Migration governance in civil war: The case of the Kurdish conflict

Abstract This article examines the management and instrumentalisation of migration and mobility as an area of contested governance in civil wars. Building on work in migration studies and rebel governance, it shows how migration and mobility regimes form part of the structure of violent armed conflicts, as both states and non-state actors seek to control processes and consequences of mobility and migration to their advantage. Governance of migration during conflict involves the strategic use of mechanisms of migration governance for the purposes of achieving conflict aims. This article develops a framework for understanding how migration governance is instrumentalised in civil war as a means of managing and controlling populations. The framework is then applied to the case of the Kurdish conflict in Turkey and beyond through an analysis of three areas of migration governance that have played significant roles in this extended regional conflict: forced migration and refugee governance; border management; and diaspora engagement. The analysis provides a challenge to dominant state-centric, securitisation and humanitarian approaches to migration and security by pointing to the political and spatial complexity of contested migration governance in situations of protracted conflict.


Introduction
Migration governance is usually associated with states and international organisations, but the control of mobility is also a key priority for non-state armed organisations involved in conflict. 1 Whereas much of the policy literature on migration governance has focused on formal structures of governance and state policies, in many regions aspects of migration management -such as border control, refugee governance, and diaspora politics -are also carried out informally by non-state and clandestine actors and form part of the overall dynamic of protracted violent conflicts. Armed organisations can seek to secure control over people, territory, and resources as part of their overall conflict strategy, and the governance of mobility and migration can therefore emerge as part of the conflict battleground and as an arena of contestation between state authorities and non-state challengers.
In this article I bring together insights from the fields of migration studies and civil war studies to examine the central role that migration governance plays in violent conflicts. Migration governance Drawing on examples from the Kurdish conflict in Turkey, but also extending out to include the regional dimensions of the conflict, including its dynamics in Northern Syria, Iraq, and beyond, I show how the use of forced migration; the control of diaspora populations and refugees; and the management of cross-border movement and mobility can all become areas of contestation between states and non-state rebel organisations in protracted conflicts.
The Kurdish conflict in Turkey, and its regional and global dynamics, is a complex conflict configuration that involves a number of actors in different states that are loosely united in their affinities with the ideology and leadership of AbdullahÖcalan, as well as their affiliation with the Kurdistan Communities Union (KCK).Öcalan is the jailed former leader of the Kurdistan Workers' Party (PKK), a rebel organisation that he founded in Turkey in the 1970s with the aim of achieving an independent Kurdish state. 9 The KCK is a broad transnational organisational structure that was established in 2005 byÖcalan, who remains its leader, and that includes the PKK as well as other regional actors such as the People's Defense Units (YPG), which is the primary component of the Syrian Democratic Forces (SDF). The members of the KCK are all affiliated with the 'Öcalanist' philosophy of democratic confederalism. The relationship between democratic confederalism, which formally seeks to transcend the state, and Kurdish nationalism, which references the nation-state, is nonetheless complex and often contradictory -both in thought and in practice. 10 The alreadycomplex conflict was complicated further with the outbreak of the Syrian civil war in 2011, which led to regional refugee flows and the setting up by the YPG of an autonomous Kurdish-dominated quasi-state in northeastern Syria, which since 2018 carries the name of the Autonomous Area of North and East Syria (AANES), but is still widely known as Rojava. 11 This case study is used for illustrative purposes and is designed as an exercise in exploratory research. 12 The case study uses a synthetic and descriptive methodological approach as a form of grounded theorising based on an integrative review and analysis of a range of sources, including the large secondary literature on the Kurdish nationalist movement and Kurdish diaspora politics; primary documents such as press and NGO reports; memoirs and writings of key actors within the Kurdish movement, as well as observations derived from original data. 13 The primary contribution 9 There is a vast literature on the Kurdish nationalist movement, the PKK, and the Kurdish conflict in Turkey. See  of the piece is, however, conceptual and theoretical, rather than the introduction of new data, in that it seeks to bring together insights from the literatures on migration governance and rebel governance, and apply them to the case of the Kurdish conflict in Turkey and Northern Syria as a means of stimulating further debate and research. The rest of the article is structured in the following manner: first, I examine the existing literature on migration, conflict, and security, suggesting that both the literature on civil wars and the literature on migration governance have undertheorised the strategic use of migration governance by states and non-state actors in situations of protracted violence. I discuss the centrality of migration management to modern statehood and how a rebel governance perspective can help shed light on the significance of migration-and mobility-management to rebel organisations and other non-state actors. I then use this lens to examine the case of the Kurdish conflict in Turkey, including its regional and global dimensions, by focusing on three areas of contested migration governance in the conflict: the role of forced displacement and refugee governance; the management of borders; and the governance of diaspora populations. I conclude with a discussion of the broader theoretical and policy implications of the argument and some directions for future research.

Migration governance and civil war
Migration can be both a cause and consequence of conflict, but the question of how conflict actors strategically instrumentalise the management of migration and mobility as a tool in conflict has received less attention in the scholarly literature. 14 Violent conflicts are clearly one of the most significant causes of forced displacement and refugee flows in the world, as populations attempt to escape from ongoing violence. 15 Moreover, there is an established literature that examines how migration and refugee movements can lead to the spread of violence by transnationalising conflict dynamics, changing local power balances, or creating increased competition for scarce resources in receiving states. 16 For example, an exodus of refugees to neighbouring countries following the 1994 Rwandan genocide is widely seen as contributing to the destabilisation of the entire Great Lakes region of Central Africa. 17 The return of refugees to Burundi arguably contributed to the emergence of postconflict violence. 18  Guatemala, Ecuador, Nepal, Sri Lanka, and Thailand, although the causal relationship is always mediated by other factors. 19 Yet, analysing migration merely as a cause or consequence of conflict misses the significant ways in which migration and conflict are more deeply intertwined. For example, migration flows are not always simply externalities or by-products of conflict, but can also be part of actors' conflict strategies. Similarly, the spread of conflict via migration is not always unintentional; conflict actors can have active interests in spreading or reshaping conflict via the strategic use of migration flows. Thus, migration flows are not simply external factors that states respond to and manage in conflict situations, but are often part of a 'cat and mouse game' and form of strategic interaction that emerges between state authorities and rebel groups as part and parcel of ongoing violent conflicts. In such situations, the control of mobility and migration -including the control of refugee, migrant, and diaspora populations -becomes a central aspect of conflict dynamics and strategy.
Both states and rebel groups can instrumentalise mobility and migration in various ways. For example, forced population displacement has figured as a strategy in both insurgency and counterinsurgency campaigns. 20 In some cases, rebel groups may operate within or take control of refugee camps or use them as a base for the recruitment of refugees into armed conflict. 21 The term 'refugee warrior' has been used to describe highly politicised refugee groups who are both victims of violence, but also contribute to its perpetuation by taking up arms or supporting conflict. 22 Indeed, violent conflicts are often characterised by repeated cycles of violence and displacementforcibly displaced populations may well hold political grievances against the governments that have displaced them. These grievances can in turn be drawn upon by actors engaged in conflict-based forms of political mobilisation, thus further fueling a cycle of conflict and violence. 23 The dynamics of many violent conflicts around the world have also been shaped by the involvement or mobilisation of transnational diaspora populations. 24 In places as diverse as Northern Ireland, Kosovo, Sri Lanka, and Eritrea, the combination of large-scale emigration and transnational mobilisation by political entrepreneurs has meant that many 'local' conflicts have a 'global' dimension, with diaspora communities viewed by conflict actors as sources of external funding and political support. 25 In the United States in the 1970s and 1980s, for example, organisations close to the Provisional Irish Republican Army (IRA) maintained fundraising networks in Irish-American communities across the United States. 26 While such trends have been noticed in the literature, however, they have not been systematically theorised as forms of migration governance. As others have noted, there is a disconnect  between the migration governance and civil war literature such that scholars of migration governance have focused largely on the roles and activities of states and international organisations, whereas scholars of civil war have rarely conceptualised the above activities as forms of migration governance. 27 In fact, both states and non-state actors can engage in similar types of governance activities, including exercising control over diaspora and refugee populations, imposing and controlling borders, or cooperating with third party actors in forms of mobility management. Such activities are not new. In the 1950s, for example, the anti-colonial rebel group Algerian National Liberation Front (FLN) worked directly with the newly formed United Nations High Commissioner for Refugees (UNHCR) to protect Algerian refugees in Tunisia during the Algerian War (1954-62). Recognition by the UNHCR was a boost to the legitimacy of the organisation's government-in-exile in its conflict with the French colonial state. 28 Additional examples of nonstate rebel groups controlling and governing refugee camps include Sahrawi refugee camps in Algeria, which are formally run by the non-state Polisario Front or the central role that the Palestinian Liberation Organization (PLO) historically played in providing services in refugee camps in Lebanon. 29 Moreover, non-state armed organisations can be involved in the running of detention centres, as occurs in parts of Libya, where the European Union has come under fire for its collaboration with non-state rebel organisations in its migration control externalisation policies. 30 The same reasons that states seek to control and assert sovereignty over migration and mobility apply to non-state challengers to the state -although it should be noted that not all non-state violent actors aspire to statehood. 31 The power to control migration, however, is viewed as a key aspect of state sovereignty which governments jealously guard. 32 Control over national borders; mobility documentation, such as the power to issue passports and visas; 33 the power to engage in 'migration diplomacy' or bargain with other actors over mobility management; 34 control over emigrants and diaspora populations; 35 and the governance of refugee populations are all means of asserting state sovereignty over populations.
Thus, the ability of a rebel organisation or violent non-state actor to gain control over some aspects of migration governance also boosts their overall legitimacy and capacity -as well as their access to material resources -in their ongoing conflict with state authorities. Rebel governance  structures emerge when rebel groups construct institutions of governance that function either in competition with, parallel to, or in absence of formal state institutions. 36 They can vary in their degree of control from minimal to hegemonic and exist in contexts as diverse as Colombia, Liberia, Greece, the Ivory Coast, and Sri Lanka. 37 As scholars of rebel governance have shown, and as mentioned in the previous section, non-state actors frequently set up systems of taxation, local justice, security, and other functions performed by states. The governance of migration and mobility thus can be viewed as part of the overall "toolbox" of strategies in violent conflict, as well as a perquisite for asserting other forms of control over populations. Scholars of migration have analysed the development of state borders and admission policies as key to understanding the modern state. 38 Just as a central component of the Weberian state is its monopoly over the legitimate use of violence, theorists of the state have extended this to include control over migration and mobility. As John Torpey writes, 'modern states, and the international state system of which they are a part, have expropriated from individuals and private entities the legitimate "means of movement". ' 39 The emergence of the modern state involved the embracing of populations for the purposes of consolidating control over territory and collecting revenue, but also for providing welfare and services. 40 It has also involved the violent exclusion, removal, and expulsion of peoples -for example, through forced population movements and population exchanges. 41 Modern states use migration policy strategically to shape their populations and to project power -in this sense, migration management is both about facilitating movement and embracing new populations, such as diasporas, as it is about restricting movement. 42 Understood as a central feature of the modern state, it is unsurprising that challengers to the state would seek to contest a state's monopoly over the legitimate means of movement as part of their overall conflict strategy, and that migration governance emerges as an area of contestation in protracted violent conflicts. Both states and non-state actors exert control over territory and populations by facilitating or restricting movement (including through the strategic use of forced migration and refugee governance); by managing borders and checkpoints and by conferring or withdrawing legal membership in a polity -including remote membership, via diaspora engagement policies.
In the rest of this article I examine how these various dynamics can operate in a particular conflict. The ongoing conflict between the Turkish state and Kurdistan Workers' Party (PKK)-linked organisations provides a useful example of the role played by the governance of migration and mobility in civil wars, and how forced migration, refugee governance, border management, and diaspora governance all become part of the overall conflict strategies of both state and non-state actors. In the case of the Kurdish conflict, which stretches over multiple decades and has grown 36  substantially more complex since the outbreak of the Syrian war in 2011, non-state actors have engaged in forms of rebel migration governance within Turkey, Syria, and Iraq, as well as within the diaspora. The interplay between formal and informal migration governance presents ongoing challenges for human security in the region, and also calls into question conventional understandings of 'migration governance' as being limited to states and international organisations. Indeed, in this ongoing conflict, we see both replications of statist forms of migration governance by non-state actors, as well as innovations, such as non-territorial forms of diasporic citizenship, membership, and belonging that transcend state boundaries and seek to provide an alternative to the state. In the following section, I provide an overview of the conflict and illustrations of how these various mechanisms of migration governance have operated over time.

Contested migration governance in the Kurdish conflict
In the long conflict between Turkish state authorities and armed organisations associated with the Kurdistan Workers' Party (PKK), the control of migration and mobility has played a central role. The Kurdish conflict in Turkey is a particularly complex conflict, which is defined by cross-border, regional, and transnational dimensions in which the instrumentalisation and control of migration, mobility, and membership have been significant factors. The conflict has deeper historical roots in the marginalisation of the Kurdish population in the region, but in its current form can be traced back to the founding of the PKK in the 1970s, which was formed with the aim of Kurdish independence, and the subsequent start of armed conflict in Turkey in the 1980s. In Turkey, the conflict was at its most intense during the 1990s, and has since gone through various iterations, with a major shift in the dynamic occurring with the outbreak of the Syrian conflict in 2011. In 2012, an autonomous region in north and northeast Syria, known as Rojava (the land of the setting sun or the West, indicating the western portion of a broader Kurdistan) was established by the Kurdish People's Protection Units (YPG) and has operated as a functioning, if unrecognised, quasi-state since then. Thus, there is a strong and organic link between the conflicts in Turkey and Syria via the common membership of the PKK and the YPG in the KCK, with both conflicts demonstrating elements of rebel governance. 43 As of 2023, there is a complex and entangled relationship between the Turkish state, Turkish-controlled cantons in Northern Syria, the AANES, and the Syrian state, with the Turkish state openly framing its military intervention and occupation of Northern Syria as part of its fight against Kurdish actors in Turkey. 44

Forced displacement and refugee governance
In both the Kurdish conflict in southeastern Turkey, as well as its regional dimensions and its extension into northern Syria in the context of the Syrian war, forced migration has been actively used by both states and non-state actors as part of their overall conflict strategies, as have forms of refugee protection and refugee governance. State strategies of forced displacement of Kurdish populations go back to the late Ottoman Empire and Turkey's early Republican period. For example, there were forced cross-border displacements of minorities -including Kurdish minorities -from the newly formed Turkish Republic to Syria (under French mandate rule) in the 1920s and 1930s. Additionally, the 1934 Law of Resettlement was used to authorise the displacement and relocation 43 See, for example, 'Rebel governance and gender in northeast Syria: Transformative ideology as a challenge to negotiating power' , Third World Thematics, 6:1-3 (2021), pp. 69-87. As noted in the introduction, the quasi-state was formally renamed the Autonomous Administration of North and East Syria (AANES) in 2018 but is still often referred to as Rojava. 44  of people internally within Turkey during the early Republican period, and was applied to Kurdish minority populations during the Dersim massacre of the 1930s. 45 Forced displacement was a central aspect of the conflict between Turkish authorities and the PKK during the 1980s and 1990s. At the height of the armed conflict in the 1990s in southeastern Turkey, it is estimated that more than three thousand villages were destroyed, with figures of internally displaced over the course of the conflict ranging from several hundred thousand to three million. 46 Displacement was carried out largely by the Turkish state, operating via an auxiliary village guard system, and under the context of emergency rule, which allowed local governors to remove populations 'for security reasons' . The programme of displacement became part of the military strategy of the Turkish armed forces in response to the successes the PKK had in dominating rural areas of southeastern Turkey via their control of networks of villages across the region throughout the 1980s. The Turkish military countered this by establishing a village guard system in the 1990s, often destroying or evacuating villages that refused to participate or that were seen as sympathetic to the PKK. 47 Such tactics were reportedly also used by the PKK, which targeted villages participating in the village guard system, with civilians often caught in the middle of the conflict between the Turkish military and the PKK. Additionally, systematic displacement tied to large-scale economic development projects, such as the construction of networks of dams, was used by the government to further exercise control over the region and led to further displacement. 48 Displaced populations fled internally to urban areas in Turkey, creating an internal Kurdish diaspora, but also across the border to refugee camps in Greece and Iraq, as well as to destinations in Europe. 49 Following a military coup in 1980, Kurdish intellectuals, activists and militants had arrived in Western Europe as part of the approximately 60,000 political exiles who fled Turkey for political reasons at the time. Across all of Western Europe, almost 350,000 Turkish citizens applied for political asylum in various European countries between 1983 and 1994, with the number of foreigners seeking admittance to Germany under its asylum policies rising by almost 8,000 per cent. 50 At the same time, the PKK largely moved its strategic operations to Syria and Iraq, with the leadership based in Damascus throughout the 1990s.
Within this context of displacement, the governance of the internally displaced populations (IDPs) and refugees came to be part and parcel of the conflict dynamics, with both the Turkish state and the PKK politicising displaced populations. The Turkish state has arguably used policies of social welfare and economic development as a tool in the conflict, including specifically targeting displaced Kurds in urban centres in Turkey with welfare assistance, such as free health care, subsidised housing, education, food stamps, disability aid, and other forms of social support. Nevertheless, the common experience of violent displacement and dislocation helped forge a Kurdish political identity in areas outside of the conflict zone, and also provided the basis for some local mobilisation by the PKK. 52 At the same time, outside of Turkey, PKK-related groups were also involved in service delivery, with refugee camps in Greece and northern Iraq becoming defacto PKK-affiliated and self-governed camps where 'strict allegiance to the PKK and resistance to the Turkish state, and the two -the camp's identity and its politics -cannot be separated' . 53 Furthermore, the extensive transnational network of Kurdish diaspora organisations means that non-state forms of refugee governance also take place within communities of the displaced who have fled to Europe, sometimes in the form of self-help groups, and sometimes in conjunction with organisations sympathetic to or connected with the PKK. 54 An example of how the PKK was able to use displacement to its strategic advantage can be found in the case of the Makhmour refugee camp located in northern Iraq/Kurdish Regional Government (KRG), which was established by PKK-supporting Kurdish refugees escaping the Turkish state's campaign of displacement in southeastern Turkey in the 1990s. It was strategically located in an isolated area in which the PKK could establish control and create links with networks in Syria. The camp became a space for the establishment of independentÖcalanist governance structures, schools, and legal institutions, such as 'Justice and Peace Committees' that provide a collective means of resolving conflicts within the camp. 55 The camp, however, has also at times been a highly militarised space and has been accused by both Iraqi and Turkish officials of being used by the PKK for recruiting and training soldiers and launching attacks on Turkey. 56 Displacement and refugee governance have been strategically utilised by both state and nonstate actors in the latest phase of the conflict, which intersects with the dynamics of the Syrian War. When the conflict in Syria led to a massive outflow of refugees and displaced persons beginning in 2011, with Turkey hosting the largest number of refugees in the region by 2018, the situation was already intertwined with the dynamics of the Kurdish conflict in Turkey. For example, following the Battle of Kobane in northern Syria in 2014, both the Turkish state and Kurdish political leaders and organisations within Turkey set up refugee camps across the border in the town of Suruç. The Kurdish camps were served by Kurdish NGOs and the Kurdish Red Crescent, and were, according to observers, 'highly politicized' . 57 Whereas in Turkish government camps refugees had to learn Turkish, the Kurdish camps organised makeshift schools with Kurdish lessons. The Kurdish camps had politically salient names, such as the names of Kurdish martyrs. One camp was named 'Rojava' , thus effectively creating a 'satellite settlement of Rojava within the borders of the Turkish Republic' . 58 Kurdish camps were autonomously governed using theÖcalan-inspired system of democratic confederalism and residents of the camp openly 52  expressed support for YPG forces. As one observer noted, 'It is clear that the camps of each actor serve to support its respective political narrative: the state emphasizes its role as primary service provider, while the Kurdish nationalist movement asserts itself as the legitimate custodian for its ethnic kin. ' 59 Thus, although much attention in the literature on forced migration and migration governance has been paid to the domestic political, humanitarian, and migration diplomacy aspects of Turkey's role in the Syrian 'refugee crisis' , and its domestic policies towards refugees from Syria, Turkey's 'refugee governance' policies in Syria need to also be understood in the context of this larger ongoing conflict -a dimension that is often understudied in more state-centric approaches. 60 In the Kurdish-dominated areas of southeastern Turkey, for example, there was concern that the influx of Syrian Arabs was linked to government policies designed to change the ethnic balance in the region, thus undercutting bids for increased Kurdish regional autonomy. 61 At the same time, Turkey became directly involved in the conflict, in large part to prevent Kurdish groups affiliated with the PKK from establishing dominance on the Turkish-Syrian border and gaining direct access to the Mediterranean. 62 Turkey's own activities in the conflict have also led to internal displacements of populations in the region, with significant numbers of the displaced having fled to the Kurdish-controlled regions of Rojava in northern Syria. The strategic connection between Turkey's response to the Syrian refugee crisis and its conflict with Kurdish actors in the region can furthermore be seen in Turkey's linking of its policy of returning Syrian refugees to northern Syria with its multiple military interventions designed to create a buffer zone and prevent the expansion of a Kurdish polity, which it views as having the potential to shift the balance of power in its own Kurdish conflict. Thus, the creation of a 'safe zone' on the Syrian border with Turkey is both a form of refugee management and governance, but also plays a strategic role in preventing the acquisition of territory by what are viewed to be PKK-related actors. 63 Non-state refugee governance activities reached a new level in the Kurdish autonomous zone of northern Syria, which has been estimated to host up to 200,000 refugees from elsewhere in Syria in a network of 13 camps in Rojava run by the AANES. 64 Additionally, there are long-standing networks that connect Kurdish communities in northern Syria with communities in southeastern Turkey. Displaced Kurdish populations in northern Syria are therefore often dependent on humanitarian assistance from networks across the border in Turkey, but such cross-border mobility between Kurdish communities in Syria and Turkey is viewed as a security threat by the Turkish state, thus exacerbating the at times dire conditions in refugee camps in the Kurdish-dominated autonomous zone. 65 Within northeastern Syria, refugee camps in the region operate under control of the internal Rojava security and police forces, and camp residents include a mix of civilians and former 59 Ibid., p.  66 The relatively high presence of IS families has also periodically led to clashes and security operations in the camps. For example, 41 people were reported killed in the first three months of 2021 in the al-Hol Camp in the Al-Hasakah district of northern Syria. 67 The unrecognised status of the camps raises broader issues of humanitarian access, and the delivery of official aid from international organisations, as well as whether autonomous and unrecognised entities can be treated as legal entities capable of offering protection to those who reside within their borders -an issue that also applies to the neighbouring (but separate) Kurdish Regional Government (KRG) in Iraq. 68

Border control
The preceding discussion points to the significant cross-border dimensions of the Kurdish conflict. The management of borders is key to a state's overall migration governance strategy, but borders also play significant roles in violent conflicts. For example, the ability to cross a national border is key for a rebel group or other non-state actor to sustain links with translocal social networks and bases of support, but also plays a role in offering protection. In the 1990s, when the fighting was at its most intense in Turkey, the PKK largely operated from Syria and Iraq, but also extended its operations to the development of governance and mobilisation networks across migrant and diaspora communities in Europe, which provided an additional source of both material and political support. Thus, cross-border mobility has been an essential aspect of the Kurdish conflict dynamics, as it provides the basis for non-state actors to stay connected with and mobilize networks that are dispersed across multiple states. 69 In the early 2000s, following the capture and imprisonment of PKK-founder AbdullahÖcalan, the rise to power of the Justice and Development (AKP) party in Turkey, and a ceasefire in the active conflict in Turkey, there was a liberalisation of borders between Turkey, Syria, and northern Iraq in the context of Turkey's 'zero problem with neighbours' foreign policy. Visa liberalisation facilitating free movement between Syria and Turkey, along with domestic political liberalisation, allowed for the building and strengthening of translocal ties across the borders of Turkey, Syria, and the Kurdish Regional Government (KRG) a region that obtained increasing autonomy following the 2003-11 Iraq war. 70 This was designed to facilitate increased trade and circular migration, but was also utilised by PKK-related groups as an opportunity to build up translocal governance structures, under the umbrella of the Kurdish Communities Union (KCK), which had been founded in the mid-2000s. Meetings of the KCK were taking place in Turkey and elsewhere in the region throughout the late 2000s and early 2010s. 71 The translocal strategy included the establishment of a Syrian branch of the PKK, the Democratic Union Party (PYD), which later, with the onset of the Syrian conflict, established its own armed units, the People's Protection Units (YPG) and Women's Protection Units (YPJ), which are the primary components of the Syrian Democratic 66 For more on the role of the Syrian Democratic Forces and the Asayish internal security forces, see European Asylum Forces (SDF). All of these groups come under the auspices of the KCK, which promotesÖcalan's principles of democratic confederalism and exercises control of the regional armed forces.
The outbreak of conflict in Syria and its links to the conflict in southeastern Turkey meant that the Turkish-Syrian border took on increased significance across multiple dimensions, including Turkey's own domestic conflict with the PKK. 72 On the one hand, Turkey emerged as a gateway for foreign fighters entering the conflict in Syria -both for Islamist fighters supporting the Islamic State and anti-Islamist fighters fighting on the side of Kurdish groups. 73 Following an initial open border policy, Turkey gradually shifted to a policy of border closure, especially following the spillover of the conflict in Syria including the spread of IS activities in Turkey. Between 2013-16 Turkey suffered numerous suicide bombings and terror attacks across the country, many of which were attributed to IS-linked groups. 74 During this time there were also contradictory pressures coming from the international community with respect to the Turkish-Syrian border. 75 Keeping the border open potentially facilitated the activities of the Islamic State and other armed groups in the region, allowing for IS to bring in new recruits and supplies, sell oil and other commodities, and to receive medical treatment and other assistance in the border regions of Turkey. Yet closing the border prevented refugees fleeing the conflict from crossing into Turkey.
At the same time, actors connected with the PKK and KCK had an interest in maintaining an open border, and the ability to cross back and forth between Syria and Turkish-controlled regions of northern Syria in order to facilitate trade. 76 The issue of Turkey's border control policies became deeply contentious, and embroiled in the Kurdish regional conflicts, with a prevalent view in Turkey's Kurdish population that Turkey was manipulating its border policies to the detriment of Kurds, many of whom had strong cross-border social and familial networks. 77 The initiation of construction of a border wall between Kurdish regions in Turkey and Syria in 2013 was accompanied by protests and hunger strikes, and may have contributed to the breakdown of peace talks between the Turkish government and the PKK, with Kurdish actors in Turkey accusing the government of supporting ISIS by keeping Turkey's western border with Syria open while closing its eastern border. 78 By 2017, however, Turkey had completing a 700-kilometre wall along most of its 900-kilometre border with Syria at a cost of US $400 million. A further wall on the Iranian border was started in August 2017 as part of an Integrated Border Security System, with both walls funded in large part by the European Union, including funds from the 2016 EU Turkey deal targeted at strengthening border controls in the region. 79 From the EU perspective, the walls were designed to deter irregular migration, although from the Turkish state's perspective they helped it achieve its aim of disrupting cross-border operations by Kurdish militants. 80 On the ground, however, 72

Diaspora engagement: Mobilisation and counter-mobilisation
Diaspora governance is an additional area of migration-related competition in civil wars, and one that has been especially significant in the Kurdish conflict. 82 From the earliest days of the conflict in the 1980s, the PKK made links with Kurdish migrant communities in Libya and Lebanon as part of its conflict strategy, and its first armed attacks in 1984 included recruits from Kurdish communities in Germany. 83 At the same time, a Turkish state policy of promoting emigration to Europe in the 1960s as a form of economic development interacted with its policy of domestic repression of Kurdish political actors in ways that strongly affected the dynamics of the Kurdish conflict by transnationalising it. 84 As mentioned above, the 1980s and 1990s saw an exodus of exiles, refugees, and asylum seekers from Turkey to Europe, many of whom settled in areas where there were already existing populations of economic migrants from Turkey. At the same time, when the PKK went into exile in Syria in the 1990s, it simultaneously developed a European strategy centred around the mobilisation of the Kurdish diaspora and the establishment of a PKK presence in Europe. Throughout the 1990s, the PKK established a Parliament-in-Exile in Europe as well as branches of its political wing, the National Liberation Front for Kurdistan (ERNK), until it was dissolved in 1999 with the arrest and capture ofÖcalan. 85 In effect, the PKK set up a formal 'diaspora engagement' policy aimed at Kurds in Europe that was also coordinated with and designed to support the armed conflict in southeastern Turkey. 86 PKK members circulated back and forth between Europe and Syria, and engaged in significant political and cultural activities in Europe, including engaging in fundraising in the diaspora through the use of a tax system, political lobbying, raising awareness about human rights abuses in Turkey, organising public demonstrations and festivals and even recruiting fighters and volunteers to support the armed conflict in Turkey. 87 The dominance of the PKK in the diaspora extended to all areas of political, social, and cultural life, with a strong focus on cultural expression, such as publishing in the Kurdish language, which was banned in Turkey during the 1990s. Within the context of ongoing conflict in the region, the promotion of Kurdish culture promotes solidarity around Kurdish political agendas and raises the visibility of the nationalist movement. 88 A significant development was the setting up of a Kurdish satellite television station, MED-TV, from Europe. The station could also be picked up by viewers in southeastern Turkey, and thus played a significant role in challenging official state narratives. 89 The PKK and related groups in Europe in the 1990s sought to exercise political hegemony and control over the diaspora, including using violence and intimidation against rival Kurdish organisations, as well as tactics of intimidation and threats to secure support from the community in the form of donations and recruits. 90 At the same time, they brought forms of rebel governance into the diaspora, such as establishing autonomous forms of community dispute resolution and other informal legal practices. 91 Since the 2000s, many organisations in Europe have become closely affiliated to the broader KCK, which treats the diaspora as an integral part of its organisational structure. 92 Alongside an already-existing network of PKK-linked Kurdish organisations in Europe, new entities tied more closely to Rojava and the YPG have emerged. The largest umbrella organisation of Kurdish groups in Europe is KCD-E (European Kurdish Democratic Societies Congress), which has numerous national affiliates in different European states and is generally considered to be sympathetic to the PKK in its orientation.
Organisations in the diaspora connected with Rojava follow models of earlier forms of the PKK using the diaspora as a base for its strategy rebel diplomacy in Europe in the 1990s. 93 The declaration of autonomy by Rojava in 2013 was also accompanied by a strategy of internationalisation, with campaigns to gain support in Europe, Russia, and North America. The PYD was able to effectively draw on resources in the diaspora and 'utilise its access to global communications and advocacy networks to pursue a sophisticated program of public diplomacy' . 94 In 2016 Rojava offices were opened in Moscow, Paris, Prague, Stockholm, Berlin, The Hague, and Copenhagen. The YPG has also sent delegations to various countries throughout Europe, where they have been received by public officials or addressed parliamentary bodies. As part of its overall transnational strategy, the offices are used to gain legitimacy and status, and to secure financial and political resources from abroad. 95 The PKK and KCK's overall diaspora strategy therefore plays a central role in the conflict -in effect challenging territorial forms of political belonging by bringing symbols of Kurdish nationalism into multiple spaces and locales in Europe and beyond, with diaspora activists mobilising solidarity networks, engaging in campaigns and supporting educational training onÖcalan's principles of democratic confederalism. Such organisations relocate aspects of Kurdish nationalism and identity to spaces and places beyond the borders of Turkey, Syria, or other states that directly form part of the Kurdistan region. They do so by both providing support to the local community of refugees and migrants and replicating and localising symbols of nationalism in the diaspora. 96 At the same time, however, the Turkish state has also long recognised the political significance of the diaspora and began to see the transnational diaspora as part and parcel of the ongoing Kurdish conflict, using various forms of state control to monitor and deter the political activities of Kurds in Europe. It has for many years engaged in the surveillance and 'long-distance policing' of political activists in Germany, France, and elsewhere and, similar to other states, it developed systems and strategies of 'transnational repression' designed to suppress political opposition overseas. 97 For example, Kurds living in Europe returning to Turkey were often harassed or would find themselves held for questioning or, in some cases, arrested. Another technique of repression was to cancel or refuse to renew the passports of Kurdish political activists living abroad, thereby compelling them to either return to Turkey, live abroad illegally, or apply for asylum. 98 In addition, the Turkish state placed pressure on European countries in the 1990s to ban the PKK and its associated organisations. The PKK was banned in Germany in 1993, and PKK-related organisations and media groups, including MED-TV in the 1990s, and organisations such as Firat News Agency in the Netherlands; ROJ-Groupa and Denge Mezopotamya Radio in Belgium; ROJ TV and MMC TV in Denmark; Newroz TV in Norway; the House of Kurdish People in Marseille, France; and the newspaper Yeni Ozgur Politika in Germany were all at times banned abroad under pressure from the Turkish state. 99 There have been several instances of members and supporters of the PKK being targeted for assassination in Europe, with suspicion that such assassinations were carried out by the Turkish Intelligence Service (MIT) or organisations close to the state. The most reported incident was in 2013, when Sakine Cansiz, one of the co-founders of the PKK, was executed, along with two other women, Fidan Dog an and Leyla Sö ylemez, in the Kurdistan Information Center in Paris. It is widely assumed that those murders were carried out by the Turkish state. In 2016 an assassination plot was allegedly uncovered that involved the Turkish Intelligence Service targeting the leaders of two Kurdish organisations, Kongra-Gel co-chair Remzi Kartal and European Kurdish Democratic Societies Congress (KCD-E) co-chair Yüksel Koç. In response the German police in Hamburg detained a Turkish intelligence services agent in December 2016. 100 This means that many Kurds in Europe have had to continue to navigate between the Turkish state and the PKK, just as they had to within parts of southeastern Turkey during the military conflict -the diaspora continues to be, in effect, a 'contested constituency' in the ongoing conflict.

Conclusions
The analysis in this article suggests a number of theoretical and policy implications. First, our understanding of the nature of migration and mobility governance in civil wars would benefit from a greater dialogue between scholars of migration governance (operating largely within a migration studies framework) and scholars of civil war and rebel governance (operating largely within a security studies framework). Understandings of migration governance in the migration studies and public policies literature focus overwhelmingly on states, local policymakers and international organisations, thus often missing the complex interplay between state-and non-state actors, including dynamics of contestation and competition across multiple areas of migration governance.
Secondly, understandings of the relationship between migration and security remain somewhat limited due to a methodological nationalist bias in IR and security studies. 101  to security in the field focus overwhelmingly on formal state migration policies and state national security interests, or treat migration as an area of global governance by state-dominated international organisations, thus under-theorising the role played by non-state actors and their effects on broader structures of migration governance. Critical approaches to migration, which place greater emphasis on questions of human security, also have a blindspot in their focus largely on states and state borders in the Global North, ignoring the complex interplay between formal and informal migration governance in many parts of the Global South -and the impacts this also has on migration governance regimes in the Global North. For example, securitisation approaches that focus largely on discursive constructions of the migrant display a marked Global North bias and often miss how migration governance regimes are strategically utilised by multiple actors in ongoing conflicts. 102 The political economy of formal migration management is connected to broader conflicts that form alternative subaltern and counterhegemonic regimes of migration management.
As such, policies of migration governance often have consequences that shift the balance of power in civil wars, exacerbate ongoing violent conflicts, or have other unintended side effects. Migration control policies in Europe reverberate across different contexts, often affecting the local balance of power in local conflicts and creating additional challenges for victims of conflicts. These dynamics have led some to argue that EU migration diplomacy in this area has in effect been akin to engaging in 'proxy wars' . 103 For example, the EU funding of enhanced border control in Turkey via the 2016 EU-Turkey deal has supported the building of border fences that disrupt local Kurdish networks on the ground. Another illustration of the complex interlinkages between European asylum policy and EU-Turkey relations emerged when Turkish President Recep Tayyip Erdog an suggested in May 2022 that Turkey would not support Finnish and Swedish membership in NATO due to them being 'guesthouses for terrorist organisations' -a clear reference to Kurdish organisations operating in Sweden. 104 Thus, both scholars and policymakers could benefit from examining the complex linkages that exist between migration governance and violent conflict. Migration governance is not simply a policy arena in which states and international organisations react to migration flows, migration governance is also used by armed rebel groups and other non-state actors in civil wars, and can be viewed as an area of strategic importance in rebel governance and rebel groups' interactions with government authorities. These dynamics can be seen in the case of the Kurdish conflict, in which practices of displacement, refugee governance, border control, citizenship and diaspora governance have all played central roles in the ongoing protracted conflict between the Turkish state, PKK-related groups, and actors in Syria.
However, such dynamics are not limited to the Kurdish conflict, but can arguably be found in a range of violent conflicts around the world and, indeed, are characteristic of civil wars and situations of protracted violence that involve armed non-state actors. Our understanding of how migration governance operates in civil wars would benefit from more comparative research across different conflicts, as well as a greater attention to the different elements of migration governance that have been identified in this article, such as the strategic use of forced migration, refugee governance, border management, and diaspora engagement as part of the conflict toolbox of both state and non-state actors.