2. Two types of universal jurisdiction defendants: Living with impunity versus living in fear

Most crimes around the world are prosecuted by the state in whose territory they took place. But under the principle of universal jurisdiction, any state may prosecute certain crimes, even if the state in question did not have any territorial, nationality, or national-interest link with the crime when it was allegedly committed. Piracy is a classical universal jurisdiction crime.Footnote ³ Since pirates are highly mobile, any state that catches pirates may apply its laws to try and punish them. In recent decades, states have also claimed universal jurisdiction over the so-called core international crimes – including crimes against humanity, genocide, torture, and war crimes.Footnote ⁴ These core international crimes are established by customary international law and have jus cogens status. As such, any state may apply its laws to try and punish them.Footnote ⁵

The predominant justification for the exercise of universal jurisdiction of the core international crimes is preventing impunity. Since crimes against humanity, genocide, torture, and war crimes are typically committed by state officials, the state whose officials committed these crimes is often unwilling or unable to prosecute and punish them. To prevent these crimes from going unpunished, any state should then be entitled to apply its laws to prosecute them.Footnote ⁶

Under this predominant account of universal jurisdiction, then, the typical universal jurisdiction defendant is a current or former state official living with impunity in their home state or in a friendly state that has no interest in prosecuting them. In other words, it is someone who has allegedly committed an international crime but whose state of nationality is unwilling or unable to hold them accountable. The exercise of universal jurisdiction against Augusto Pinochet fits into this account.

Augusto Pinochet was the dictator of Chile from 1973 until 1990. During his regime, thousands of people were kidnapped, tortured, and killed. After he left office, he remained as commander-in-chief and then senator-for-life under the same constitution he had enacted during his regime. His hold on the Chilean military and an amnesty that had been passed during his regime enabled him to live with impunity in Chile for most of the 1990s. But in October 1998, when he was in London for a medical procedure, he was arrested by Scotland Yard based on an arrest request by the Spanish investigating judge Baltasar Garzón. The justification for such an arrest was again preventing Pinochet from living with impunity. However, after 15 months in house arrest pending his extradition proceedings, the United Kingdom released him and let him return to Chile.Footnote ⁷

Today, defendants who are ‘Living with Impunity’ are typically high-level government officials that hold positions of power in their home state’s regime. They are often targeted for prosecution when they visit a foreign state on official business. For example, Bouguerra Soltani was a leader of the Movement for the Society of Peace (an Islamist party) and a former minister in the Algerian government. In 2009, TRIAL International, a human rights NGO, filed a criminal complaint against Soltani after learning that he was speaking at an event in Switzerland. TRIAL International accused Soltani of ordering, authorizing, and inciting public officials to commit acts of torture. After learning of the charges, Soltani immediately fled back to Algeria where he did not face judicial consequences.Footnote ⁸

These defendants often hold immunity, which enables them to leave the state where the complaint is filed and travel safely back to their home state.Footnote ⁹ In the rare cases in which a defendant is arrested and held, he often receives legal and diplomatic support from his home state. For example, Khaled Nezzar, the former Minister of Defense of Algeria, was detained for 48 hours in Switzerland in October 2011, before being released under the condition that he attended subsequent hearings.Footnote ¹⁰ He could then return to Algeria without being bothered.Footnote ¹¹ After proceedings against him dragged on for over a decade and included discussions over immunity and other legal issues, he was indicted by Swiss prosecutors for alleged crimes against humanity and war crimes committed in Algeria, and a trial start date was set for his case for June 2024.Footnote ¹² But there were serious doubts that he would come to trial because Algeria protected him and would most probably refuse to force him to come to Switzerland.Footnote ¹³ On 29 December 2023, Mr. Nezzar died in Algiers at the age of 86.Footnote ¹⁴

Another example is the case of Hamid Nouri who was a prison official and prosecutor in Iran. In 2019, he was arrested while traveling for personal reasons to Sweden. He was accused of committing crimes against humanity and war crimes in connection with mass killings of prisoners in Iran in 1988. Iran described the trial as illegal and formally protested the trial to the Swedish envoy to Tehran. Despite these complaints, Nouri was found guilty and sentenced to life in prison in May 2022.Footnote ¹⁵ However, Iran did not give up in its efforts to protect Nouri, and he was released in a prisoner exchange with Sweden in June 2024.Footnote ¹⁶

In contrast, defendants who are ‘Living in Fear’ of their home state have typically fled their home state and sought asylum in a host state. Some of these defendants have found themselves on the losing side of a civil war and have fled their home state in fear for their lives. For example, Pascal Senyamuhara Safari – who is primarily known by his alias Pascal Simbikanwa – was a Chief Intelligence Officer for the Habyarimana regime that held power during the genocide in Rwanda in 1994. After the genocide, he fled Rwanda first to the Democratic Republic of the Congo, then east Africa, the Comoros Islands, and finally in 2005 Mayotte – a French department in the Indian Ocean – where he requested refugee status using a false name. He was arrested there by French authorities in 2008 for his participation in the production of false documents. At that point, the authorities discovered his true identity, as well as the fact that Rwanda wanted him for his participation in the genocide and that there was an Interpol red notice out against him. But France rejected Rwanda’s extradition request and instead tried him first in relation to the false documents and then for his participation in the genocide in Rwanda.Footnote ¹⁷ Four NGOs participated as civil parties in this second trial. He was convicted by the Court of Assises in Paris for genocide and complicity in crimes against humanity and sentenced to 25 years imprisonment on 14 March 2014. The conviction and sentence were later confirmed by a second French court of assises and by the French Cassation Court.Footnote ¹⁸

Similarly, Ousman Sonko was the Minister of Interior of The Gambia for about a decade during the presidency of Yahya Jammeh, whose regime was accused of committing very serious human rights abuses. In the final months of Mr. Jammeh’s rule, in September 2016, Mr. Sonko applied for residency in Sweden but was denied.Footnote ¹⁹ He was arrested in Switzerland at a centre for asylum seekers in January 2017.Footnote ²⁰ On 15 May 2024, he was convicted in Switzerland for crimes against humanity committed in The Gambia.Footnote ²¹

Defendants who are living in fear may even become citizens of the prosecuting state. Alemu Eshetu came to the Netherlands as a refugee and eventually acquired Dutch citizenship. Meanwhile he was tried in absentia in his home state of Ethiopia, and sentenced to death for the killings of many persons during the Derg regime. After the Netherlands created its war crimes unit, he was arrested in 2015. On 15 December 2017, a Dutch court found him guilty of war crimes, including arbitrary detention, inhumane treatment, torture, and murders. He was sentenced to life in prison in the Netherlands.Footnote ²²

Other defendants who are living in fear of their home state include those who held positions of power in their home state but chose to defect from the ruling regime. For example, Eyad al-Gharib was an official in Syria’s notorious General Intelligence Directorate. He was part of Branch 251, which is responsible for internal security in the Damascus region. In 2013, he defected from the Assad regime and spent five years in hiding before entering Germany and seeking asylum. In 2019, he was arrested on charges of aiding and abetting crimes against humanity because of his involvement in transporting and burying the bodies of detainees from Syrian detention facilities. He was sentenced to four and a half years in prison in 2021.Footnote ²³

In each of these cases, the defendants were not living with impunity. To be sure, when they were subjected to universal jurisdiction prosecutions, they had not been punished in their home state yet. But Rwanda and The Gambia wanted to try Simbikanwa and Sonko respectively. Eyad al-Gharib was at risk of being extrajudicially executed and/or tortured if returned to Syria, and Ethiopia had convicted Eshetu in absentia and sentenced him to death. Knowing the fates that awaited them in their home states, these defendants were leaving in fear, claiming refugee status in their receiving countries, and hiding from authorities. In these cases, the justification of universal jurisdiction trials was then not only or even mainly preventing impunity, since deportation or extradition back to their home state would have subjected these defendants to punishment or other harsh consequences. Rather, the main justification of these universal jurisdiction trials could be understood as providing a fair trial and sentence for these defendants.

Some prior scholars have examined how the incentives of the political branches of the prosecuting state can affect the prosecuting state’s willingness to pursue universal jurisdiction cases.Footnote ²⁴ Factors that can affect the exercise of universal jurisdiction include the benefits and costs that individual universal jurisdiction prosecutions would bring to the legislature and the executive branch of the prosecuting state.Footnote ²⁵ In this article, we consider a complimentary question: how politics in the defendant’s home state affects whether a universal jurisdiction case will occur.Footnote ²⁶ Defendants living in fear and defendants living with impunity have different relationships with their home government. While defendants living in fear are disfavoured by the sitting government of their home state, defendants living with impunity are either serving in their home state’s sitting government or openly or tacitly supported by the government. In this regard, the distinction between defendants living with impunity and defendants living in fear is a political distinction since it relies on the political relationship of the defendant with his/her/their home state government.Footnote ²⁷ But as this article demonstrates, this distinction has important implications for universal jurisdiction as a legal concept and for the legal cases that rely on such a concept.

Others have noted the opposition between ‘small versus big fish’ to distinguish between universal jurisdiction defendants.Footnote ²⁸ But the difference between ‘small versus big fish’ does not mirror the distinction we introduce in this article either. After all, Nezzar, Nouri, and Soltani (who we classify as living with impunity), and Simbikanwa, Sonko, and Alemu (who we classify as living in fear) would all have been considered ‘big fish’ when they (allegedly) committed international crimes. And in Germany, the ‘small fish’ Eyad Al-Gharib was tried jointly with the ‘big fish’ Anwar Raslan, a colonel who led an intelligence unit in Syria but that, like Al-Gharib, was a universal jurisdiction defendant living in fear.Footnote ²⁹

The following sections explore important descriptive, predictive, and normative implications of our analytical distinction, demonstrating that classifying defendants as living with impunity or living in fear provides important insights about universal jurisdiction theory and practice.

3. The evolution of universal jurisdiction complaints over time

This section shows that defendants living in fear make up a substantial portion of universal jurisdiction defendants, and that their proportion has increased since the early 2000s. We begin by describing how we collected our database of universal jurisdiction defendants. We then describe how we classified each defendant in the database based on attributes of the defendant into one of the two categories: ‘Living with Impunity’ and ‘Living in Fear’. Finally, we show how the frequency of these two types of defendants has evolved over time.

3.2 Measuring defendants attributes

To classify the defendants, we considered their life circumstances at the time that the complaint was filed against them and in subsequent case proceedings. Namely, we sorted them into two mutually exclusive categories. The first category – ‘Living with Impunity’ – includes individuals like Augusto Pinochet who fit into the classic understanding of universal jurisdiction defendants. The second category – ‘Living in Fear’ – includes individuals who migrated to escape legal liability and/or violence in their home state.

This sorting process required a holistic assessment based upon the best available information about the defendant. However, we inductively developed consistent criteria to guide this assessment. These criteria are available in Table 1.

Table 1. Classifying the Defendants

In our view, one of the strongest indicators that a defendant was living with impunity is that he or she was a government official who was able to remain in his or her office, even after the complaint was filed. For example, Khaled Ben Saïd was convicted in absentia in France and sentenced to 12 years of imprisonment for torture committed in Tunisia against a Tunisian citizen. He was a Tunisian Vice Consul in France and when the complaint was launched he returned to Tunisia where he was said to continue working in the Tunisian Ministry of the Interior.Footnote ³² Similarly, home states often publicly condemned universal jurisdiction cases and publicly pledged to support the defendants, like when a Spanish investigating judge issued formal charges against 40 Tutsi former members of the Rwandan Patriotic Front and Rwandan officials.Footnote ³³ President Paul Kagame condemned the Spanish investigation and the Spanish Supreme Court dismissed the case against all of them.Footnote ³⁴ Other more subtle indicators of living with impunity include the ability of a defendant to return to his home state and/or maintain a residence, even after being named in a universal jurisdiction complaint.

In contrast, the strongest indicator that a defendant was living in fear was that a state or institution with territorial or active nationality jurisdiction over the crime requested extradition of the defendant. For example, in the 1980s the UK received numerous universal jurisdiction complaints against Soviet nationals who allegedly committed war crimes during the Second World War and were hiding in the UK. The USSR asked the UK to extradite these suspects so that they could face trial in their home country, but the UK refused to do so.Footnote ³⁵ Similarly, many of the defendants in our dataset are Rwandan Hutus who were living in hiding in various European states while fearing extradition to either Rwanda or the International Criminal Tribunal for Rwanda (or both).Footnote ³⁶

Another common indicator that defendants are living in fear is that they apply for (and are often granted) status as refugees, often in the state that ultimately prosecutes them.Footnote ³⁷ Under the 1951 Refugee Convention and 1967 Protocol, to receive refugee status, an individual must, owing to a well-founded fear of being persecuted for reasons of race, religion, nationality, membership of a particular social group or political opinion, be outside the country of his nationality, and be unable, or, owing to such fear, be unwilling to avail himself of the protection of that country; or who, not having a nationality and being outside the country of his former habitual residence as a result of such events, be unable or, owing to such fear, be unwilling to return to it.Footnote ³⁸ But the status of refugee does not apply to any person who has committed a crime against peace, a war crime, or a crime against humanity.Footnote ³⁹ In some cases, defendants who were living in fear became the subject of universal jurisdiction complaints because evidence of their crimes came to light during the process of their refugee application. In other cases, defendants received refugee status before evidence of their alleged crimes was reported or came to light.

Third, defendants who are living in fear are often prosecuted in other states in absentia. These trials are sometimes the basis for arrest warrants and requests for extraditions.Footnote ⁴⁰

Fourth, defendants who are living in fear are often members of rebel or terrorist groups like the Free Syrian Army or Islamic State. Given their past histories of challenging the government of their home state, these individuals are inherently fearful of political retribution.

Fifth, a handful of defendants explicitly said they were living in fear during court proceedings. These defendants explicitly argued that deportation to their home state would violate the host state’s legal obligations, because of the risk that they would face persecution or harm in their home state.

And finally, we felt confident that a small group of defendants were living in fear based on the details of their case, such as the timing of their flight, the details of their alleged crimes, and their former professions in their home states. These individuals were primarily Hutus in Rwanda and pro-Assad defectors in Syria. However, we could not classify these individuals using the other criteria. We accordingly added the final criterion to our list: ‘The defendant fled his home state after the crimes were committed.’

We were unable to classify 52 complaints into one of these two categories because there was insufficient information about the defendant’s circumstances when the complaint was filed. This coding process left us with 1,013 observations in our final dataset.

Across all observations, we classified 78.1 per cent of complaints as applying to defendants who were living with impunity and 21.9 per cent of complaints as applying to defendants who were living in fear. In other words, about one out of every five universal jurisdiction defendants was living in fear when the universal jurisdiction complaint against them was filed.

In addition to identifying which defendants were ‘living with impunity’ versus ‘living in fear’, other variables that we use for this article are the name of the defendant, the defendant’s home state (i.e., the state of the defendant’s nationality when the crime was allegedly committed, which is often also where the alleged crimes took place), the prosecuting state (i.e., the state filing the universal jurisdiction complaint or before whom the universal jurisdiction complaint was filed), and proceeding variables that we discuss later in the article. We also use trial outcome (i.e., whether an accused was convicted for all, some or no charges at trial) and sentence (i.e., which sentence the accused received if convicted).

3.3 ‘Living with Impunity’ versus ‘Living in Fear’ over time

Is ‘Living in Fear’ a relatively new phenomenon, or has it been a persistent feature of universal jurisdiction cases? To answer this question, we examined a cross-section of our data, first by comparing ‘Living in Fear’ rates by decade, and then by year.

3.3.1 Comparisons by decade

Table 2 breaks down our data by decade beginning in 1980. At the start of this research project, we initially hypothesized that the percentage of defendants living in fear would increase over time due to the growth of international migration,Footnote ⁴¹ the use of extradition requests and Interpol’s red notices by states against political and military opponents,Footnote ⁴² and the ratification of new treaties by states and the expansion of refoulment prohibitions with them in response to extradition requests.Footnote ⁴³ But our data by decade show instead a ‘V’ trend with higher percentages of defendants living in fear in the 1980s and the 2010s, and lower percentages in the 1990s and 2000s. The variation in the percentage of defendants living in fear seems to be related to the nationals of which states are subjected to universal jurisdiction rather than to the factors we initially considered.

Table 2. Variation in Defendant-Type by Decade

From 1980 to 1989, nearly 69 per cent of complaints were against defendants living with impunity, and about 31 per cent were against defendants living in fear. The defendants living in fear during this time period were mostly people who had allegedly been Nazis or Nazi collaborators. These individuals were wanted by the Soviet Union for allegedly participating in Nazi-era international crimes. Between 1990 and 1999, attention shifted to Latin America. Human rights activist and victims began filing universal jurisdiction complaints to seek justice for alleged atrocities committed in places like Argentina, Chile, and Guatemala. Most of these defendants were living in impunity. However, the number and percentage of complaints against defendants living in fear began to grow dramatically in the 2000s and beyond with the rise of cases against Rwandan Hutus, Syrian migrants, members of Islamic State, and other similar individuals who could not be returned to their home state for prosecution without risking their basic human rights.

3.3.2 Comparisons by year

Figure 1 visually shows comparisons in the data by year. Once again, we see that complaints against individuals who are living in fear are a regular and persistent feature of universal jurisdiction cases. These cases are not new, unique, or extraordinary.

Figure 1. Variation in Defendant-Type by Year

Careful readers may question the large spike in living in fear complaints in 2007. This outlier was caused by a major campaign to launch a case in Spain against individuals from Western Sahara. The defendants named in these complaints were members of the Polisario Front rebel group and/or the Sahrawi Arab Democratic Republic government, which is not recognized by Morocco as a legitimate government.

Also note that the decline in cases in recent years is likely a result of our data collection process since the filing of many cases is not immediately publicly known or easy to determine and of the fact that we are excluding John Doe cases from our analysis for the reasons already explained supra.Footnote ⁴⁴ Governments rarely publicly announce every complaint they receive (or case they consider by their own motion). We are thus reliant on media, NGO publicity, other scholars, and the legal process itself to reveal over time what complaints have been filed. As complaints generate investigations and hearings over time, we will have more knowledge about the documents that initiated these processes.

4. The likelihood that each type of defendant will be arrested and tried

We can now examine whether defendant-types differ in their likelihood of being investigated, having a formal procedure opened, arrested and tried following a universal jurisdiction complaint. We expect that proceedings against defendants living in fear would move forward towards arrest and trial at higher rates than defendants living with impunity for multiple reasons.Footnote ⁴⁵ First, the home state of the defendant will typically be supportive of prosecuting defendants who are living in fear because these defendants are often opponents of the home state regime. Often, the home state will put pressure on the receiving state to prosecute these defendants through an Interpol red notice, an extradition request, or in a more informal fashion. In contrast, defendants who are living with impunity are usually supported by their home state, meaning that the prosecuting state will face pressure from the home state to end the proceedings.

4.1 Arrests

Table 3 presents data on arrests that are consistent with this prediction. Arrests may take place at any time after a complaint is filed.Footnote ⁴⁶ An arrest does not necessary mean that the defendant stays in jail pending prosecution and trial, since defendants may be released on bail and may even flee the prosecuting state before facing trial.Footnote ⁴⁷ But arrests constitute deprivations of physical freedom for defendants and are much more invasive than the mere filing of a complaint. As such, arrests impose substantial burdens on defendants. Arrests also require that the prosecution meets a burden of proof of probable cause or a similar standard that the defendant committed the offense. Arrests thus convey that there are evidentiary grounds for the allegations against the defendant. Consequently, arrests of defendants living with impunity may trigger protests or other ways to oppose the arrest by the defendant’s sending state. In contrast, the sending state may welcome an arrest when the defendant is living in fear.

Table 3. Arrests by Defendant-Type

It is worth emphasizing two points about the data on arrests. First, notice that defendants living in fear are more than ten times more likely to be arrested than defendants living with impunity, confirming our prediction that the former are much more likely to be arrested than the latter.

In addition, defendants living in fear have almost a 50 per cent chance of being arrested if a complaint is filed against them. This is remarkable since most criminal complaints around the world do not end up with an arrest as defendants are often not identified or found; cases get dismissed for insufficient elements of proof, for public policy considerations, or for formal reasons; and in the cases that are charged or go to trial many states summon the defendant instead of arresting him/her.

4.2 Proceedings

Criminal procedure varies substantially among states. States establish different requirements and procedural steps that a case has to meet to move towards a verdict and punishment, use different terminology and concepts that do not find an exact equivalent in other legal systems, and have different conceptions of what the goals of criminal procedure are. States also differ on which should be the right balance among these goals when they conflict with each other, and have different understandings of what the role of prosecutors, defense attorneys, judges, and other actors is. Overall, the institutional, cultural, political, and economic settings where criminal procedure operates varies widely between states.Footnote ⁴⁸

However, every criminal process includes a sequence of potential or actual procedural steps towards adjudication. To analyse universal jurisdiction cases across states, we thus created a common criminal procedure matrix of proceedings variables from the filing of a complaint towards adjudication. Besides arrests already discussed in the prior subsection, these proceedings variables include complaints, investigative measures, formal proceedings, and trials. A ‘complaint’ includes two types of situations: (i) public authorities of the prosecuting state consider whether a certain person should be investigated for the commission of a criminal offense; or (ii) a private individual, NGO, another state, or other legal entities file a report before the authorities of the prosecuting state alleging that a person has committed a criminal offense. ‘Investigative measures’ is the next procedural step and it applies if the prosecuting state investigates the allegations against the defendant or inquires whether other states are investigating or prosecuting the defendant. Next, ‘Formal proceedings’ applies when the prosecuting state issues an indictment, information or other formal charge against the defendant, when one of its investigating judges or prosecutors moves the proceedings further forward against the defendant, or when the authorities arrest the defendant to move forward with the charges against him. ‘Trial’ is the next procedural step and refers to the adjudication stage at which a court or other adjudicating body decides whether the defendant is guilty or not guilty.

Figure 2 represents the relationship between these procedural steps as a temporal sequence towards adjudication. Every case against a defendant starts with a complaint. This complaint may be dismissed or investigated. If the complaint against the defendant is investigated, the case may again be dismissed, or formal charges may be issued against a defendant. At the formal proceedings stage, once again the case may be dismissed, or the case may be brought for adjudication at trial. At any of these stages, the case against the defendant may be dismissed due to insufficient evidence, for public policy reasons (e.g., the prosecution decides that pursuing the case forward is not in the public interest), or for formal reasons (e.g., the defendant has immunity).

Figure 2. Procedural Steps of a Universal Jurisdiction Case Against a Defendant

The further a case against a defendant moves forward in this temporal sequence of procedural steps, the closer the defendant is to getting convicted for the commission of a core international crime. One would thus predict that cases against defendants living in fear are more likely to move through this temporal sequence of procedural steps, since in these cases the home state often puts pressure on the receiving state to prosecute.

Consistent with our prediction, Tables 4(a) and 4(b) show that a case against a defendant living in fear is much more likely to move from one procedural step to the next than the case of a defendant living with impunity. A complaint against a defendant living in fear is about 30 times more likely to reach the trial stage than a complaint against a defendant living with impunity. Also, almost one out three complaints against defendants living in fear end up with a trial. This is remarkable given the challenges to build universal jurisdiction cases for alleged international crimes that happened hundreds or thousands of miles away of the prosecuting state.

Table 4. The Impact of Living in Fear on the Stages of Legal Proceedings

5. Do universal jurisdiction trials fulfill their due process and proportionate punishment functions?

In Section 2, we explained that in the case of defendants living in fear, the justification for universal jurisdiction prosecutions and trials is not only or even mainly preventing impunity, since many defendants will be punished or be subjected to other negative consequences – such as being tortured, extra judicially executed, subjected to a show trial, or sentenced to death – if they were returned to their home state. Rather, in many of these cases, the justifications for the exercise of universal jurisdiction are providing due process to defendants so that they are not wrongly convicted, and their procedural rights are respected, and, in case of conviction, applying a proportionate punishment to them.

Based on the universal jurisdiction database already described, we assess whether universal jurisdiction exercised against defendants living in fear has lived up to these justifications. To provide broader context, we also include data on trials for defendants living with impunity. But we make no predictions about whether each type of defendant would have a higher likelihood of being convicted or more heavily sentenced, as this is heavily context-dependent.

Table 5 shows that the percentage of total and partial acquittals in universal jurisdiction trials has been substantial.Footnote ⁴⁹ These acquittal rates suggest that universal jurisdiction trials are, on average, true adjudicatory processes, – i.e., processes in which the determination of guilt or innocence is based on the available elements of proof, rather than predetermined by political expediency. There has also been a substantial percentage of total and partial acquittals at trials against defendants living in fear, many of whom would have otherwise been subjected to torture, extrajudicial execution, a show trial, or the death penalty, had they been extradited to their home country. Instead, these defendants were fully acquitted in ten per cent of universal jurisdiction living-in-fear trials, and fully or partially acquitted in forty per cent of these trials. In addition, two convictions – one against a defendant living with impunity, another against a defendant living in fear – were reversed on appeal, making the percentage of defendants fully acquitted at trial or on appeal even larger.Footnote ⁵⁰ In the small number of trials against defendants living-with-impunity, the substantial acquittal rates suggest that these trials have also been true adjudicatory processes – though it is not possible to make strong conclusions with confidence on these cases given the low number of trials in this category.

Table 5. Trial Outcomes in Universal Jurisdiction Cases

* The total number of trials is higher than the addition of living-with-impunity and living-in-fear trials because we did not have sufficient information to classify five defendants in either category

** Acquittal on some charges and conviction on some charges

Table 6 shows the range of sentences against defendants living with impunity and defendants living in fear who were convicted in universal jurisdiction trials for crimes against humanity, genocide, torture, and war crimes.

Table 6. Sentences Against Universal Jurisdiction Defendants

* The total number of sentenced defendants is higher than the addition of living-with-impunity and living-in-fear sentenced defendants because we did not have sufficient information to classify four defendants in either category

** There are only five defendants living with impunity against whom a sentence was ultimately imposed because the conviction against a 6^th defendant was reversed on appeal.

*** There are only 62 defendants living in fear against whom a sentence was ultimately imposed because the conviction against a 63^rd defendant was reversed on appeal.

About a third of convicted defendants living in fear were sentenced to life imprisonment – a sentence not necessarily disproportionate or otherwise inadequate given that some core international crimes are among the most serious crimes imaginable. Again, given the potential punishments or other negative consequences these defendants would have faced had they been returned to their home country, even a sentence of life imprisonment may be playing a protection function for them.

More than half of convicted defendants living in fear received a sentence of 14 years of imprisonment or less, about 40 per cent less than nine years of imprisonment, and remarkably about a fifth less than five years of imprisonment. Since these are sentences for the commission of core international crimes, these sentences are, on average, not disproportionately harsh – though in some cases, they may be too lenient for the crime in question.Footnote ⁵¹ At the same time, it is important to notice that core international crimes include war crimes, some of which may not be among the most serious crimes. For instance, there have been convictions for war crimes such as degrading treatment against defendants who appeared in a photograph next to the corpse of a person protected under humanitarian law. Though this behaviour may be reprehensible, it is not nearly as serious as, for instance, a killing of a protected person or one or more killings committed as part of a widespread or systematic attack against a civilian population, or killings committed with the intent to destroy, in whole or in part, a national, ethnical, racial or religious group, as such. In any case, since many defendants living in fear would have faced unfair treatment and trial, the death penalty, and dire prison conditions had they been returned to their home country, sentences lower than 14 years of imprisonment may be playing a protection function for them.

The number of living-with-impunity defendants who have received sentences is too low to make any strong conclusions regarding these cases.

Our interpretation of these data is tentative because, in contrast with our data on universal jurisdiction trials and sentences, we only have anecdotal information about how defendants living in fear are treated when they are caught, prosecuted, tried, convicted, and sentenced by their home state. We also do not have coded data about certain characteristics of each case to assess whether a sentence is proportionate for it, – e.g., on the number of victims a convicted defendant’s actions affected. However, based on our data and this anecdotal information, it is plausible to infer that universal jurisdiction trials and sentences may be advancing their goals of protecting defendants living in fear against wrongful conviction, unfair treatment, and disproportionate sentencing.

6. Should universal jurisdiction democracies do anything against being used by autocracies to prosecute the latter’s opponents?

The previous section showed that universal jurisdiction states seem to have generally respected minimum due process standards and to have generally not applied disproportionately harsh punishments against defendants living in fear. However, there are other normative and public policy issues in relation to the prosecution of these defendants. One of these issues is that authoritarian states may be using universal jurisdiction to make other states prosecute and punish their political or military opponents by formally or informally reporting the presence of these opponents in the territory of the other state or by requesting the extradition of opponents from the other state. For instance, extradition requests by the authoritarian government of Rwanda have triggered many universal jurisdiction prosecutions in European states. It is also possible – though it is hard to find reliable evidence on it given the nature of intelligence operations – that authoritarian governments like Syria have done intelligence operations in other states to prompt or facilitate the universal jurisdiction prosecution of defectors or opponents to their regimes. And even if these intelligence operations have not taken place, these prosecutions by third states are welcome news for these authoritarian regimes.

This is not a problem per se – provided, of course, that the accused person is actually guilty. The motivations of informants and complainants are generally irrelevant when the decision is whether to prosecute and punish perpetrators of heinous crimes. Using a domestic analogy, if a national of state X commits a murder in the territory of state X against a national of state X, it is generally irrelevant that the authorities of state X learn about the murder through an informant or complainant who is himself engaged in problematic behaviour and who is not looking for justice, and who rather reports the murderer because he is the murderer’s competitor in another realm, such as business, politics, personal relations, or illegal activities. The murderer should still be punished.

Similarly, if a universal jurisdiction state learns about a defendant living in fear in its territory through a formal or informal report or other prompting by the defendant’s authoritarian home state, it would still be justified to punish her or him for the commission of a serious core international crime based on the goals of punishment, such as retribution, deterrence, or the enforcement of a social norm against the commission of these crimes.Footnote ⁵² In other words, it would still be just to punish the perpetrator, and the punishment could advance the goal of creating incentives against the commission of international crimes for other defendants, and of strengthening the social norm around the world against the commission of such crimes.

However, the instrumental prompting of or pressure for these prosecutions by authoritarian states may create problems in some situations. For example, when both sides of a conflict commit international crimes, but the prompting or pressure by the authoritarian state leads to the prosecution of only one side by universal jurisdiction states. For instance, in Rwanda, on top of the genocide committed by many Hutus, there were allegations of crimes against humanity and war crimes committed by the predominantly Tutsi members of the Rwandan Patriotic Front that have been running the country since they took power in 1994.Footnote ⁵³ In Syria, there have been allegations that both members of Assad’s regime and opponents to it have committed international crimes. But if only Hutus and opponents of the Assad’s regime were forced to emigrate to universal jurisdiction states, only the crimes committed by one side are likely to be prosecuted by universal jurisdiction states prompted or pressured by the home state. This pressure is partially enabled by international treaties and organizations that are designed to promote judicial co-operation, such as extradition treaties and Interpol.

These situations create three types of problems. The first is a problem of equality before the law. One of the ideals of any law, including international criminal law, is that it must be equally applied across people and cases.Footnote ⁵⁴ But if two sides commit international crimes and only the members of one side are prosecuted, tried, and punished, international criminal law is not equally applied.Footnote ⁵⁵ This may provide grounds for claims by individual defendants on tu quoque, selective prosecution or other similar grounds.Footnote ⁵⁶ It may also provide grounds for criticism by the broader audiences to whom these prosecutions, trials, and punishments are directed – such as members of the societies where the international crimes occurred or the international community as a whole.Footnote ⁵⁷

The second problem is that, since Nuremberg, one of the functions of international criminal jurisdictions has been setting a historical record about the commission of atrocities.Footnote ⁵⁸ But if, in a given situation, only the crimes committed by one side are tried, universal jurisdiction prosecutions, trials, and punishment would have an inter-state or inter-situation distortive effect –, i.e., they would provide a distortive historical record that would only partially reflect what happened.Footnote ⁵⁹ Besides their transitional justice dimensions, historical grievances may also set the basis for current and future claims to political power.Footnote ⁶⁰

The third problem is the denial of the right to access to justice and of the right to reparations to victims of defendants living with impunity.Footnote ⁶¹ If prosecuting states focus all their resources on prosecuting living in fear defendants because their home states support these prosecutions, prosecuting states may skew resources away from equally serious cases against defendants who are living with impunity.

Are there measures that universal jurisdiction states could take to avoid these problems when prompted by authoritarian states to prosecute defendants living in fear? If the international crimes in question are not as serious as other international crimes – such as appearing posing in a photograph next to a corpse, or a head of a person – the receiving state could simply choose not to prosecute. The goals of respecting the principle of equality before the law or of not distorting the historical record may be sufficient public policy grounds for the authorities of the receiving state to not pursue punishment when a crime is not very serious.

But if a crime is serious – as most core international crimes are – the principle of equality before the law or accuracy in setting the historical record may not provide enough public policy grounds for the authorities of the receiving state not to pursue the case. Going back to a domestic analogy, if two murders are committed in the territory of state X, each by a member of two competing gangs, the authorities of state X should prosecute and punish both. But if one of the murderers were beyond the reach of state X – for instance, because he escaped outside of the boundaries of the state – the principle of equality before the law would not provide enough grounds not to punish the other murderer. In this situation, the authorities of state X should still punish the murderer within their reach, while doing their best to still find and hold accountable the murderer who escaped outside of their territory.

If this analogy held in the context of universal jurisdiction, a prosecuting state should punish a perpetrator that was or is a member of one of the sides in a conflict, while giving their best effort to still find and hold accountable perpetrators of the other side outside of the state’s boundaries. This is what Germany has been doing regarding the situation in Syria. It has launched not only prosecutions against defendants living in fear in its territory, but also against defendants living with impunity in Syria.Footnote ⁶² This policy has advanced, at least to some extent, access to justice for victims, regardless of where the perpetrators are located.

Germany’s approach requires that universal jurisdiction states embrace a global enforcer conception of universal jurisdiction – under which states prevent and punish core international crimes committed anywhere in the world. However, many universal jurisdiction states have instead embraced a no safe haven conception of universal jurisdiction, limiting their role in the universal jurisdiction regime not to be a refuge for participants in core international crimes.Footnote ⁶³ For the latter states, prosecuting perpetrators outside of their territory would not be an option. But these states could still take measures to try to advance the principle of equality before the law, to not distort the historical record, and to advance the right to access to justice and the right to reparations of those harmed by defendants living with impunity in their home states or elsewhere. For instance, no safe haven states could file communications before the International Criminal Court against members of the other side of the conflict or ask the UN Security Council to refer the situation to the ICC so that perpetrators from both sides are prosecuted. No safe haven states could also use tools different from criminal prosecutions, such as issuing sanctions against the authoritarian regime, publicly denouncing or exposing the commission of international crimes by members of the ruling authoritarian regime, supporting the work of human rights international bodies or nongovernmental organizations doing such a work, or getting the international community more involved in documenting ongoing and historical atrocities even if no legal remedies are attached, – e.g., through UN Fact-Finding Missions.