Introduction
Authoritarian governments increasingly reach across borders to threaten dissidents and opponents residing in other countries.Footnote 1 The 2018 assassination of exiled journalist Jamal Khashoggi in Saudi Arabia's consulate in Istanbul reveals the great lengths repressive rulers will go to silence opponents abroad. Regimes engaging in transnational repression rely on a range of tactics, from Interpol listings, renditions, and assaults to pressure on home country relatives and online harassment.Footnote 2 Given their border-blurring qualities, digital technologies are essential components in the toolkit of extraterritorial coercion. Practices of digital transnational repression (‘DTR’) include surveillance and hacking attacks, online harassment, and disinformation campaigns against migrants with ties to authoritarian countries.Footnote 3 They provide repressive regimes with ways to monitor and respond to the activities of exiles with greater scope and speed.Footnote 4 Moreover, these digital threats are often intertwined with traditional methods of extraterritorial coercion, preparing or triggering an escalation of threats. In the Khashoggi case, the Saudi regime decided to go ahead with the operation against the journalist after it had penetrated the smart phone of one of his close associates living in Canada and apprehended details of the projects the two dissidents were planning.Footnote 5
Despite the increasing occurrence of DTR and the potentially severe consequences for the personal lives and political activities of those targeted, its normative implications and security dimensions remain underexplored, particularly with regard to the states hosting the affected exiles. Digital threats against civil society are typically considered as human rights violations, infringing on the rights to privacy and freedom of expression, among others.Footnote 6 However, the human rights approach primarily shines a spotlight on the relation between the authoritarian state as rights violator and its ‘subjects’ abroad, thereby obscuring the role and interests of the host state. Only recently research has turned to investigating host governments’ obligations under international humanitarian law and their responses to transnational repression.Footnote 7
While we believe that it is necessary and urgent to continue pursuing the issue under a human rights framework, in this article we explore an alternative approach and consider practices of DTR as sovereignty violations against the state hosting the targeted migrants, both in a juridical and a political sense. We argue that acts of DTR can violate state sovereignty in that they constitute extraterritorial enforcement jurisdiction; distort public debate and interfere with national self-determination; and impede the host state's adherence to fundamental norms of international humanitarian law. In addition to these three normative arguments grounded in international law, we stress that DTR is also contrary to states’ self-interest: by weakening government institutions, rule of law, and social cohesion in the receiving countries of political exiles, practices of DTR undermine the host state's internal authority, domestic sovereignty, and capacity to successfully integrate immigrants.Footnote 8
Framing transnational repression as a violation of sovereignty has two advantages: First, it spells out more clearly the ways in which practices of extraterritorial authoritarian rule interfere with the interests and authority of other states, underlining that transnational repression does not threaten ‘only foreigners’, but also raises security issues for the host state. Even Western democracies are at times reluctant to apply the language of human rights to cases of extraterritorial repression, given their own use of force in counterterrorism operations abroad.Footnote 9 Foregrounding the aspect of sovereignty violation highlights the normative and strategic relevance to counter DTR. This focus might push host country governments more effectively to meet their obligations for protecting the targeted individuals as well as society more broadly against such practices. Second, emphasising sovereignty violations caused by DTR meets the main perpetrators on their own discursive grounds. Regimes engaging in transnational repression and human rights violations, such as China and Russia, typically stress the normative importance of state sovereignty and condemn external interference, including in the context of digital technologies.Footnote 10 Among other instances where these regimes have violated state sovereignty in a more drastic sense, their extraterritorial practices in the context of DTR are clearly at odds with this rhetoric.Footnote 11 As the ‘only generally acceptable and practical normative basis of world politics’,Footnote 12 the notion of sovereignty could help establish an inclusive discourse for condemning and curtailing DTR across political divides.
Our arguments should be read with the following clarifications in mind. First, our emphasis on the integrity of state sovereignty is not a return to a brand of realism that conceives of national security as an end in itself. On the one hand, stressing that the principles associated with sovereignty are frequently violated, a purely realist view would likely register DTR as a rather low-intensity transgression and subordinate potential answers to the interest of stability and other strategic priorities.Footnote 13 On the other hand, any response to DTR primarily focusing on national security and ignoring the needs of civil society would run the risk of furthering the securitisation of the digital sphere that is already unfolding in relation to other, similar phenomena, such as the spreading of misinformation across borders.Footnote 14 Instead, we follow a broadly liberal approach that considers sovereignty as essential to guarantee the rule of law, particularly individual rights and human security within a given territory.Footnote 15 In the context of digital technologies, this perspective corresponds to a ‘human-centric approach to cybersecurity’.Footnote 16
Second, while focusing on sovereignty, we acknowledge that the transnational nature of digital networks is principally in tension with the concept of territorially bounded sovereignty. In cases of DTR, we will show, conventional notions of sovereignty mean that perpetrators remain unpunished because they never physically enter another state's geographical territory when committing these acts. However, we reject assumptions that sovereignty is particularly threatened or obsolete because of the Internet's global diffusion.Footnote 17 Such arguments ‘represent classical “Westphalian” sovereignty as far more complete than even the most powerful state ever managed to achieve historically’.Footnote 18 In contrast, we build on ‘post-territorial’Footnote 19 or ‘elastic’Footnote 20 conceptualisations of sovereignty that adapt and reconfigure sovereign authority in order to meet the challenges of the digital age.
Third, and following from the above, we concur with approaches that view sovereignty as a social construct; as a result of practices that ‘produce, reform, and redefine sovereignty and its constitutive elements: population, recognition, authority, and territory’.Footnote 21 The construction of state sovereignty is a process of drawing boundaries – not only to demarcate a specific territory or determine political authority, but also to impose ‘meanings about who belongs and who does not belong to the nation’.Footnote 22 Research in critical migration and security studies has analysed this inclusionary/exclusionary function of sovereignty as a mechanism for ‘othering’ and securitising immigrants.Footnote 23 As a consequence, exiles and diasporas often find themselves in a grey zone between the authoritarian home state that ‘treats its subjects abroad as if they were still under territorial jurisdiction’Footnote 24 and the host state failing to accord them full access to a political community ‘that defends and upholds one's right to have rights’.Footnote 25 Instead of marginalising immigrants and their security needs, we suggest stretching the protective boundaries drawn by sovereignty to exclude the authoritarian practices that often follow them across borders.
Our article brings research on globalised authoritarian repression in dialogue with international law and scholarship on digital sovereignty and cyber security. After summarising the relevant debates, our empirical section provides a more detailed picture of DTR, its methods and effects on diasporas and their host societies. We then present three legal arguments and one political argument clarifying the ways in which DTR can be understood as a sovereignty violation. We also outline possible pathways for countering DTR, focusing on distributed cyber security, law enforcement activities, sanctions, and international norm development. In concluding, we discuss the political conditions that would facilitate a more stringent countering of DTR and outline a horizon for further research.
Extraterritorial authoritarian rule and transnational repression
Transnational repression is considered a response of authoritarian governments to intensified cross-border flows of migration and information.Footnote 26 A fundamental aim of regimes who coerce populations abroad is ‘to enhance regime security’ and to strengthen their position, both domestically and on the international level.Footnote 27 Authoritarian rulers seek to contain exiles whose positions risk to gain traction and mobilise challengers inside the country. They also try to silence critics whose claims may increase international pressure and shape the international environment in ways opposing regime interests.Footnote 28 Among the targets of transnational repression are exiled human rights defenders, journalists, and political opponents as well as former regime insiders, ethnic and religious minorities; they range from recent refugees and emigrants to second-generation diaspora members.Footnote 29
Extending domestic coercion across borders, ‘contemporary authoritarian rule structures socio-political space in ways that partially transcend both territorial jurisdiction and geographical distance.’Footnote 30 At the same time, authoritarian extraterritorial rule highlights a power shift in the international system. Global freedom declined for the fifteenth consecutive year in 2020, continuing a trend of democratic recession.Footnote 31 As emboldened authoritarian rulers persecute their opponents across borders, they challenge the norms and values of liberal democracies and the rules-based multilateral order.Footnote 32 Acts of transnational repression are typically embedded in practices that include ‘media and disinformation campaigns, the co-optation and corruption of host country officials and elites, building alliances with antiliberal parties and movements, and sponsoring cyberattacks’.Footnote 33
While the literature shows how transnational repression reproduces the coercive power of the state outside its territorial boundaries, it does not raise the question of how these practices interfere with the security and sovereignty of the host state in which the targeted migrants reside. Mainstream state-centric approaches in international relations and security studies have difficulties to capture state repression against individuals in other countries because these practices do not match established categories of interstate competition.Footnote 34 Shifting from a national to an ‘entangled global’ perspective on security, however, opens up ‘new possibilities for understanding a range of security relationships and issues, bringing new actors and spaces into our understanding of global security’.Footnote 35 Considering security as a sociospatial practice permits disaggregating how host states are affected by and responding to acts of transnational repression. Digital threats, in particular, unfold in constellations of actors, networks, and infrastructures that span across divides between democratic and authoritarian governments, private and non-state actors.Footnote 36 Examining acts of DTR as sovereignty violations is thus closely linked to ongoing debates about the conceptual tension between the Internet as a globally distributed network and the sovereignty of states, commonly conceived of as territorially bounded.
Territory and sovereignty in cyberspace
While respect for state sovereignty in cyberspace constitutes a broad consensus across geopolitical fronts, it remains a contested issue, both conceptually and practically. The contrast between the networked character of the Internet and the territorially bounded nature of the nation-state even led to assertions that cyberspace establishes a form of transnational technological sovereignty distinct from the nation-state.Footnote 37 This idea has been labelled cyberspace exceptionalism.Footnote 38 In popular culture, it found expression in John Perry Barlow's famous ‘Declaration of the Independence of Cyberspace’.Footnote 39 Others argued that cyberspace poses a ‘threat to sovereignty’Footnote 40 or could bring about ‘cyber anarchy’.Footnote 41 In practical terms, the relation between sovereignty and cyberspace raises a number of concrete jurisdictional conflicts concerning the regulation of domains as diverse as law enforcement, taxation, hate speech, and product piracy.Footnote 42
Diagnosing a ‘jurisdictional paradox’, Milton Mueller argues that if states enforced their sovereignty in the context of digital technologies, they would need to act extraterritorially, thus destroying ‘the whole model of national sovereignty.’Footnote 43 While this view is obviously exaggerated and grounded in a traditional notion of territorial sovereignty, it is correct that states can hardly protect the rights of their citizens online without resorting to extraterritorial jurisdiction. Relying on a so-called destination approach, the European Union's General Data Protection Regulation (GDPR), for instance, regulates websites accessible from within the Union's borders rather than only those hosted on servers located in the EU. In contrast, the ‘data-controller approach’ of the CLOUD act in the United States allows law enforcement to access data stored by US-based cloud services, regardless of where the data is physically located. Another example is taxation that touches upon the economic essence of sovereignty: the government of France currently leads an initiative to tax companies offering services in the country's digital market but not necessarily based on French territory.Footnote 44
What these emerging approaches to the regulation of digital data and services have in common is that they move beyond the ‘monopolistic spatiality of territorial sovereignty’.Footnote 45 Adapting state authority and jurisdiction to a new technological environment, they create ‘regulatory territories of extraterritorial reach’.Footnote 46 The flexible exercise of sovereignty is not new and unique to data governance. States frequently modify their spatial reach, bending and stretching their authority in relation to geographical space.Footnote 47 The geographical understanding of territory itself is only a simplifying reification of a far more complex reality of social practices that actively construct territory by communicating boundaries, asserting power, and claiming authority.Footnote 48 In the context of digital technologies, sovereignty is conceptualised as ‘elastic’Footnote 49 and ‘post-territorial’Footnote 50 because the limits of state jurisdiction are not based on the mere factuality of geographical territoriality, but on specific normative principles, such as the protection of the fundamental rights of ‘data subjects’ in the case of the GDPR.
Sovereignty violations and countermeasures in the digital age
The demarcation of state sovereignty in the digital realm is closely linked to questions of how to define sovereignty violations and legitimate countermeasures in cyberspace. There is broad consensus on the legitimacy of direct reprisals to cyberattacks destroying physical infrastructure on foreign soil, such as the Stuxnet-attacks on Iran's nuclear facilities. More relevant for the discussion of DTR, however, are sovereignty violations below the threshold of physical impact, such as Russia's interference in the campaign for the 2016 US presidential elections. Although the NATO's Tallinn Manual on the application of international law to cyber conflict affirms that ‘a State may not intervene, including by cyber means, in the internal or external affairs of another State’, it does not elaborate in more detail on sovereignty violations without physically violent effects.Footnote 51 While some consider the Russian meddling as part of an espionage operation without a coercive element, others argue that such interference still constitutes a violation of self-determination because ‘the election process is the ultimate expression of a people's sovereign will.’Footnote 52 Nicholas Tsagourias goes even further and points to the potential coerciveness of election interference in that it allows for external control over a state.Footnote 53
Assessing digitally enabled non-violent sovereignty violations is not only complicated by the wide scope of international law for interpretation, but also the problem of attribution. The anonymity that the Internet offers facilitates false-flag operations. States also deliberately send ambiguous signals, when being accused of a cyberattack in order to shape perceptions of their capabilities and resolve.Footnote 54 Moreover, governments are often unwilling to publicly reveal evidence leading to attribution when it is obtained in intelligence operations. In general, the public attribution of cyber intrusions is a complex process with considerable risks, which requires ‘a clear understanding of the attributed cyber operation and the cyber threat actor, but also the broader geopolitical environment, allied positions and activities, and the legal context’.Footnote 55
Current scholarship on cyber security broadly agrees that the attribution problem as well as the proliferation of state and non-state threat actors, low thresholds for executing potentially damaging attacks and other factors thwart the effectiveness of conventional deterrence mechanisms that seek to prevent attacks by threatening retaliation with comparable means and severity.Footnote 56 Most cyberattacks resemble less a military confrontation than intelligence and subversion operations.Footnote 57 Much like traditional campaigns of espionage and sabotage, cyber operations exploit vulnerabilities to ‘secretly infiltrate a system of rules and practices in order to control, manipulate, and use the system to produce detrimental effects against an adversary’.Footnote 58 Consequently, some authors recommend strategies of ‘deterrence by denial’ aiming ‘to increase an adversary's difficulty in conducting its own offensive cyber operations’.Footnote 59 Means range from improved resilience and defence mechanisms to the exposure of adversary tactics and the active disruption of their ability to carry out attacks.
Complementary approaches to cyber deterrence emphasise the necessary contribution of the private sector and civil society.Footnote 60 Taking orientation from concepts of civilian-based deterrence developed during the Cold War, a system of ‘distributed cyber deterrence’Footnote 61 could involve even individuals with particular resources and impact, such as hackers and whistleblowers, to convince ‘potential attackers not to commit an aggressive or hostile act because certain consequences would follow which they would prefer to avoid’.Footnote 62 Comparable ideas are also reflected in policy approaches such as the ‘whole-of-society-approach’ promoted in the EU, which emphasise the involvement of the technology sector and civil society to counter cyber operations and other hybrid threats.Footnote 63 Although these approaches stress the role of civil society, they primarily focus on attacks on public and private infrastructure and nation-state conflicts and thereby largely ignore state-sponsored transnational attacks targeting individual actors in civil society. Our investigation of how digital repression against exiled dissidents and diasporas undermines the security and sovereignty of their host states addresses this gap.
Silencing across borders: Digital transnational repression
Digital threats have become a core component in the repertoire of authoritarian governments engaging in transnational repression. Of the 31 states a Freedom House investigation documented having committed acts like assaults, renditions, and assassinations against exiles between 2014 and 2020, 21 also used some form of digital threat and at least 17 relied on spyware to target individuals outside their territory. Perpetrators included global players such as Russia and China, regional powers like Saudi Arabia, and smaller states, such as Kazakhstan and Vietnam.Footnote 64 For these regimes, digital threats against exiled opponents reduce the costs of extraterritorial political control. They no longer need to send agents abroad to spy on and intimidate critics in the diaspora. With minimal costs and risk of consequences, a successful hacking attack against a single activist in the diaspora can expose a trove of confidential communications and unravel entire networks, including home country associates in direct reach of regime authorities.Footnote 65 The combination with other methods of transnational repression increases the effects of digital threats so that they often succeed in fostering uncertainty, fear, and mistrust within diaspora communities, reinforcing dynamics of self-censorship and silencing.Footnote 66
DTR also impacts the position, relations, and activities of individuals with ties to authoritarian contexts in their society of residence. Members of the Uyghur diaspora across the world, for instance, are reluctant to speak openly about China's extensive repression campaign against their people for fear of severe repercussions for their families in Xinjiang. Uyghur exiles reported that their calls and social media communications were being monitored; they even received threatening phone calls from police agents using the accounts and devices of family members. Chinese authorities have used surveillance and threats against relatives to silence publicly outspoken Uyghur activists in countries like Germany, France, and the United States.Footnote 67 These cases highlight that, in addition to infringing on the privacy and autonomy of targeted exiles, practices of transnational repression curtail their ability to partake in the social and political life of host countries, sabotaging public debate and accountability processes.Footnote 68
The authoritarian reach across borders is not limited to exile and diaspora communities. The tools and techniques that regimes deploy against their populations abroad can also spill over into host countries. In fact, cyber threat actors attacking civil society and diaspora groups are often identical or overlap with those going after foreign targets. Vietnamese human rights defenders and journalists based in Germany, for instance, were persistently attacked with malware attributed to the ‘Ocean Lotus’ group which is suspected of working in alignment with the Vietnamese government since at least 2012 and has also targeted foreign officials and companies.Footnote 69 Iranian hacking groups are known to hone their methods on civil society before reaching for more complex targets abroad.Footnote 70 The revelations around the widespread abuse of the NSO Group's Pegasus surveillance tool provide the most striking illustration of how closely digital threats against individual dissidents are linked to the security of their broader host society. A number of authoritarian governments used the powerful spyware not only against their own nationals, at home and abroad, but also a range of foreign targets, including journalists, lawyers, and high-ranking politicians.Footnote 71 The Saudi regime infiltrated the smartphone of Jeff Bezos, CEO of Amazon and owner of the Washington Post, the newspaper in which Khashoggi published his articles before being murdered.Footnote 72 Moroccan authorities are suspected to be behind digital surveillance operations against leading French journalists and government officials.Footnote 73
Digital transnational repression as a sovereignty violation
Building on our outline of the conceptual scope of state sovereignty in cyberspace and the empirical reality of DTR, we present three arguments grounded in international law on why these practices can be considered as violating the sovereignty of the countries in which the targeted individuals reside. Next to these normative claims, we outline one political argument that concerns the immediate self-interest of host states affected by DTR.
First argument: DTR is extraterritorial enforcement jurisdiction
Per definition, DTR constitutes governmental action on another state's territory. While states generally benefit from significant latitude regarding intelligence operations abroad, extraterritorial enforcement jurisdiction is tightly restricted under international law: ‘The exercise of enforcement jurisdiction is an exercise of State sovereignty, and the rule that governs it is simple. No State may exercise its enforcement jurisdiction in the territory of another State without that State's permission.’Footnote 74
Although a significant overlap between DTR and intelligence activities cannot be denied, DTR appears to be closer to enforcement as it is not motivated by issues of interstate competition, but rather by the perpetrating state's domestic security interests. The prohibition of extraterritorial enforcement jurisdiction is non-controversial regarding non-digital modes of enforcement as it ‘violates the fundamental principle of territorial sovereignty that restricts the reach of the police forces of one sovereign into the territorial jurisdiction of another’.Footnote 75 For instance, even in the case of the morally justifiable abduction of former Nazi official Adolf Eichmann by Israeli security officers from Argentina, it was concluded that the operation violated Argentine sovereignty.Footnote 76
While DTR has not been discussed as extraterritorial enforcement yet, the 2014 case Kidane v. Federal Democratic Republic of Ethiopia, dealing with an act of DTR in the United States, was dismissed on the grounds of sovereign immunity. In the trial, an Ethiopian dissident with US citizenship living in Maryland sued the government of Ethiopia for infecting his computer with spyware produced by the German-British FinFisher company. The plaintiff made the case that the Foreign Sovereign Immunities Act (FSIA) did not apply because the Ethiopian government had committed a tort on US territory, which is one of the Act's exceptions to immunity. However, the Court argued that the Ethiopian government had planned, prepared, and executed the hacking of the computer from abroad: ‘all of the acts by Ethiopia or its agents that allegedly precipitated the tort occurred outside the United States’.Footnote 77 Therefore, the exception to immunity was considered not applicable, highlighting the tension between a successful persecution of DTR and territorial jurisdiction.Footnote 78 A similar argument might be brought forward to refute our claim that DTR constitutes extraterritorial enforcement jurisdiction. However, digital modes of enforcement can be compared to physical enforcement and should be treated equally. In criminal law it is acknowledged that obtaining e-evidence from servers abroad constitutes a physical intrusion in a different jurisdiction:
Contrary to the current metaphor often used by Internet-based service providers, digital information is not actually stored in clouds; it resides on a computer or some other form of electronic media that has a physical location.Footnote 79
Such intrusions usually require a Mutual Legal Assistance Treaty (MLAT) or similar bilateral agreements. In the cases of DTR outlined above, such as Saudi Arabia's hacking into the phones of exiled dissidents, the enforcement aspect is much stronger than in the transborder obtaining of e-evidence as the obtained information is used to further threaten, pressure and punish targets. Therefore, DTR amounts to extraterritorial enforcement jurisdiction prohibited by international law, if committed without the knowledge or agreement of the host state. This is also the case if the information accessed or manipulated in acts of DTR is actually stored in a distributed manner in the cloud so that a link to a specific jurisdiction cannot be established. Similar to the destination approach in the GDPR, the place where the rights violation through DTR occurs should be decisive, rather than the place where the breached or manipulated information is stored.Footnote 80
Second argument: DTR interferes with national self-determination
DTR can be regarded as an illegal interference with a state's domaine réservé, ‘the areas of State activity that are internal or domestic affairs of a State and are therefore within its domestic jurisdiction or competence’.Footnote 81 Self-determination is usually understood as regarding the constitution, that is, the founding of a state. However, as Jens David Ohlin argues, this emphasis is owed to an outdated state-centred understanding that ignores the continuous constitutive role of deliberative processes in society.Footnote 82 Especially in democracies, self-determination through deliberation is a permanent and open-ended process. Even in authoritarian societies, such deliberative practices can play an important role in stabilising and legitimising sovereignty.Footnote 83
Political emigrants often participate actively in host state processes of public deliberation, seeking to influence foreign policymaking with regard to their home country.Footnote 84 As outlined above, DTR can have ‘chilling effects’ on diaspora activists, muting or distorting their voices. By intimidating and silencing exiled dissidents who often possess detailed knowledge on political developments in their home countries, therefore, practices of DTR indirectly interfere in the self-determination of the host state. This close link between transnational repression and the national self-determination of the host state was already emphasised by Jack I. Garvey when referring to the activities of the Iranian secret police against Iranians living in the US before the revolution of 1979. He argues that the repression of exiled dissidents was so effective that it obscured the extent of dissatisfaction with the Shah's rule and the likelihood of imminent revolution. Hence, the Iranian state's practices of extraterritorial coercion affected the foreign policies and geopolitical interests of the US government.Footnote 85 In the same way, it can be argued that the surveillance and repression against the Uyghur diaspora impedes European governments to apprehend the full scale and scope of China's extensive efforts to control the Uyghur population in its Xinjiang region. In this case, transnational repression against Uyghurs is clearly part of the attempt to shape political decision-making in Europe and elsewhere.
The prohibition of cyber intervention, as outlined in the Tallinn Manual, is not only based on the interference with another state's domaine réservé, but also on the coerciveness of this interference: ‘The term intervention … is limited to acts of interference with a sovereign prerogative of another State that have coercive effect.’Footnote 86 Coerciveness is defined as being ‘designed to influence outcomes in, or conduct with respect to, a matter reserved to a target State’.Footnote 87 Further, ‘the coercive act must have the potential for compelling the target State to engage in an action that it would otherwise not take (or refrain from taking an action it would otherwise take).’Footnote 88 Given these qualifications, DTR against individual exiles does not, per se, amount to coerciveness. However, it could be of coercive nature if undertaken with the intention to alter the host state's political agency, rather than the political agency of the individual dissident. This might require a certain number or prominence of exiles targeted by DTR, in relation to the host state's public sphere. In response to this caveat, however, it can be argued that transnational repression often targets individuals precisely for their outreach and public presence in host societies – consider the example of Jamal Khashoggi, a regular contributor to the Washington Post where he published his criticism of the Saudi government. The Chinese campaign against Uyghurs abroad certainly fulfils the criterion of extensive scope. Moreover, surveillance and digital threats, especially when coupled with other more extreme methods of transnational repression, have ripple effects promoting uncertainty and fear throughout entire diaspora communities.
Third Argument: DTR's violation of fundamental and human rights interferes with a state's sovereign to uphold these norms
As argued by human rights approaches, DTR interferes with core fundamental and human rights, such as the right to privacy and free speech.Footnote 89 The popular sovereignty and legitimacy of a state are closely connected with its capacity to guarantee these very rights on its territory so as to safeguard an open public debate and flourishing civil society. In addition, DTR interferes with provisions in the EU Charter of Fundamental Rights and the Universal Declaration of Human Rights that stipulate the right to seek and enjoy political asylum.Footnote 90 Many national legislations include similar articles. Particularly in Western democracies the right to seek asylum is an integrative part of political culture and identity.Footnote 91 The commitment to upholding such fundamental norms is clearly the expression of a state's sovereign decision making. Given the high importance of the right to asylum and other fundamental and human rights, any form of transnational repression against political emigrants, including its digital forms, must be regarded as an interference with the political will and the sources of legitimacy of a state and, hence, a violation of its sovereignty.
Appealing to the self-interest of host states: Political reasons to consider DTR a challenge to sovereignty
While the preceding arguments proceed normatively on the basis of international law, they do not necessarily speak to a state's self-interest. Particularly if facing a powerful authoritarian state committing DTR, less influential host states might prefer to turn a blind eye to norm violations because they do not consider threats to individual migrants as existential for their own pragmatic interests. Therefore, complementing our three principal arguments, we outline an additional rationale focusing primarily on the self-interest of a state. We argue that any challenge by another state to the host state's ability to guarantee the safety and rights of political exiles on its territory also signifies a challenge to its ability to maintain domestic sovereignty which refers ‘to the organisation of public authority within a state and to the level of effective control exercised by those holding authority.’Footnote 92 We illustrate this point with two empirical examples:
First, surveillance in combination with threats against the home country families of exiles interfered in the first criminal trial on atrocity crimes of Syrian regime officials in Koblenz, Germany. Witnesses withdrew or altered their testimonies after their participation in the hearings became known and family members in Syria were threatened by supporters and agents of the Assad regime. The judge in court expressed resignation as to their limited capability to provide protection for witnesses testifying against a repressive regime still in power.Footnote 93 Potential witnesses in investigations against former Syrian regime officials in Sweden and the Netherlands were intimidated through similar tactics.Footnote 94 Here, practices of DTR obstructed the host state's judicial processes, which is a central element in the exercise of domestic sovereignty.
In a second example, the Turkish government encouraged the denunciation of government opponents among the Turkish diaspora in Germany, providing even a digital tool to streamline surveillance and reporting. The smartphone application provided by the Turkish Interior Ministry allowed loyal diaspora members to upload photos and contact details of alleged dissidents for reporting them to law enforcement agencies. Those reported risked arrest and interrogation when travelling to Turkey. A briefing of the German parliament saw potential violations of German law by this application, including laws on data protection, against foreign espionage, and political defamation.Footnote 95 With its strategies of diaspora engagement and instrumentalisation, the government of President Erdogan was thus actively inciting its supporters to act against legislation of the host country.
In such cases, practices of DTR go beyond single acts of extraterritorial enforcement jurisdiction, as discussed under our first argument above. Their effects reach further than constituting an obstacle to the capacity of migrants to participate in the host state's public sphere, as outlined in argument two. Their implications for the security dimension of sovereignty are also much clearer than in human rights violations affecting the targeted individuals and a host state's legitimacy, which we laid out in our third argument. Rather, these practices undermine the credibility of the host state as the holder of effective control over the monopoly on enforcement jurisdiction and the guarantee of fundamental rights. Thereby, they challenge the host state's authority. In this regard they resemble other types of foreign interference and hybrid threats, such as election manipulation and disinformation campaigns that seek to undermine trust in public institutions and sow societal division.Footnote 96
The effects of weakened authority are arguably too complex to be measured in empirical terms so that it seems difficult to set a threshold for when they would constitute a sovereignty violation in and of themselves. Yet, from a perspective of the mere self-interest of a state, the potential risks for the social cohesion and security of the host country are obvious. A lack of belief in functioning state institutions and the rule of law might hamper the adherence of affected migrant communities to the social and political norms of the host country and their successful inclusion as equal members of society. This, in turn, could have negative effects on migrants’ social and economic participation with implications for the security within diaspora communities and their broader host society. In their research on Eritrean refugees in Canada, Berhane and Tyyskä highlight that intimidation, threats, and surveillance not only ‘put the peace and mental stability of some refugees at risk’, but also contravene the efforts of the Canadian government to ‘create a secure system to protect refugees and ensure that their integration process moves smoothly.’ Similar to the example of Turkey mentioned above, some Eritrean refugees are pushed to break Canadian law as embassy staff force them to donate money for military activities in their country of origin. Such coerced allegiance to the state in the origin country that is contrary to the law of the host state will likely lower the host societies’ capacity to integrate the targeted migrants. An action plan of the European Commission, for example, underlines the importance of the rule of law and security for integration processes.Footnote 97
Host state responses to DTR
Regimes engaging in DTR benefit from a considerable extent of impunity. The political reactions of host states to even the most flagrant acts of physical transnational repression are often subordinated to economic and geostrategic priorities, as evidenced in the international responses to the murder of Jamal Khashoggi.Footnote 98 Digital threats against diasporas are executed at still lower costs and consequences. As we have shown, practices of DTR not only threaten the personal security of diasporas with ties to authoritarian contexts, but can also violate the sovereignty of their host states. To deter perpetrators from engaging in digital threats against diasporas, some form of legal, political, or operational consequences is crucial for ‘making them believe that the costs to them will exceed their expected benefits.’Footnote 99 In response to DTR, host states could take a number of steps to change the opportunity structures and cost-benefit calculations of perpetrating regime actors. The suggestions outlined below are not meant as an exhaustive programme for defending state sovereignty against digitally enabled interventions of authoritarian rulers. Rather we indicate four possible pathways of practices and policies that appear instrumental to curtail DTR.
First Pathway: Strengthen the digital resilience of civil society and build distributed cyber deterrence
Strengthening a consistently under-resourced civil society against the powerful state actors relying on DTR corresponds to the logic of distributed deterrence and deterrence by denial outlined above. Enhanced capacities for privacy protection and data security within civil society will not prevent attacks altogether, but make them more difficult. As potential targets of DTR become more resilient, capable of defending against and recovering from attacks, digital threats will be less appealing in the first place. As well as effective interventions educating on digital hygiene and information security, inclusive networks for emergency support and security advice will help diaspora activists to maintain agency under conditions of evolving sociotechnical risks.Footnote 100
In addition to these defensive strategies, civil society can also be part of offensive measures aiming to curtail the capabilities of perpetrators. Exposing the tools, methods and enablers of digital repression are important steps to patch vulnerabilities and disable attacks. In September 2021, for instance, researchers at the Citizen Lab discovered that the NSO Group's Pegasus spyware used a vulnerability in Apple products to infect devices without targets even opening a compromised link. Identified on the device of a Saudi activist, this rare ‘zero-click’ method exposed more than 1.65 billion Apple products to spyware infections. In response, Apple not only issued an emergency software update but also pledged financial support to organisations pursuing cybersecurity research.Footnote 101
There is still limited awareness on and documentation of cyber operations against civil society, including cases of DTR. To a large extent, knowledge on cyberthreats in policy circles and cybersecurity scholarship is shaped by commercial threat reporting, which ‘primarily focuses on cybercrime, economic espionage and sabotage of critical infrastructure’.Footnote 102 Threats against civil society organisations (and even less so individual political exiles) are largely underreported. The increasing scope and detrimental effects of DTR are known only as a result of investigations by a handful of independent researchers, non-governmental organisations, and journalists. More participatory and cross-sectoral mechanisms are needed for documenting and investigating cyber operations against civil society. Private sector resources could provide more accurate intelligence on the tools and techniques of threat actors, facilitating attribution and risk mitigation. In turn, the involvement of civil society in oversight and attribution processes increases their legitimacy, eventually supporting government decisions on policy responses. Public attribution combined with a ‘naming and shaming’ of perpetrators could work as an effective deterrent and even help reversing some of DTR's ‘chilling effects’. As Ronald J. Deibert points out, independent Computer Emergency Response Teams (CERTs) that combine the expertise and resources of different actors could form an important backbone for such approaches to a more distributed and human-centred security for the digital environment.Footnote 103
Second Pathway: Use counterintelligence and law enforcement activities to protect against DTR
Practices of DTR represent a distinct threat within a broader array of foreign interference tactics that aim to influence the political process and undermine the rule of law in target countries. Given the similarity and overlaps with espionage operations, host country law enforcement and counterintelligence agencies play a key role in detecting, constraining, and disrupting the activities of perpetrators and, wherever possible, bring them to accountability. In the United States, the Department of Justice has repeatedly pressed charges against Chinese nationals for acting as foreign agents in operations of transnational repression with methods that included digital surveillance and disrupting online meetings.Footnote 104 Two former Twitter employees were charged for spying on behalf of Saudi Arabia by using their access to collect private, identifying information on Twitter users critical of the Saudi government.Footnote 105 In Sweden, foreign intelligence activities targeting dissidents in exile, so-called ‘refugee espionage’, represent a distinct crime. The German domestic intelligence agency in its annual public reports documents activities which count as acts of transnational repression from countries like Turkey, China, and Iran.
While these established counterintelligence and law enforcement mechanisms may capture attempts of digital repression, they are not specifically aimed at DTR. Concurrently, dedicated cyber operations could be used to disable infrastructure used for digital threats against diasporas, curtailing perpetrators’ capabilities and increasing their operational costs. To prevent interference in the 2019 midterm elections, for instance, the US Cyber Command blocked Internet access to a ‘troll factory’ in St Petersburg that played a key role in the Russian influence operation against the presidential elections in 2016. Similar offensive measures could target individuals and entities known to engage in DTR. These efforts should also include working with the technology sector. In 2021, for instance, Facebook was able to disrupt Chinese espionage operations against Uyghur activists overseas.Footnote 106 Moreover, in order for such measures to serve the purpose of deterrence, they need to be communicated clearly as actions countering practices of DTR.Footnote 107
Finally, law enforcement strategies against DTR will have to include measures of outreach to and support of targeted communities. Research for the European Parliament on countering hybrid threats and foreign interference identifies diasporas at risk of ‘being used as proxies’ and recommends targeted programmes for awareness-raising, media literacy, and strengthening cybersecurity skills.Footnote 108 Authorities in Germany warn refugees, especially from Syria and Iran, of foreign government pressure and espionage within immigrant communities.Footnote 109 For such efforts to be credible and have the desired effects, however, it is important to establish trusted relationships and refrain from framing migrants as a national security risk instead of acknowledging them as persons at risk of cross-border repression.
Third Pathway: Sanctions against perpetrators and enablers of DTR
Sanctions are an important tool to constrain and punish the perpetrators and enablers of DTR, while highlighting their inappropriate behaviour to a wider international audience. Sanctions could target individuals, organisations, and companies involved in or facilitating significant acts of DTR. For instance, the ‘Khashoggi Ban’, introduced in 2021 in the United States, allows imposing visa restrictions on ‘individuals who, acting on behalf of a foreign government, are believed to have been directly engaged in serious, extraterritorial counter-dissident activities’.Footnote 110 The measure implicitly brings up the principle of territorial sovereignty as perpetrators ‘should not be permitted to reach American soil’.Footnote 111 The European Union has yet to introduce a similar instrument. Its global human rights sanctions regime targets individuals and entities involved in serious human rights violations and abuses worldwide, including if these are ‘of serious concern as regards the objectives of the EU common foreign and security policy’, but has not been applied to acts of transnational repression yet.Footnote 112
Other than human rights related restrictions, sanctions in response to malicious cyber activities could be adapted to capture severe cases of DTR. Under its newly established cyber sanction regime, the EU has imposed two rounds of restrictive measures against individuals and legal entities in China, Russia, and North Korea for their engagement in cyberattacks against the Union and its member states.Footnote 113 A more comprehensive and consistent application of this instrument, however, seems to have been hampered by a lack of intelligence sharing and coordination among member states.Footnote 114 To be used in cases of DTR, the thresholds for triggering these restrictions would have to be lowered as the sanction regime currently captures only severe national security threats, such as attacks on critical infrastructure and election interference. Yet, the actors targeting diasporas and those conducting malicious cyber activities against foreign targets often overlap. A more stringent application of these sanctions could also impact the capabilities of DTR perpetrators.
Judicial scrutiny, political pressure, and sanctions will also affect the companies providing the tools and infrastructure for DTR. Authoritarian regimes increasingly rely on commercial spyware furnished by a burgeoning and largely unregulated private industry. In these partnerships, authoritarian states and the private sector are drawn into a dangerous spiral of mutual incentivisation as companies competing for lucrative contracts answer the demands of intrusive regime agents, raise the market's standards by constantly improving the effectiveness of their products, and shun public scrutiny to protect their client relations. As already outlined, in addition to targeting their nationals abroad, authoritarian regimes have not shied away from using the newly purchased capabilities against foreign targets too. The commodification of surveillance technologies thus ultimately threatens the security interests and rule of law in the countries they originated from. Former operatives of the US National Security Agency (NSA), for instance, who were contracted by the United Arab Emirates intelligence agency through a private consultancy, admitted to violating US hacking laws and prohibitions on selling sensitive military technology, after an investigation of the federal police.Footnote 115 To curtail the capabilities of authoritarian regimes for engaging in DTR, it is important to bring more oversight into the business practices of the surveillance industry and, whenever possible, sanction their servicing of rights-abusing regimes. In June 2021, for instance, judges in France indicted executives of the companies Amesys and Nexa Technologies, which supplied surveillance equipment to the Egyptian and Libyan regimes, for complicity in torture and enforced disappearances.Footnote 116 In November 2021, the US government blacklisted the NSO Group, effectively barring the company from receiving American technologies, ‘for engaging in activities that are contrary to the national security or foreign policy interests of the United States’. Such steps send important signals to companies in the surveillance sector and their investors.
Fourth Pathway: International agreements and norm development
Our arguments highlight several normative blind spots that facilitate a proliferation of DTR and hamper its condemnation and curtailment. A first issue that needs tackling is the unclarity of frameworks applying international law to cyberconflict, such as the Tallinn Manual 2.0. The Manual took a step in the right direction by classifying cyberattacks with violent effects as a use of force. However, as discussed above, it is far from clear on sovereignty interference below the use of force threshold. This problem mostly stems from the imprecise criterion of ‘coerciveness’. In addition, the fact that the attribution of cyberattacks is complex whereas legitimate reprisals require a certain degree of swiftness still constitutes a conundrum that needs to be solved, for instance by defining rigorous standards for attribution and the maximum time lapse of a legitimate reprisal. To address cyberthreats and enforce binding rules at a global level, it has been suggested to create an independent institution for the examination and attribution of cyberattacks, modelled after the International Atomic Energy Agency.Footnote 117
A second issue are regulations like the Foreign Sovereign Immunities Act and the very similar European Convention on State Immunity that connect state immunity to territoriality. Accordingly, legitimate exemptions from state immunity depend on whether a rights violation by a state occurred on the territory of the state where the trial is held. As discussed with the example of Kidane v. Federal Democratic Republic of Ethiopia, such an understanding of territory solely focusing on the physical location of actors no longer fits the reality of transnational digital technologies. Rather, the physical location of the acts that occur should be decisive, for example the intrusion in digital networks or access to stored information on the host state's territory.
Third, international agreements and controls are needed to contain the proliferation of offensive cyber capabilities and bring more transparency into the largely unregulated private market for surveillance technologies.Footnote 118 In 2019, the UN Special Rapporteur on freedom of opinion and expression concluded his review on the surveillance industry with a call for an immediate moratorium on the sale, transfer, and use of surveillance technology until human rights-compliant regulatory frameworks were in place.Footnote 119 Governments need to tightly regulate the export of equipment and knowledge for surveillance to other countries where it is likely to be used for transnational repression. The revised EU export controls, for instance, aim to bring more transparency into the spyware trade. However, civil society organisations criticise that restrictions foreseen in previous drafts were watered down following pressure from private industry and member states with stakes in the exports.Footnote 120
A final fundamental issue that undergirds the reach of state repression across borders is the current weakness of sovereignty. In fact, Western democracies, led by the US government, contributed significantly to this normative decline with measures under the ‘global war on terror’ after 11 September 2001. The renditions, targeted killings, and designations of ‘ungoverned spaces’ for anti-terror operations on foreign territory effectively ‘conditioned’ sovereignty and paved the way for a more assertive behaviour of states in extraterritorial space.Footnote 121 Law enforcement and security agencies extended their reach across national borders in complex set-ups of international surveillance and security collaboration. Particularly, Russia and China were close partners of the US in the ‘war on terror’.Footnote 122 Initiated in the Global North, these developments ‘have given rise to understandings and technologies of security with global impacts’.Footnote 123 Authoritarian regimes now make ample use of the label of terrorism to persecute their opponents abroad. They also benefit from a burgeoning private industry for purchasing advanced surveillance capabilities.Footnote 124 The ‘war on terror’ and its consequences exemplify that often ‘[p]ower, more than legal concepts, seems to determine the weight given to sovereignty.’Footnote 125 Although sovereignty is recognised as a fundamental cornerstone of the international order by governments across the political spectrum, there is no higher authority to arbitrate and sanction any interference. In order to strengthen sovereignty's protective functions under the realities of globalisation and digitalisation, states thus need to restrain themselves, for instance, by imposing strong mechanisms of oversight and public transparency onto their security agencies and surveillance practices.Footnote 126
Conclusion
Digital technologies have given authoritarian governments new tools for political control and coercion beyond borders. With surveillance, malware attacks, online harassment, and other forms of digital transnational repression, these regimes aim to prevent, constrain, and punish dissent among exiles and diaspora populations. DTR is an essentially globalised form of repression: it is committed by state agents and their affiliates against individuals on other territories; perpetrators rely on platforms and infrastructures that do not necessarily overlap with the geographical territory of neither home nor host state; and they use the services of private companies and non-state actors enabling digital threats and augmenting their effectiveness.
Examining these extraterritorial authoritarian practices as sovereignty violations against the state hosting the targeted diasporas, we presented three arguments grounded in international law to show potential interference. First, as governmental action on another state's territory, acts of DTR represent extraterritorial enforcement jurisdiction which is illegal under international law. Second, DTR interferes in sovereign self-determination by silencing diaspora voices in the public debate of host societies, which risks, among others, distorting deliberations on foreign policy. Third, DTR violates host state sovereignty because it affects the government's capacity to protect fundamental and human rights on its territory, which goes against the sovereign choice of a state to adhere to such norms. In addition to these normative arguments, we appeal to the self-interest of host states by arguing that DTR can be understood as a violation of domestic sovereignty because it counteracts the authority of host state institutions in the maintenance of the rule of law and its integrative functions. This can have a range of ripple effects in terms of security and the societal participation of individuals with ties to authoritarian countries.
In practice, many everyday incidents of DTR will likely not cross the line of formal sovereignty violations. Yet the recurrent silencing of critical exiles, the spreading of fear and uncertainty within diaspora communities are stretching limits. Gradually, these practices may amount to the erosion of host state authority and sovereignty we describe. Our arguments thus highlight DTR's subversive effects on the broader host societies – beyond the immediate threat to the personal security of exiles.
To counter DTR, we outlined a strategy of distributed deterrence that will involve public, private, and civil society actors to deny perpetrators the opportunity for attacks and signal that costs outweigh the benefits. This includes, among other measures, investigative research to expose threat actors and their methods, interventions for building the digital resilience of civil society, and counterintelligence activities extending protection to communities at risk. In addition, targeted sanctions and other punitive instruments will constrain the perpetrators and enablers of DTR, stressing their rules-violating actions. Finally, we highlighted several issues of necessary norm development: the rules for cyberconflict that are too state-centred to be applicable to low-intensity attacks on civil society; a global regime for the use and sale of offensive cyber capabilities; and the adaptation of current notions on sovereign immunity to a transnational digital environment.
For host governments to actually act upon this threat and protect targeted migrants under the premise of sovereignty enforcement, a shift in perception and practice is required. Migrants are frequently perceived as a threat to host states’ security and therefore subjected to processes of securitisation and exclusion. Instead, host states need to come to an understanding of migrants as referent objects of security and recognise the authoritarian practices that follow them as the actual threat.Footnote 127 Recent policy documents reveal an increasing awareness of the issue among governments who tie the curtailment of transnational repression to strategies of democratic resilience.Footnote 128 In fact, political exiles from authoritarian contexts are often persecuted for upholding their liberal rights and could be considered as strategic allies in the fight against disinformation and authoritarian interference.Footnote 129
In her critical examination of current Western migration regimes and their ‘shifting’ borders, Ayelet Shachar highlights that the reconfiguration of the boundary-drawing functions of state sovereignty in response to global migration flows needs to be ‘matched by a corresponding expansion of responsibilities’.Footnote 130 What she describes as a ‘fixation on territorial access as a precondition for securing refuge and protection’Footnote 131 frequently results in the exclusion of political refugees who are unable to reach the soil of countries promising asylum. But this fixation also means that those who made it to the territory of liberal host states are falsely assumed to be safe, and often left with a minimum of rights and protection. The rights-extending dimensions of sovereignty, therefore, need to be detached from geographical territory in order to shield individuals from the digitally enhanced transnational reach of authoritarians. In this sense, the ‘sovereignty effect’ of governance practices that ‘seek to reaffirm the foundational elements of belonging to one group as opposed to another’Footnote 132 could be turned into a means for guaranteeing exiles’ safety and fundamental rights, free from internal and external coercion.
Our article highlights some of the manifold ways in which the border-blurring nature of cyberspace challenges established notions of territory, jurisdiction, and sovereignty. Malicious actors exploit gaps in a still emerging normative order for cyberspace, undermining rule of law and security across territorial borders and in transnational space. Cyberspace clearly is a domain for political competition and conflict. However, as research in this field primarily focuses on interstate dynamics and cyberattacks on critical state assets, it has left aside the issue of transnational threats against civil society, and how these might be linked to broader questions of security and sovereignty. This is the gap our article addresses. In doing so, we open up avenues for further research into the security implications of digital transnational repression for host states. Empirical investigations need to further disaggregate the corrosive effects of transnational repression on host societies, such as hampering migrants’ successful integration and social inclusion or interfering in public debate and political decision making. On the conceptual and normative level, further reflections are needed about how to exercise and protect state sovereignty in the digital age while safeguarding fundamental rights and human security.
Acknowledgements
The authors are deeply grateful to Paul De Hert (Vrije Universiteit Brussel) for encouraging the formation of this article and commenting on an earlier draft. We also thank the three anonymous reviewers whose constructive comments helped to significantly strengthen the manuscript. The research of Marcus Michaelsen was funded by the European Union's Horizon 2020 Research and Innovation Program under the Marie Skłodowska-Curie Grant Agreement No. 845988 (DIGIACT). During this research, Johannes Thumfart received funding from Gerda Henkel Stiftung's special programme Security, Society, and the State and the European Union Horizon 2020 research programme under MSCA COFUND grant agreement 101034352 with co-funding from the VUB-Industrial Research Fund.
Marcus Michaelsen (PhD) is an independent researcher working on digital authoritarianism and transnational repression. From 2019 until 2022, he was a Marie Skłodowska-Curie Fellow in the Research Group on Law, Science, Technology and Society (LSTS) at Vrije Universiteit Brussel. Previously, he held a Senior Information Controls Fellowship of the Open Technology Fund for a research project on digital transnational repression against exiled Middle Eastern activists. Author's email: email@marcusmichaelsen.eu
Johannes Thumfart (PhD) is a Fellow in the Research Group on Law, Science, Technology and Society (LSTS) at Vrije Universiteit Brussel where he currently works on a project about Internet shutdowns within the Horizon 2020 Marie Skłodowska-Curie COFUND Action IMPACT. From 2020 to 2022, his research on digital sovereignty at LSTS was funded by the Gerda Henkel Stiftung. He is also teaching ethics of international security management at the Berlin School of Economics and Law. Author's email: Johannes.Thumfart@vub.be
Open access
