capital punishment. It was concentrated in a small group of states. Iran and Saudi Arabia together accounted for 93% of all known executions. That concentration weakens any argument that the death penalty is regaining wide international legitimacy.

Other increases were still significant. Egypt rose from 13 executions in 2024 to 23 in 2025. Singapore rose from 9 to 17. Kuwait nearly tripled its figure, moving from 6 to 17. Yemen rose from at least 38 to at least 51. The United States recorded 47 executions, its highest number since 2009 (Amnesty International, 2026, pp. 7–8).

The United States deserves separate attention because it was again the only executing state in the Americas. That regional isolation matters. It shows that the continued use of executions by one powerful state does not represent the practice of the region as a whole. It also places the United States in tension with a wider Inter-American trend away from capital punishment.

Drug-related executions are the most legally exposed category in the 2025 data. Amnesty recorded 1,257 executions for drug offences, representing 46% of all known executions. These executions were recorded in China, Iran, Kuwait, Saudi Arabia, and Singapore (Amnesty International, 2026, p. 7). Under the Human Rights Committee’s interpretation of Article 6 of the ICCPR, the death penalty may be imposed only for crimes of extreme gravity involving intentional killing (Human Rights Committee, 2018, para. 35). Drug offences may be serious, but they do not ordinarily meet that threshold.

The report also shows that executions remained numerically concentrated while abolition remained widespread. By the end of 2025, Amnesty recorded 113 countries as abolitionist for all crimes, while 17 countries were known to have carried out executions (Amnesty International, 2026, pp. 7–8). This contrast is the core factual lesson of the report. Capital punishment became more intense in some states, but not more accepted across the international system.

1.4 The Problem of Penal Sovereignty

States that retain the death penalty usually defend it through the language of criminal justice. They refer to murder, terrorism, drug trafficking, espionage, treason, corruption, or public security. The argument has political force because punishment is traditionally linked to sovereignty.

International law does not deny that states may punish serious crime. Governments have the authority to define offences, prosecute offenders, protect victims, and maintain public order. The issue is whether that authority includes the power to take life without meeting international safeguards.

Once a state accepts human rights obligations, domestic criminal justice is not insulated from external legal review. A sentence may be valid under national legislation and still violate international law. That point is crucial. The question is not only whether the execution was authorized by domestic law. The question is whether the offence, procedure, evidence, sentence, and method comply with the state’s international obligations.

The word “terrorism”, for example, does not automatically justify execution. Some terrorism offences involve intentional killing. Others are drafted broadly enough to include conduct far removed from lethal violence. The same problem arises with treason, espionage, drug trafficking, or national security offences. Labels do not decide legality. The substance of the conduct and the fairness of the proceedings do.

Amnesty’s report suggests that some governments used capital punishment as a tool of political control, not only as criminal punishment. It refers to “tough on crime” narratives, restrictions on civic space, silencing of dissent, and disregard for international human rights protections (Amnesty International, 2026, p. 6). That is legally significant because the risk of abuse increases when death sentences are linked to dissent, minority identity, vague security offences, or public fear.

The deeper conflict is clear. Penal sovereignty remains real, but it is limited by the right to life, human dignity, fair trial guarantees, and the prohibition of torture or cruel treatment. The death penalty sits at the hardest edge of that conflict because an unlawful execution cannot be repaired in any meaningful way.

2. Article 6 ICCPR and the Right to Life

Article 6 of the International Covenant on Civil and Political Rights is the main global treaty provision governing capital punishment. It does not abolish the death penalty for every state party. Instead, it protects the inherent right to life and permits executions only under narrow conditions for states that have not abolished them (United Nations, 1966, art. 6).

That structure is important. Article 6 does not treat execution as a normal punishment. It treats it as an exceptional practice that survives only inside strict limits. The Human Rights Committee describes the right to life as the supreme right because without it other rights cannot be enjoyed (Human Rights Committee, 2018, para. 2).

2.1 The Right to Life as the Starting Point

Article 6(1) states that every human being has the inherent right to life. It also says that this right must be protected by law and that no one may be arbitrarily deprived of life (United Nations, 1966, art. 6(1)).

The word “arbitrarily” does much of the work. It means that an execution is not automatically lawful just because a national court imposed the sentence. Domestic legality is necessary, but it is not enough. International law asks a further question: was the deprivation of life compatible with human rights standards?

A person may be executed under a statute that exists, after a trial that produced a judgment, and by officials acting under state authority. Yet the execution may still be arbitrary if the trial was unfair, the offence did not meet the required gravity threshold, the sentence was mandatory, the confession was obtained by torture, or the person was under 18 at the time of the offence.

This is where Amnesty’s 2025 findings become legally important. The report does not merely count executions. It identifies practices that raise right-to-life concerns: drug-related executions, secrecy, unfair proceedings, special courts, military courts, mandatory death sentences, and allegations of torture or ill-treatment (Amnesty International, 2026, pp. 6–10). Those features affect the legality of executions under Article 6.

For non-specialist readers, the point can be stated plainly. International law does not ask only whether a state followed its own criminal law. It asks whether the state respected the minimum conditions required before it may take a person’s life through punishment.

2.2 The Retentionist Exception

Article 6(2) applies to states that have not abolished the death penalty. It says that a death sentence may be imposed only for the most serious crimes, in accordance with the law in force at the time of the offence, not contrary to the Covenant or the Genocide Convention, and only after a final judgment by a competent court (United Nations, 1966, art. 6(2)).

This provision should not be read as a broad permission to execute. It is a restrictive exception. The Covenant was drafted at a time when many states still retained capital punishment, but Article 6 limits that practice rather than normalizing it.

The most important phrase is “most serious crimes”. The Human Rights Committee has interpreted it narrowly. It refers only to crimes of extreme gravity involving intentional killing. Crimes that do not directly and intentionally result in death cannot serve as the basis for capital punishment under this standard (Human Rights Committee, 2018, para. 35).

That interpretation has immediate consequences for the 2025 report. Drug trafficking, drug transportation, corruption, bribery, political offences, religious offences, consensual sexual conduct, and broad security crimes generally fall outside the permitted scope unless the specific conduct involved intentional killing. A state cannot avoid this limit by using severe labels or by declaring a social problem to be a national emergency.

The report’s figure for drug-related executions makes the point concrete. Nearly half of all known executions in 2025 were for drug offences (Amnesty International, 2026, p. 7). If Article 6 is applied according to the Human Rights Committee’s interpretation, many of those executions stand in direct tension with the international standard.

2.3 Final Judgment by a Competent Court

Article 6(2) also requires a final judgment by a competent court. This connects the right to life with fair trial protection. A court must be more than a formal institution with the power to issue sentences. It must be independent, impartial, legally established, and capable of giving the accused a genuine opportunity to defend themselves.

The requirement is especially strict in capital cases. A prison sentence imposed after an unfair trial can sometimes be corrected, reduced, or compensated. An execution cannot. That is why procedural safeguards carry exceptional weight when the accused faces death.

Special courts, emergency courts, security courts, and military tribunals can raise serious concerns, especially when they try civilians or operate under political pressure. The danger increases when proceedings are closed, lawyers are denied access to evidence, appeals are rushed, or defendants cannot properly challenge the prosecution’s case.

A final judgment must also be final in a meaningful sense. It should follow a process in which the accused had access to counsel, time and facilities to prepare a defence, the ability to examine evidence, an effective appeal, and a genuine opportunity to seek clemency. Without those safeguards, finality becomes a formality.

Amnesty records death sentences linked to unfair proceedings, alleged torture or other ill-treatment, in absentia trials, mandatory death sentences, and the use of special or military courts in 2025 (Amnesty International, 2026, pp. 9–10). In Article 6 terms, these are not minor defects. They can turn an execution into an arbitrary deprivation of life.

The practical lesson is direct. In death penalty cases, fair trial guarantees are not a procedural decoration. They are part of the legal boundary between punishment and unlawful killing by the state.

2.4 Article 6 as an Abolition-Oriented Rule

Article 6 ends with a provision that points beyond regulation. Article 6(6) states that nothing in Article 6 may be invoked to delay or prevent abolition of capital punishment by any state party (United Nations, 1966, art. 6(6)).

This clause gives Article 6 its direction. The Covenant tolerates retention only under narrow conditions. It does not place abolition and retention on equal footing. The right to life provision leaves limited room for states that still use executions, while making clear that abolition remains the preferred legal endpoint.

The Second Optional Protocol to the ICCPR confirms that movement. Its object is the abolition of the death penalty, allowing only a narrow wartime reservation for the most serious military crimes committed during wartime (United Nations, 1989, arts. 1–2). It transforms the abolitionist direction of Article 6 into a direct treaty obligation for states that join it.

This matters when reading Amnesty’s 2025 report. A numerical rise in executions does not prove that international law is moving back toward acceptance. Practice and legality are not the same. If the increase is driven by drug offences, secrecy, unfair trials, broad security charges, or mandatory death sentences, it may show defiance of international standards rather than renewed legitimacy.

The report itself reflects this mixed reality. Executions surged in 2025, but abolitionist developments continued. Amnesty notes that Viet Nam abolished the death penalty for eight offences, the Gambia abolished it for murder, treason, and other offences against the state, and 113 countries were abolitionist for all crimes by the end of the year (Amnesty International, 2026, pp. 7, 39–40).

The best reading is not that the death penalty has disappeared. It plainly has not. The stronger conclusion is that its lawful space has narrowed. Article 6 leaves retentionist states with a limited exception, and Amnesty’s 2025 findings show how often current practice presses against that exception or moves beyond it.

3. The “Most Serious Crimes” Threshold

The “most serious crimes” rule is the doctrinal centre of the modern law on capital punishment. It determines which offences may still be punished by death in states that have not abolished it. Without this threshold, Article 6 of the ICCPR would leave governments free to classify almost any conduct as grave enough for execution. That is not how the provision operates.

For the death penalty under international law, the decisive issue is not how strongly a government condemns an offence. The legal question is narrower: did the crime involve intentional killing? If it did not, capital punishment is difficult to reconcile with contemporary human rights standards.

3.1 The Restrictive Meaning of the Rule

Article 6(2) of the ICCPR allows the death penalty only for the “most serious crimes” in states that have not abolished it (United Nations, 1966, art. 6(2)). The Human Rights Committee has interpreted that phrase restrictively. In General Comment No. 36, it stated that the term must be read as referring only to crimes of extreme gravity involving intentional killing (Human Rights Committee, 2018, para. 35).

This interpretation sharply limits retentionist discretion. It excludes offences that are serious, socially harmful, or politically sensitive but do not involve intentional killing. The Committee expressly identifies several offences that cannot justify capital punishment, including drug offences, attempted crimes, corruption, political offences, economic offences, adultery, blasphemy, apostasy, and consensual sexual relations between adults (Human Rights Committee, 2018, para. 35).

The rule also prevents governments from using labels to evade legal scrutiny. A state may call an offence terrorism, treason, sabotage, espionage, drug trafficking, or corruption. Those words may carry political weight, but they do not settle the Article 6 question. The real test is the conduct for which the person was convicted and whether it involved an intentional taking of life.

This matters because death penalty laws are often drafted broadly. Some statutes cover direct killings. Others include assistance, membership, financing, possession, transportation, conspiracy, or speech-related conduct. Treating all of these categories as equally eligible for execution would collapse the “most serious crimes” threshold into a domestic policy choice. General Comment No. 36 prevents that result.

The threshold also supports legal certainty. People must know which conduct can expose them to the gravest sanction. Vague offences, elastic security laws, and broad morality provisions create a risk that capital punishment becomes a tool of intimidation rather than justice.

3.2 Drug Offences

Amnesty’s 2025 report records one of the clearest conflicts between state practice and international standards: 1,257 executions for drug-related offences, amounting to 46% of all known executions worldwide. These executions were recorded in China, Iran, Kuwait, Saudi Arabia, and Singapore (Amnesty International, 2026, p. 7).

Drug trafficking can cause serious social harm. It may strengthen criminal networks, fuel violence, damage public health, and undermine institutions. None of that changes the Article 6 test. Drug transportation, possession, smuggling, sale, or trafficking do not ordinarily involve intentional killing by the accused person. For that reason, they fall outside the “most serious crimes” threshold as interpreted by the Human Rights Committee.

This distinction is not a technicality. Criminal law often punishes harm at different levels. A person who intentionally kills another person commits a different kind of offence from a person who transports narcotics, even if the wider drug trade contributes to violence elsewhere. International law requires attention to the individual conduct proved in court, not the government’s general description of the social problem.

The 2025 data make this point urgent. Nearly half of the known executions were connected to drug offences. If the Human Rights Committee’s standard is applied, a large share of the year’s executions sits outside the permitted scope of Article 6. That does not mean drug crimes are trivial. It means that death is not a lawful or proportionate response under the modern right-to-life framework.

Singapore shows the problem in a concentrated form. Its continued use of executions for drug offences reflects a strict domestic deterrence model. Yet deterrence cannot override the treaty limit. Iran and Saudi Arabia show a different pattern: drug-related executions occur within broader systems where fair trial concerns, confession evidence, minority impact, or secrecy may also arise. The legal defect is not only the offence category. It is the combination of non-lethal crimes with irreversible punishment.

A human rights-based approach to drug policy does not require impunity. States may prosecute traffickers, dismantle organized criminal groups, seize assets, and impose proportionate prison sentences. What they may not do, under the prevailing interpretation of Article 6, is execute people for drug crimes that do not involve intentional killing.

3.3 Terrorism and National Security Offences

Terrorism and national security offences require more careful analysis than drug crimes because some terrorist acts do involve intentional killing. A deliberate bombing of civilians, a massacre, or a planned attack causing death may meet the gravity threshold if the accused intentionally participated in the lethal act and received a fair trial. The legal problem begins when states extend capital punishment to broad security offences that do not involve intentional killing.

Espionage, collaboration with a foreign state, treason, armed rebellion, sabotage, membership in a banned group, financing, propaganda, or political opposition may be treated as existential threats by governments. Yet Article 6 does not allow execution merely because conduct is framed as a threat to national security. The required inquiry remains concrete: what exactly did the accused do, and did that conduct involve intentional killing?

Amnesty’s report notes that Iran intensified executions for espionage or collaboration with Israel after Israeli military strikes in June 2025. It also records executions connected with protest-related charges and broadly framed national security narratives (Amnesty International, 2026, pp. 6–7). These examples show why the label attached to an offence cannot control the legal assessment.

National security law is especially vulnerable to abuse because it often operates under secrecy. Evidence may be classified. Hearings may be closed. Lawyers may be restricted. Families may not receive timely information. Courts may defer to executive claims. In capital cases, those features are dangerous because they increase the risk of wrongful conviction and make later correction impossible.

The same concern applies to terrorism laws drafted in vague or mandatory terms. A statute that imposes death for “terrorism” without distinguishing intentional killing from non-lethal support, association, expression, or preparatory conduct violates the logic of the “most serious crimes” rule. The state must prove the kind of conduct that international law permits to be punished by death, not merely invoke a category that frightens the public.

This does not weaken the duty to protect victims of terrorism or serious violence. It means that even the most serious security threats must be addressed through lawful punishment. The right to life is not suspended by fear, anger, or political emergency.

3.4 Economic Crimes and Corruption

Economic crimes and corruption may cause enormous public damage. Large-scale bribery, embezzlement, fraud, and abuse of office can destroy public trust, drain state resources, weaken hospitals, schools, infrastructure, and courts, and deepen poverty. Domestic law may treat these crimes severely. International law still does not permit execution for them when they do not involve intentional killing.

Amnesty’s report refers to publicly available reports in China showing new death sentences for bribery and other financial offences, while also stressing that China’s overall death penalty figures remain secret (Amnesty International, 2026, pp. 5–7). This creates a double problem: the offence category falls outside the intentional-killing threshold, and the lack of transparent data prevents meaningful assessment of how often such sentences are imposed or carried out.

The Human Rights Committee is clear that economic and political crimes cannot serve as a basis for capital punishment when they do not involve intentional killing (Human Rights Committee, 2018, para. 35). Corruption may be grave, but gravity alone is not enough. The rule requires a specific type of gravity: extreme harm involving an intentional death.

There is also a proportionality issue. Execution for corruption treats financial wrongdoing as equivalent to intentional killing. That equivalence is legally unsustainable under Article 6. It also risks turning criminal justice into public theatre, especially where anti-corruption campaigns are used to display state power or remove political rivals.

A rights-compliant anti-corruption system has many tools: imprisonment, asset recovery, disqualification from public office, fines, restitution, international cooperation, and institutional reform. Capital punishment adds irreversible severity without satisfying the right-to-life threshold. It may look strong politically, but legal strength is measured by compliance with binding safeguards, not by the harshness of the sentence.

3.5 Religion, Morality and Sexual Conduct

The clearest violations arise where death penalty laws punish conduct that should not be criminalized at all or does not involve violence. Apostasy, blasphemy, adultery, consensual same-sex relations, and other morality-based offences fall outside the “most serious crimes” threshold. They also interfere with rights such as freedom of thought, conscience, and religion, privacy, equality, and non-discrimination.

The Human Rights Committee identifies blasphemy, apostasy, adultery, and consensual sexual relations between adults as offences that can never justify the death penalty (Human Rights Committee, 2018, para. 35). The reason is not only that these acts do not involve intentional killing. In several instances, criminalizing the conduct itself conflicts with human rights protections.

This category exposes the strongest limit on penal sovereignty. A state cannot avoid international scrutiny by saying that its criminal law reflects religious doctrine, public morality, or cultural tradition. Human rights treaties do not allow states to extinguish life for protected belief, expression, private conduct, or non-violent moral offences.

There is also a discrimination concern. Morality-based capital offences often fall hardest on women, religious minorities, sexual minorities, political dissidents, and socially marginalized groups. Even when executions are rare, the existence of the death penalty for such offences can create fear, silence, blackmail, forced confession, and unequal enforcement.

The legal conclusion is direct. Offences based on religion, morality, or consensual adult conduct do not meet the “most serious crimes” threshold. Where they are punished by death, the violation is not borderline. It goes to the core of the right to life and the basic principle that criminal punishment must not destroy human dignity.

4. Fair Trial Guarantees in Capital Cases

Fair trial guarantees are not ordinary procedural rules when the accused faces execution. In capital cases, procedure protects life directly. A flawed trial may lead to imprisonment, reputational damage, or wrongful conviction in any criminal system. When the sentence is death, the same flaw can produce an irreversible state of killing.

Article 14 of the ICCPR must be read together with Article 6. The right to life sets the outer boundary; fair trial rights control the process that leads to the sentence. The Human Rights Committee has made clear that carrying out a death sentence after a trial that fails to meet Covenant standards amounts to arbitrary deprivation of life (Human Rights Committee, 2018, para. 41).

4.1 Capital Punishment and Strict Due Process

The basic rule is simple: the more severe the punishment, the stronger the procedural protection must be. Death is the most severe punishment a state can impose. It leaves no room for later correction if the court relied on false evidence, coerced confession, discriminatory prosecution, ineffective defence, or political pressure.

Article 14 of the ICCPR protects equality before courts, a fair and public hearing, the presumption of innocence, adequate time and facilities to prepare a defence, legal assistance, examination of witnesses, interpretation where needed, and review by a higher tribunal (United Nations, 1966, art. 14). In capital cases, these rights cannot be applied weakly or treated as formalities.

A trial may look lawful on paper while failing in practice. A defendant may have a lawyer who is denied access to the file. An appeal may exist but be too narrow to correct serious errors. A hearing may be public in name but shaped by intimidation, secrecy, or executive pressure. International law looks beyond the existence of institutions and asks whether the accused had a real chance to defend their life.

Amnesty’s 2025 report records death sentences imposed after unfair proceedings, including alleged torture or other ill-treatment, in absentia trials, mandatory sentences, and proceedings before special or military courts. These findings matter because due process failures in death penalty cases are not secondary defects. They strike at the legality of the sentence itself (Amnesty International, 2026, pp. 9–10).

4.2 Independent and Impartial Courts

A competent court must be independent, impartial, and legally established. Those requirements are demanding in capital cases because courts must resist political pressure, public anger, security narratives, and executive influence. A judge who cannot act independently cannot provide the protection required before the state takes life.

Military courts, revolutionary courts, special criminal courts, terrorism courts, and emergency tribunals are not automatically unlawful. The problem is institutional design and actual operation. If a court is tied to the executive, uses secret evidence, restricts lawyers, limits appeals, tries civilians without adequate safeguards, or operates inside a political campaign, it becomes deeply problematic in a death penalty case.

The danger is clearest in national security proceedings. Governments often claim that secrecy is necessary to protect intelligence sources, witnesses, or public order. Some confidentiality may be justified in exceptional situations. Yet secrecy cannot deprive the accused of the ability to know and challenge the case against them, especially when conviction may lead to execution.

A court that functions as part of a security apparatus cannot cure the problem merely by issuing a judgment. The requirement of a “final judgment by a competent court” under Article 6(2) of the ICCPR means more than finality. It requires a judicial process capable of independent review, reasoned decision-making, and protection against arbitrary punishment (United Nations, 1966, art. 6(2)).

Amnesty’s report is relevant here because it links several death penalty practices to special or military jurisdictions and to offences framed through public security or state-protection narratives. That combination requires careful scrutiny. Where courts operate under emergency logic, the risk of wrongful or politically motivated death sentences increases sharply (Amnesty International, 2026, pp. 6–10).

4.3 Torture-Tainted Confessions

A confession obtained through torture or ill-treatment cannot support a lawful death sentence. Article 7 of the ICCPR prohibits torture and cruel, inhuman, or degrading treatment or punishment. The Convention against Torture adds a specific evidentiary rule: statements made as a result of torture must not be invoked as evidence in proceedings, except against a person accused of torture as proof that the statement was made (United Nations, 1966, art. 7; United Nations, 1984, art. 15).

The reason is not only reliability, although coerced statements are often unreliable. The deeper issue is legality. A justice system that uses torture to produce evidence destroys the fairness of the trial and violates human dignity. If that evidence leads to execution, the violation becomes even more serious because the state turns unlawful coercion into a death sentence.

This point is crucial for readers who treat torture as a separate abuse. In capital cases, torture does not remain outside the trial. It contaminates the conviction, the sentence, and any later execution. The final act may be carried out in a prison, but the legal defect begins earlier, during interrogation, detention, or prosecution.

Amnesty’s 2025 report identifies cases where death sentences were imposed after proceedings involving allegations of torture or other ill-treatment. Under Article 6, such cases raise a right-to-life issue. If a state executes a person after relying on torture-tainted evidence, the execution may amount to arbitrary deprivation of life, not merely punishment after an imperfect trial (Amnesty International, 2026, pp. 9–10).

Courts and reviewing bodies must examine these claims seriously. A confession should not be accepted because it appears in a case file or because the defendant later repeats it under pressure. The state must investigate credible allegations of coercion, exclude unlawful evidence, and provide effective remedies before any death sentence can be treated as legally safe.

4.4 In Absentia Death Sentences

Trials in absentia require careful distinction. In rare circumstances, a defendant may voluntarily and clearly waive the right to be present after receiving notice of the proceedings and access to legal representation. That is different from a process where the accused is absent because they were never properly notified, were unable to attend, were denied counsel, or were tried as a political or security target.

The right to be present at trial is central to defence. It allows the accused to hear the evidence, instruct counsel, challenge witnesses, explain facts, and respond to the prosecution’s case. When the sentence may be death, absence becomes far more serious because the accused may lose the only meaningful opportunity to prevent execution.

A lawful in absentia process must include strict safeguards. The defendant must have adequate notice, effective representation, access to appeal, and the right to a retrial if later apprehended or if absence was not a genuine waiver. Without these protections, the trial becomes an administrative route to conviction rather than an adversarial process.

Amnesty’s report records death sentences imposed in absentia in 2025. The legal concern is not the label alone. The issue is whether the absence reflected a voluntary waiver or structural denial of participation, defence, and review. If the latter, carrying out the sentence would be incompatible with the heightened safeguards required in capital cases (Amnesty International, 2026, pp. 9–10).

This is especially dangerous in national security and political cases. Authorities may use absence to convict exiles, dissidents, members of opposition groups, or persons accused of collaboration with a foreign state. A death sentence imposed without genuine participation should be treated with exceptional suspicion under Articles 6 and 14 of the ICCPR.

4.5 Mandatory Death Sentences

Mandatory death penalties are especially difficult to reconcile with the right to life. They require courts to impose death once a defendant is convicted of a particular offence, without allowing the judge to assess the person, the facts, the level of participation, mitigating circumstances, age, disability, intent, duress, mental condition, or prospects of rehabilitation.

The Human Rights Committee has stated that mandatory death sentences are arbitrary when they leave domestic courts with no discretion to decide whether the offence and the circumstances of the offender warrant death (Human Rights Committee, 2018, para. 37). The problem is not only severity. It is the removal of individualized justice.

A murder committed after careful planning is not the same as a killing committed under coercion, impaired judgment, secondary participation, or extreme personal circumstances. Drug trafficking at the direction of an organized network is not the same as low-level courier conduct. Mandatory sentencing erases these differences. In a capital case, that erasure may cost a life.

Amnesty records mandatory death sentences in several states in 2025, including Iran, Nigeria, Pakistan, Saudi Arabia, and Singapore. It also notes that Singapore’s 2025 executions for drug offences were connected to mandatory death sentences, with some cases involving people judicially found to be “couriers” but denied sentencing discretion because prosecutorial certification was not issued (Amnesty International, 2026, pp. 9, 23–24).

That structure creates a severe legal problem. If life or death turns on a mandatory rule, or on prosecutorial discretion that the court cannot fully review, the judiciary is prevented from performing its core protective function. International law requires individualized sentencing before the gravest punishment. A system that blocks such assessment risks arbitrariness by design.

5. Protected Persons and Categorical Exclusions

Some people may not be sentenced to death or executed at all, regardless of the offence. These categorical exclusions reflect a basic principle: even in retentionist states, capital punishment cannot be applied to persons whose age, condition, or vulnerability makes execution incompatible with human dignity and the right to life.

These protections are not acts of mercy. They are legal limits. They restrict punishment because certain persons require heightened protection against irreversible state violence.

5.1 Children at the Time of the Offence

International law absolutely prohibits the death penalty for crimes committed by persons who were below 18 at the time of the offence. Article 6(5) of the ICCPR states that a sentence of death shall not be imposed for crimes committed by persons below 18 years of age. The Convention on the Rights of the Child contains the same prohibition (United Nations, 1966, art. 6(5); United Nations, 1989, art. 37(a)).

The rule focuses on age at the time of the offence, not age at trial or execution. A person who is 25 when executed may still be protected if the alleged crime occurred before they turned 18. This distinction is essential because many people spend years on death row after conviction.

The prohibition reflects reduced culpability, developmental vulnerability, and the special status of children in international law. It also recognizes that children may lack the maturity, social power, or legal capacity to defend themselves effectively in criminal proceedings.

Amnesty’s 2025 report states that at least three people were executed for alleged crimes committed when they were below 18: one in Iran and two in Saudi Arabia. It also indicates that other people in this category remained on death row in those countries (Amnesty International, 2026, p. 10).

These executions are not borderline cases under international law. If the person was under 18 at the time of the alleged offence, the death sentence is prohibited. No claim of deterrence, public order, national security, or domestic criminal policy can override that rule.

5.2 Disputed Age and Benefit of the Doubt

Age disputes are common where birth registration is weak, records are missing, documents are unreliable, or conflict and displacement have disrupted family life. In such cases, states may be tempted to treat uncertainty as a reason to proceed. International standards require the opposite approach.

Amnesty notes that the actual age of a prisoner is often disputed because clear proof, such as birth registration, may not exist. It states that governments should apply a full range of appropriate criteria, including physical, psychological, and social development, and should give the individual the benefit of the doubt where age remains uncertain (Amnesty International, 2026, p. 10).

That approach is consistent with the best interests of the child, a core rule of the Convention on the Rights of the Child. Age assessment must be child-sensitive, evidence-based, and protective. It cannot rely only on appearance, police claims, confession records, or administrative assumptions.

The benefit of the doubt is not a loophole for offenders. It is a safeguard against irreversible error. If a state executes someone and later evidence shows that the person was a child at the time of the offence, no remedy can restore the life taken.

Courts should require prosecutors to establish age beyond a reasonable doubt before allowing a capital sentence to stand. Where documents are missing or conflicting, the state should use multidisciplinary assessment and resolve uncertainty against execution. The burden should not fall on a vulnerable defendant to prove childhood with records that may never have existed.

5.3 Pregnant Women

Article 6(5) of the ICCPR also prohibits carrying out a death sentence on pregnant women (United Nations, 1966, art. 6(5)). This is an express treaty rule. It protects the woman and recognizes the additional human consequences of executing a pregnant person.

The rule is minimum protection, not the full extent of humane treatment. Broader human rights practice has moved toward protection for new mothers and women with dependent infants. The ECOSOC Safeguards Guaranteeing Protection of the Rights of Those Facing the Death Penalty state that capital punishment shall not be carried out on pregnant women, new mothers, or persons who have become insane (Economic and Social Council, 1984, safeguard 3).

This broader approach matters because pregnancy, childbirth, and early caregiving cannot be separated from dignity, health, family life, and the interests of the child. Executing a new mother or a person caring for an infant inflicts harm beyond the condemned person and raises serious questions of humanity and proportionality.

Amnesty’s report notes reform discussions and commutation rules in some jurisdictions that address pregnant women or women raising very young children. Viet Nam, for example, issued guidance including commutation possibilities for death sentences imposed on pregnant women or women raising children under 36 months old. Nigeria also considered reform proposals concerning pregnant women facing capital punishment (Amnesty International, 2026, pp. 25, 36).

These reforms are limited, but they point to a wider principle. Retentionist systems must, at a minimum, comply with the express prohibition concerning pregnancy. A more rights-respecting approach extends protection to the period after childbirth, where execution would severely affect dependent children and family integrity.

5.4 Mental or Intellectual Disability

International human rights law has increasingly rejected the execution of persons with serious mental or intellectual disabilities. The concern is not only compassion. It is legal culpability, the ability to participate in the defence, the reliability of proceedings, and the cruelty of executing a person whose condition limits understanding, judgment, or communication.

The Human Rights Committee has stated that states must refrain from imposing the death penalty on individuals who face special barriers in defending themselves on an equal basis, including persons whose serious psychosocial or intellectual disabilities impede effective defence. It has also recognized limits connected to reduced moral culpability and severe mental suffering (Human Rights Committee, 2018, para. 49).

A person with serious intellectual disability may not fully understand proceedings, assist counsel, evaluate plea choices, remember events accurately, or withstand interrogation. A person with severe psychosocial disability may experience distorted perception, impaired decision-making, or extreme vulnerability in detention. These conditions affect both trial fairness and the justice of the sentence.

Amnesty records that people with mental or intellectual disabilities were under sentence of death in several countries in 2025, including Japan and the United States. It also notes the Alabama clemency granted to Rocky Myers, a Black man with an intellectual disability who had spent more than 30 years on death row after a trial marked by serious concerns (Amnesty International, 2026, pp. 10, 16).

The legal problem has several layers. First, executing a person with serious disability may violate human dignity and the right to life. Second, the disability may have impaired the defence, making the trial unreliable. Third, diminished culpability may make death grossly disproportionate. Fourth, prolonged death row conditions may intensify suffering in ways that approach cruel, inhuman, or degrading treatment.

A rights-based system must identify these conditions early, provide expert assessment, ensure effective representation, and exclude capital punishment where disability undermines defence capacity or moral culpability. In capital cases, failure to assess disability is not a medical oversight. It can become a pathway to arbitrary execution.

6. Methods of Execution and Human Dignity

The method of execution is not a secondary issue. It affects the assessment of cruelty, transparency, medical involvement, judicial review, and state responsibility. Even where a state has not abolished capital punishment, it remains bound by the prohibition of torture and cruel, inhuman, or degrading treatment or punishment.

For the death penalty under international law, the question is not only whether a sentence was imposed by a court. It is also how the state carries out the sentence, how much suffering the method creates, how transparent the protocol is, and whether the condemned person has a real chance to challenge the procedure before death.

6.1 Execution Methods in 2025

Amnesty’s 2025 report records several execution methods: beheading, hanging, lethal injection, shooting, and nitrogen gas asphyxiation. Beheading was recorded in Saudi Arabia; hanging in Egypt, Iran, Iraq, Japan, Kuwait, Singapore, and South Sudan; lethal injection in China, the United States, and Viet Nam; shooting in Afghanistan, China, North Korea, Saudi Arabia, Somalia, Taiwan, the United Arab Emirates, the United States, and Yemen; and nitrogen gas asphyxiation in the United States (Amnesty International, 2026, p. 9).

A legal analysis should not describe these methods sensationally. The point is not shock value. The question is whether each method exposes the condemned person to severe physical pain, intense psychological suffering, secrecy, discriminatory application, or arbitrary decision-making. A method may be presented by the state as clinical or orderly, but that presentation does not settle its compatibility with human dignity.

Execution protocols also raise transparency concerns. States often restrict access to information about drugs, equipment, witnesses, medical personnel, training, timing, and contingency plans. In some systems, secrecy is defended as necessary to protect suppliers or prison staff. That argument becomes weak when secrecy prevents courts and lawyers from assessing the risk of severe suffering.

Medical ethics add another layer. Lethal injection and some gas-asphyxiation procedures may involve technical knowledge associated with medicine, even when doctors do not perform the execution. International professional ethics have long treated medical participation in executions as incompatible with the healing role of the profession (World Medical Association, 2019). A state cannot solve the dignity problem by making killing look medical.

State responsibility remains attached to the whole process. If officials choose a method that causes severe suffering, conceal the protocol, deny effective review, or continue after credible evidence of pain emerges, the issue becomes more than domestic penal administration. It may implicate the state’s obligations under Article 7 of the ICCPR and the Convention against Torture (United Nations, 1966, art. 7; United Nations, 1984, arts. 1 and 16).

6.2 Public Executions

Public executions intensify the human rights problem because they turn punishment into a spectacle. The condemned person is no longer treated only as a prisoner subject to sentence. They are used as an instrument to frighten others, display state power, and reinforce public submission.

That public dimension changes the character of the act. An execution carried out before a crowd humiliates the person at the point of death and exposes families, witnesses, and the broader community to state-staged violence. It also risks encouraging revenge-based justice rather than lawful accountability.

The Human Rights Committee has treated public executions as incompatible with the prohibition of cruel, inhuman, or degrading treatment or punishment. The reason is clear: public killing adds degradation to death. It communicates that the state may use the body of the condemned person as a public warning (Human Rights Committee, 2018, para. 40).

Public executions are also dangerous for due process culture. They shift attention away from evidence, appeal, mercy, and judicial restraint. The central message becomes deterrence through fear. In a rights-based system, punishment must remain accountable to law, not public theatre.

This is why the issue cannot be reduced to local tradition or public demand. Human dignity limits how states punish, even after conviction. A crowd cannot legitimize a punishment that degrades the person being killed.

6.3 Nitrogen Gas Asphyxiation

The United States is the key example of nitrogen gas asphyxiation in Amnesty’s 2025 report. Alabama and Louisiana were the only states recorded as using nitrogen hypoxia that year. Alabama used it in four executions, while Louisiana used it in one. The report also notes that several US authorities amended protocols or sought alternative equipment to continue executions despite legal challenges to existing lethal injection procedures (Amnesty International, 2026, pp. 15–16).

The legal issue is not novelty alone. A new method is not unlawful merely because it is new. The proper questions are more concrete. Does the method create a substantial risk of severe suffering? Are the protocol and equipment open to meaningful review? Can the prisoner challenge the method before execution? Are witnesses, courts, and medical experts able to assess what happened?

Nitrogen gas procedures create particular concern because the state may describe the method as painless while limiting independent access to the protocol, medical evidence, or execution records. If courts cannot examine the real risk of distress, panic, prolonged consciousness, or physical suffering, judicial review becomes incomplete.

The problem also connects to experimentation. When a state adopts a method with a limited execution history, the condemned person may become the test case. That raises a dignity concern separate from the final result. A person facing death must not be used to validate an unproven penal technique.

In the United States, litigation often focuses on constitutional standards such as cruel and unusual punishment. For international law, the analysis is broader. Article 7 of the ICCPR and the Convention against Torture require attention to severe pain, mental suffering, transparency, and the state’s duty to prevent cruel treatment. A procedure cannot be treated as humane just because officials say it is controlled.

6.4 Death Row Conditions

The method of execution is only one part of the suffering associated with capital punishment. The time spent under sentence of death can itself raise serious human rights concerns. This is often described as the “death row phenomenon”: prolonged detention under the threat of execution, combined with harsh conditions, uncertainty, isolation, and psychological deterioration.

The leading European case is Soering v United Kingdom. The European Court of Human Rights held that extraditing a person to face a real risk of the death row phenomenon in Virginia would violate Article 3 of the European Convention on Human Rights. The Court focused on the long delay before execution, harsh detention conditions, the applicant’s age and mental state, and the intense psychological strain of living under a death sentence (European Court of Human Rights, 1989, paras. 104–111).

The principle has broader significance. Death row suffering is not limited to the final minutes of execution. Years of uncertainty can produce anxiety, depression, trauma, and social destruction. Family members also live under the same suspended violence: a sentence has been imposed, but the moment of death remains unknown.

This matters for Amnesty’s report because high execution numbers often conceal the larger population under sentence of death. Many prisoners remain on death row for years, including people with serious mental or intellectual disabilities, disputed age claims, inadequate representation, or pending appeals. The harm is not only statistical. It is lived over time.

A state may argue that delay results from appeals and safeguards. That argument has force in some cases, because review is necessary. Yet delay created by systemic dysfunction, repeated warrant-setting, poor prison conditions, or prolonged isolation cannot be excused so easily. The right question is whether the state has maintained a system that predictably produces severe mental suffering before death.

7. Drug Policy and Capital Punishment

Drug-related executions are one of the clearest points of conflict between Amnesty’s 2025 findings and international human rights standards. The report records 1,257 known executions for drug-related offences in 2025, representing 46% of all recorded executions worldwide. These executions occurred in China, Iran, Kuwait, Saudi Arabia, and Singapore (Amnesty International, 2026, p. 7).

This issue deserves separate treatment because governments often present drug control as a matter of public survival. The claim is politically effective, but it does not override Article 6 of the ICCPR. Drug offences do not ordinarily involve intentional killing by the accused person, and the Human Rights Committee has made clear that they do not meet the “most serious crimes” threshold (Human Rights Committee, 2018, para. 35).

7.1 The Punitive Drug-Control Model

The punitive drug-control model treats harsh punishment as the central response to drug markets. In its most extreme form, it uses execution to send a message: the state will protect public order by eliminating traffickers, couriers, suppliers, or repeat offenders.

That message may sound decisive, but it is legally weak. Article 6 requires more than seriousness, social harm, or political urgency. It restricts the death penalty to crimes of extreme gravity involving intentional killing. Drug offences can damage communities, but most do not satisfy that standard.

The “war on drugs” narrative also tends to blur individual guilt. It connects one defendant to a wider social crisis: addiction, violence, organized crime, border insecurity, corruption, and public fear. Courts must resist that collapse. A person should be punished for what the prosecution proves they did, not for the whole damage caused by the drug economy.

Another problem is unequal enforcement. Drug laws often fall hardest on poor defendants, migrants, ethnic minorities, foreign nationals, and low-level couriers. These defendants may lack money, language access, legal knowledge, or bargaining power. When the sentence is death, those inequalities become fatal.

A rights-based system can punish serious drug crime without using the death penalty. Long prison sentences, asset confiscation, organized crime prosecutions, anti-corruption measures, treatment programmes, and cross-border cooperation all remain available. The issue is not whether states may respond to drug trafficking. It is whether they may take life for offences that do not meet the Article 6 threshold.

7.2 Iran, Saudi Arabia, Kuwait and Singapore

Iran, Saudi Arabia, Kuwait, and Singapore show different models of the same core problem: drug-related executions in tension with the right to life. They differ in scale, procedure, legal structure, and transparency, but the central Article 6 issue remains the same.

Iran recorded the largest number of executions in 2025. Close to half of its executions were for drug-related offences, with 998 people executed for such crimes. The report links this rise to the resumption of a heavily punitive and lethal drug policy since 2021. It also notes that the death penalty disproportionately affected oppressed minorities, including Baluchi, Kurdish, and Afghan communities (Amnesty International, 2026, pp. 29–30).

Saudi Arabia presents another pattern. The report records high use of executions, including for drug-related offences, and links the 2025 figure to a broader increase in the country’s use of capital punishment. Where drug offences are punished by death, the main problem remains the absence of intentional killing. Added concerns may arise where proceedings lack transparency or where defendants face barriers to effective defence (Amnesty International, 2026, pp. 31–32).

Kuwait is important because the report records a sharp increase in executions and notes legislative efforts to expand the scope of the death penalty for drug-related crimes. Expansion is legally troubling because Article 6 points toward restriction, not enlargement. When a state broadens capital punishment for non-lethal offences, it moves against the direction of contemporary human rights law (Amnesty International, 2026, pp. 7, 27).

Singapore illustrates the role of mandatory penalties and restricted sentencing discretion. Amnesty International records executions for drug offences in 2025 and criticizes the continued use of capital punishment in drug cases. The concern is not only that the offence category falls outside the intentional-killing threshold. It is also that mandatory or quasi-mandatory structures can prevent courts from assessing personal circumstances, level of participation, coercion, disability, or mitigation (Amnesty International, 2026, pp. 23–24).

Across these systems, the same conclusion emerges. Drug-related executions are not made lawful by strict domestic statutes, public anxiety, or deterrence claims. International law asks whether the individual offence involved intentional killing and whether the process met strict safeguards. Many drug cases fail at the first step.

7.3 International Cooperation Risks

Abolitionist states face a separate but serious problem: cooperation with retentionist drug enforcement systems. Police assistance, intelligence sharing, extradition, deportation, mutual legal assistance, controlled deliveries, evidence transfer, and joint investigations may help a foreign prosecution that ends in a death sentence.

The risk is not theoretical. Drug enforcement is often transnational. One state may collect intelligence, intercept communications, identify suspects, provide laboratory results, transfer financial data, or assist with arrest. If that support foreseeably contributes to a capital prosecution, the assisting state may become connected to the outcome.

Human rights law requires states to avoid exposing individuals to a real risk of execution through transfer or cooperation. The clearest setting is extradition: an abolitionist state should not surrender a person without reliable assurances that the death penalty will not be imposed or carried out. The same logic applies to other forms of assistance when the link to a death sentence is foreseeable (Human Rights Committee, 2018, para. 34).

Diplomatic assurances must be specific and credible. A vague promise is not enough. The requested state should examine who gives the assurance, whether it is binding, whether courts will respect it, how compliance will be monitored, and what remedy exists if the assurance is breached.

Drug cases heighten the risk because some states impose death for trafficking, possession above threshold quantities, transportation, or importation. Abolitionist governments should build safeguards into cooperation agreements. They should condition assistance on non-use of the death penalty and suspend cooperation when the receiving state refuses adequate guarantees.

This point is often overlooked. A state may oppose capital punishment at home while indirectly supporting it abroad. That is a serious inconsistency. Human rights commitments require abolitionist policy not only in domestic criminal law, but also in foreign law-enforcement cooperation.

7.4 Human Rights-Based Drug Policy

A human rights-based drug policy begins with proportionality. Punishment must reflect individual culpability, actual harm, role in the offence, and personal circumstances. Death is not a proportionate response to drug offences that do not involve intentional killing.

This approach does not deny the seriousness of organized drug markets. It rejects the assumption that severity equals effectiveness. Extremely harsh penalties may push markets underground, increase violence, discourage treatment, and strengthen the power of criminal organizations over vulnerable couriers. The law should reduce harm, not only display force.

Public health measures are also relevant. Treatment, prevention, harm reduction, social support, and targeted action against organized criminal networks can address drug-related harm without violating the right to life. The International Guidelines on Human Rights and Drug Policy reject the death penalty for drug offences and connect drug control with dignity, health, equality, and proportionality (United Nations Development Programme, International Centre on Human Rights and Drug Policy, UNAIDS, and World Health Organization, 2019).

Non-custodial alternatives may be appropriate for low-level or non-violent drug offences. For serious organized trafficking, imprisonment and asset-based measures may be justified. The central point is calibration. A legal system that treats all drug offenders as deserving death abandons the individualized assessment required by human rights law.

Amnesty’s 2025 report shows why this matters. Drug-related executions were not marginal. They formed nearly half of all known executions. That makes drug policy one of the most important battlegrounds in the contemporary law of capital punishment. If Article 6 is applied seriously, the execution of people for non-lethal drug crimes should have no place in a lawful criminal justice system.

8. Regional Patterns in 2025

The 2025 report does not show a uniform global return to executions. It shows concentration. A small group of states carried out most known executions, while many regions either remained abolitionist, avoided executions, or moved through moratoria and commutations. That distinction is central to understanding the death penalty under international law: the practice became more intense in some jurisdictions, but not more broadly accepted.

Regional analysis also helps avoid a weak global narrative. Iran, Saudi Arabia, China, Singapore, and the United States raise different legal issues. Some problems concern mass execution. Others concern secrecy, drug offences, national security charges, racial or minority impact, federal fragmentation, or resumption after years of non-use. The legal standards are common, but the factual patterns are not.

8.1 Middle East and North Africa

The Middle East and North Africa region dominated the 2025 execution figures. Iran and Saudi Arabia were the central states. Amnesty records that Iran alone carried out at least 2,159 executions, while Saudi Arabia also exceeded its already high 2024 figure. Together, Iran and Saudi Arabia accounted for 93% of all known executions worldwide (Amnesty International, 2026, pp. 6–8, 27–32).

Iran raises several overlapping legal concerns. The first is scale. When one state accounts for around 80% of all known executions, the issue is no longer isolated sentencing. It becomes a structural feature of criminal governance. The second concern is the offence category. Amnesty records a major number of executions for drug-related offences, which fall outside the “most serious crimes” threshold when they do not involve intentional killing.

National security charges add another layer. The report notes executions linked to the 2022 Women, Life, Freedom protests and an increase in executions for alleged espionage or collaboration with Israel after Israeli military strikes in June 2025. These cases require strict scrutiny because security language can conceal weak evidence, political pressure, restricted defence rights, or punishment of dissent (Amnesty International, 2026, pp. 6–7, 29–30).

Minority impact is also relevant. Amnesty’s regional findings point to the disproportionate effect of executions on oppressed minorities, including Baluchi, Kurdish, and Afghan communities in Iran. International law does not assess capital punishment only through the offence charged. It also asks whether the punishment is applied in a discriminatory manner or through procedures that fail to protect vulnerable groups.

Saudi Arabia presents a related but distinct pattern. Its use of the death penalty remained high, including for drug-related offences and broadly framed terrorism-related charges. Amnesty also links some terrorism cases to persons from the Shia minority who supported anti-government protests between 2011 and 2013 (Amnesty International, 2026, pp. 6–7, 31–32).

The doctrinal issue is clear. A state may punish intentional killing after a fair trial. It may not use broad categories such as terrorism, national security, drug control, or public order to bypass Article 6 of the ICCPR. Where proceedings are opaque, confession evidence is contested, or minority communities are overrepresented, the risk of arbitrary deprivation of life becomes acute.

8.2 Asia-Pacific

The Asia-Pacific region shows why secrecy is as legally important as execution numbers. China remains the largest unresolved issue. Amnesty excludes Chinese executions from its global total because official data are treated as state secrets, while available information indicates that thousands of people are executed and sentenced to death there every year (Amnesty International, 2026, pp. 5, 19–24).

This secrecy prevents meaningful review. Without reliable public figures, it is difficult to assess the offences punished by death, the quality of trials, the number of appeals, clemency practice, regional disparities, minority impact, or the use of capital punishment for financial and corruption offences. A hidden execution system does not become lawful because its details are unavailable. Lack of transparency weakens accountability and obstructs international supervision.

Viet Nam raises a similar transparency problem. Amnesty could not determine credible minimum figures for executions there, although the country is believed to continue using the death penalty extensively. At the same time, the report records an important legislative development: Viet Nam abolished capital punishment for eight offences, including drug transportation, bribery, and embezzlement (Amnesty International, 2026, pp. 7–8, 19–25).

Singapore illustrates the drug offence problem in a highly structured legal system. Amnesty records 17 executions there in 2025, many connected with drug offences. The legal concern is not simply the number. It is the combination of non-lethal offences, mandatory or restricted sentencing discretion, and the Article 6 requirement that capital punishment be confined to crimes involving intentional killing (Amnesty International, 2026, pp. 7–8, 23–24; Human Rights Committee, 2018, para. 35).

Japan and Taiwan are significant because they resumed executions after hiatuses. Amnesty records one execution in Japan and one in Taiwan in 2025. Resumption matters legally and politically because a state that has not executed for years may be moving toward abolition in practice. Restarting executions interrupts that trajectory and can weaken the protective value of long-term non-use (Amnesty International, 2026, pp. 7–8, 19–24).

Myanmar shows the opposite danger: expansion. Amnesty notes that the scope of the death penalty was expanded ahead of elections. Expansion of capital offences is especially difficult to reconcile with the abolition-oriented reading of Article 6. The Covenant tolerates a narrow retentionist exception; it does not support the enlargement of capital punishment for political or security convenience.

8.3 Americas

The Americas present a concentrated regional anomaly. Amnesty records that the United States was the only country in the region to carry out executions for the seventeenth consecutive year. The national total reached 47 executions in 2025, the highest figure since 2009, with Florida driving much of the increase (Amnesty International, 2026, pp. 7–8, 13–18).

The United States does not fit the same pattern as Iran, Saudi Arabia, or China. It is not a unitary mass-execution system. It is a federal system in which the death penalty depends heavily on state-level law, prosecutors, courts, governors, clemency boards, and prison protocols. That fragmentation creates uneven protection. A person’s exposure to execution can depend heavily on the state in which the prosecution occurs.

Florida’s role is especially important. When one state drives a national increase, the regional picture can be distorted. The United States appears as one executing country, but the practical reality is narrower: a limited number of states use executions actively, while many others have abolished capital punishment, imposed moratoria, or have not carried out executions for long periods.

The legal issues in the United States often concern fair trial safeguards, racial bias, intellectual disability, lethal injection litigation, nitrogen gas asphyxiation, long death row delays, and clemency. Amnesty’s report also notes the historic clemency granted in Alabama to Rocky Myers, a Black man with intellectual disability who had spent more than 30 years on death row (Amnesty International, 2026, pp. 15–17).

The regional isolation is important for legal interpretation. The Americas are not moving toward broader use of executions. The United States remains the exception. That weakens any argument that capital punishment is an ordinary feature of criminal justice in the region.

8.4 Sub-Saharan Africa

Sub-Saharan Africa showed both retentionist practice and abolitionist movement. Amnesty International records executions in Somalia and South Sudan. Somalia accounted for at least 17 executions, while South Sudan carried out one known execution after a hiatus. These figures show that capital punishment remains active in some parts of the region (Amnesty International, 2026, pp. 7–8, 33–36).

The legal problems in executing states often concern conflict, weak institutions, military justice, transparency, and limited defence capacity. Where criminal justice operates under conditions of insecurity or fragile judicial administration, the risk of wrongful conviction and arbitrary execution increases. Capital punishment requires the strongest procedural safeguards; fragile systems are often least able to provide them.

At the same time, the region cannot be described only through execution. Zimbabwe commuted all existing death sentences in 2025. The Gambia abolished the death penalty for murder, treason, and other offences against the state. Bills to abolish capital punishment were also pending before legislative bodies in Nigeria (Amnesty International, 2026, pp. 7, 33–36).

These developments matter because abolition often occurs in stages. A state may first commute existing death sentences, then adopt a moratorium, then remove capital punishment from ordinary crimes, and later abolish it for all crimes. Legal change is not always immediate or linear, but commutation and legislative narrowing can reduce the risk of execution and move the system toward abolition.

Sub-Saharan Africa’s 2025 pattern is mixed, not static. Executions continued in a small number of states, yet abolitionist reform remained visible. That mixture reflects the global pattern more broadly: active retention by some governments, combined with a wider legal direction away from the death penalty.

8.5 Europe and Central Asia

Europe and Central Asia offer two useful examples: Belarus and Kyrgyzstan. They show different ways in which the death penalty can be restrained. One concerns the non-use or absence of recorded death sentences. The other concerns constitutional and legal barriers against reintroduction.

Belarus has long been the major exception in Europe. Amnesty’s report notes that for the first year since President Alexander Lukashenko assumed office in 1994, no death sentences were recorded as imposed or implemented in Belarus. That is a significant factual development, even though it does not equal abolition (Amnesty International, 2026, pp. 7, 26).

The legal meaning should not be overstated. Absence of recorded executions is not the same as repeal. A state can remain retentionist even during a period of non-use. Yet non-use can still matter. It may support a move toward abolition in practice, reduce immediate risk, and create political space for reform.

Kyrgyzstan presents a different example. The Constitutional Court declared attempts to reintroduce the death penalty unconstitutional. That is important because abolition can be protected not only through treaties, but also through domestic constitutional law. Once abolition becomes constitutionally entrenched, political demands for restoration face a legal barrier (Amnesty International, 2026, pp. 7, 26).

The regional lesson is straightforward. Non-use, abolition, and non-reintroduction are distinct legal positions. Belarus shows the limits of non-use where formal retention remains. Kyrgyzstan shows how constitutional law can help prevent regression.

9. Abolition and Non-Reintroduction

International law no longer treats abolition as a marginal policy preference. It has become the direction of travel in global and regional human rights law. Some treaty rules impose abolition directly on states that accept them. Other instruments promote moratoria, restrictions, transparency, commutation, and non-reintroduction.

Abolition is not achieved only when a state removes every capital offence at once. It may proceed through stages: reducing capital crimes, ending mandatory death sentences, commuting existing sentences, adopting an execution moratorium, abolishing the death penalty for ordinary crimes, ratifying abolition treaties, and creating constitutional barriers against restoration.

9.1 Second Optional Protocol to the ICCPR

The Second Optional Protocol to the ICCPR is the main global treaty on abolition. It was adopted in New York on 15 December 1989 and entered into force on 11 July 1991. UN Treaty Collection data list 92 parties to the Protocol (United Nations Treaty Collection, 2026).

The core obligation is direct. Article 1 provides that no one within the jurisdiction of a state party shall be executed and that each state party shall take all necessary measures to abolish the death penalty within its jurisdiction (United Nations, 1989, art. 1).

This matters because the Protocol changes the legal position of states that ratify it. Under the ICCPR alone, a retentionist state is tightly restricted but not universally barred from executions. Under the Second Optional Protocol, abolition becomes a binding treaty obligation. The state is no longer merely regulated in its use of capital punishment. It is legally committed to ending it.

The Protocol also strengthens non-reintroduction. A state party cannot abolish, ratify, and later revive executions as an ordinary penal policy. Ratification turns abolition into an international commitment, not only a domestic legislative choice.

9.2 Limited Wartime Reservation

The Second Optional Protocol permits only one narrow reservation. At the time of ratification or accession, a state may reserve the right to apply the death penalty in time of war after conviction for a most serious military crime committed during wartime (United Nations, 1989, art. 2).

The narrowness of this reservation is legally important. It excludes ordinary crimes. It excludes peacetime offences. It excludes broad public order arguments. It is tied to wartime, military crimes, conviction, and seriousness. That structure confirms that abolition is the Protocol’s object, while the reservation is exceptional.

A state cannot use the reservation to preserve a general death penalty system. It cannot invoke drugs, terrorism in peacetime, murder, corruption, espionage, or political violence unless the specific reservation conditions are met. The reservation must also be made at the moment of ratification or accession; it is not an open-ended option to be revived later when domestic politics change.

This design reflects the broader direction of human rights law. Retention is not treated as equal to abolition. The legal system increasingly treats executions as a practice to be eliminated, with any remaining allowance kept narrow, supervised, and exceptional.

9.3 Regional Abolition Treaties

Regional systems have reinforced abolition. In Europe, Protocol No. 6 to the European Convention on Human Rights abolishes the death penalty in peacetime, while Protocol No. 13 abolishes it in all circumstances. Protocol No. 13 is especially strong because it removes the wartime exception and does not permit derogation (Council of Europe, 1983, arts. 1–2; Council of Europe, 2002, arts. 1–3).

The European model shows how regional law can turn abolition into a condition of shared legal identity. Capital punishment is not merely unused. It is legally excluded as incompatible with the human rights order of the region.

The Inter-American system also contains a regional abolition treaty: the Protocol to the American Convention on Human Rights to Abolish the Death Penalty. It requires states parties not to apply the death penalty in their territory, while permitting a narrow reservation for wartime application for extremely serious military crimes (Organization of American States, 1990, arts. 1–2).

Regional abolition treaties matter because they consolidate legal expectations. They help transform abolition from national policy into regional commitment. They also make reintroduction harder by creating external obligations, judicial standards, and diplomatic costs.

These instruments also affect interpretation. Even for states that have not joined abolition protocols, regional practice can shape the legal environment. It becomes harder to defend executions as ordinary when neighbouring states have renounced them as a matter of human rights law.

9.4 Moratoria as Legal and Political Tools

Abolition, moratorium, commutation, and abolition in practice are related but different. Abolition removes the death penalty from the law. A moratorium suspends executions, usually through executive, legislative, or political commitment, while the penalty remains on the books. Commutation replaces death sentences with lesser penalties. Abolition in practice describes states that retain capital punishment legally but have not executed for a long period and appear to follow a settled policy of non-use.

These distinctions matter. A moratorium can save lives immediately, but it may be reversed. Commutation protects specific prisoners, but it may not prevent new death sentences. Abolition in practice reduces risk, but it leaves the legal machinery available. Full abolition is the strongest safeguard because it removes both sentence and execution.

UN General Assembly resolutions on a moratorium are not binding treaties. Their importance lies in political and normative pressure. They ask retentionist states to suspend executions, restrict capital offences, respect safeguards, make information available, and move toward abolition. Resolution 79/179, adopted on 17 December 2024, continued that process and requested further reporting on implementation (United Nations General Assembly, 2024).

Moratoria also help test claims of necessity. If a state suspends executions and public order does not collapse, the deterrence argument weakens. Non-use can create practical evidence that criminal justice can function without killing prisoners.

The main weakness is reversibility. Governments may end moratoria after elections, high-profile crimes, public pressure, or security crises. That is why moratoria are best understood as transitional tools, not final solutions. They are useful when they reduce immediate risk and create a pathway to legal repeal.

9.5 Human Rights Council Resolution 60/17

The Human Rights Council’s 2025 resolution on the question of the death penalty confirms continuing institutional pressure against capital punishment. Resolution 60/17 was adopted on 7 October 2025 by 31 votes to 7, with 8 abstentions (Human Rights Council, 2025).

The vote matters because it shows strong cross-regional support for scrutiny of the death penalty within the UN human rights system. It also demonstrates that abolition and restriction are not confined to one legal tradition or one region. The death penalty remains contested, but the institutional direction is increasingly restrictive.

Human Rights Council resolutions do not operate like treaties. They do not by themselves, abolish capital punishment in national law. Their value lies in interpretation, monitoring, political pressure, and standard-setting. They keep the issue on the international agenda and reinforce the link between executions, transparency, discrimination, fair trial rights, and human dignity.

Resolution 60/17 should be read alongside Amnesty’s 2025 report. Amnesty documents a sharp rise in executions. The Council’s resolution shows that the institutional response is not acceptance, but continued scrutiny. This contrast is central to the article’s argument: execution numbers rose in 2025, yet the legal and institutional environment did not move toward normalization.

The strongest conclusion is not that abolition is complete. It is not. The point is more precise: contemporary international law has made retention harder to defend, expansion harder to justify, secrecy harder to excuse, and reintroduction harder to reconcile with human rights commitments.

10. State Responsibility for Unlawful Executions

Unlawful executions engage the responsibility of the state. The issue is not limited to the individual judge, prosecutor, prison officer, or execution team. Under the law of state responsibility, conduct carried out by state organs is attributable to the state, even when domestic law later treats the act as wrongful or unauthorized (International Law Commission, 2001, arts. 4 and 7).

This matters for the death penalty under international law because capital punishment is implemented through official institutions. Courts impose the sentence, prosecutors seek it, prison authorities hold the prisoner, executive officials may control clemency, and execution personnel carry out the final act. If the process breaches international obligations, the state cannot avoid responsibility by isolating blame at the level of individual officials.

10.1 Attribution to the State

Executions ordered by courts and carried out by prison authorities are classic examples of state conduct. Judges, prosecutors, police, prison officials, medical officers acting under state authority, and execution personnel exercise public power. Their conduct is attributable to the state when they act in an official capacity or under colour of official authority (International Law Commission, 2001, arts. 4–5).

This remains true even if the official violates domestic rules. A police officer who tortures a suspect, a prosecutor who knowingly relies on coerced evidence, or a prison official who executes a defective notice may act unlawfully under national law. That does not remove the act from the state. International responsibility attaches because the official used state power.

Attribution also covers judicial conduct. Domestic courts are organs of the state for the purposes of international law. A wrongful death sentence imposed after an unfair trial, a denial of appeal, or a refusal to examine torture allegations is not insulated by judicial independence. Independence protects courts from political interference; it does not shield the state from international responsibility when courts violate treaty obligations.

The same reasoning applies where an execution follows a chain of decisions. A death sentence may involve investigation, interrogation, prosecution, trial, appeal, clemency, detention, and execution. International law does not examine only the final official who carries out the sentence. It looks at the state process as a whole.

10.2 Breach of International Obligations

An execution becomes internationally wrongful when it breaches an obligation binding on the state. The most direct breach is arbitrary deprivation of life under Article 6 of the ICCPR. This may occur when a person is executed for an offence that does not meet the “most serious crimes” threshold, after an unfair trial, under a mandatory death sentence, or despite belonging to a protected category such as a person who was below 18 at the time of the offence (United Nations, 1966, art. 6; Human Rights Committee, 2018, paras. 35–37 and 41).

Fair trial violations can also produce responsibility. Article 14 of the ICCPR requires independent and impartial courts, adequate defence rights, presumption of innocence, access to counsel, examination of witnesses, interpretation where needed, and review by a higher tribunal. In capital cases, those safeguards carry exceptional weight because an unfair process can lead to an irreversible death (United Nations, 1966, art. 14).

Torture or ill-treatment creates another layer of breach. If a death sentence rests on a confession extracted through torture, the state may violate Article 7 of the ICCPR, the Convention against Torture, and the right to life if the sentence is implemented. The Convention against Torture also requires states to exclude torture-tainted statements from proceedings, except when used against the alleged torturer as evidence that the statement was made (United Nations, 1984, arts. 2, 12, 15 and 16).

Discrimination can make capital punishment unlawful even where the offence itself is serious. If the death penalty is imposed disproportionately on racial, ethnic, religious, political, national, or social groups because of discriminatory law, prosecution, sentencing, or clemency practice, the state may breach equality and non-discrimination obligations. Amnesty’s 2025 report is relevant here because it identifies minority impact concerns in some execution patterns, especially where drug offences, national security charges, and weak safeguards overlap (Amnesty International, 2026, pp. 29–32).

Child rights violations are among the clearest breaches. Article 6(5) of the ICCPR and Article 37(a) of the Convention on the Rights of the Child prohibit capital punishment for offences committed by persons below 18 years of age. If a state executes someone for an offence committed as a child, the breach is not debatable under contemporary treaty law (United Nations, 1966, art. 6(5); United Nations, 1989, art. 37(a)).

Abolition treaty obligations create an even stronger bar. A state party to the Second Optional Protocol to the ICCPR must not execute anyone within its jurisdiction and must take all necessary measures to abolish the death penalty. For those states, any execution would breach a direct treaty commitment, apart from any additional violations concerning trial, offence category, or method (United Nations, 1989, art. 1).

10.3 Reparation After Execution

Reparation after an unlawful execution is uniquely inadequate. International law requires full reparation for injury caused by an internationally wrongful act, including restitution, compensation, satisfaction, and guarantees of non-repetition where appropriate. Yet restitution is impossible after death. No court, payment, apology, or reform can restore the life taken (International Law Commission, 2001, arts. 31 and 34–37).

That impossibility does not remove the duty to repair. It changes its form. The state may owe acknowledgment of wrongdoing, public disclosure of the facts, compensation to relatives, correction of the record, return of remains, formal apology, and measures to protect the family from intimidation or stigma.

Investigation is also essential. If torture, fabricated evidence, discriminatory prosecution, enforced disappearance, denial of counsel, or official misconduct contributed to the execution, the state must investigate effectively. Where criminal offences are established, responsible officials may have to be prosecuted. This is especially important in torture cases because the Convention against Torture requires investigation and prosecution or extradition of alleged torturers under defined conditions (United Nations, 1984, arts. 4–7 and 12).

Legal reform may be the most important remedy for future protection. A state may need to abolish the death penalty, narrow capital offences, remove mandatory death sentences, exclude military courts from civilian capital cases, improve access to counsel, strengthen appeal and clemency procedures, publish execution data, and create safeguards for children, persons with disabilities, pregnant women, and foreign nationals.

Guarantees of non-repetition must be concrete. A general statement that mistakes will not happen again is not enough. In the death penalty context, non-repetition requires institutional changes that reduce the risk of unlawful execution before the state reaches the point of no return.

10.4 Responsibility Beyond the Executing State

Responsibility may also arise for states that assist another state’s capital prosecution. The problem appears in extradition, deportation, intelligence-sharing, police cooperation, mutual legal assistance, transfer of evidence, and joint drug operations. If assistance foreseeably contributes to a death sentence or execution, the assisting state may breach its own human rights obligations.

The law of state responsibility also recognizes responsibility for aiding or assisting another state in the commission of an internationally wrongful act, where the assisting state acts with knowledge of the circumstances and the act would be wrongful if committed by that state itself (International Law Commission, 2001, art. 16).

Human rights law frames the issue more directly. A state must not expose a person to a real risk of arbitrary deprivation of life, torture, cruel treatment, or the death row phenomenon through transfer or cooperation. The Human Rights Committee has stated that states parties must not deport, extradite, or otherwise transfer individuals where there are substantial grounds for believing that there is a real risk of irreparable harm, such as that contemplated by Articles 6 and 7 of the ICCPR (Human Rights Committee, 2018, para. 34).

This is especially relevant in drug cases. An abolitionist state may help identify suspects, intercept shipments, transmit intelligence, provide financial records, or transfer forensic evidence. If the receiving state punishes drug offences by death, cooperation without safeguards may help build a capital case.

The practical rule should be strict. A state that opposes the death penalty at home should not help produce it abroad. At a minimum, it must seek reliable assurances that the death penalty will not be imposed or carried out. In high-risk systems, even assurances may be insufficient unless they are specific, binding, monitored, and supported by a record of compliance.

11. Extradition, Deportation and Transfer

Extradition and deportation are central to the external reach of capital punishment. A person may be physically outside the executing state, yet still face death if another state sends them back. That is why abolitionist and human rights-bound states must examine the death penalty risk before transfer.

The question is practical, not abstract. Will the person face a real risk of execution, death row suffering, torture, unfair trial, or a capital charge that does not meet international standards? If the answer is yes, the transferring state must refuse transfer or obtain reliable protection against those outcomes.

11.1 Non-Refoulement to Execution Risk

Non-refoulement means that a state must not transfer a person to another state where they face a real risk of serious human rights harm. In death penalty cases, that risk may concern execution itself, a capital trial lacking basic safeguards, prolonged death row conditions, or exposure to torture or ill-treatment.

The European Court of Human Rights gave this rule major force in Soering v United Kingdom. The Court held that extraditing the applicant to the United States, where he faced a real risk of the death row phenomenon in Virginia, would violate Article 3 of the European Convention on Human Rights (European Court of Human Rights, 1989, paras. 104–111).

The Human Rights Committee applies a similar logic under the ICCPR. A state party must not remove a person where substantial grounds show a real risk of irreparable harm under Articles 6 or 7. This covers risks to life and risks of torture or cruel, inhuman, or degrading treatment (Human Rights Committee, 2018, para. 34).

Abolitionist states should be especially careful. If a state has rejected capital punishment as incompatible with its legal order, it cannot ignore the predictable consequences of sending a person to face execution elsewhere. The prohibition is not satisfied by saying that the death sentence would be imposed by another sovereign state. Transfer is itself an exercise of state power.

The safest course is refusal unless the receiving state gives credible, case-specific assurances that the death penalty will not be sought, imposed, or carried out. In some systems, assurances against execution alone may not be enough if the person still faces torture, an unfair trial, or severe death row conditions.

11.2 Diplomatic Assurances

Diplomatic assurances are promises by the receiving state about how the person will be treated after transfer. In death penalty cases, they may state that prosecutors will not seek death, courts will not impose it, or the executive will commute any sentence if imposed.

Not all assurances are credible. A generic statement of goodwill is weak. A reliable assurance should be specific to the person, issued by an authority capable of binding prosecutors and courts, clear in legal effect, communicated before transfer, and open to monitoring after surrender.

The requested state must examine the receiving state’s record. Has it honoured similar assurances before? Can its executive control or bind the relevant prosecution authority? Are courts independent in a way that makes executive promises uncertain? Is there a mechanism for consular or diplomatic monitoring? Can a breach be detected before execution?

Assurances are especially fragile where the death penalty is mandatory. If domestic law requires death upon conviction, an executive promise may not be enough unless the receiving state has a legally effective mechanism to prevent the sentence. The same concern arises where prosecutors have independent authority to seek death or where courts cannot accept foreign assurances as binding.

Abolitionist states should not use assurances as a formal device to approve risky transfers. The question is not whether a document exists. The question is whether the assurance actually removes the real risk of execution and related ill-treatment.

11.3 Death Row Phenomenon

The death row phenomenon links the extradition law with human dignity. It refers to the psychological suffering caused by prolonged detention under sentence of death, often combined with isolation, harsh conditions, repeated execution dates, uncertainty, and fear of the final act.

In Soering, the European Court did not hold that every death penalty case automatically bars extradition. It examined the expected delay, conditions on death row, the applicant’s personal circumstances, and the severity of mental suffering. On those facts, transfer would have violated Article 3 of the European Convention (European Court of Human Rights, 1989, paras. 104–111).

Under the ICCPR, the same problem is examined through Articles 6 and 7. Article 6 protects life; Article 7 prohibits torture and cruel, inhuman, or degrading treatment or punishment. A transfer that exposes someone to years of severe psychological suffering before possible execution can engage both provisions (United Nations, 1966, arts. 6–7; Human Rights Committee, 2018, para. 40).

This issue remains important even where the receiving state promises that execution will be lawful under domestic law. A death sentence may follow a statute and still expose the person to treatment that human rights law prohibits. The focus is not only the final killing; it is the entire path toward it.

The death row phenomenon also affects prisoners with mental illness, intellectual disability, youth at the time of the offence, trauma history, or serious medical conditions. For such persons, prolonged uncertainty and isolation may be especially damaging. A transferring state must consider those individual factors before extradition or deportation.

11.4 Consular Protection

Foreign nationals facing the death penalty require special attention. Article 36 of the Vienna Convention on Consular Relations gives detained foreign nationals the right to have their consulate notified if they so request, and it gives consular officers the right to communicate with and assist them (United Nations, 1963, art. 36).

This right is practical. A foreign national may not understand the legal system, language, evidence rules, appeal deadlines, or available lawyers. Consular assistance can help secure representation, contact family, gather mitigation evidence, challenge procedural violations, and monitor detention conditions.

The International Court of Justice has confirmed the importance of Article 36 in capital cases. In LaGrand, the Court held that the United States had breached obligations owed to Germany by failing to notify two German nationals of their consular rights. In Avena, the Court required review and reconsideration of convictions and sentences of Mexican nationals affected by consular notification violations (International Court of Justice, 2001, paras. 77–91; International Court of Justice, 2004, paras. 121–153).

Consular notification does not automatically invalidate every conviction. The legal consequence depends on the system, the prejudice suffered, and the remedy required. In death penalty cases, however, the violation is especially serious because lost consular assistance may affect defence preparation, mitigation evidence, appeal strategy, and clemency.

A state that sentences or executes a foreign national without respecting consular rights risks breaching both consular law and human rights safeguards. The issue is not diplomatic courtesy. It is protection against wrongful and arbitrary execution.

For the broader analysis, consular protection shows one more way in which capital punishment cannot be treated as an isolated domestic matter. Once a foreign national is on death row, the case may implicate treaty law, diplomatic protection, international adjudication, and the responsibility of both the executing and sending states.

12. How to Evaluate Amnesty’s Evidence

Amnesty International’s report is highly useful, but it must be used with legal discipline. It should not be treated as a substitute for treaties, courts, UN bodies, domestic legislation, or official records. Its role is evidentiary. It helps identify factual patterns that require legal assessment.

For an article on the death penalty under international law, this distinction is essential. The report shows what happened in 2025 as far as Amnesty could verify. International law then determines which practices are lawful, restricted, or incompatible with human rights obligations.

12.1 NGO Reporting and Legal Analysis

NGO reports can play an important role in public international law research. They often document facts that governments suppress, understate, or publish only selectively. In death penalty cases, this matters because official information may be unavailable, delayed, politicized, or incomplete.

Amnesty’s report draws on official figures, judgments, information from people sentenced to death, families, representatives, media reports, and civil society sources. It includes figures only where there is reasonable confirmation. That makes it a serious evidentiary source, but not a final legal determination (Amnesty International, 2026, p. 5).

A careful legal analysis should compare Amnesty’s findings with other materials. These include national criminal codes, execution protocols, official prison data, UN Secretary-General reports, Human Rights Committee documents, Special Rapporteur communications, regional human rights decisions, domestic court rulings, and treaty ratification records.

This does not mean that NGO reporting is weak. It means that legal writing must separate fact-gathering from legal authority. Amnesty may record that a person was executed for a drug offence. The legal conclusion that such an execution violates Article 6 depends on the ICCPR, the Human Rights Committee’s interpretation, and the facts of the offence.

The best method is triangulation. Use Amnesty to identify the pattern, official records where available to confirm the state’s own position, and international legal sources to test compliance. That produces stronger analysis than simply repeating Amnesty’s conclusions or rejecting them because they come from an advocacy organization.

12.2 The Risk of Undercounting

The greatest evidentiary risk in death penalty research is not exaggeration; it is undercounting. Amnesty repeatedly states that its figures are minimum recorded figures. In countries where execution data are secret or access to courts and prisons is restricted, the real number is likely higher than the published total (Amnesty International, 2026, p. 5).

This matters most for China, Viet Nam, and North Korea. China is excluded from Amnesty’s global execution total because official figures remain a state secret, even though available information indicates that thousands of people are executed and sentenced to death there each year. Viet Nam also treats death penalty data as secret, and Amnesty could not identify a credible minimum figure for executions in 2025 (Amnesty International, 2026, pp. 5, 8).

Legal analysis should not confuse the absence of evidence with the absence of executions. A state that refuses to publish data should not benefit from that secrecy. If anything, secrecy increases the need for scrutiny because it prevents independent assessment of offence categories, fair trial safeguards, clemency practice, discrimination, and wrongful conviction risk.

Undercounting also affects regional comparisons. A region may appear to have fewer executions than another because the most active executing states hide figures. The legal writer must state this clearly. Numerical rankings can be misleading when transparency differs sharply between countries.

The safest formulation is precise: Amnesty recorded at least 2,707 executions in 2025, excluding thousands believed to have taken place in China. That wording avoids false certainty and keeps the evidentiary limits visible.

12.3 The Danger of Overclaiming

The opposite risk is legal overclaiming. It would be inaccurate to say that all use of the death penalty is prohibited under universal treaty law. The ICCPR still contains a restrictive retentionist clause for states that have not abolished capital punishment. Article 6(2) allows the death penalty only for the most serious crimes and only under strict safeguards (United Nations, 1966, art. 6(2)).

The stronger argument is more precise. International law has narrowed the lawful use of capital punishment so severely that many executions recorded in 2025 appear incompatible with contemporary human rights standards. That is especially true for executions after unfair trials, executions for drug offences, executions of persons who were children at the time of the offence, mandatory death sentences, and executions based on torture-tainted confessions.

This distinction improves credibility. A serious article should not collapse treaty law, abolitionist advocacy, and moral condemnation into one statement. It should say exactly what the law does: it restricts, supervises, delegitimizes expansion, and binds abolitionist treaty parties to non-use.

The Second Optional Protocol to the ICCPR does prohibit executions for its parties. Regional abolition treaties do the same within their membership. But states outside those instruments remain governed by the ICCPR’s restrictive framework, customary safeguards, and other human rights obligations, not by a universal abolition rule applicable in identical terms to every state (United Nations, 1989, art. 1).

The legal conclusion must be disciplined. Retention still exists. Some executions may occur within domestic law. Yet many practices documented in 2025 fall outside the lawful space left by international human rights law. That is the accurate argument, and it is stronger than an overbroad claim that the present law cannot support.

Also Read

• Soering v United Kingdom Case

• The Principle of Non-Refoulement

• Human Rights in Iran: Law Violations and Accountability

• Duterte ICC Trial and the Law of Accountability

• Religion and International Law: Conflicts and Coexistence

• The Right to Protest in International Law

• Common Article 3 in Non-International Armed Conflict

• Human rights violations in Venezuela: legal analysis

13. Central Legal Findings for the Article

Amnesty’s 2025 report should lead the article to five core findings. These findings are not merely descriptive. They explain why the report matters legally and why the 2025 execution surge should be read as a challenge to international human rights law, not as evidence of renewed global acceptance.

13.1 The Death Penalty Is Legally Exceptional

Capital punishment is no longer an ordinary expression of penal sovereignty. A state cannot simply say that criminal punishment belongs to domestic jurisdiction and end the inquiry. Once human rights obligations apply, the death penalty becomes subject to international limits.

Article 6 of the ICCPR places the right to life at the centre. It allows retention only under narrow conditions for states that have not abolished the penalty. It also states that nothing in Article 6 may be invoked to delay or prevent abolition. That structure makes capital punishment exceptional, restricted, and abolition-oriented (United Nations, 1966, art. 6).

This is the article’s first major finding. The death penalty survives in some legal systems, but it survives under pressure. Its use must be justified offence by offence, trial by trial, and safeguard by safeguard. Domestic legality is not enough.

13.2 The 2025 Surge Is Legally Concentrated

The 2025 increase in executions does not show broad global acceptance. It shows intensified use by a small number of states. Iran alone accounted for around 80% of known executions, while Iran and Saudi Arabia together accounted for 93% of the recorded global total. China remained outside the total because its figures are secret (Amnesty International, 2026, pp. 5–8).

That concentration changes the interpretation of the data. The world did not collectively return to the death penalty. A limited group of states used it more aggressively, often in areas where international law is most restrictive: drug offences, national security charges, unfair trials, secrecy, and vulnerable defendants.

The regional picture confirms this. The United States remained the only executing state in the Americas. Executions in Sub-Saharan Africa were confined to Somalia and South Sudan. In the Middle East and North Africa, the known figures were dominated by Iran and Saudi Arabia. The practice was severe, but not widely distributed (Amnesty International, 2026, pp. 7–8).

13.3 Drug Executions Are the Clearest Violation

Drug-related executions are the most direct doctrinal conflict in the report. Amnesty recorded 1,257 executions for drug offences in 2025, representing 46% of all known executions. These executions were recorded in China, Iran, Kuwait, Saudi Arabia, and Singapore (Amnesty International, 2026, p. 7).

The legal problem is straightforward. The Human Rights Committee interprets “most serious crimes” as crimes of extreme gravity involving intentional killing. Drug transportation, trafficking, possession, importation, or sale do not ordinarily satisfy that standard (Human Rights Committee, 2018, para. 35).

This does not minimize the harm caused by drug markets. It clarifies the legal boundary. International law permits strong punishment for serious drug crime, but it does not permit execution for non-lethal offences under the prevailing interpretation of Article 6.

This finding should be one of the article’s strongest points. If nearly half of recorded executions were for drug offences, then a large part of the 2025 execution surge stands in direct tension with the core right-to-life framework.

13.4 Fair Trial Failures Convert Sentences into Violations

A death sentence imposed after an unfair process is not merely procedurally weak. If carried out, it may become an arbitrary deprivation of life. Articles 6 and 14 of the ICCPR must be read together in capital cases because fair trial protection is part of the safeguard against unlawful execution (United Nations, 1966, arts. 6 and 14).

The most serious defects include torture-tainted confessions, denial of effective legal assistance, military or special courts trying civilians, in absentia proceedings without genuine waiver, mandatory death sentences, denial of appeal, and clemency systems that function only on paper.

Amnesty’s report records several of these concerns in 2025. It refers to unfair trials, alleged torture or other ill-treatment, in absentia death sentences, mandatory death penalties, and special or military courts. These are not isolated procedural details. In capital cases, they go to the legality of the final punishment (Amnesty International, 2026, pp. 9–10).

This finding should be expressed clearly for general readers. The issue is not only what crime was charged. It is also how the conviction was obtained. A state that executes after torture, secrecy, political pressure, or denial of defence rights does not deliver justice. It takes life through an unsafe process.

13.5 Abolition Is the Normative Direction

The final finding is that abolition remains the normative direction of international law. Retention continues, but the legal space for it is contracting. Article 6 narrows use. The Second Optional Protocol abolishes it for parties. Regional abolition treaties strengthen non-use. UN moratorium resolutions and Human Rights Council activity keep institutional pressure on retentionist states.

State practice also supports that direction. Amnesty records 113 countries as abolitionist for all crimes by the end of 2025. The same report notes reform steps in states such as Viet Nam, the Gambia, Zimbabwe, and Kyrgyzstan, including abolition for certain offences, commutation of existing death sentences, and barriers against reintroduction (Amnesty International, 2026, pp. 7, 39–40).

The conclusion should not overstate the law. The death penalty has not disappeared. Some states still execute, and some did so at alarming levels in 2025. The better conclusion is that execution now operates against a dense legal framework that restricts offences, procedures, categories of persons, methods, transfers, and reintroduction.

That is the central meaning of Amnesty’s report on the death penalty under international law. The 2025 surge is not proof of renewed legitimacy. It is proof of a widening gap between aggressive retentionist practice and an international legal order that increasingly treats capital punishment as exceptional, suspect, and moving toward abolition.

Conclusion

Amnesty International’s 2025 report matters because it records more than a numerical increase in executions. It shows a sharp rise in state killing at a time when international human rights law has made capital punishment harder to justify, narrower in scope, and more exposed to legal challenge. The report’s figure of at least 2,707 executions is alarming, but the deeper issue is the type of executions documented: drug-related cases, national security charges, opaque proceedings, unfair trials, mandatory death sentences, and continued secrecy in major executing states (Amnesty International, 2026).

The central legal point is that the death penalty under international law is no longer treated as an ordinary expression of criminal sovereignty. Article 6 of the ICCPR permits retention only within strict limits for states that have not abolished it. It does not give governments a general licence to execute. The right to life remains the starting point, and any deprivation of life by criminal sentence must avoid arbitrariness, comply with fair trial guarantees, and remain confined to the narrow category of the “most serious crimes” (United Nations, 1966, art. 6).

The clearest conflict appears in drug-related executions. Amnesty’s report records 1,257 executions for drug offences in 2025. Under the Human Rights Committee’s interpretation, crimes eligible for capital punishment must involve intentional killing. Drug trafficking, transportation, possession, and related offences may be serious, but they do not ordinarily satisfy that threshold (Human Rights Committee, 2018, para. 35). This makes a large share of the 2025 execution surge legally vulnerable.

Fair trial failures deepen the problem. A death sentence imposed after torture-tainted evidence, denial of effective defence, military or special court proceedings against civilians, in absentia trials without genuine waiver, or mandatory sentencing cannot be treated as a reliable act of justice. In capital cases, procedure is not a technical formality. It is the barrier between lawful adjudication and arbitrary deprivation of life.

The regional picture also weakens any claim of renewed global acceptance. The increase was driven by a limited group of states, especially Iran and Saudi Arabia, while many countries remained abolitionist, observed moratoria, commuted death sentences, or restricted capital punishment. The United States remained isolated in the Americas, China continued to conceal figures, and some states expanded or revived capital punishment despite the abolition-oriented direction of modern human rights law (Amnesty International, 2026).

The final conclusion must be precise. The death penalty has not disappeared. It remains active in several jurisdictions and was used at disturbing levels in 2025. Yet its lawful scope under international law has become so narrow that many contemporary executions no longer look like protected choices of domestic criminal justice. They reveal violations of the right to life, fair trial guarantees, human dignity, and the international movement toward abolition.

References

1. Amnesty International (2026) Death Sentences and Executions 2025. London: Amnesty International. Available at: https://www.amnesty.org/en/documents/act50/0778/2026/en/ (Accessed: 18 May 2026).

2. Council of Europe (1983) Protocol No. 6 to the Convention for the Protection of Human Rights and Fundamental Freedoms concerning the Abolition of the Death Penalty. Strasbourg: Council of Europe. Available at: https://www.coe.int/en/web/conventions/full-list?module=treaty-detail&treatynum=114 (Accessed: 18 May 2026).

3. Council of Europe (2002) Protocol No. 13 to the Convention for the Protection of Human Rights and Fundamental Freedoms concerning the Abolition of the Death Penalty in All Circumstances. Vilnius: Council of Europe. Available at: https://www.coe.int/en/web/conventions/full-list?module=treaty-detail&treatynum=187 (Accessed: 18 May 2026).

4. Economic and Social Council (1984) Safeguards Guaranteeing Protection of the Rights of Those Facing the Death Penalty, Resolution 1984/50. New York: United Nations. Available at: https://www.ohchr.org/en/instruments-mechanisms/instruments/safeguards-guaranteeing-protection-rights-those-facing-death (Accessed: 18 May 2026).

5. European Court of Human Rights (1989) Soering v United Kingdom, Application no. 14038/88, Judgment, 7 July 1989. Strasbourg: European Court of Human Rights. Available at: https://hudoc.echr.coe.int/eng?i=001-57619 (Accessed: 18 May 2026).

6. Hood, R. and Hoyle, C. (2015) The Death Penalty: A Worldwide Perspective. 5th edn. Oxford: Oxford University Press.

7. Human Rights Committee (2018) General Comment No. 36 on Article 6 of the International Covenant on Civil and Political Rights, on the Right to Life, CCPR/C/GC/36. Geneva: United Nations. Available at: https://docs.un.org/en/CCPR/C/GC/36 (Accessed: 18 May 2026).

8. Human Rights Council (2025) The Question of the Death Penalty, A/HRC/RES/60/17. Geneva: United Nations. Available at: https://docs.un.org/en/A/HRC/RES/60/17 (Accessed: 18 May 2026).

9. International Court of Justice (2001) LaGrand (Germany v United States of America), Judgment, I.C.J. Reports 2001, p. 466. The Hague: International Court of Justice. Available at: https://www.icj-cij.org/case/104 (Accessed: 18 May 2026).

10. International Court of Justice (2004) Avena and Other Mexican Nationals (Mexico v United States of America), Judgment, I.C.J. Reports 2004, p. 12. The Hague: International Court of Justice. Available at: https://www.icj-cij.org/case/128 (Accessed: 18 May 2026).

11. International Law Commission (2001) Draft Articles on Responsibility of States for Internationally Wrongful Acts, with Commentaries. Yearbook of the International Law Commission, 2001, Vol. II, Part Two. New York: United Nations. Available at: https://legal.un.org/ilc/texts/instruments/english/commentaries/9_6_2001.pdf (Accessed: 18 May 2026).

12. Organization of American States (1990) Protocol to the American Convention on Human Rights to Abolish the Death Penalty. Washington, DC: Organization of American States. Available at: https://www.oas.org/juridico/english/treaties/a-53.html (Accessed: 18 May 2026).

13. Peng, L-Y. (2026) ‘The customary prohibition of the mandatory death penalty in international law’, Human Rights Law Review, 26(1), ngaf047.

14. Schabas, W.A. (2002) The Abolition of the Death Penalty in International Law. 3rd edn. Cambridge: Cambridge University Press.

15. United Nations (1963) Vienna Convention on Consular Relations. Vienna: United Nations. Available at: https://treaties.un.org/pages/ViewDetails.aspx?src=TREATY&mtdsg_no=III-6&chapter=3 (Accessed: 18 May 2026).

16. United Nations (1966) International Covenant on Civil and Political Rights. New York: United Nations. Available at: https://www.ohchr.org/en/instruments-mechanisms/instruments/international-covenant-civil-and-political-rights (Accessed: 18 May 2026).

17. United Nations (1984) Convention against Torture and Other Cruel, Inhuman or Degrading Treatment or Punishment. New York: United Nations. Available at: https://www.ohchr.org/en/instruments-mechanisms/instruments/convention-against-torture-and-other-cruel-inhuman-or-degrading (Accessed: 18 May 2026).

18. United Nations (1989a) Second Optional Protocol to the International Covenant on Civil and Political Rights, Aiming at the Abolition of the Death Penalty. New York: United Nations. Available at: https://www.ohchr.org/en/instruments-mechanisms/instruments/second-optional-protocol-international-covenant-civil-and (Accessed: 18 May 2026).

19. United Nations (1989b) Convention on the Rights of the Child. New York: United Nations. Available at: https://www.ohchr.org/en/instruments-mechanisms/instruments/convention-rights-child (Accessed: 18 May 2026).

20. United Nations Development Programme, International Centre on Human Rights and Drug Policy, UNAIDS and World Health Organization (2019) International Guidelines on Human Rights and Drug Policy. New York: United Nations Development Programme. Available at: https://www.undp.org/publications/international-guidelines-human-rights-and-drug-policy (Accessed: 18 May 2026).

21. United Nations General Assembly (2024) Moratorium on the Use of the Death Penalty, A/RES/79/179. New York: United Nations. Available at: https://docs.un.org/en/A/RES/79/179 (Accessed: 18 May 2026).

22. United Nations Treaty Collection (2026) Second Optional Protocol to the International Covenant on Civil and Political Rights, Aiming at the Abolition of the Death Penalty: Status of Treaties [online]. Available at: https://treaties.un.org/pages/ViewDetails.aspx?src=TREATY&mtdsg_no=IV-12&chapter=4 (Accessed: 18 May 2026).

23. World Medical Association (2019) WMA Resolution on Prohibition of Physician Participation in Capital Punishment [online]. Available at: https://www.wma.net/policies-post/wma-resolution-on-prohibition-of-physician-participation-in-capital-punishment/ (Accessed: 18 May 2026).