EU peacebuilding’s new khaki: Exceptionalist militarism in the trading of good governed for military-capable states

This article explores how European Union (EU) peacebuilding is being reconfigured. Whereas the EU was once a bulwark of liberal peacebuilding, promoting a rule of law–based international order, it is now downplaying the goal of good governance and placing military capacity as central for international peace and security. Several works have analysed these changes but have not theorised militarism, despite war-waging and war-preparation have marked EU peacebuilding’s direction. The article argues that EU peacebuilding continues to expose elements of liberal militarism since its origins but is now changing from what Mabee and Vucetic call a nation-statist to an exceptionalist militarism. This shift implies that peace has ceased to be served by the intervention of sovereignty with a discourse based on the link between order, good governance, and human rights and is now premised on the upholding of sovereignty, even if that means the suspension of rights. The research draws on thematic analysis of EU documents and interviews undertaken with EU and G5 Sahel officials and managers of EU-funded peacebuilding programmes. It also briefly analyses the case of the Sahel as an example of how the build-up of states’ military capacity is strengthening states’ capacity to override human rights and repressing dissent.

European Union (EU) peacebuilding is being reconfigured. Whereas the EU was once a bulwark of liberal peacebuilding, promoting peace and development as a way to foster a rule of law-based international order, good governance is now a desirable but distant goal, with military capacity positioned centrally for international peace and security. Such reconfiguration is relevant to understanding new interpretations and practices of EU's foreign policy and where liberal peace might be going. References to liberal peacebuilding's end are becoming commonplace, based on a downturn of liberal goals and a rise of peace enforcement (Chandler, 2017;Karlsrud, 2015Karlsrud, : 1215Sloan, 2011). For the EU, the new focus has implied few references to good governance, an increase in military operations to train and equip, and a new regulation, called Capacity-Building for Security and Development (CBSD), that grants non-lethal military capacity through one of the main EU peacebuilding's instruments. Whereas peacebuilding was used to address a 'problematic relationship between [Southern] states and their armed forces', it is now working to boost those forces to secure sovereignty and territorial control (Abrahamsen, 2019: 543).
Several works have explored how pragmatism, resilience, local ownership, and capacity-building are transforming EU peacebuilding (Bargués, 2020;Chandler, 2017;Edmunds and Juncos, 2020;Ejdus and Juncos, 2018;Juncos, 2017). According to this literature, a pragmatic turn has propelled a move away from large-scale liberal reforms and a push to place 'locals' at the forefront. However, it has not fostered inclusivity and has continued to assert liberal values and the need for long-term interventions, even engaging in forms of governmentality. These are important aspects and contradictions of EU peacebuilding's transformations. Nonetheless, that building war capacity has become central to peacebuilding reveals the need to study the role of militarism in defining continuities and changes.
Militarism has constituted and transformed EU peacebuilding all along. The latest shift partly evokes a truism among policy-makers -that states need to be secured to undertake development, democracy, or reform projects -but it is problematic. In the Sahel, a region that has since 2012 seen a combination of coups, armed rebellions, and terrorism, the EU has supported the creation of G5, a military force of five Sahelian countries (Burkina Faso, Chad, Mali, Mauritania, and Niger). This approach has, however, reproduced the context of war and human rights abuses in the region while not curbing terrorism or armed groups (Zyck and Muggah, 2013). It thus shows a match of security and strategic interests between these countries and the EU, and that 'local' solutions can be as problematic as those from the 'top'. Examining EU peacebuilding's reconfiguration requires a sustained critical inquiry about militarism and its consequences.
Militarism is understood here as a discursive and material practice that influences social relations and decision-making in the preparation and waging of war (Shaw, 2013). Insofar the EU has engaged those practices and embedded military production and warpreparation in its own development, its peacebuilding has a different khaki tonality, but the khaki itself is not new. Following Mabee and Vucetic (2018), EU's new khaki denotes a shift from nation-statist militarism of the early liberal peace to a new exceptionalist militarism -that is, from a militarism based on building state security and defence alongside social and economic development, to one boosting sovereignty and military capacity to exercise control and coercion. This exceptionalist turn coexists, simultaneously, with liberal militarism, maintaining moral justifications for the use of force, even if not as front banner; with the commitment to war-related production and preparation as a modernising project (Mabee, 2016); and with civilian and humanitarian goals and practices, premised on the security-development nexus (Abrahamsen, 2019).
The article makes two interrelated arguments. First, it explores how EU peacebuilding is changing, arguing that the most defining change is the transposition of the goal of good governed for that of military-capable states. It then conceptualises the role of militarism underpinning these changes, arguing that EU peacebuilding has gone from a nation-statist to an exceptionalist militarism while maintaining a liberal militarist framework in which the security-development nexus and humanitarian and development actors are central. The article first analyses the evolution of EU peacebuilding as linked to militarism and explores types of militarism. This is followed by an empirical examination of changes in Common Security and Defence Policy (CSDP) missions, the Instrument contributing to Stability and Peace (IcSP), with the new CBSD initiative and a brief analysis of the case of the Sahel. 1 EU peacebuilding and militarism: A co-constituted relationship EU peacebuilding has developed as part of EU's ethos to promote peace and the rule of law inside and outside its borders. Although the EU is not guided by any particular definition of peacebuilding (Stamnes, 2016: 3), having the end of the Cold War, the rise of international humanitarian intervention, and the breakup of Yugoslavia as the background, EU peacebuilding goals became defined by the liberal peace agenda and by liberal peace's own militaristic character.
Liberal peace is a term designated by its intellectual critics (Duffield, 2001), broadly referring to post-Cold War interventions of Northern states to reform Southern states in the context of conflict management (Sabaratnam, 2011). The 'liberal' label identified how interventions became based on the idea that 'the rule of law and market economics would create sustainable peace in post-conflict and transitional states and societies, and in the larger economic order they were part of' (Campbell et al., 2011: 1). Its militaristic aspect comes from its commitment to seek peace through war and militaristic institutions (Basham, 2018). Liberal peace can be seen as a form of 'liberal militarism' (Basham, 2018;Stavrianakis, 2018). It features moral and legal justifications to restore and maintain international peace and order; technology-based preparations for war both in intervening and in intervened states with a modernising aim; and is intrinsically linked to security (Mabee, 2016;Stavrianakis, 2018). These elements identify how EU peacebuilding evolution is co-constitutive of its own militarism.

The evolution of EU peacebuilding
EU peacebuilding could be seen as having three distinct phases: an initial one from the early 1990s to the early 2000s, an expansive phase from about 2003 to the late 2000s, and the current one, which could be seen as a reconfiguring phase. Common to these three phases are the elements of militarism just formulated.
The initial phase, in the midst of the Balkans' post-Cold War conflicts, was triggered by EU's recognition that 'it needed to assume its responsibilities in the field of conflict prevention and crisis management' (EEAS, 2016: para. 2). The Maastrich Treaty introduced the first provisions for the EU's Common Foreign and Security Policy (CFSP), which explicitly aimed to 'safeguard' the values, interests, and security of the Union, as well as to 'preserve peace', 'strengthen international security', 'promote international cooperation', and 'develop and consolidate democracy and the rule of law' (Art. J.1). Manners (2006: 186 and 189) argues that this early approach was based on building civil capacity for reconciliation under the idea of 'sustainable peace' and that only from 2003, the EU took a militarised 'turn [. . .] towards a full spectrum of instruments for robust intervention'. Yet, as mentioned, militarism already underpinned ideas around conflict prevention and crisis management.
In the early 1990s, the Western European Union (WEU) -a defensive alliance -first developed the St. Petersburg tasks, which outlined how military troops would engage in combat for crisis management and peace-making, aside from other humanitarian, peacekeeping, and post-conflict stabilisation activities (EEAS, 2016: para. 5). These tasks, formally integrated in the CFSP when the 1999 Treaty of Amsterdam developed the European Security and Defence Policy (ESDP -later CSDP) and later expanded in the Treaty of Lisbon (EEAS, 2016), point to the use of war and militaristic institutions to seek peace. At that time, the breakup of Yugoslavia had already questioned strict adherence to a civilian non-interventionist policy (Barbé and Kienzle, 2007). By the end of the 1990s, the EU was concerned with 'developing capacity for peacekeeping intervention abroad in situations where NATO declined to get involved' and where Western countries were reluctant to contribute to United Nations (UN) troops (Chandler, 2017(Chandler, : loc. 1216. As explained in detail below, this 'capacity' meant technological preparations for war with a modernising aim both for the EU and for the intervened states. EU peacebuilding was consolidating and entered an expansionist phase. The second phase was marked by Javier Solana's (2003) European Security Strategy (ESS). It focused on linking state failure to threats like terrorism, regional conflicts, and organised crime. Peacebuilding then acquired what below will be defined as nation-statist militarism for two reasons: issues of security became tied to state failure and the securitydevelopment nexus, and peacebuilding became statebuilding. Interventions between 2003 and 2008, as shown later, expanded in quantity and concerns to address states' institutional infrastructure, encompassing all security, military, political, economic, and social arenas. Operation Artemis in the Democratic Republic of the Congo (DRC) -the first autonomous EU-led military operation launched in 2003 -showed this new approach, setting the benchmark for statebuilding linked to good governance, democratisation, and development (EU Parliament, 2006: 17).
Several authors note how Solana's ESS reflects how EU peacebuilding and militarism interrelate. Kronsell (2012: 69) shows that during this period, the EU consolidated as a 'postnational' defence organisation beyond its members, concentrating military organisation on security issues overseas. For Kronsell (2012: 11 and 13), this military organisation is both intrinsic to seeing war as a means for conflict resolution and global norms promotion, and built upon particular constructions of masculinities and gender relations. Manners (2006: 183) identifies a 'quantitatively different arms dynamic' brought in by the 'drive towards "martial potency" and the growth of a Brussels-based "military-industrial simplex"'. For peacebuilding, this meant an agenda geared towards the imposition of EU's will by force, defined by the military and technology-industrial lobby and intentions to place 'the benefits of technological and industrial defence and aerospace research and investment' at its centre (Manners, 2006: 193).
In this early to mid-2000s period, the EU concentrated on security and stability, amid a broad reform agenda (Richmond et al., 2011). Not surprisingly, the first peacebuilding instrument, the Instrument for Stability (IfS), created in 2007 dropped the term 'peace', focusing on building government institutions, supporting elites, and engaging in large reconstruction projects, related to state security, and economic and industrial bases (Natorski, 2011).
The third and current phase has been triggered by the 2008 economic crisis, changes in the international security climate with terrorism and migration rising in Europe, and the inherent contradictions of liberal peacebuilding. It is reflected in Mogherini's Global Security Strategy (GSS), which, contrary to the shifting point marked by Solana, encapsulates a well-entrenched trend in EU foreign policy. This trend makes 'resilience' and 'local ownership' central to peacebuilding, tied to the necessity to uphold rather than intervene sovereignty. Under a self-identified approach of 'principled pragmatism', the GSS states that 'soft and hard power go hand in hand' (EU Commission, 2016: 4 and 9). The IcSP, which replaces the IfS in 2014, brings 'peace' back, maintaining conflict prevention and peacebuilding goals alongside tackling global and trans-regional threats to peace and international security (EU Commission, 2016: 4). Since 2017, it provides nonlethal military capacity. If for Solana the EU was not only capable but also responsible for international peace and security, for Mogherini the EU cannot achieve this by itself and requires self-defence and building partner military capacity. These broader changes have implied not only the need for military institutions and technological advancement but also the need to outsource war-capacity to national armies alongside development and humanitarian programmes (Frowd and Sandor, 2018: 5).
The new 2021-2027 multiannual financial framework reflects these priorities, moving from growth and structural cohesion in the EU to migration, border management, security, defence, research, and innovation (Kolek, 2018;Parry and Sapala, 2018: 2-3). Although the 'Neighbourhood, Development and International Cooperation' envelop, where peacebuilding will be located, is foreseen to grow 11%, it is far from the foreseen 207% increase for the Migration and Border Management Heading or the 1138% increase for Security and Defence (Parry and Sapala, 2018: 21-23). EU's new khaki represents a u-turn in regard to which values and strategies best serve international order.

Conceptualising the latest EU shifts
The latest shifts should be seen as a response to several intertwined crises. First, the crisis of liberalism -felt in the EU with the loss of the United States as an ally in liberal internationalism, Brexit, and the rise of anti-liberal politics -has marked 'an end to the long postwar project of building a greater union' and to its role as the 'quiet bulwark of the wider liberal international order' (Ikenberry, 2018: 7). Second, the crisis of liberal peace has questioned the paradigm grounding EU peacebuilding. Third, the 'migration-crisis', with the arrival of about 1 million migrants in 2015 and the latest experiences of Islamist terrorism in Europe, has put the EU in a defensive stance. Articulating a response to these issues has generated contestation within the EU (Johansson-Nogués et al., 2020). But though France and Germany's push to boost defence capacity has clashed with Nordic countries' traditional understandings of development (as rather requiring civilian infrastructure), a militarised security-first approach in peacebuilding represents a majority view (Iñiguez de Heredia, 2020a).
Scholars working on the EU have explored these issues through pragmatism, resilience, and governmentality. Pragmatism captures the mood of 'resignation', 'disillusionment', and 'frustration' between the theory and practice of peacebuilding (Bargués, 2020;Chandler, 2017). It identifies policy-makers' concern with the consequences of applying top-down approaches and universalist understandings of what peace and international order should be (Bargués, 2020;Chandler, 2017;Juncos, 2017: 5). It also accounts for how solutions have become premised on bottom-up, resilience, and local-ownership approaches.
Resilience is welcomed for its potential to create new opportunities to think creatively about conflict, development, security, and humanitarianism (Wagner and Anholt, 2016). It has made the 'demand' side of peacebuilding and the impact of 'local' agency on EU policies' outcomes more visible (Ejdus and Juncos, 2018: 5-6;Müller and Zahda, 2018). However, resilience has also operated as a way to commit less resources under a similar top-down approach. Military capacity-building is a long-time claim from third states. Since 2004, the African Peace Facility (APF) provides military equipment and training for African Union troops. The new turn is, however, boosting political and military elites, outside a multilateral framework, and the requirements for political reform (Fisher and Anderson, 2015;Frowd and Sandor, 2018). From this perspective, the 'local turn' looks neither innovative nor unproblematic.
The 'resilience', 'bottom-up', and 'local-ownership' turn has ultimately been largely 'unfulfilled' and contradictory (Joseph and Juncos, 2020). According to Bargués (2020: 14), both in Bosnia and in Kosovo, the EU stepped in with an aim to engage in more locally driven and less invasive practices, yet the EU has maintained a long-term presence, and staff members claim simultaneously 'that they should intervene less but stay longer; [. . .] offer recommendations and facilitate peace'; and 'let local actors lead the way'. 'Local ownership' like 'capacity-building' has been perceived as a form of governmentality (Edmunds and Juncos, 2020;Ejdus, 2018). With 'capacity-building', argue Edmunds and Juncos (2020: 3 and 5), the EU has turned into a more 'technical, problemsolving endeavour' seeking to promote local actors' resilience to deal with different challenges and insecurities by supplying 'material assistance, knowledge and technology' as a result of an identified 'lack' thereof. However, this approach continues to impose externally derived understandings of what proper capacity is; it implies the creation and reproduction of power and hierarchical relations between the EU and the target states that are deemed 'incapable'; and no actual capabilities end up being built, because of both the flaws in the strategy and the contingencies and resistances found in its application (Edmunds and Juncos, 2020). Pragmatism has not completely displaced the universalist liberal framework but is providing a way to 'deny the moral imperatives of those universal categories' (Juncos, 2017: 2).
These analyses are reflected in changes in CSDP missions and IcSP programmes as well as with the case of the Sahel discussed below. Yet the question that emerges is not just how EU peacebuilding has changed, nor what theory and practice contradictions have emerged, but how has militarism shaped recent transformations and what are its consequences. This is because the perceived value of the military and the institutionalisation of war and war-preparation are central to pragmatic, resilience, and capacity-building shifts in EU peacebuilding.

Militarism and its types: Nation-statism and exceptionalism in the EU
Historical sociological and feminist approaches to militarism are among the most useful, as they pinpoint interactions of social, ideological, and material elements in the discourses and practices of liberal peace's militarism. Historical sociologists emphasise (Shaw, 2013: para. 2;Stavrianakis and Selby, 2013) discursive and material practices that influence social relations and prescribe ways of thinking and doing around the capacity, preparation, and waging of war. Feminists emphasise the normalisation of 'war-related, war-preparatory, and war-based meanings and activities' as well as 'the blurring or erasure of distinctions of war and peace, military and civilian' (Sjoberg and Via, 2010: 7). This also highlights peacebuilding's inclusion of the military in civilian, humanitarian, and development affairs and vice versa (Abrahamsen, 2019;Gelot and Hansen, 2019).
While conceptualising militarism is important to understand EU peacebuilding, thinking in terms of types unearths continuities and changes (Stavrianakis, 2018: 10). The most prolific literature on types of militarism has come from historical sociology. Some typologies do not travel well because of their historical boundedness (e.g. 18th-century militarism vs feudal and post-absolutist militarism (Speier, 1936); autocratic and capitalist-liberal militarism in Russia, the United States, and Europe (Mann, 1993: 686); Mann's (1996) typology of Cold War militarism in the USSR and Western nations. Others have a degree of flexibility, despite their contexts, but do not capture EU peacebuilding well enough (e.g. Shaw's (2013) typology by mode of warfare).
Liberal militarism frames many constitutive practices of EU peacebuilding and liberal peace as a whole. The term, coined by David Edgerton to identify a distinct 'British way of warfare', has been systematised as a broader phenomenon, which includes the moral and legal justifications of the use of force and the embeddedness of a capital and technology-intensive approach to military production and war-preparation into the political economy of modernisation both across wartime and peacetime (Mabee, 2016: 3;Stavrianakis, 2018: 12). As shown in the previous section, the EU has premised different aspects of peacebuilding, and military intervention in particular, on the need to defend human rights and a rule-based order. This has been underpinned by investments in warrelated technology, whether as part of an incipient conflict management strategy in the 1990s, as part of statebuilding in the early 2000s, or the current focus on building military capacity. But if liberal militarism highlights continuities, Mabee and Vucetic's typology accounts for important changes.
Mabee and Vucetic (2018: 100-103) identify four varieties: (1) exceptionalist militarism, related to the constitution of sovereignty and the political as theorised by Carl Schmitt; (2) nation-state militarism, related to the formation and development of the nation-state, including in its social and economic arenas; (3) civil society militarism, related to non-state but state-supported violence, as seen in colonial imperialism, paramilitary violence, and some contemporary criminal and insurgent groups; and (4) neoliberal militarism, related to the marketisation and privatisation of personnel, arms, equipment, logistics, and organisation of the military.
This typology has several advantages. First, as will be shown in this article, the varieties of nation-statist and exceptionalist militarism account for changes in EU peacebuilding. Second, without an aim to be 'transhistorical', this typology has a more generalistic application outside the specific historical contexts of other typologies (Mabee and Vucetic, 2018: 99). Third, these varieties have to be understood as co-constitutive of security, thus capturing well EU peacebuilding practices for the central role given to state sovereignty and its intertwinement with security strategies (Mabee and Vucetic, 2018: 97 and 104). In that regard, despite the supranational character of the EU, these varieties apply as they reflect the interaction between micro-and macro-dynamics that have affected different actors at different times (Mabee and Vucetic, 2018: 99). Finally, both nation-statist and exceptionalist types are compatible with liberal militarism.
Although Mabee and Vucetic (2018: 102) see liberal militarism as part of their nationstatist type and as surpassed by neoliberal militarism, where the marketisation of defence procurement and the rise of military private actors make it a differentiated category, as Stavrianakis (2018) argues, these varieties overlap. Mabee and Vucetic take note of the 'brittle' difference there is between neoliberal and nation-statist militarism, given that markets and liberalisation have not weakened the nation-state. In addition, exceptionalist militarism as seen from EU peacebuilding is exercised following what Abrahamsen (2019) calls 'feel good' militarism, whereby the upholding of sovereignty and military capacity is conveyed by military elites and government officials as much as by humani-tarian and development actors. Using these categories can help us account for what material and discursive practices persist and which ones are transforming.

Liberal peacebuilding and nation-statist militarism
Liberal peacebuilding and EU's peacebuilding during the 2000s could be seen as both pursuing and being based on nation-statist militarism. Mabee and Vucetic (2018: 101) draw on Michael Mann to see nation-statism as 'the default ("normal") setting for militarism in international and global life' and as defined 'by some form of civilian control over the armed forces and a state-led economic and social mobilization of "destructive" forces'. Under this approach, 'capacity-building' was already important during the 2000s, but it was institutional, bureaucratic, and governance capacity (Chandler, 2017(Chandler, : loc. 1790Edmunds and Juncos, 2020: 5).
As Chandler (2017Chandler ( : loc. 1825) states, 'capacity-building' was in essence a process where international institutions, and EU in particular, were building states by 'undermining sovereign autonomy', pushing 'these states to cede their sovereign powers to external institutions'. In Bosnia, for instance, sovereignty was transferred to Brussels (Chandler, 2017(Chandler, : loc. 1921). Programmes centred on securing states, deploying a variety of Security Sector Reform (SSR), development, and democratisation programmes that without dismissing the centrality of the security and military apparatus of the state were embedded in a broader political and economic project of state reform.
This basic framework is still present as the EU continues to assert liberal values, but it is turning towards an exceptionalist form, requiring the upholding rather than the intervention of sovereignty and a suspension rather than the granting of the package attached to good governance.

The turn to exceptionalist militarism
In between these three decades since the 1990s, several adjustments have taken place. The EU has turned to build military capacity, while the UN has shifted to peace enforcement (Sloan, 2011). The agency of the use of force has moved from Western to non-Western states and to regional organisations (Chandler, 2017(Chandler, : loc. 1197(Chandler, -1248De Coning, 2017: 146-150). Interventions have gone from a focus on stabilisation and recovery of state authority to intervene at the sub-provincial local level with an increasing engagement of civil society while simultaneously fostering a rise of military capacity-building (Abrahamsen, 2019;Chandler, 2017: ch. 5;Mac Ginty and Richmond, 2013).
Mabee and Vucetic relate exceptionalist militarism to the revival of Carl Schmitt's thought, through scholars like Giorgio Agamben. As they point out, '[m]uch of this thought revolves around the political and legal constitution of sovereignty through the suspension of regular legislative and judicial rules and procedures, via unified civil society support, for the purposes of dealing with enemies and security threats' (Huysmans cited in Mabee and Vucetic, 2018: 100-101). This suspension is becoming explicit in how military capacity is prioritised regardless of good governance. In the Sahel, this strategy is reproducing violence and strengthening the position of military and police actors, accused of human rights violations, vis-à-vis civilian actors. But if the Sahel has seen the deployment of exceptional military measures with Operation Barkhane -a French counter-insurgency operation with 4500 troops -and the G5, the broader significance of this turn is its concern with building military capacity and defending sovereignty as a simultaneous military, civilian, and humanitarian endeavour.
While the elements of liberal militarism have remained fairly constant, the main differences between the nation-statist and the new exceptionalist approach rely on the status of sovereignty. If the 'suppression' of sovereignty was first key to secure states around good governance (Chandler, 2017), its defence and upholding has become most important. These two forms of militarism have not developed neatly one after the other, but thinking in their terms signals the centrality of militarism in constituting and transforming EU peacebuilding.

Military-capable states: From a compromise to the ultimate goal
The following section illustrates empirically the two interrelated claims this article makes: first, that a prominent shift in EU peacebuilding has implied a focus on military capacity, downgrading good governance and the most expansive institutional reform agenda of the liberal peace; second, that this could be understood as a shift to exceptionalist militarism in that these military-capable states are concerned with defending sovereignty and have the potential to suppress rights and democracy.
The analysis focuses quantitatively and qualitatively on the most important and current peacebuilding tools -CSDP missions and the IcSP, including the CBSD initiative. It also explores briefly the case of the Sahel. The CSDP data come from the original mission mandates from 2003 to 2019. 2 The IcSP analysis is based on a self-compiled database from the full list of projects from 2014 to 2019, with particular attention to the distribution of number of contracts and themes. The qualitative analysis has been done with NVivo software for a deeper thematic analysis. This is expanded with 26 interviews with relevant EEAS officers, Council members, G5 members, and IcSP project managers undertaken in May 2017 in Brussels and followed up by phone throughout 2018 and 2019. 3 The section finishes with a brief exploration of the case of the Sahel, which features the latest EU intervention approach and illustrates the exceptionalist militarist turn.

CSDP missions
As previously seen, CSDP missions have by and large consistently prioritised the strengthening of states' governing and security apparatus. Military capacity is the latest variation in this priority, which as shown by Graph 1 starts to take place from 2010 onwards.
At the end of the 2000s, missions change in quantity, form, and content. Between 2003 and 2008, 23 missions were launched, 6 of them military (26%), 1 civil-military, and 16 civil. There are no missions in 2009. From 2010, missions drop in number (14 overall), but military missions rise to make 42% of the total. Yet whereas, as observed below, in the early 2000s all military operations had an enforcement mandate, only two of them have it after 2010, most of them being military training missions. This means that although CSDP missions have been consistently concerned with maintaining order and stability, and military means have been an important instrument, the approach has moved towards building states' capacity to manage war. This includes the boosting of armed forces with skills on 'infantry, force organisation, sniper, mortar firing, leadership, engineering logistics, tactical air control and intelligence gathering', and equipment such as four-wheel drives, motorcycles, patrol boats, drones, detectors, and IT, as well as with programmes towards resilience and stabilisation (EU, 2019: 5).
As Frowd and Sandor (2018: 74) argue, the militarism of this approach does not come from the EU's glorification of war but from the fetishisation of state's coercive power. The EU has in fact reduced its troop deployment (going from 1000 in Ituri, DRC, and 4300 with EUFOR-Tchad/CAR up until 2008 to an average of just over 300 troops from 2010 (e.g. EUFOR CAR (700) and EUMAM RCA (60), except for EUNAVFOR Sophia with 2260). Yet while the EU has not purely relied on an aggressive martial approach, and has proclaimed the need to restrain military power in the name of security, development, and human rights, it has been a driver for sustaining states' monopoly of the means of violence and its coercive capacities, facilitating the social acceptance of military means, values, and solutions (Abrahamsen, 2019;Frowd and Sandor, 2018: 77;Stavrianakis, 2018).
The rise of military capacity is also seen from the mandates. Graph 2 shows that up until 2008, the greatest number of missions have a rule of law and, most significantly, a police and SSR mandate. Between 2003 and 2009, police and SSR missions make up 45% of the total. From 2010 to 2020, there are no new Rule of Law missions, and EUPOL and SSR missions transform into EUCAP missions. EUCAP missions integrate many aspects of Police and SSR missions but frame reform under capacity-building with an explicit view to grant states capacity to independently respond to security and armed threats by 'train and equip' means. For instance, EUCAP Niger and EUCAP Mali focus on the restoration of order. Even if this order is defined as constitutional and democratic, the mission is not linked to an extensive reform of state administration and democratisation, but to a military force, the G5 Sahel, explored below. In the early 2000s, EUPOL Proxima, EUPOL DRC, and EUPM Bosnia, though having an important element of training and capacity-building and a focus on security and transnational crime issues, saw significant budget and personnel commitment towards broader reform goals linked to good governance. Whereas 'training', as an explicit goal, is only seen in 4% of missions at that time, it now defines 71% of them, being 38% civilian and 30% military. This new 'capacity' has to do with addressing threats to sovereignty, which require the build-up of the security and military apparatus of the state. The security and military 'capacitating' of states comes alongside the downgrading of good governance goals. Table 1 shows the shifts in the language of mandates. Whereas, in general, CSDP missions have the ultimate ethos of contributing to the stability of a particular region, in which some form of state reform would take place, it is remarkable how average references to 'good governance' and 'rule of law' have predominated until about 2008 but are lost thereafter. In contrast, references to training have doubled and those to security and stability have increased.
The displacement of good governance for military capacity and training suggests changes in discursive and material practices that have to do with the goals and means the EU deploys for peacebuilding. Speaking to delegates from France and Germany, they reiterate that good governance goals 'go hand in hand' and 'cannot be seen as detached' (e.g. Interviews 1,2,and 3). Indeed, as Palm and Crum (2019: 527) also note, 'valuebased considerations' are still relevant for justifying military missions, but 'utility-based considerations have become critical ' (2019: 517-518). Yet whereas for these authors this change produces 'no militarisation', the fact that military missions are now directed to military training reflects a certain type of militarism and not absence of it. Other instruments in the EU signal the same pathway.

IcSP and the capacity-building for security and development
As noted above, the IcSP renewed EU's actions towards crisis response and preparedness, conflict prevention, peacebuilding, and security threats (EU Parliament, 2014: art. 1). The comparison between programmes funded by the 2007-2013 instrument, the IfS, and those by the IcSP illustrates the transformations discussed above. Exploring the distribution of funded projects by theme, Graphs 3 and 4 show that, in the IcSP, dialogue, conflict prevention, and mediation come first, followed by security, counter-terrorism, and capacity-building. Graph 4 illustrates the same distribution once CBSD is consolidated. These programmes represent a shift away from institutional and governance concerns of the IfS, illustrated by Graph 5, which matches EU's statist approach of earlier years, and a turn to programmes focused on securing state authority, without the broader institutional reform package. The IcSP focuses on building security and military capacity and working on challenges to state authority. Speaking to managerial personnel of the most funded projects and exploring the projects in detail illustrate a trend to secure state authority through assistant to migrants, counter-terrorism, deradicalisation, and youth programmes often independently of states' human rights record. These programmes operate under a human security or security-development nexus perspective, which, as Stavrianakis (2018) and Abrahamsen (2019) argue, have been a driver for sustaining states' monopoly of the means of violence and its coercive capacities. The first and third most funded projects relate to the Strengthening [of] operational capacities of the Turkish Coast Guard in managing migration flows in the Mediterranean Sea (contracts 406305 and 374782 in 2018 and 2016, respectively), totalling €50 million. This is part of a larger €3 billion project aimed at 'stopping the flow of irregular migration' (EU Parliament, 2019). Not only has this deal been reached irrespective of Turkey's own human rights record in the treatment of migrants and refugees (Interview 4), it cannot be disconnected from the militarised border approach that the EU has taken to secure its borders ever since 'the migration crisis' in 2015 (Akkerman, 2018). Projects in Niger and Lebanon that build police and border guard skills going from psychological to removal skills have raised the same issues (Interviews 5 and 6). Niger has instrumentalised EU's funds to clamp down on dissent, while the EU has instrumentalised Niger to externalise EU's border (Davitti and Ursu, 2018: 3;Venturi, 2017).
The Central African Republic (CAR) is the IcSP's fourth most funded country. As a 'laboratory of EU strategies', it hosts all EU instruments including CSDP, IcSP, and fiduciary and development funds in what is conceived as 'truly comprehensive approach' (Interview 7). This approach has fostered the goal of good governance and democratisation more than, for instance, in the Sahel. Yet two EEAS officers coincide in noting that in the CAR 'governance was a driver' and 'we have now shifted to security' (Interviews 7 and 8). In a country where '60% of the territory is under armed groups control' (Interview 8), these funds aim to extend and secure state authority by simultaneously building hospitals and schools, and deploying a myriad justice and SSR programmes as well as military barracks, deradicalising youth and 'making threats to counterparts as light as possible' (Interviews 9 and 10).
Appeasing social relations in conflict areas, reducing security threats to the EU and neighbours, especially in regard to migrants, 'radical youth', and counter-terrorism, is a Graph 5. IfS' short-term funded projects by theme (Natorski, 2011: 20). main priority of IcSP projects. In Syria and Jordan, projects facilitate peaceful coexistence not just among communities, but also between communities and security forces so that security forces can operate in a 'non-hostile environment' (Interview 11). In Cameroon (e.g. contract 360817), a partner country in the struggle against Boko Haram, projects have been initiated to 'appease relations between government and citizenry' despite Paul Biya being the longest serving leader in Africa. A significant amount of mediation and conflict prevention projects work to uphold the value of the military in society and to enhance civil-military relations. In this context, the CBSD initiative consolidates well-entrenched practices.

CBSD
In the spirit of the security-development nexus, CBSD brings into perspective the need to support the military capabilities of states, albeit the non-lethal ones. The CBSD initiative supports capacity-building programmes in third countries aimed at training and mentoring, the provision of non-lethal equipment and assistance with infrastructure improvements, and help with strengthening the capacity of military actors. (EPRS, 2017: 1) It is controversial because such support is not allowed by the Treaty of the EU and can potentially take away development money towards more strictly security-related cooperation (Interviews 12 and 13; see also plenary debate in Committee on Foreign Affairs, 2017). The final green light was given a budget of €100 million over the period of (EU Commission, 2016. Although this figure represents a mere 4.27% of the total budget, it is substantial if put in comparison with other themes. For instance, 'Women, peace and security and gender mainstreaming' spent about €9.5 million for the entire period 2014-2019. In addition, it represents a new outlook in peacebuilding policy where peace rests increasingly on state military capabilities. As the initial CBSD proposal made by the Commission states, '[f]or any country to ensure its security and development, it must have or acquire adequate capacities in all critical sectors, including security and defence' (EU Commission, 2015: 3). The experience of an IcSP-related official shows that this initiative denotes a transformation: I can see a tendency to securitise the instrument. We are pushed on that a lot. It started with the training and equipment, after 3-4 years we are increasingly pushed to work hand by hand with the CSDP missions, mostly civilian, but also military, and to see that the IcSP can be complementary to the CSDP missions. We have done that in Mali, Niger, Somalia, and in a certain way in the DRC with the SSR activities complementary to EUSEC. There is a tendency to push the IcSP because there are no other financial instruments. (Interview 14) CBSD programmes generally include, on the one hand, military training and equipment for maintaining law and order (e.g. Somalia, Burkina Faso, Mali, Nigeria). 5 On the other hand, as previously mentioned, they promote the social value of the military amid the civilian population while supporting societal resilience in armed conflict. 6 Security and military EU officers tend to see no issues in using these instruments for these purposes in places where there is a conflict or fragile situation and where the army already plays a civilian police role (Interviews 15 and 16). In the CAR, a recent project supports the deployment of military forces as a 'temporary equivalent' for the deployment of state authority (Interview 16). The result however, as experienced in the DRC, where both the government and donors have fostered the deployment of the military as semblance of state authority, is a gradual militarisation of civilian life, where the army supports the civilian population in a range of issues and becomes a source of conflict for its coercive capacities (Verweijen, 2015: 126 and 183).
While these activities are conceived as working towards a peaceful environment, and they do so in some contexts, how the boosting of military and security actors affect power relations in the societies is taken uncritically. As in the case of the Sahel, discussed below, a militaristic approach to conflict has reproduced multiple sources of violence.

Exceptionalism in the Sahel
Actions in the Sahel region have been tied to the Sahel strategy, which requires strengthening security to enable development (EEAS, 2011: 1). For that, military training is provided through CSDP missions and the G5 Sahel. The strategy is perceived as contributing to a future of stability and rule of law, which requires a previous military victory of the states in the region. The problem is that these military operations have become attached to human rights violations (Interview 17). The EU has engaged in an 'action-reaction' strategy that is reproducing violence (Interview 17). In addition, many countries in the region have declared the state of emergency, have boosted their defence budget, and have reinforced their authority, not by democratic means but often by clamping down on dissent.
This is not an EU-imposed strategy, but one that has been led by Sahelian states themselves. The G5 force is an initiative of the heads of five Sahelian states involved in fighting armed groups throughout their territories (EU Commission, 2018). The EU quickly endorsed the initiative by supporting it economically, inscribing its operations within a human rights and humanitarian law framework and facilitating international donors' coordination (EU Commission, 2018: 3).
However, minimal reporting has been done so far. Interestingly, in the spirit of 'local ownership' and a security-first approach, the Head of Office of the High Commissioner for Human Rights (OHCHR) in New York 'acknowledged that donors feel more comfortable if human rights violations are not committed, but he warned the compliance framework "must not be seen as a donor-driven exercise, and we are very keen to make sure that it isn't"' (IPI, 2019: para. 14). While the demands of human rights still frame the actions of donors, they mark the tail-end of the actions rather than the driving force. Moreover, the actions of the G5 and those of individual countries in the Sahel, which are part of the broader strategy, have been notable for human rights abuses.
Since 2014, military budget has tripled in Mali and doubled in Burkina Faso, Chad, Niger, and Mauritania (SIPRI, 2018: 15). Burkina Faso has also maintained the state of emergency since December 2018 in the Northern provinces (Reuters, 2019). Although these groups have attacked security forces and have killed dozens, Human Rights Watch (2019) have reported at least 130 extra-judicial killings by security forces between 2017 and 2019. In Mali, human rights violations by security forces have been increasing since 2014 in the course of attacks, interrogations, and persecution of terrorists as part of counter-terrorist operations (Human Rights Watch, 2019; ICC, 2019). Mali has also been notorious for the prevalence of sexual-related violence (Guterres, 2019: 22). The UN Secretary General points out that although sexual violence by armed groups and Malian armed forces has been systematically reported since 2014, no investigation has been carried out. Counter-terrorist activities have also often been used to repress dissent (Human Rights Watch, 2018; Nozawa and Lefas, 2018: 2 and 9).
The Sahel is unique but not an exception to EU peacebuilding. The Sahel illustrates some of the consequences of the practice of placing military capacity front and centre whether through the CSDP, IcSP, or CBSD. While this is accompanied by projects in multiple aspects of development, they are the necessary extension of a central policy that places sovereignty and means of coercion as a requirement for development, without the impetus for institutional capacity and good governance of earlier years. The turn to exceptionalist militarism lies in the reinforcement of 'authoritarian and non-democratic power structures supported by better trained and better supported security forces' (Abrahamsen, 2019: 551). However contradictory and problematic, exceptionalist militarism is defining EU peacebuilding and is likely to continue doing so in the upcoming years.

Conclusion
The transformation of EU peacebuilding has multiple dimensions, but a central one revolves around its militarism. As shown in this article, militarism has shaped and transformed EU peacebuilding throughout its evolution. Yet despite military capacity-building and the downplaying of good governance defining the last shift, militarism has been absent from its analyses. Scholarship accounting for this transformation has noted a turn to pragmatism, resilience, and local ownership, identifying a step back from the most invasive aspects of liberal peacebuilding, which has ended up being contradictory and reproducing power relations.
By analysing militarism in EU peacebuilding, this article has contributed to these analyses by identifying continuities and changes in the interaction between discursive and material practices in war-related production, war-preparation, and war-waging, which are central to peacebuilding. In so doing, it has shown how both pragmatism and bottom-up approaches can be a conveyor belt for, and not a turn away from, militarism (Moe, 2018). Working through several types of militarism, the article has examined the continuities of liberal militarism in the EU, identifying the relevance of liberal moral values to justify different peacebuilding strategies and of the commitment to military production across peace and wartime as a modernisation project. This has been central to the granting of military training and equipment to support third states' use of force. In addition, following Mabee and Vucetic (2018), this article has conceptualised the latest changes as denoting a transformation from nation-statist to exceptionalist militarism. It was argued that whereas the practice of liberal peacebuilding already implied giving priority to security and stability, the current militaristic approach denotes the suspension of the package attached to good governance, prioritising states' military capacity. Building militarycapable states has gone from being a compromise and a means to an end, to the actual goal. But in so doing, it has contributed to locate exceptionalism not just in the context of the War on Terror but also in the security-development nexus, where not only military actors and political elites but also humanitarian and development actors have been the engines of a new militarism (Abrahamsen, 2019).
The article provided an examination of EU's most significant peacebuilding tools, including CSDP missions, IcSP programmes, the CBSD initiative, and the case of the Sahel, using thematic analysis of relevant documents as well as interviews. It illustrated how military capacity-building has become a central priority for the EU; it has identified the continued relevance of liberal values, but a loss of priorities regarding good governance and the rule of law as a front banner; it has also shown how EU and local goals can match on strategies that are not necessarily conducive to peace. This analysis has problematised how military support is provided without questioning its impact on recipient societies' balance of forces, nor the engagement in activities that put down dissent and different challenges to state authority. The Sahel case shows how these policies reproduce violence and a context where human rights can be suspended for the purpose of upholding state authority.
EU peacebuilding's new khaki may be conceived as a consequence of the policies and approaches long used in security and peacebuilding policies. But if for peacebuilding's post-Cold War agenda security and development cooperation entailed the promotion of liberal values as a way to redeploy a particular kind of state, by military means if necessary, the contemporary context throws a more inward-looking approach, where the EU, acknowledging its limited capacities, boosts states' military capacity. This is not therefore a transition from a non-militaristic to a militaristic approach, but a transition to a different variety of militarism, which also indicates new and important developments in the understanding and practices of security and peace in the EU and beyond.