EASO—Support Office or Asylum Authority? Boundary Disputes in the European Field of Asylum Administration

With the establishment of the European Asylum Support Office (EASO) in 2010, a new collective actor entered the European asylum policy arena. Although the agency commands limited financial and personnel resources, and has no formal powers to directly interfere in the decision-making practices of asylum authorities across Europe, its mandate and reach of influence are a subject of recurrent dispute. We consider the struggles surrounding the EASO’s role and position as manifestations of enduring conflicts of recognition, valuation and distribution in a Europeanised asylum administrative field. By analysing civil servants’ position-takings vis-à-vis the EASO, we demonstrate that officials from a variety of member states are united by a shared belief in disinterested, apolitical bureaucratic rules of procedure, based on a notion of ‘expertise’ that transcends national boundaries and supersedes national concerns. At the same time, the discursive boundary work invested by the interviewees draws on spatial, temporal and procedural categories of differentiation, highlighting complex processes of ongoing relational positionings and practices of distinction. The corresponding hierarchies and inequalities are further indications of a transnational administrative field with its own principles of valuation.


Introduction
With the establishment of the European Asylum Support Office (EASO) in 2010, a new collective actor entered the European asylum policy arena. The EASO is charged with coordinating, supporting and integrating the activities of national and supranational bodies involved in implementing the European asylum acquis. Like other agencies in the EU's Area of Freedom, Security and Justice (AFSJ), it commands only limited financial and personnel resources, and has no formal powers of its own to interfere in asylum decision-making. Nevertheless, on the level of discourse, its mandate is a recurring subject of dispute, and the agency has by now become directly or indirectly involved, in a number of formal and informal ways, with asylum procedural tasks that entail administrative discretion (Tsourdi 2016a). Following Bourdieu's concept of social fields (Bourdieu 1998;Bourdieu and Wacquant 2006), we consider the struggles surrounding the EASO's role as manifestations of enduring conflicts of recognition, valuation and distribution in the asylum administrative field. On the one hand, these conflicts revolve around the distribution of asylum seekers, administrative personnel and financial resources-between as well as within EU member states. On the other hand, negotiations and struggles take place over the question of by whom, and according to which valuation standards, the sovereign right should be exercised to control territorial borders and to grant and organise the access of 'third-country nationals' to the European space of social positions. The latter struggle, revolving as it does around the monopoly of the legitimate use of physical and symbolic violence, in other words state power (Bourdieu 1994, p. 3), is central to the present paper.
By focusing on national civil servants' position-takings vis-à-vis the EASO, we inquire into the conflicts and possible changes regarding claims to authority within a transnational field of asylum administration. In the shadow of political debates around the EASO's formal mandate and the future of the Common European Asylum System (CEAS), bureaucratic actors are engaged in complex relational positionings which are indicative of the dynamics of power and counter-power surrounding the institutionalisation of a European asylum administrative field.
The paper is structured as follows: After a description of the EASO's position in the CEAS and a summary of the discursive struggles that accompanied its establishment, we elaborate on our conceptualisation of boundary disputes within a Europeanised asylum administrative field. In the empirical section of the article, we briefly describe our data and methods before presenting our findings concerning civil servants' position-takings regarding the EASO's role and mandate. Two central insights emerge from the analysis: a) interviewees' position-takings varied according to their home authorities' position within the CEAS and may be differentiated along an axis of more or less insistence on national sovereignty when it comes to discretionary powers. Nevertheless, b) they shared a conception of autonomous bureaucratic practice, not bound to the nation-state but rather horizontally distributed through relational networks. In both respects, they were engaged in ongoing processes of boundary work (Gieryn 1983). In discursive practices of distinction, interviewees differentiated between groups of actors in the field and their respective relations to the EASO. The assignment of positions and the proposed boundaries of authority were legitimated with reference to spatial, temporal and procedural classifications, and were related to unequal power-relations within the field. In conclusion, our analysis shows that the struggles and conflicts within a Europeanised field of asylum administration cannot be fully understood by focusing only on national interests and intergovernmental negotiations. Rather, the interactions organised through and around the EASO are indicative both of internal classificatory struggles within the field and of bureaucratic immunisation strategies above and beyond a formal transfer of sovereign powers to the supranational level.

Betwixt and between-the EASO as a new actor in the European field of asylum administration
Asylum as a policy area has become increasingly institutionalised and judicialised at the European level over the last two decades. Contrary to initial expectations of a lowering of standards and a race to the bottom (e. g., Guiraudon 2000;Lavenex 2001), refugee rights have-at least on paper-been strengthened rather than restricted by the Europeanisation of policy formulation and legislation (Kaunert and Léonard 2012;Thielemann 2012, pp. 29-32;Kaunert 2010;Zaun 2016Zaun , 2017. While the set of directives and regulations agreed upon during the first phase of the CEAS (completed in 2005) defined minimum standards for handling claims for international protection, the recast process of the second phase of the CEAS (completed in 2013) aimed to define common, as opposed to minimum, standards, albeit without substantially changing the core of the legislative principles agreed upon during the first phase (Ripoll Servent and Trauner 2014). These developments have been accompanied by an increasing structuration and formalisation of administrative cooperation and inter-agency exchange (Costello 2006;Tomei 2000). While asylum authorities initially worked together in more or less informal ways and mainly to solve particular problems and to increase the efficiency of asylum procedures (Engelmann 2014), they are now acting within a common working context and subject to similar constraints (Lahusen 2016). However, neither judicialisation nor cooperation has resolved the central conflicts in the asylum administrative field. Restrictive border controls impede access to asylum procedures and national asylum systems still differ substantially in their standards, caseloads and outcomes (ECRE 2015;Parusel 2015;Toshkov and de Haan 2013;Trauner 2016). At the same time, the CEAS itself is far from coherent and is still "halfway between a bric-a-brac and a true system" (Chetail 2016, p. 586).
The considerations around installing a European agency to assist member states in implementing European asylum legislation need to be considered in the light of this conflictual and yet deeply integrated character of the asylum administrative field. The idea of establishing a 'support office' had already been brought forward in the Den Hague Programme of 2004, but it took until 2011 for the EASO to become fully operational (Comte 2010). From the beginning of the negotiations there were disputes between member states concerning the EASO's role and function (Angenendt and Parkes 2010). 'Traditional asylum countries' or 'strong regulators' in the north and west of Europe (particularly Germany, the United Kingdom, France and the Netherlands) considered diverging asylum standards as important pull factors leading to an unequal distribution of asylum seekers across Europe. They used their regulatory expertise to shape the construction of the overall framework of the CEAS (Zaun 2016(Zaun , 2017 and demanded an agency that would push forward the harmonisation of administrative standards in countries that were considered noncompliant with EU asylum rules (Angenendt and Parkes 2010, p. 1). Conversely, member states in the south and east of Europe that had just begun to establish their national asylum systems, such as Italy, Greece, Malta and Cyprus, argued that diverging standards could offer a sort of 'compensatory mechanism' for the heavier burdens carried by countries located at the external borders (Angenendt and Parkes 2010, p. 2).
The EASO regulation, as it was finally adopted in May 2010, is an expression of these conflicts. It rules out discretionary powers by stating that the EASO "should have no direct or indirect powers in relation to the taking of decisions by Member States' asylum authorities on individual applications for international protection" (Regulation (EU) No 439/2010, p. 12). Member states' concern to safeguard national K sovereignty is also reflected in their dominance on the agency's management board and the limited resources initially provided to the EASO. In 2011, it started out with a budget of 8 million euros and 42 staff; by the end of 2016 it had a budget of 69 million euros and approximately 175 staff. Compared to Frontex, however, with its 238 million euro budget and 417 staff, the EASO's resources are still rather limited (Trauner 2016, p. 317-318). 1 At the outset, the agency's activities consisted mostly of 'soft' modes of regulation. Like other European agencies in the area of Justice and Home Affairs, the EASO mainly functions as a producer of 'expert knowledge' in a context characterised by epistemic uncertainty and high political salience (Boswell 2008;Parkin 2012). It has become an import service provider in the areas of information and analysis, training and quality standards, and operational support. By assuming an active role in taking over and extending existing networks and creating new ones, it has successfully established itself as a sort of 'interface' between national asylum authorities. At the same time, the agency has until recently attracted comparatively little public attention. It was mainly concerned with how the member states, and specifically the heads and high-ranking officials of the national asylum authorities, viewed its role. As Robert Visser, Executive Director of the EASO until November 2015, put it It was not our focus to make sure every citizen knows EASO. You have to explain what you're doing, since you're doing it with public money, but the priority on communication is with the member states. [...] I've heard people say "We expected more of [the agency]", and I understand this, but there is nothing we could do about that. [...] We have interpreted our mandate to the fullest. This is a step-by-step process in building trust (cited in Cerulus 2015).
Since 2015, however, there have been notable changes in the focus and reach of the agency's activities, particularly in the volume and nature of operational support, with EASO-coordinated deployments of 'asylum experts' at particular locations in Greece and Italy (Tsourdi 2016a). While the first and second phases of the CEAS were driven by the assumption that legislative harmonisation would lead to an eventual harmonisation of administrative practices, the measures taken since 2015 mark a shift from the sharing of norms to "crisis-induced people-sharing arrangements" (Tsourdi 2016b). In the framework of relocation and hotspots, EASO experts are now carrying out tasks involving administrative discretion, such as interviewing and evidence-assessment. Formally, it is still the national authorities that take the decisions, but given the importance of the substantive interview in assessing the merits of a case, it seems "reasonable to assume that their role could amount to rubberstamping a decision whose merits were decided by the deployed experts" (Tsourdi 2016a(Tsourdi , p. 1024. Simultaneously, the role of the EASO in political discourse has changed remarkably. In a report drafted by the LIBE Committee, the EASO was envisaged as the new "principal coordinator of the CEAS", which "needs to develop from a collection of experts from Member States into a fully-fledged Union Agency providing opera-1 See EASO (2016, pp. 21-22) and http://frontex.europa.eu/pressroom/faq/about-the-agency/. tional support to Member States and at the external borders." (European Parliament 2016, p. 22). The European Commission even went one step further and considered the possibility of transferring responsibility for the processing of asylum claims from the national to the EU level, for instance by transforming EASO into an EU-level first-instance decision-making Agency, with national branches in each Member State, and establishing an EU-level appeal structure (European Commission 2016a, p. 8).
The ensuing proposal, in May 2016, for a European Union Agency for Asylum (EUAA)-which was still being discussed at the time of writing 2 -was far less ambitious. Nevertheless, it provided for a considerable increase in the agency's staff and financial resources and envisaged competences beyond the EASO's current mandate. These would be, inter alia, to steer and monitor, on its own initiative, the asylum administrative practices of a particular member state, and to intervene and address identified 'shortcomings' through the deployment of asylum experts if the member state concerned does not comply with the agency's recommendations (European Commission 2016b, p. 28). In the context of the legitimacy problems of the CEAS, the agency had thus become a focal point of calls for 'more Europe' (Guild 2016, p. 588).

Boundary disputes in the asylum administrative field
The political debates around the EASO's mandate can be conceived as "constitutional struggles" (Adler-Nissen 2014, p. 662) over who is to exercise sovereignty. In line with the subsidiarity principle, the CEAS entails a Europeanisation of legislative and judicative functions but operates on the assumption that executive functions, and administrative discretion in the taking of individual asylum decisions in particular, should remain with the competent authorities of the member states. As the section above has shown, recent developments exhibit signs of more direct forms of supranational administration even though the EASO's formal mandate has as yet remained unchanged. This points to gradual and rather informal changes in sovereignty which, following Adler-Nissen, we understand "as a concrete claim to authority [...] constituted by a set of historical practices, influenced by tacit understandings of legitimate action in particular social contexts" (Adler-Nissen 2012, p. 179). The ways in which the EASO is practically engaged in asylum decision-making on the ground show that the boundaries of authority may be differentially defined, maintained, defended or shifted. Borders between states-as objectifications of the reach of the state's sovereign authority-and the division of responsibilities between the supranational and the national levels are not the only boundaries at stake here. Disputes may also 2 In June 2017, a preliminary agreement was reached between the Maltese Presidency of the Council and the European Parliament (European Council 6/29/2017) which is subject to endorsement by Coreper. An overall agreement depends, however, on resolving the relevant linkages with the other legislative proposals on the third revision of the CEAS (see http://www.consilium.europa.eu/en/press/press-releases/2017/06/ 29-eu-agency-for-asylum/, accessed 04/10/2017).
K involve the relations between politics, bureaucracy, the law and different types of legal, technocratic or scientific expertise. What constitutes legitimate action is fought out not only in official statements and policy recommendations, but also in concrete local settings, involving diverse groups of actors and organisations.
We propose to conceive of these kinds of struggles as ongoing processes of boundary work (Gieryn 1983). In general terms, boundary work refers to practices of establishing, maintaining or challenging distinctions between groups of actors and between what counts as legitimate practice and what does not. Boundaries establish entities (Lamont and Molnár 2002;Abbott 1995) and hierarchies by controlling entries and access to material and symbolic resources (Bourdieu 1996, pp. 223-227;Bourdieu and Wacquant 2006, pp. 130-131). They have a history in the sense that they build on-and possibly redefine-existing distinctions and may acquire a more or less enduring spatial quality when they become attached to a particular territory, state borders being a case in point. The concept of boundary work draws our attention to the constant and collective reproductive effort that is needed to maintain even state borders and jurisdictions. Initially developed in the fields of sociology of science (Gieryn 1983), the professions (Abbott 1988), and science and technology studies (Star and Griesemer 1989;Star 2010), it puts a particular focus on jurisdictional authority, legitimacy and symbolic power as relationally created.
In order to understand why, how and on what basis agents engage in boundary work, Pierre Bourdieu's conception of field, capital and habitus is helpful. Bourdieu views social space as being differentiated into relatively autonomous, societal fields with their own specific stakes and interests. A field is essentially structured by a hierarchy of domination, by the stakes or the object of struggle peculiar to the field, and by doxa, that is, the taken-for-granted knowledge of the rules of the field (Bourdieu and Wacquant 2006, pp. 126-130). Agents' position-takings (as a particular form of practice) are a result of both their habitus as a system of relatively durable dispositions and their position in structured social space, defined by the distribution and composition of different forms of capital (ibid., p. 136). Neither field nor habitus are static. Rather, the ways in which agents perceive, categorise and interpret the world around them, are formed and continuously adjusted in accordance with the requirements of the field(s) in which they are located. Conversely, a field's boundaries and the principles of its internal stratification are contentious and constantly modified in struggles around status, positions and prestige.
Since the reach of the EASO's influence is meant to end where administrative discretion begins, boundary work within the bureaucratic field deserves particular attention. In his discussion of the state, Pierre Bourdieu ascribes to the bureaucratic field a special relation to the field of power. He defines the state as a kind of metapower in that it has successfully claimed the monopoly of legitimate physical and symbolic violence and reaches across fields by defining norms that affect all other fields (Bourdieu 2014, p. 638). It is the site of sedimented historical struggles and ongoing societal conflicts. The power of the bureaucratic field essentially lies in the 'magic' of naming, certifying and accrediting, of imposing "principles of vision and division" (Bourdieu 1994, p. 15). Thereby, the bureaucratic field not only produces and accumulates knowledge instrumental to its own functioning but provides the categories through which other actors perceive and understand the world. Its practices are legitimated with reference to a universal or common good-something which exerts a deep and lasting influence on those who do the naming, too. In situations of transnational cooperation, civil servants encounter others with quite different "minds of state" (Bourdieu 1994, p. 12) which poses a potential challenge to established definitions of the common good. While Bourdieu's conception of societal fields was initially developed with reference to fields dominated by the French interventionist state, any field-theoretical construction of the object of inquiry demands an initial openness regarding a field's boundaries and reach. In this endeavour, entities like 'nation', 'state' or 'Europe' should not pre-supposed but rather relationally reconstructed and understood as field-effects in their own right (Gengnagel 2014, p. 291;Vauchez 2011, p. 340). In a similar vein, the agents involved cannot be treated as representatives of a particular nation-state or the EU from the outset. Rather, the actions and statements of bureaucratic actors must be conceptualised as positiontakings within a 'space of possibles' that contribute to the continuous production and reproduction of state sovereignty as a particular action problem (Adler-Nissen 2012. This is of particular relevance in the case of the asylum administrative field, since it occupies a special position at the interstices both of different societal fields and of differentially defined social spaces, due to the global reach of both migration phenomena and the legal norms of an international refugee regime and its enduring and close connection with issues of national sovereignty. Over the last three decades, the European Union has witnessed a proliferation of decentralised agencies, essentially created to cope with complex transborder challenges and growing interdependence by integrating national authorities and providing structures for networks of coordination and information exchange between supranational and national bodies (Chiti 2004;Hofmann 2008;Coen and Thatcher 2008;Dehousse 1997;Eberlein and Newman 2008;Majone 1997). In recent years, agencies have also gained importance in the contentious and sensitive area of Justice and Home Affairs (Kaunert et al. 2013(Kaunert et al. , 2015Parkin 2012;Carrera et al. 2013). While agencies differ in form and purpose, they all occupy an ambiguous position, in that it remains unclear which 'public interest' they are actually designed to serve (Everson and Joerges 2006). In many ways, they seem to function to insulate 'technical' issues of regulation from political conflicts of interests and contentious policies (ibid.). Research into agencies has often focused on their formal mandates, competences and institutional design (Thatcher 2011) and on patterns of autonomy, accountability and control in the interactions between agencies and their principal(s) (Egeberg et al. 2015;Busuioc 2009), particularly concerning policy-autonomy (Bach and Ruffing 2013;Bach et al. 2015). In assessing agency development over time, several authors have pointed out the need to study changes not only in their de jure but also their de facto autonomy (Busuioc et al. 2011;Carrera et al. 2013). This includes informal policy innovations on the ground that are formalised only at a later stage, which raises a number of accountability issues. Since European agencies answer not to a single but to multiple 'principals' with heterogeneous interests (Dehousse 2008), they are under a particular pressure to demonstrate their credibility and impartiality. Hence, their legitimacy depends not only on their actual output but above all on their professional reputation (Majone 1997;Carpenter and Krause 2012;Busuioc 2016), that is, on their recognition by other relevant actors in the field. The willingness of member state officials to cooperate is central for an agency's reputation and it is on their loyalties that the implementation of bureaucratic rules ultimately rests. However, in interactions with European agencies, it is also the reputation of these officials' home authorities that is at stake (Busuioc 2016). Attempts to encroach upon core state powers are a particularly sensitive issue in this regard and may raise bureaucratic concerns with 'turf' (ibid.). The ongoing negotiations and contentions around the division of responsibilities between European agencies and national bodies after delegation are therefore of particular relevance for an analysis of the relational distribution of recognition and possible changes in sovereign claims.

Data and Methods
This article is based on extensive field-work conducted in the context of a research project on the Europeanisation of asylum administrative practices. We started out with a focus on the EASO Training Curriculum (ETC) and analysed both the conceptual development of selected training modules and their practical implementation. Particular attention was devoted to an analysis of existing administrative practices in Sweden and Germany (Schneider and Wottrich 2017;Schittenhelm 2015;Schittenhelm and Schneider 2017). More recently, we have begun to focus more broadly on the development and gradual consolidation of cross-national, inter-agency cooperation, for example in the framework of the Dublin procedure. As part of our research, we conducted semi-structured interviews with experts involved in the development of the ETC, with senior officials of national asylum authorities, the European Commission and the EASO itself, and with representatives from international organisations (IOs) and EU-level non-governmental organisations (NGOs) which play an active part in monitoring and further developing the EASO's activities. For the purposes of the analysis presented below, we concentrate on six interviews with senior officials from different national asylum authorities. 3 The choice of this subsample was based on temporal and occupational criteria: the interviews were conducted between November 2015 and February 2017, when the discourse on the extension of the EASO's mandate was gathering momentum. Most of the interviewees possessed lengthy experience in the field, having worked in the area of asylum administration for 15 years or more. At the time of interviewing, these senior officials were all involved in issue-specific networks coordinated by the EASO. They were thus functioning as a 'hinge' between the EASO and their home authority and were deeply involved in the discourse and negotiations on the EASO itself, its current activities and its future role within the CEAS.
We analysed the interview data using the documentary method (Bohnsack 2010(Bohnsack , 2014, a qualitative data analysis procedure with a particular stance of observation and analysis inspired by Karl Mannheim's sociology of knowledge (Mannheim 2003(Mannheim [1980). Generally, the aim of the documentary method is to reconstruct the implicit, 'tacit knowledge' that underlies everyday practice and thus forms a 'framework of orientation' for habitualised actions. However, it has also been used to analyse the special kinds of knowledge that are relevant in organisational settings (Liebig 2007) and highly specialised occupational fields such as the asylum administration (Schittenhelm 2015; Schittenhelm and Schneider 2017). Here, we adapt the documentary method to analyse the reference frames and the types of knowledge that transnationally active bureaucrats draw on during the social-scientific interview. Due to its explicit differentiation between 'what' is said and 'how' it is said, this method is particularly well suited to an analysis of the implicit references and distinctions underlying more explicit position-takings. By focusing on the taken-for-granted, seemingly self-evident classifications that civil servants use when demarcating legitimate practices and distinguishing themselves symbolically from other groups of actors, we aim to generate insights into the dynamics of power and counter-power regarding claims to authority in the asylum bureaucratic field. Approaching the subject from such a vantage-point means leaving the initial analysis open as to the question of which reference-frames dominate the agents' position-takings. In line with our field-theoretical approach, we treat interviewees' statements as manifestations of the thinkable and the unthinkable, the sayable and the unsayable. Accordingly, we are not interested in the specificities of each case, but aim at a reconstruction of the different forms of boundary work at play in interviewees' position-takings, both within a single case and across cases. Thereby, we seek to arrive at an exploratory analysis of the space of (im)possibles structuring the statements of a very particular group of actors: national civil servants who were involved in a transnational discourse on the EASO and its future role.

Support Office or Asylum Authority?
Our analyses indicate that the EASO was of growing relevance for the structuration of the interviewees' occupational activities. There was a shared understanding that the EASO had become an important actor in the asylum administrative field. Perceptions of the EASO's role within the CEAS ranged from framing it as a pool of practical tools and neutral experts, with no power or intention to influence politics, to hoping it might take over the role of a new 'super-' or 'mega-'authority. While some interviewees would only selectively pick and choose from what the EASO had to offer and drew quite strict boundaries as to where, and under what conditions, it should be 'allowed in', others would like it to take over responsibility for the asylum procedural process as a whole and seemed to have little problem with giving up national discretionary powers. The positions taken during the interviews may be distinguished into three broad categories: 1) the EASO should take over as a supranational authority with decision-making powers; 2) the EASO should fulfil supervisory functions with the power to reprimand; 3) the EASO should continue to assist in the practical cooperation among member states and to support those under particular pressure. In the following paragraphs, we first describe these three position-takings in more detail and then present findings regarding their underlying commonalities. For all three positions, it is important to note that the senior officials in our sample did not speak purely as representatives of national interests and that their positions cannot simply be equated with the political preferences of their national governments or ministries. Rather, the interviews showed that the EU level and the interviewees' involvement in the EASO as a "networked organisation" (Levi-Faur 2011) could also serve as a venue to gain autonomy from their principal at home (Bach and Ruffing 2013).

EASO as a supranational authority with discretionary powers
The first position-taking we would like to describe is the one most closely associated with a de-coupling of claims to authority from conceptions of national sovereignty. It is characterised by a remarkable indifference towards the notion that discretionary powers should remain in the hands of the nation-state. This position was mainly taken by officials from authorities of member states which had joined the European Union when the regulative framework of the CEAS was already relatively advanced. The support for a transfer of responsibilities for asylum decision-making to the EASO complemented the wish to reduce a perceived uncertainty and ambiguity in the handling of asylum through increased and centralised guidance, in short, to have "straight instructions" (Senior Official F, 1513). Informal agreements on the proper interpretation of central regulatory components of the CEAS were criticised as non-binding, leaving too much room for member states to shape inter-agency cooperation according to particularistic interests. The Dublin system, for example, was portrayed as an unpredictable game in which member states only pursued one objective: "who will win and transfer more asylum seekers" (Senior Official F, 129). The installation of a supranational administrative authority was supposed to remedy diverging implementation practices caused by too much political interference (Senior Official F, 123-139 and 264-313) my vote is to make EASO as a umbrella agency to control everything [...] not support agency but agency who will supervise all asylum systems in EU so it means no none authority or political authority could influence on asylum system except EU (Senior Official F, 1487-1506 In the future scenario imagined here, national asylum authorities would be responsible for 'pure' implementation; the present work situation is characterised as being 'contaminated' by politics and divergent interests originating from the national level. This position-taking went hand in hand with a perception of the CEAS as something that pre-existed and to which one needed to adapt not only organisational structures, administrative routines and procedures but also ways of thinking and categories of valuation. In this context, EU-level meetings seemed to constitute an important arena both for learning about pre-existing ways of thinking and for negotiating relational field positionings. The interviewees represented themselves and their home authorities as 'newcomers' to a field that was pre-structured by both explicit and tacit understandings of legitimate and appropriate action, which other agents had successfully imposed during past struggles and which they now had to 'buy into'. The crucial point here is that the existing regulative framework (e. g., the Dublin system) in itself was not perceived as institutionalising asymmetric power relations, privileging some and disadvantaging others (for example, concerning the distribution of caseloads). Rather, a picture of more dynamic and constantly changing positions was drawn.
with Dublin III we started getting involved on first contact committees but it was all a new I didn't maybe six months after I didn't recognise I realised how why some member states address some stress some problems because the situation in [country 6] was quite different it means some articles in Dublin regulation we never apply because we are not a destination country [...] time after time you must start to think not only [...] from the point of view of [country 6] but because maybe some situation that is already happening in Sweden maybe tomorrow will happen in [country 6] it was challenging for me actually (Senior Official F, Here, thinking in 'containers' of nation-states still forms an important point of orientation in defining one's position. At the same time, relational aspects, interdependencies and shared problem definitions are emphasised. According to this interviewee's account, the classification of member states as either 'transit' or 'destination country' is no longer self-evident, since a situation that another asylum authority has to cope with on a given day might become acute for one's own authority on the following day. Civil servants from other member states are portrayed as 'partners', sharing an interest in coping with the irritations and challenges accompanying the (common) administration of asylum.
In this context, the EASO was conceived by interviewees as an actor with as yet limited material resources, but with the necessary symbolic capital to take over responsibility for the 'management' of the CEAS in the long run: "It's the most relevant agency so they can get control on everything [...] in my opinion they need more staff [...] in order to cover everything and to manage the situation because with this staff at this level they cannot do it in a proper way" (Senior Official B, 310-317). Above all, the anticipated shift in the relational position of the EASO was expected to reduce uncertainties and ambiguities. Responsibility for the future failures and successes of the CEAS was attributed entirely to the supranational level: "they have to be responsible for it in my opinion together with the Commission of course" (Senior Official B, 583-584).
This position-taking was thus characterised by a belief in bureaucratic practice as the neutral execution of tasks agreed upon elsewhere. The relation between politics and bureaucracy was conceived of as strictly hierarchical. The perceived problem consisted of the proliferation of too many diverging 'public interests'. Importantly, and in contrast with narrations dominating political discourse, the incumbents, i. e., the more established national asylum authorities, were the ones to be criticised and taken to task for their disregard of European rules.

EASO as a supervisor with powers to reprimand
Another position taken during the interviews was that the EASO should play a bigger role in monitoring and supervising the implementation of European asylum procedural rules and that its 'operational activities' should be enhanced. In this scenario, K responsibilities for ex-post oversight of administrative practices would be shifted to the agency, although a federally organised European asylum agency with discretionary powers was still outside the realm of the thinkable. The interviewees who were inclined to take this position worked for more established national authorities in north and central Europe. Here, the problem of the CEAS was not perceived as consisting of too much political interference and the incompatibility of member states' interests-as in position 1-but as an administrative problem involving implementation failures, inadequate staffing and a lack of resources. This was based on the assumption that the common administration of asylum requires permanent monitoring in order to detect 'problems' in advance and to ensure that national asylum systems are prepared and well-equipped to 'manage the situation'.
we would strongly object [...] the agency becoming a kind of federal asylum authority for the whole of Europe [...] but a strong monitoring role for the asylum agency is something that we are very much in favour of cause the asylum crisis one of the main reasons for the asylum crisis is the lack of implementation and therewith we really think that the new European asylum agency can have a strong added value both in detecting problems at an earlier stage but also in addressing and remedying these problems quicker than we can now (Senior Official C, 967-981) The CEAS was framed as something which was gradually taking shape and of which the interviewees had been a part from the very beginning of its establishment. The field of asylum administration was likened to an evolving organism that was gradually refining its organisational structures, standards, and operating procedures (Senior Official C, 21-24, Senior Officials G_I/G_II, 420-423). The establishment of the EASO was perceived as having prompted new dynamics in the further integration and development of transnational administrative procedures: it was finally "speeding up" (Senior Official C, 139) cooperation and the realisation of ideas that bureaucratic experts had invested much time and energy into developing, drafting and negotiating at the EU level.
it always seems like nothing's moving you keep on discussing commas and details etc. but if you take one step back and then see where we stood in 2005 and where we stand now it's a completely different world EASO is a very clear example of the way in which member states cooperate has changed completely since EASO was developed in 2010 the idea of hotspots the idea of relocation that was something that was absolutely impossible in 2014 and is reality in 2015 (Senior Official C,(108)(109)(110)(111)(112)(113)(114)(115)(116)(117)(118)(119)(120)(121) This second position-taking most clearly distinguished between member states' differential positioning according to their (perceived) adherence to the 'rules of the game'. The latter were perceived as a 'common good', not dominated or shaped by particular national interests but the outcome of the formation and formalisation of expert knowledge. Categorisations according to levels of experience and 'expertise' were relevant for drawing the boundary as to where the EASO's influence should end and for how support and monitoring would be legitimated. The EASO was characterised as a pool into which expertise, generated by the more experienced asylum authorities, was 'uploaded' and made available for the common, European 'good' by using this combined expertise of the most experienced member states we also help the less experienced member states because we can together develop training develop COI develop quality instruments which can be used by 28 member states so the whole standard at EU level is being improved slowly but steadily (Senior Official C,(164)(165)(166)(167)(168)(169)(170) The agency was welcomed as an actor that could coordinate the generation and spread of specialised knowledge, work that formerly had to be carried out by officials from a group of five to six 'experienced member states' (Senior Officials G_I/G_II, 231-250, 602-621). The individual experts' positions were not substituted or threatened by the EASO, but rather allowed to unfold their full potential in cooperation with it. While many activities had formerly taken place "outside the EASO" (Senior Official C, 129), they now continued their work "inside EASO" (Senior Official C, 130), importing and further elaborating the knowledge they had previously generated in more informal networks (e. g. in bilateral information exchange or joint-processing projects). The crucial point here is that this was conceived as generalised and shared knowledge. It was portrayed as being universally applicable for identifying and solving problems across a Europeanised asylum administrative field. This position-taking supported an agency that would systematically observe and evaluate the implementation of the asylum acquis, while a strict line was drawn concerning any potential 'takeover' of the actual 'core' of the asylum procedure: casework and decision-making. This demarcation also became obvious in the importance that interviewees attached to their own institutional anchoring within their home authorities. By emphasising their role as specialised officials in departments for "EU and international affairs", responsible for issues like EU legislation, funding for common pilot projects and "practical cooperation" (Senior Official C, 31-49; Senior Officials G_I/G_II, 15-32), they portrayed themselves as gatekeepers, as "the entrance" to their respective home authorities (Senior Official G_I/G_II, 473-485), negotiating the potential reach of the EASO's influence. What is at stake in the CEAS was defined as a separate and specialised area of professional activity, which also provided new opportunities for individual careers. Regarding actual asylum casework practices, however, rival claims to authority from the European level were strongly opposed and, contrary to position 1, divergent political interests did not seem to be of any importance. Consequently, this position-taking argued for an agency that would evaluate administrative practice using the 'disinterested expertise' generated by 'experienced officials' as legitimate evaluation standards. Ultimately, it aimed at tighter regulation of 'weaker' or 'newer' authorities and maintenance of the status quo in one's own authority. 4

EASO as a support office
The third position-taking aimed at maintaining the status quo by continuing to ascribe merely supportive and coordinating functions to the EASO. This position was taken by staff from both well-established or 'traditional' asylum authorities in the 'north' and less-established authorities in the 'south' of Europe, but from different vantage points. Similarly to the second position, the EASO was conceived as a pool of expertise and a box of practical tools that might be accessed and used as needs arise. In contrast to the second position, however, the definition and identification of situations in which support would be necessary was perceived to be a matter of local or national responsibility. While being similar in their general opposition to 'federalist' intrusions into 'the national way things are run', the position-takings belonging to this category involved different kinds of boundary work, depending on whether interviewees construed their authorities more as recipients or more as providers of "help" (Senior Official G_II, 705-713). In principle, EASO missions were conceived of as deployments of impartial, specialised officials to particular locations in the EU (e. g. the Dublin unit in Athens or Rome), which functioned as temporary job sites where particular staff shortages or other needs could be handled. Where the EASO was already involved in the processing of asylum applications, however, it became obvious that the experts found themselves in a 'grey zone', at sites where their role and position was ambiguous and subject to constant negotiations. From the perspective of 'providers of help', "those issues are quite delicate because you're always stepping on someone's toes which you don't want to do because you want to help but it's important that you actually get to use your expertise" (Senior Official G_II, 1043-1047). Interviewees were aware that their presence might be perceived as undermining or interfering in another official's domain, but the provided expertise in itself was described as something neutral that should be applied, not only in the interest of the 'beneficiaries of help' but also in one's own interest, in order to "get to use your expertise" (Senior Official G_II, 1046-1047).
From the perspective of recipients of this kind of support, the presence of EASO experts had been perceived as an "intrusion" (Senior Official H, 533) in the beginning, but had come to be welcomed as a solution for dealing with staff shortages (Senior Official H, 531-554). Thus, 'factual constraints' were used as rationalisations for accepting the presence of EASO experts in the first place. At the same time, regarding their concrete involvement in casework, various forms of symbolic boundary work came into play. The interviewee cited below was at pains to emphasise how her authority, as a 'recipient of help', tried to ensure that sovereignty was not being encroached upon.
we have the supervision we always have a supervision on all the operations [...] and so we have taken care of the fact that they have a clearance specific clearance for processing very personal data (Senior Official H,(445)(446)(447)(448)(449)(450)(451) the call for experts made by EASO is based on our specific requests our specific needs so it depends on what we ask for and yeah EASO the request from EASO our are tailor-made on our needs (Senior Official H,(874)(875)(876)(877)(878)(879) For this interviewee, it was of paramount importance that civil servants from other member states came as EASO-experts and were responsible for well-defined, "specific operations" (Senior Official H,(426)(427)(428)(429)(430)(431)(432). As long as this was ensured, they and their expertise might be used for and inserted into her own authority's work routines. Their support was appreciated as a contribution to the more efficient handling of cases. However, sovereign authority (conceived in terms of the nationstate) had to be maintained on the level of symbols and rituals-it is not for the EASO to determine when and where to second experts, the member state requesting support must define the conditions under which EASO experts might be employed and provide the relevant clearances.
This position-taking was characterised by the absence of a clear picture of what the future of the CEAS might look like and by a distinction between situations or periods in which the CEAS operates in 'default mode', when national authorities have the authority to exercise administrative discretion, and those in 'emergency mode', when the EASO might temporarily take over. In periods of 'emergency', the EASO was meant to function as a trouble-shooter, enabling national authorities to eventually return to 'default mode' again, while a constant monitoring of decisionmaking practices and possible reprimand by a European authority were clearly rejected. With its emphasis on 'states of exception', this position-taking questioned the universal suitability and adequacy of the expertise generated at EU level. Standards and best practices agreed upon at the European level were subjected to reevaluation and negotiation in concrete encounters; accordingly, hierarchical relations between individuals according to expertise and professional experience and a situational claiming or reclaiming of authority were of greater importance here.

Shared beliefs in bureaucratic expertise and discursive practices of distinction
The analysis presented above has shown how national civil servants are engaged in discursive boundary work on a number of different levels. The interviewees position themselves and their home authority in relation to other authorities and draw on spatial (e. g., local, national, regional, European polity levels and/or geographical spaces), temporal (e. g., long-term vs. exceptional situations, established authorities vs. newcomers), and procedural (e. g., cooperation in Dublin procedures vs. asylum decision-making proper) classifications for distinguishing between groups and for differentially defining and legitimating the proposed reach of the EASO's influence. While interviewees' loyalties to a particular member state or national authority and its position within the CEAS go some way towards explaining their respective position-takings, the common frames of reference underlying their reasoning can only be explained by their being subject to the effects of a specific, transnational bureaucratic field that is more than the sum of the EU's member states. Boundary work based on temporal and procedural criteria provides an example of how claims to authority may become partially de-coupled from geographical or jurisdictional spaces and re-articulated on the basis of other, transnational hierarchies. Rather than constituting just another instance of intergovernmental power-struggles, intervie-wees' position-takings are manifestations of relational struggles within a common, internally differentiated space of positions. Across cases, the formulation of law and policy was distinguished from their seemingly 'practical', 'apolitical' implementation. The work invested into maintaining this symbolic boundary contributes to immunising the bureaucratic field from heteronomous influences. At the same time, it is what is at stake in the field itself. What precisely counts as 'purely' bureaucratic is a matter of contention in actors' struggles around power and positions, but no matter which of the three positions was taken, it was legitimated with reference to bureaucratic expertise as inherently disinterested, universal, and oriented towards a common good. Hence, while the analysis clearly revealed different position-takings that went hand in hand with divergent positions within the administrative field, "both orthodox and heterodox positions share a taken-for-granted 'doxa'" (Swidler and Arditi 1994, p. 317). but that is a high level cooperation in-between the persons responsible for the policies [...] the council for example is organising the high level meeting in the field of migration and asylum policy and they are discussing their priorities and so on [...] but this is another level of we are just experts we are involved in in the action (Senior Official B,(882)(883)(884)(885)(886)(887)(888)(889)(890) With regard to this involvement "in action", interviewees attached particular importance to channelling and structuring communication via networks and working groups (Senior Official G, 395-413) under the leadership of the EASO. "Practical" cooperation via the EASO was depicted as something detached from politics (Senior Official B, 310-317); it involved the construction of a group of colleagues and of a shared action problem which was mainly based on a perception of (forced) migration as inherently unforeseeable and fraught with uncertainties and ambiguities. "Pressure" and "shortcomings" were construed as emanating from the outside and potentially affecting any member state or any "colleague" from another asylum authority (Senior Official H, 1207-1220. Against this background, the EASO was conceived as a venue for cooperation among 'equal partners' and field-internal power-relations according to states were de-emphasised. "International cooperation" itself, however, was conceived as something that could and should be accomplished by specialists and specialist departments. The interviewees distinguished themselves from other specialised officials, as well as from caseworkers responsible for processing asylum applications on the ground: "People know that there is international cooperation but not everybody's gonna know all the operations for EASO GDISC IGC ICMPD we are nerds sitting around this table" (Senior Official G_II,(847)(848)(849)(850)(851). 'Knowing about the EASO' as a specialised form of knowledge and field of activity was something they shared with their colleagues from partner authorities. EASO networks thus have the potential to integrate and stabilise a transnational field of 'Euro-professionals', even if these professionals are still deeply anchored in national contexts of action (Büttner and Mau 2014). Across cases, the EASO was perceived as an actor whose influence was gradually spreading across the different departments and hierarchical levels within the respective asylum authorities in indirect ways because of its focus on "practical issues" (Senior Official G_I, 380-391). Being part of the EASO networks opened up new career paths and opportunities to specialise and to improve one's position within the home authority. In this sense, interviewees were not distinguishing so much between countries but rather between the formulation of standards by experts and their implementation by specialists with no power to define what is considered as legitimate practice.
These tendencies notwithstanding, classifications according to member state are still of relevance for internal differentiations within the field. Interviewees conceived of length of involvement and experience as important in structuring the opportunities to engage in other forms of cooperation.
we still have bilateral cooperation as well and we do value these bilateral or limited multi-lateral cooperations cause from time to time it's easier more flexible just to call our Dutch colleague and say well can we work together on this and then you set up a meeting we travel there back and forth cause working within the EASO structure means you always have to work for all EU member states and that can sometimes be slowing the process (Senior Official C, 219-227) These forms of cooperation outside of the EASO point to the continuing importance of more informal ties between particular groups of countries. Interviewees differentiated between groups of colleagues from "like-minded" countries "facing the same challenges" (Senior Official G_II, 279-281) and others, who were perceived as 'less experienced' or less interested in stepping up cooperation. Both processes-the formation of shared understandings and the integration of a field of EU-professionals through the EASO and through cooperation outside it-have consequences for the EASO's current and future field position and the ways in which conflicts are negotiated in and around the agency. While some rely on it as the sole or main form of cooperation, others can draw on additional networks as a form of social capital accumulated over time.

Conclusion
The empirical analysis presented above indicates that transnationally active bureaucrats use conceptual distinctions and classifications which not only reveal different position-takings in discursive disputes about divergent problem definitions but also facilitate a common production of knowledge, separating what counts as 'bureaucratic' from the 'non-bureaucratic' across state-defined jurisdictions. At the same time, the analysis highlights internal differentiations and new transnational hierarchies structuring power-relations within the field.
Although the position-takings vis-à-vis the EASO might vary, the 'space of possibles' is limited by the logic of the bureaucratic field, inducing agents to present both their own and the EASO's role as disinterested and detached from day-to-day politics. With reference to the 'expertise' it allegedly supplies, political, moral, or legal arguments are disarmed and subsumed under the logic of the neutral, efficient execution of tasks. In this way, the challenges and ambiguities of asylum administrative practices (Lahusen and Schneider 2017), which arguably render the field K of asylum administration particularly vulnerable to heteronomous influences, are downplayed, even silenced. Expert knowledge not only forms an important resource in the EASO's strategies to strengthen its own position but can be used by other actors to substantiate and legitimise policy preferences (Boswell 2008;Parkin 2012). From the perspective of national civil servants, cooperation with or via the EASO provides important symbolic resources for navigating the grey zone between politics and 'pure' administration (Egeberg and Trondal 2017, p. 677) in their own particular work contexts. As an effect of the resulting information asymmetry, participation in EASO networks might empower national agencies vis-à-vis their respective parent ministries (Egeberg and Trondal 2017, p. 680;Bach et al. 2015) and also contribute to shielding them from public scrutiny. This may be illustrated by the fact that a central concern of activists and NGOs is peculiarly absent from each of the three position-takings, namely the question of whether or not a central European asylum authority would have a strong and clear 'protection mandate' Guild 2016, p. 600) and when, how and to whom it should be held accountable in the case of unlawful actions.
Furthermore, the EASO's reach of influence becomes an object of struggle in agents' attempts to improve their position within the asylum administrative field in the narrow sense. With increasing centralisation and formalisation under the umbrella of the EASO, an integration of unequal partners takes place that has considerable effects on the overall structure of the field. Previous studies have shown that newer member states have an interest in the formalisation of transnational cooperation-above all, because they cannot contribute the expertise that is exchanged in older and more informal networks on the basis of trust and reciprocity (Engelmann 2014;Zaun 2017). Newer member states will arguably be more dependent on the EASO and also more willing to accept an extended mandate to improve their own position, while more established authorities will continue using informal structures of cooperation alongside the more formal ones. In this regard, the latter will always be one move ahead. Insofar as this inner circle of strong regulators is able to invent and test new modes of cooperation, they are also in a position to assess in advance the potential consequences that new modes of cooperation and regulation might entail (Zaun 2016). Hence, they are more likely to be successful in further immunising their own domain and sphere of administrative practice from heteronomous influences-of the EASO as well as from the field of national or European politics. However, this does not imply that less-established authorities would have no capacity to challenge field-specific ordering principles. Since the CEAS consists of rules that are not of their own making, an insistence on adherence to the rules provides an important resource in strategies of 'counter-power'.
The growth and formalisation of EASO-coordinated networks, as well as the formal and informal ways in which the agency is already involved in administrative practice on the ground, are developing dynamics of their own. As we have demonstrated above, unequal power relations on the level of states are crisscrossed by networks of self-acclaimed experts involved in the distribution of recognition regarding the EASO's role. The fact that EASO experts are primarily engaged in operational tasks in weak regulating states such as Greece and Italy is evidence of the power of strong regulators to determine the rules of the game. However, the lack of a clear legal basis for such operations and the corresponding issues of accountability and legal remedy can have a boomerang effect, in the sense that they force the agents involved to reformulate the legitimating principle behind such 'intrusions'. In this regard, national sovereignty concerning policy implementation has lost its taken-for-grantedness and expertise seems to have taken its place. To the extent to which such expertise, as a specific form of cultural capital, is acquired in transnational networks, this affects the structural relations between positions and has consequences for subjective position-takings. Thus, where and how national civil servants work on maintaining or shifting boundaries depends not only on their belonging to a particular national authority but also on the specific types of capital they have accumulated in the course of transnational cooperation. The symbolic struggles surrounding the EASO's position therefore have tangible consequences for the dynamics of power and counter-power within and across the boundaries of the administrative field. In this sense, the boundaries themselves constitute an embattled object of interest (Bourdieu and Wacquant 2006, p. 135).