Doing Europe: agency and the European Union in the field of counter-piracy practice

ABSTRACT The practice turn provides new avenues for core questions of international relations and European Studies. This article draws on a practice theoretical account to shed new light on the constitution of agency in global politics. An understanding of agency as achievement that requires significant practical work and the participation in international fields of practice is developed. Drawing on the case of the field of counter-piracy practice and the European Union’s (EU’s) work to counter piracy off the coast of Somalia, it is shown how the EU achieved the position as a core actor in the field. A detailed discussion of the EU’s work in interrupting and knowing piracy, in building capacity, and in governing piracy is provided.

Contributing to this agenda of studying Europe through practice theoretical lenses, my objective is to elaborate on a core claim of practice theory, namely that agency does not exist prior or outside of practice. Agency is a property of practice. This implies to understand agency as a relational achievement. To understand how actorhood is achieved, we need to investigate the efforts of participating in practices. In consequence, the core question of this article is how does the EU participate in international fields of practice and how is thereby a specific form of agency produced or done?
Since theorising practice in the absence of actual practices does not make much sense, I draw on the case of a distinct field of practice. My case is the field of counter-piracy practice. I study the efforts of the EU in the fight against Somali piracy starting from 2008. Counter-piracy offers a particularly exciting case. It has become an interesting playing field for international actors including the EU. 1 As a field of practice, counter-piracy is characterised by experimentation: international actors have introduced and tried out a range of new means and ways of cooperating with each other. It is a field of international activity that is relatively novel and characterised by high complexity. It was only from 2008 when piracy attacks off the coast of Somalia started to escalate that it became an issue that international political actors paid attention to, and in consequence started to initiate activities to fight piracy. Given this novelty, no existing international institution could make a legitimate claim to have authority over the issue; given the complexity of the issue, no agency was able to deal with it on its own. When the field of counter-piracy evolved, roles and responsibilities were hence undefined, and no routines of addressing piracy were settled. With these characteristics, and in contrast to long-established and institutionalised fields, counter-piracy provided an issue area in which roles, authority, and responsibilities had to be negotiated. This, in turn, provided a particular space for the participation of emergent international actors, such as the EU. Moreover, the field of counter-piracy was from the beginning one of strong convergence. Counter-piracy actors agree on a common problem definition in legal, territorial, and political terms. In consequence, the field is less driven by rivalries, clashing interests, or competition. For European states, there was a strong convergence of national agendaswhether these are general concerns over the safety of trade (e.g. Germany, France, and the UK), the shipping industry (Denmark, Greece, and the UK), the fishing industry (France and Spain), or over the stability of the Eastern African region and Somalia in more general terms (France, Italy, and the UK).
As I demonstrate in the following, throughout the evolution of the counter-piracy field, the EU increasingly performed a leadership role and moved from the periphery to the centre. Gradually the EU became a lead actor in all of the major counter-piracy practices. With the launch of EUNAVFOR Atalanta in 2009, the EU became one of the "big three" military providers and vital in maritime situational awareness. Through programmes, such as the MARSIC, or EUCAP Nestor projects, the EU became one of the core agencies in capacity building. When the European External Action Service (EEAS) assumed in 2014 the chairmanship of the Contact Group on Piracy off the Coast of Somalia (CGPCS)the central site of governance in the counter-piracy fieldthe EU also became a core diplomatic and governance actor in the field.
In recasting the journey of the EU from the periphery to the centre, my objective is hence both theoreticalto further investigate how practice theory can shed light on fundamental questions of IR theoryand empiricalto describe and understand the EU's role in the fight against Somali piracy. The discussion draws on data and observations gathered in a wider praxiography of counter-piracy governance. 2 The primary interest of this praxiography was in describing and understanding the practices by which the international community responds to Somali piracy.
I start in offering a brief discussion of my assumptions and concepts. This clarification is important since my understanding of what follows from centring theory in practice differs from other variants of the family of practice theories. I start out from a strong relationalist position on practices. In the spotlight of my discussion are however the actual practices, that is how the EU participated in counter-piracy practice. I discuss Europe's contribution to the four practices that constitute the field: (1) interruption practices, (2) epistemic practices, (3) capacity-building practices, and (4) governing practices. In the conclusion, I intend to scale up and come back to the implications of my empirical discussion for European Studies and the debate on international agency.

Practice
Practice theories have become a strong contender in the repertoire of international relations theory. They promise new insights by starting out investigations of global politics from the study of practices (Adler and Pouliot 2011, Bueger and Gadinger 2014, Adler-Nissen 2016. The primacy of practice implies that explanations for social and political order are not sought in the intersubjective coordination of (interest-driven or norm-following) individuals, but in shared understandings, that is, culture. Culture, in turn, is not understood as based in human minds or in discursive structures, but in practical activities (Reckwitz 2002). The majority of practice theories conceive of practices as "embodied, materially mediated arrays of human activity centrally organized around shared practical understandings" (Schatzki 2001, p. 2). A widespread conceptualisation of practices sees them as composed of relations between three core elements: bodily movements, material artefacts, and shared practical understandings (Reckwitz 2002, Bueger 2014a. Much philosophical ink has been spilled on how to define these elements and their relations. Various conceptual apparatuses have been worked outof which perhaps Bourdieusan praxeology, Wenger's communities of practice approach, and Actor-Network Theory are the most well known (see Bicchi 2016, Bremberg 2016, Graeger 2016, for examples drawing on the community of practice approach). This has led to a quite heterogeneous family of practice theories (Bueger and Gadinger 2014). Here I follow an understanding of practice theory which mainly appreciates it as an empirical project of describing and deciphering practices. As I have argued elsewhere, such a project can be best described as praxiography (Bueger 2014a), that is, as an account that conceives of "theory" as a set of sensitising concepts useful for empirical analysis. This understanding follows many of the ideas as they are discussed under the label of Actor-Network Theory elaborated by, among others, Bruno Latour, John Law, or Annemarie Mol (see Bueger and Gadinger 2014, pp. 43-50).
The move from practice theory's concern about identifying general and systematic patterns to praxiography implies that rather than searching for general abstractions, "we have to attend carefully to practices and ask how they work" (Law 2012, p. 157). In consequence, if we want to understand how Europe is done on an international stage and how international agency is produced, we need to peer to specific sites and their practices.
Investigating the sites of counter-piracy and their practices does not lead us to general conclusions of how Europe is performed in world politics, but provides insights into how such practices work. Such practices are highly context dependent, and hence might be fundamentally different if one investigates other sites.
Saying that we need to proceed empirically does not imply that we can survive without a parsimonious set of basic assumptions and sensitising concepts. I here follow the lead of John Law that "practices are detectable and somewhat ordered sets of material-semiotic relations" (Law 2012, p. 157). As Law (2012, p. 157) describes it, to study practices is […] to undertake the analytical and empirical task of exploring possible patterns of relations, and how it is that these get assembled in particular locations.
[…] Realities, objects, subjects, materials, and meanings, whatever form they take these are all explored as an effect of the relations that are assembling and doing them. Practices, then, are assemblages of relations.
Practices form together a distinct activity system and are geared at achieving a specific effect. They centre on the production of an "object of activity". Objects of activity might relate to a certain state or object, such as an epistemic object (Knorr Cetina 2001), or a "matter of concern" (Latour 2004), that is, a distinct problematic issue or situation. A useful research strategy for praxiography is hence to explore practice by identifying an object of activity that these practices produce or aim at coping with. The practices I discuss in the following have a common object: they are directed at stopping piracy off the coast of Somalia. These practices shall be described by the term "counter-piracy practice".
Practices do not have a definite scale. While they are patterns and relations, they do not exist in a natural scale. In consequence, we can zoom in and zoom out as it makes analytical sense. With the term "field of practice", I like to describe a bundle of practices, that is, a distinct practical configuration consisting of several practices which together achieve effects. I take the global field of counter-piracy practice to be such a configuration. As has been argued elsewhere, 3 and as further elaborated below, the field consists of a set of four practices which each address the object of Somali piracy. Yet, each of the four practices takes a different direction and purpose: Interruption practices are concerned about stopping piracy activities. They are composed of various patterns of activity and include the surveillance of the operational terrain of pirates, patrolling these areas, interrupting piracy attempts, arresting pirates, collecting evidence, and prosecuting them. Epistemic practices are concerned with knowing piracy. This includes the maintenance of the collection and circulation of data on piracy, studying the organisation of piracy, and making threat assessments and trends analyses. Capacity-building practices are concerned with building the capacity in the piracy-infested region in order to stop and prevent piracy in the future. Various activities can be clustered around this notion, including law drafting, mentoring, training courses, or exercises. Governing practices involve scripting, orchestrating, and aligning practices so they pursue a common interest. Activities such as coordination meetings, negotiating the text of a communique, or drafting a report belong to this cluster. These four practices should not be seen in isolation from each other. It is their relations that form together the field of counter-piracy practice. They form functional parts; none of them could be reduced to the other; and there is no clear hierarchy between them.
Within the practice theoretical ontology, actors are an effect of relations. Actors are made in and through these relations and in the way they enact certain practices. The primary status of actors is as participants in practices; they enact practices; they might simply reproduce them, or reinvent them. In consequence, we need to start out with getting a sense of what the practices are that drive a distinct field before we go to the question of how actors contribute or play in it. As Passoth et al. (2012, p. 5) remark, such a position is based on, "the twist not to begin with a definitive answer concerning what agency is, but to look at the diverse situations from where different forms of agency emerge".
In consequence, to gather insights about EU actorness on the global stage, the practice theoretical perspective poses the following research question: How does the EU participate in international fields of practice, such as the field of counter-piracy practice?
The field of counter-piracy practice Fields of practice are constituted by patterns of practices geared at an object of activity. The field of counter-piracy practice started to substantially form in 2008 when (Somali) piracy was recognised as a problem requiring international action. The UN Security Council held its first debate on piracy in that year and UN-sanctioned naval action was initiated. In earlier years, piracy was an issue debated frequently within the Maritime Safety Committee of the International Maritime Organization (IMO). This led to a range of recommendations, in particular for shipping companies, but not to immediate actions. In 2008 the IMO decided to send a request to the United Nations Security Council to deal with the matter. Piracy is an issue on the agenda of the council since; a series of resolutions authorise international action, and the UN Secretary-General reports on the issue biannually. To coordinate those governments affected by piracy or active in addressing it, in spring 2009 the CGPCS was established. 4 Initially only a limited forum, participation in it grew quickly to over 60 states, international organisations, and non-governmental organisations. The CGPCS meets biannually to discuss the state of counter-piracy and coordinate activities. Very quickly it became the institution at the centre of the emerging field since it provided the forum in which all actors engaged in counter-piracy would meet. Other core institutions of the field include the Shared Awareness and Deconfliction mechanism, known as SHADE, likewise initiated in 2009. SHADE is an informal forum in which the naval forces active in counter-piracy meet quarterly to inform each other about their plans and intentions and to organise joint operations, such as group transits or surveillance. This brief review of core sites and actors already gives an impression of the extensive relations that the field of counter-piracy practice established since its initial formation.
Observing what happens at these sites gives us clues about the set of practices that can be described as governing practices. Governing practices are activities that are concerned about the coordination of actions, maintaining their relations, and developing shared narratives that describe the problem, measures to its solutions, and the activities and roles of the diverse actors. "In such practices the relation and sharing of labour between organizations is negotiated and norms, rules and standards for the counter-piracy field are developed" (Bueger 2013a, p. 96). To a large degree, this set of practices is structured or informed by the routines of diplomatic practices (Pouliot and Cornut 2015) and legal practices (Latour 2010).
To understand the other practices of the counter-piracy field, we have to scrutinise the substantial issues discussed in the field's fora, and peer into the actual actions taken in the region of concern. This leads us, firstly, to actions that are concerned with stopping piracy activities. These can be clustered around the notion of interruption practices. These practices involve a combination of military, policing, and judicial activities. 5 The UN Security Council in the first instance mandated naval action and starting from 2009 all major international navies operate in the Horn of Africa and the wider Western Indian Ocean region with a counter-piracy mandate. Navies patrol the area, engage in surveillance, offer specific protection schemes for shipping vessels, and aim at interrupting piracy attempts and arresting suspects. Piracy suspects are then transferred to regional states, where they are prosecuted and potentially jailed.
The next set of practices comprises activities which are directed at improving the capacities of littoral states to prevent and handle piracy incidents. Capacity-building practices are concerned with building the capacity in the piracy-infested region in order to stop and prevent piracy in the future. These are informed by the repertoire of development assistance and state-building practices (Graef 2015). A 2009 assessment report by the CGPCS revealed the low capacities of littoral states to provide basic maritime security and prosecute and jail pirates. 6 In reaction, a substantial number of capacity-building projects were launched. These initially focussed on rule-of-law projects and ensuring human rights in prosecuting states, such as Kenya. They were increasingly broadened to also build maritime security capacities, such as coast guards or navies. Various activities can be clustered around this practice. It includes support for prisons, courts, maritime authorities, police units, coast guards, or navies by sending mentors, organising training courses and exercises, and providing offices and equipment.
The final set of practices that drive the field of counter-piracy practice is perhaps the least obvious, that is, epistemic practices. Any field of practice and governance activity draws on an episteme, that is, a certain set of knowledge, assumptions, and facts. Epistemic practices are concerned about making piracy known and creating the (epistemic) object, which is the object of activity (Bueger 2015). This implies not only to collect data about piracy behaviour and organisational structures, but also to hypothesise the origins and causes of piracy and what can be done against it. Practices that produce or distribute knowledge about piracy were instrumental to the field from its onset (Bueger 2015). In particular, incident reports and statistics documented the drastic increase in piracy activity in 2008. In later stages, once the counter-piracy measures showed success, piracy knowledge allowed to focus efforts and evaluate the success of operations.
It is important to recognise that these four sets of practices are streams of activities. They are not stable or fixed institutions, but processes. They change. Participants in the practices have throughout the evolution of the field adjusted and transformed them and included new elements. Understanding counter-piracy as a field of practice hence emphasises that we are not confronted with a stable and clearly expressed regime or institutional architecture, but with a moving and changing arrangement. If counter-piracy draws on and integrates formal institutions and processes, it is, in essence, a field that depends on, often informal, activities. This fragile and contingent character of the field is not the least observable through the recurrent changes in practices and the position of actors in enacting them. In the next step of the analysis, I shall document how the position of the EU has been changing throughout the life span of the field and how the EU achieved to become a central counter-piracy actor. I discuss the contribution of the EU to the enactment of the four central counter-piracy practices. I start with the EU's role in interruption as the domain in which it engaged first. I proceed in showing how it gradually became active in the other practices, that is, epistemic and capacity-building practices, as well as most recently in the domain of governing practices.
Interruption practices: one of the "big three" When the UN Security Council mandated states in 2008 to engage in military activity to contain Somali piracy, a range of European naval vessels was already employed in the region with a counterterrorism mandate. Several of them had already encountered pirates. With international plans crystallising to organise naval counter-piracy operations, the EU announced it could take over the protection of shipping in the area. The EU's first reaction in 2008 was to establish a naval coordination cell (EUNAVCO). Operating from Brussels, the cell was intended to coordinate the national naval activities of member states (EU Doc 2008/749/CFSP). Quickly these plans were changed and the launch of an integrated European and Security Defence mission was prepared. Hence, in November 2008 European Union Naval Force (EUNAVFOR) Somalia Operation Atalanta was established by a decision of the Council (EU Doc 2008/851/CFSP). The first naval European Security and Defense mission ever was declared to be operational in December of the same year. To get fully operational, however, took longer than expected and the EU was only ready to start in spring 2009. NATO had to step in with its mission Operation Allied Provider. NATO's operation was initially about to end once the EU was ready. It was, however, continued when it became clear that the EUNAVFOR in its own would be insufficient. Hence NATO's Operation Ocean Shield was launched.
The EU installed a naval headquarter and a situational awareness room in the UK's naval headquarters in Northwoodthe same compound which also hosts NATO's operational naval headquarters. A network of political liaison officersso called POLAD's, "political advisors"was established across the Western Indian Ocean region. These were tasked to handle local logistics and deal with the relations to the countries that were providing port access or prosecute pirates (including, e.g. Djibouti, Kenya, or the Seychelles). A logo for the mission was created which depicts Somalia as well as a naval vessel. The mission has a significant public affairs dimension. A dedicated website was created under the address eunvafor.eu, which provides information on the mission, such as the mandate or force structure, as well as pictures from the day-to-day activities. A dedicated Public Affairs Office started to publish a weekly newsletter on the mission and offered regular media support. Together with the mission, a Facebook page was launched, which started to show significant activities from 2013 when the Public Affairs Office started to provide daily updates and pictures on the mission.
As Germond and Smith (2009) have argued, Operation Atalanta represents a significant milestone in the European Security and Defence Policy (ESDP). This concerns at least three areas, firstly, the way that Atalanta relies on joint planning, a joint headquarters, and the pooling of resources, that goes far beyond the low-scale coordination of prior ESDP missions. Secondly, as Germond and Smith suggest, in contrast to other missions, Atalanta relied on a shared strategy and a European definition of threats and interests. "Atalanta thus clearly represents the EU's first ESDP operation that directly serves EU member states' interests (by protecting their maritime trade)" (Germond and Smith 2009, p. 576). "The EU decision to launch Atalanta was clearly motivated by common perceptions of not just the interests but also the grandeur of the EU and the affirmation of its values" (Germond and Smith 2009, p. 583). Thirdly, with Atalanta the EU acted for the first time as an independent international security actor. It was the first mission that "did not involve or require the use of NATO resources, including its planning capacities or operational headquarters" (Germond and Smith 2009, p. 576). As suggested by Gebhard and Smith (2015), the operation also allowed to meet NATO at the same height and to develop and explore new means of informal cooperation between the organisations.
The mission was scheduled to last for an initial period of 12 months (until December 2009). It was subsequently extended by the Council of the EU for three years (until December 2012), then for two further years (until December 2014), and most currently until December 2016, which is the projected end date of the mission. The mission's mandate is a rather peculiar one since it projects the operation as a primarily humanitarian one centred on the protection and guarantee of safe delivery of humanitarian aid to Somalia by the World Food Program. Framing the mission in such terms camouflaged the main purpose of it, that is, to protect merchant vessels and interrupt piracy activities (Bueger 2010). Yet, such a framing not only ensured national political support for the mission and parliamentary endorsement (Riddervold 2011, Peters et al. 2014. It also allowed the EU to present the mission as a full success from the onset, despite the growth in piracy activity between 2009 and 2011. The mandate has gone nonetheless through several substantial operational changes. In July 2010, the area of operation was extended from 500 nautical miles off the Somali coast to the entire Indian Ocean, reflecting the growing operational terrain of pirates. In May 2011, it was decided by the Political and Security Committee that navies should also take action against motherships. In March 2012, the operational terrain was further extended in that it now included the destruction of pirate infrastructure (ships, supplies, etc.) on Somali coastline.
Atalanta became soon known as one of the so-called big three. The big three became the colloquial term among counter-piracy practitioners to refer to the three main operations (NATO's Ocean Shield and the US-led Combined Maritime Task Force 151 being the other two). As a sort of triumvirate, the big three initiated a range of operational coordination measures. This included the SHADE mechanism in which all navies active in counter-piracy coordinate their operations. SHADE was initially chaired in a rotation between the big three. An International Recommended Transport Corridor (IRTC), a high protection area for shipping in which the deployment of naval assets is coordinated, was introduced. Part of Atalanta was also to install an information sharing and communication centre (the Maritime Security Centre Horn of Africa (MSCHoA) based in Bahrain), which was, as further discussed in the following section, designed to provide maritime situational awareness, and incident and activity alerts, as well as to handle the communication with the shipping industry. The EU also initiated an Internet-based communication platform, called MERCURY, which provides a real-time chat-enabled infrastructure for coordinating the activities between naval actors.
In the realm of interruption practices, the EU very quickly assumed the role of a leading participant. Following a slow start, it recognisably shaped how the practice is carried out by participating in the planning and implementation of measures such as the IRTC and SHADE, as well as by installing MERCURY. Indeed, the EU gained the status of one of the "big three" along with established security organisations such as NATO.

Epistemic practices: a fuzzy profile
Before the launch of counter-piracy operations from 2008, epistemic activities were primarily carried out in the frame of the reporting mechanisms of the IMO and the International Maritime Bureau of the International Chamber of Shipping, both of which continued to be major knowledge producers after 2008. With the rise of Somali piracy, the range of actors participating in epistemic practices, however, continuously expanded. Soon epistemic actors included advocacy organisations, such as the US non-governmental organisation Oceans Beyond Piracy, international organisations such as the World Bank, the UN Sanctions Monitoring Group for Somalia, as well as a growing group of academics. Also, the naval operations entered this field. They not only recorded piracy activity through a reporting system, but also implemented large-scale surveillance measures, for instance, through reconnaissance planes. The EU participated in the epistemic practices in at least three ways, by organising a reporting system, by conducting so-called strategic analyses for policy-makers, and by developing a counter-piracy strategy. Compared to the domain of interruption practices, the EU has, however, maintained a less visible presence, and its profile as an actor is fuzzier.
The EU set up and runs the MSCHoA in Bahrain. Since its inception in 2009, MSCHoA became the core backbone of the reporting of ship movements and the primary point of contact for the international shipping community to report an attack. The so-called Best Management Practices for the shipping community call upon shippers to register with MSCHoA when transiting through the piracy high-risk area. Although the main purposes of the organisation are in the distribution of knowledge, coordinating military responses, and the relation between mariners and navies, MSCHoA is also a primary data collector on piracy incidents. These data feed into strategic assessments of the big three. The information collected also constitutes the data on which the piracy statistics are based. These statistics are not only published by EUNAVFOR on its website, but also inform the so-called SHADE briefings. SHADE briefings are conducted at the CGPCS meetings. They are the main channel by which international policy actors attending the CGPCS meetings are informed about current developments. The SHADE briefings are carried out by representatives from the big three in rotation. Hence, through these briefings the EU established authority as a recognised knowledge producer about piracy. If these two forms of participation in epistemic practices (maintaining MSCHoA and conducting briefings) are on the level of producing knowledge about the object of piracy, the strategic document developed by the EU is an example of knowledge production in terms of a response. In 2011 the European Council published its so-called Strategic Framework for the Horn of Africa. As discussed by Erhart and Petretto (2014), the framework provided a major attempt to contextualise the fight against piracy in a larger set of policy measures towards rebuilding the Somali state. It provided one of the first documents which suggested that in order to tackle piracy, the larger context and the origins of piracy (the so-called root causes) require to be addressed.
In sum, the EU was able to become a core actor in the domain of epistemic practice. In comparison to interruption practices, this profile was, however, less well established. This has to be seen in the light of the observation that the overall number of actors in the domain of epistemic practices is significantly larger and that less attention was paid to promote the role of the EU in this domain.

Capacity building: muddling in a mess
Early counter-piracy activities concentrated on interruption practices. The overall emphasis in the field increasingly shifted to capacity building, when from 2012 piracy activities declined. Compared to interruption practices, capacity building is less well organised. A quite messy and uncoordinated range of activities is carried out mainly on a bilateral rather than a multilateral level. Various states and NGOs play in this practice. Two core actors, however, established a clear international profile from 2008: firstly, the IMO and its regional capacity-building project, the so-called Djibouti Code of Conduct (DCoC), and, secondly, the United Nations Office on Drugs and Crime (UNODC) Counter Piracy Programme (CPP) primarily working in the police, justice, and prison sectors. The EU was initially a largely invisible contributor to capacity building. It primarily acted as a funder of these projects without achieving significant visibility. This fundamentally changed with the launch of a new ESDP mission, EUCAP Nestor, in 2012. Since then, the EU has aimed at becoming the core actor in organising and maintaining capacity building.
In the early days of counter-piracy, capacity building was concentrated on enabling regional states to handle piracy suspects, organise trials, and ensure imprisonment following human rights standards. Initially, this work was primarily carried out by UNODC CPP. The EU funded a significant portion of the CPP, but did not act in its own right in this area. UNODC documents carried a small note saying "funded by the European Union". Apart from this barely recognisable note, the EU was hardly seen as a capacity builder. The other main project, the IMO's DCoC, entailed setting up a regional information sharing centre and running regular training courses and exercises with representatives of regional states. The EU supported DCoC with the so-called MARSIC project which was contracted out to a French consultancy. MARSIC is a project funded under the Instrument for Stability of the Development and Cooperation Directorate General and ran from August 2010 to July 2015 with a 6-million budget. Its core activities were to support DCoC by organising training and mentoring activities. MARSIC is certainly the least wellknown EU activity and the IMO was taking the main credentials for the DCoC. By 2010 the low profile of the EU in the domain of capacity building started to be recognised as problematic by EU staff. Plans arose not only to promote the EU's contribution in the area, 7 but also to set up an independent capacity-building mission.
With the creation of EUCAP Nestor, the EU explicitly aimed at creating an independent profile in the domain of capacity building. The civilian ESDP mission was launched in July 2012 with a mandate of delivering training and providing mentors in Djibouti, Kenya, Somalia, Seychelles, and Tanzania EU (Doc 2012/389/CFSP). The objective was to assist these states in "self-sustainable capacity for continued enhancement of their maritime security including counter-piracy, and maritime governance" (EU Doc. 2012/389/CFSP). When Nestor started its work, it found a rather difficult environment. For many practitioners and prospective recipients, it was unclear how the mission would deliver what was not already being done by the existing capacity-building projects. In consequence, Nestor had a rather rocky start. It struggled to identify its niche or to spend its budget, and one of the main recipients (Kenya) declined the offered support (House of Commons 2013, Bueger 2013b. With a new head of mission joining the operation, Nestor made significant efforts to step up its activities, but still faced a staffing problem and the difficult security situation in Somalia. In consequence, Nestor primarily was active in two countries: in Djibouti, where also its headquarters was located, and in the Seychelles. In both countries, an extensive training and mentoring programme was launched. Improving its public image and visibility became increasingly important for the mission. This implied to engage in outreach activities such as a Facebook page, a newsletter, or the organisation of a briefing for CGPCS participants during the 2013 CGPCS meeting in Djibouti. After a review of the mission conducted in 2014, it was decided in summer of that year to extend the mission for further 2 years. The renewed mandate implied that the mission now should focus its efforts on Somalia and also move the headquarters to Mogadishu. The increasing efforts of the EU, and EUCAP Nestor specifically, to participate in capacity building were also reflected in contributions to the governance of capacity building. As further discussed in the next section, the EU initiated a subgroup of the CGPCS, the Capacity Building Coordination Group (CBCG), to steer capacity-building efforts. In 2014 EUCAP Nestor took up the chair of this group. In the field of capacity-building practice, the EU was hence initially only a shadow player that funded and supported ongoing activities, while the glory was taken by others. With the launch of EUCAP Nestor, this however changed and the EU has made significant efforts to be recognised as a core player and implementer in this set of practices.
Governing practices: moving to the heart of the field Governing practices involve scripting, orchestrating, and aligning practices so they pursue a common interest. The main site of governing practices is the international contact group, the CGPCS, in which international actors meet to discuss and agree on a strategy. The main actors in governing practices and the CGPCS are clearly states. If other actors, such as representatives of international and regional organisations, companies, NGOs, or navies, are participants in the CGPCS, the group is run by state representatives, that is, national delegations composed of professional diplomats. Organised by diplomatic practices, chairmanship over the plenary of the CGPCS and its different working groups was hence in the hand of states. Other actors would only give presentations or updates (such as the SHADE briefing already discussed).
In the early days of counter-piracy, the EU did not have a significant voice or role in governing practices, other than through the presentations by EUNAVFOR as part of SHADE. It was not allowed to take a leading role, but participated as an observer. This status incrementally changed when the EEAS started to launch the EUCAP Nestor mission. Recognising that capacity-building practices would require more coordination and that EUCAP Nestor would be entering a crowded field, the EEAS proposed a new subgroup to one of the CGPCS working groups. When the CBCG was launched as a technical subgroup, it was a novum: for the first time, a body of the CGPCS was not chaired by states. Three international organisations, the EEAS, along with IMO and UNODC, took over the chair. After a series of meetings of the group, in 2014 EUCAP Nestor became the chair.
Having successfully proven its capacity to chair a subgroup, in 2013 the EEAS was approached to take over the chairmanship of the entire CGPCS. The EEAS was formally invited to chair by the plenary and it subsequently took over in January 2014. This was again a novum, since the CGPCS had only been chaired by states before. As a former chairperson phrased it in a discussion: "The EU is doing a lot, so why shouldn't they take the responsibility and chair?" In 2014 the EEAS organised the CGPCS meetings. The chairmanship implied that the EEAS presided over the actual meetings, designed the meeting agenda, and, most importantly, was leading on the negotiation and drafting of the main output of the group, that is, the communique. Recognising that chairing an international organisation, if only an informal one, was a significant first for the EU, the EEAS seized this opportunity in significant ways. The chairmanship was taken over by a Deputy Secretary General of the EEAS and supported by a significant team of staff. In January the EEAS organised an informal CGPCS strategy meeting hosted by the EU Institute for Security Studies. A major reform process of the the group was launched following the meeting of the CGPCS and used both to populate the agenda with significant numbers of presentations of EU missions and bodies. As the EEAS summarised its role: "the Chairmanship of the Contact Group allows the EU to assume international responsibility commensurate with its involvement in the Horn of Africa and Somalia in particular" (EEAS 2014). Having successfully chaired the CGPCS for a year, the EEAS started the search for a successor following the established routine of a rotating chairmanship. Since no other candidate stepped up, the EEAS agreed to continue its work and chaired the group for another year. A further plenary meeting was organised by the EEAS. In autumn 2015, the work was handed over to the government of the Seychelles. During its tenure as chairperson, the EEAS could clearly demonstrate its capacity as an international actor. The significant achievements of the EU, such as leading the reform process of the group, negotiating a solution to one of the main controversies in the group over the revision of core measures, as well as preparing for what was now called the "endgame of the fight against piracy", were widely recognised among counter-piracy practitioners.
In the field of governing practices hence the EU, represented by the EEAS, worked its way up from a marginal player in the early stages of the field, to the position of the core orchestrator of the field in 2014 and 2015. The fact that, contrary to established diplomatic practice, the EU was invited to chair the CGPCS demonstrates how successful the overall efforts of taking a leadership role and becoming recognised as a major counterpiracy actor were.
Doing Europe and the empirical study of "actor-work" The starting point for a praxiographic inquiry on agency is to understand the practices that maintain a distinct sociopolitical order. In this article, this order was the field of counterpiracy practice-a set of heterogeneous activities to address the international problem of (Somali) piracy. To be an actor implies to enact, to reproduce, to shape, or to reinvent the practices that maintain this order. Agency is not a property of individuals or entities, it is an achievement. It requires work. As I have shown, the EU has significantly and increasingly participated in maintaining this order. Through this work, the EU achieved the status as one of the pivotal counter-piracy actors. A diverse range of activities was required to achieve and maintain this status. This included sending out navies, managing headquarters, organising the epistemic flow of piracy knowledge, engaging in capacity building through sending staff and organising missions, staffing and maintaining offices, and organising meetings or mundane activities such as providing folders for the international community. And it included substantial efforts of presenting the EU as a counter-piracy actor through offices and various public diplomacy measures.
Practice theory and the toolbox of praxiography give us powerful ways of thinking and researching the question of agency differently. Much of international relations theory holds on to an ontology of the international that takes the agency of states for granted and projects them as the main actors. Questions of agency usually arise only when other (non-state) actors are at stake, such as the various international and regional organisations or the transnational communities, networks, or professions active in international relations. In international organisations research, for instance, there is a long line of investigations of whether these entities are actors in their own right (Barnett and Finnemore 2004). Decades of constructivist scholarship has aimed to show how other entities than states matter (Avant et al. 2010). Much of this scholarship moves into the right direction when it argues that they matter because they act on the stages of global politics. Yet, there is the constant risk to fall into the trap of defining agency in relation to states, that is, to understand someone as an (autonomous) actor if the exertion of influence (or an effect) on states can be proven. Understanding agency as acting is also the intent of a practice theoretical account. Yet, there is no reason to grant special status to the state. As has been shown in recent relational scholarship (Passoth and Rowland 2010), the state, likewise, is not a fixed entity, but an achievement. It is an entity that requires ongoing maintenance work. Also states are part of a struggle over agency on the international levels. The practice theoretical account hence starts out from parsimonious premises and turns the question of agency into an empirical puzzle as related to the maintenance of distinct sociopolitical orders. Practice theory hence, first of all, invites us to study the work that is required to become an actor. This implies to pay close attention to how practices unfold empirically and to study how they are enacted. For this, we require methods that bring us close to the actual activities, and this is the purpose of the praxiographic toolbox. As has been demonstrated in the discussion, "actor-work" should be seen in relation to particular fields of international practice. Instead of making absolutist claims about actorhood, determining who is an actor and who is not on the basis of theoretical criteria, we are invited to study the work of actorness in relation to particular domains of international activity. This provides us with a much more fine-grained and complex picture of who the actors and their identity in international relations are, and allows to appreciate divergence across fields. In short, the practice theoretical perspective invites us to think and study agency differently. If we are interested in how Europe achieves international agency, then we have to ask how it is done in specific international fields of practice. Notes 1. Compare the discussion in Bueger (2013aBueger ( , 2014b. 2. For the notion of praxiography, see Bueger (2014a). The term aims at grasping the actual study of practices, in contrast to the notion of "practice theories". As briefly elaborated below, it hence emphasises the importance of descriptive work and hence prioritises close observation of what actually goes on, over a theory of fictitious forces. The article primarily draws on the results of different research techniques including meeting observations, conversations with counter-piracy actors, as well as the analysis of recorded interviews, documents, and a range of other artefacts. 3. For a detailed discussion of each of these practices and on how far the global field is composed of them, see Bueger and Stockbruegger (2013) and Bueger (2013a). 4. For a discussion of the CGPCS, see Tardy (2014). 5. See the overview given in the chapters of Guilfoyle (2013). 6. See Houben (2014). 7. For instance, this issue was constantly flagged during conversations with EU practitioners in Nairobi in 2010.