The Never‐Ending Road Towards the CEAS: Utopia, Teleology, and Depoliticisation in EU Asylum Policies

This article explores the temporal dimension of the Common European Asylum System (CEAS) by exposing its teleological character and the effects of the latter on the governance of asylum in the European Union. Drawing on EU policy doc‐ uments, the article shows how the CEAS has been presented since its inception as a teleology, that is, a process that is inexorably unfolding towards a specific outcome to be reached in an indefinite time in the future. The outcome consists in the establishment of a common area of protection constituted by a level playing field in which asylum seekers and bene‐ ficiaries of international protection will be treated alike regardless of the place of residence. Such a teleological narrative informing the CEAS paves the way to overly optimistic expectations on the possibilities of implementation, which in turn result in an overestimation of the potential of harmonisation. By discussing the limitations of harmonisation in relation to the reception of asylum seekers, this article calls into question the possibility of a homogeneous area of protection where equivalent conditions are offered to all asylum seekers across the EU. Such a homogeneous space is utopian because harmonisation does not aim to eradicate differences but rather to mitigate them, thus tolerating diverse arrangements. The article, therefore, argues that the level playing field projected by the CEAS constitutes a promise that has two key effects: First, it depoliticises the CEAS itself by framing problems as technical issues, requiring technical solutions; second, it paves the way to further EU intervention in this field.


Introduction
The CEAS will provide better access to the asylum procedure for those who seek protection; will lead to fairer, quicker and better quality asylum decisions; will ensure that people in fear of persecution will not be returned to danger; and will provide dignified and decent conditions both for those who apply for asylum and those who are granted international protection within the EU. We have travelled a tough road to get here. But our achievement is not yet fully complete. We now need to put in a great effort to implement our legislation and ensure this common system will function well and uniformly. (European Commission, 2014a, p. 2, emphasis added) With these words, included in a promotional leaflet published by the European Commission in 2014, the then European Union's Commissioner for Home Affairs, Cecilia Malmström, celebrated the establishment of the Common European Asylum System (CEAS). The CEAS had been launched fifteen years earlier at the European Council meeting in Tampere, where the leaders of EU member states had agreed on the development of a common asylum policy "based on the full and inclusive application of the Geneva Convention" (European Council, 1999, p. 4). A process of harmonisation of member states' asylum policies followed the meeting in Tampere, covering aspects like asylum procedures, contents of protection, reception conditions, and determination of member states' responsibility for asylum applications. This process was divided into two phases. In the first one, between 1999 and 2004, the first core legal instruments of the CEAS were adopted, including the reception conditions directive, the asylum procedure directive, the qualification directive, the Dublin II regulation, and the Eurodac regulation. The second phase lasted longer because of the difficulties that characterised the negotiations of the new legislative documents. Initially scheduled to end in 2009, it was only completed in 2013 when all the recast directives and regulations were adopted.
The quote above by former Commissioner Malmström constitutes a perfect starting point from which to examine the temporal dimension of the CEAS. This article intends to do so by focusing on the conflicting temporalities characterising two pillars of the CEAS: the harmonisation of asylum policies and the Dublin system that defines the criteria for determining the state responsible for an asylum application. On the one hand, harmonisation is a future-oriented process that will eventually create "a level playing field…where all asylum seekers will be treated in the same way, with the same high-standard guarantees and procedures, wherever in the EU they make their asylum claim" (Commission of the European Communities, 2008, p. 11). On the other hand, the Dublin system has an immediate effect and allocates asylum seekers to different member states as if such a homogenous space of protection was already existing. This temporal discrepancy, it is argued, transforms the CEAS into a teleology-a process that is travelling along a well-defined path, the end of which is clear and to some extent taken for granted. It also leads to overly optimistic expectations of harmonisation, whereas the objective of the latter is not complete uniformity, but rather the alignment of differences within pre-defined common standards.
The teleological character of the CEAS is particularly evident in Malmström's words, which look like a declaration of objectives as opposed to the celebration of something that has just been created. Far from praising what the CEAS has brought about, Malmström focuses on its future outcomes and presents once again the objectives of better procedures, decisions and reception conditions as targets to be met in the future. By doing so, she defers the completion of the CEAS to a later stage, thus framing it as a process and a future-oriented endeavour. The CEAS is even equated with a road, which the EU and its member states are encouraged to travel further to finally reap the benefits of this common effort.
This article investigates the effects of the teleological discourse informing the CEAS on the overall governance of asylum policies in the EU. Two effects are discussed. The first is an effect of depoliticisation according to which the problems of the CEAS are framed as technical issues that require technical solutions, whereas the overall policy framework is never called into question. The second effect concerns the legitimation of a greater involvement of EU institutions and agencies in asylum matters, which is produced precisely by the depoliticisation of the workings of the CEAS. The article, therefore, contributes to the literature on the CEAS by drawing attention to the relevance of its temporal dimension. This aspect has been quite neglected by studies of the CEAS, which often replicate the teleological narrative informing the EU institutional discourse on this issue. In addition, the analysis of the temporal governance of the CEAS constitutes a unique contribution to debates on harmonisation. While the latter has primarily focused on the spatial dimensions of harmonisation, this article describes harmonisation as a temporal project that constitutes Europe as a space to be governed, not only through space but also through time.
The article is organised into five sections, including this introduction. The following section illustrates the conflicting temporal processes at stake in EU asylum policies, while also situating the analysis within the literature of the CEAS. The third section shows how the teleological character of the CEAS places excessive confidence on the potential of implementation and how this in turn leads to an overestimation of the scope of harmonisation. The fourth section discusses depoliticisation and the intensification of EU intervention as key effects of the teleological discourse informing the CEAS. In the conclusion, the main contributions of the article are summarised.

The Temporal Governance of the Common European Asylum System
The teleological character of the CEAS stems from a temporal discrepancy that concerns its very foundations. It is a discrepancy between the temporality of the process of harmonisation of asylum policies and the temporality of the Dublin system. These two temporalities conflict and the clash between them obliges the CEAS to be constantly forward-looking and running after its expected outcomes. On the one hand, harmonisation is a process whose outcomes are situated in an indefinite future. In October 1999, when the CEAS was launched at the European Council meeting of Tampere, the leaders of EU member states inaugurated a process of harmonisation of reception conditions, asylum procedures, and contents of protection that was expected to establish a common area of protection where similar cases are treated alike. This process was clearly future-oriented, as it is demonstrated by the opening sentence of the paragraph introducing the CEAS: "The European Council…has agreed to work towards establishing a CEAS" (European Council, 1999, p. 4).
The extended temporal horizon of the CEAS is also confirmed by the objectives included in the Conclusions, which are divided into short-term and long-term ones: This system should include, in the short term, a clear and workable determination of the State responsible for the examination of an asylum application, common standards for a fair and efficient asylum procedure, common minimum conditions of reception of asylum seekers, and the approximation of rules on the recognition and content of the refugee sta-tus….In the longer term, Community rules should lead to a common asylum procedure and a uniform status for those who are granted asylum valid throughout the Union. (European Council, 1999, p. 4, emphasis added) On the other hand, the temporality of the Dublin system is immediate. The Dublin III regulation, which identifies the first country of entry as the member state that is responsible for an asylum application, has an immediate effect. Yet, the Dublin system presupposes that the harmonised space, in which asylum seekers should receive similar treatment when it comes to reception, procedures, and contents of protection, already exists. This is the temporal paradox of the CEAS. The country of destination is imposed on asylum seekers based on the assumption that similar conditions are provided across the EU and the place of reception does not make any difference because reception conditions, procedural standards and chances of being granted asylum are equivalent. This is clearly not the case, both concerning standards and outcomes, as extensive academic and non-academic literature has shown (AIDA, 2018(AIDA, , 2019Beirens, 2018;Caponio et al., 2019;Gill & Good, 2018a). For instance, scholars have highlighted that recognition rates for asylum seekers belonging to the same nationality vary significantly across member states (Gill & Good, 2018b). Others have discussed the heterogeneity characterising the types of reception facilities and the quality of services provided to asylum seekers between and even within member states (Vianelli, 2017).
From a logical and practical standpoint, the harmonisation of domestic legislation and the approximation of national standards should have predated the identification and top-down imposition of the destination on asylum seekers. However, EU member states "proceeded backwards…prompted by their political will of excluding asylum-seekers from free movement within the EU territory" (Chetail, 2016, p. 7). The harmonised, homogenous space of the common area of protection has thus become a condition and an objective of the CEAS. As a result, the latter has lagged since its beginning, constantly expected to catch up with those very conditions that should have constituted its premise (i.e., harmonised, equivalent conditions for all asylum seekers across the EU). The whole harmonisation project has therefore existed in a continuous deferral that is exemplified by the use of the future tense in several institutional documents and communications, as showed by Malmström's quote at the beginning of this article.
The conflicting temporalities at the heart of the CEAS have been neglected by academic scholarship on asylum policy and law in the EU (see Schweitzer et al., 2018). This literature has primarily scrutinised the progress and weaknesses of the CEAS by examining the harmonisation and transposition of legislative instruments Velluti, 2014). Several scholars have focused on the flaws of the Dublin system (Den Heijer et al., 2016;Fratzke, 2015;Maiani, 2017), and how it undermines fair sharing between member states (Bauböck, 2018;Thielemann, 2018). The problems produced by the lack of an effective responsibility-sharing mechanism have also gathered much attention, alongside calls for a more equitable distribution of asylum applicants between member states (Baumgartner & Wagner, 2018;Maiani, 2017;Thym & Tsourdi, 2017;Tsourdi, 2017). Future scenarios for the CEAS and alternatives ways forward have also been explored (Gomes & Doomernik, 2019a, 2019b. In some cases, the lack of interest in the temporal dimension of the CEAS has resulted in a substantial replication of the teleological discourse that informs EU policy documents on this matter. This is evident when metaphors like "road" and "steps" are used, thereby echoing the image of the "tough road" presented by Malmström in the opening of this article. For instance, in their attempt to "assess how far the EU asylum policy has travelled on the road to supranational governance," Kaunert and Léonard (2012, p. 4, emphasis added) acknowledged that "significant steps have been taken towards establishing a 'Common Area of Protection' " (p. 20). Chetail (2016, p. 35, emphasis added) instead observed that "while the harmonisation process has been reinforced and consolidated by the recast directives and regulations, there is still a long road for a genuine CEAS to be achieved." Notwithstanding, he emphasised that second phase legislative instruments "constitute an important step towards a Common European Asylum System, which is rather a work in progress than a legal reality" (p. 35).
In this regard, the construction of the CEAS encapsulates well the overall governance of the EU, which, as Walters (2004, p. 161) argued, "is something always in progress." From a temporal point of view, the harmonisation of asylum policies shares characteristics of the broader process of European integration, which is similarly imbued with metaphors of progression, development and expansion (Walker, 2000). These features of the integration process have been emphasised by theories of neofunctionalism, according to which European integration is a cumulative process, evolving over time and resulting from mechanisms of spillover through which cooperation in some policy areas produces demand for further cooperation in other areas (see Niemann & Schmitter, 2009;Tranholm-Mikkelsen, 1991).
The CEAS offers an interesting case study for neofunctionalist theories not only because of the graduality informing its establishment but also because its very existence can be interpreted as a spillover effect of the creation of the common market. In fact, a possible explanation for the harmonisation of asylum policies can be found in the attempt to securitise the Schengen area in the wake of the abolition of internal border controls that was necessary to realise the common market. However, despite similarities with integration processes in other fields, the CEAS is somehow unique precisely because of the temporal discrepancy that was emphasised above. Whilst in the creation of the common market, graduality was achieved by organising the process into stages, the progression through which was made conditional upon the achievement of specific objectives (Walters & Haahr, 2005, p. 52), this has not been the case of the CEAS. Here, one key aspect of asylum policies-the determination of member states' responsibility through the Dublin system-has escaped any sort of graduality. The Dublin system was introduced even though the level playing field that was supposed to constitute its preliminary condition had not been created yet, thus resulting in differential treatments of asylum seekers across member states.
Although neofunctionalism captures some dynamics at stake in the process of harmonisation of asylum policies, it needs to be stressed that this theory is not just descriptive of integration processes, but also prescriptive (Niemann & Schmitter, 2009, p. 46). Neofunctionalism has been critiqued for "its investment in the teleological discourse of modernization, the way in which it imagines political change in Europe as an almost inexorable, linear process from old to new" (Walters, 2004, p. 160). One of the risks that are implicit in this approach is to assume a greater role of EU institutions as an inevitably positive thing, as if more EU would necessarily translate into more progressive decisions and better conditions for asylum seekers. This is what emerges from the conclusion drawn by Kaunert and Léonard (2012, p. 20), who admit that "there is still a considerable way to go" as "asylum matters are not governed supranationally yet." As Chamlian (2016, p. 394) highlighted, such an approach "fosters the idea that a world without the EU is unthinkable and naturalises the view that the latter is not part of the problem but part of the solution to contemporaneous challenges." The approach adopted by this article is different and can be defined as Foucauldian insofar as it focuses on the discursive dimensions of European integration, the rationalities of government underpinning it and the practices through which Europe is constructed as an object of government (see Walters & Haahr, 2005). Far from accepting the teleological inevitability of integration and its expediency, this article focuses on the ways in which the harmonisation of asylum policies produces a particular way of governing Europe through space and time. Accordingly, the purpose of this investigation is not to provide legitimation for discourses of European integration and harmonisation of asylum policies, but rather to interrogate the assumptions underpinning them and explore their effects.

Relying on Implementation, Overestimating Harmonisation
The teleological discourse informing EU asylum policies not only results in an uncritical reading of European integration in this field, but also paves the way to overly optimistic expectations on the possibilities of implementation. Again, former Commissioner Malmström's words in the brochure celebrating the CEAS are indicative in this regard: "We now need to put in a great effort to implement our legislation and ensure this common system will function well and uniformly" (European Commission, 2014a, p. 2). Implementation is presented as the missing link between the plan and its actualisation. This confidence in implementation has been particularly evident since the end of the second phase of the CEAS, when the adoption of the recast legislative instruments set off the completion of the common system. Since then, EU institutions have reiterated calls for the effective transposition and implementation of the new legal instruments-the underlying idea being that the foundations of the CEAS had been laid and the goal had become the "full implementation and enforcement of existing instruments" (European Commission, 2014b, p. 2).
For instance, in a Communication published in 2014, the European Commission (2014b, p. 6) refers to the necessity to consolidate the CEAS by putting it "in practice." In the same year, the Council also confirmed that "the overall priority now is to consistently transpose, effectively implement and consolidate the legal instruments and policy measures in place" (European Council, 2014, p. 2). Even more recently, although the weaknesses of the CEAS were laid bare by the so-called "refugee crisis" in 2015-2016, which exposed the inability of the system to cope with significant migrant arrivals and asylum applications, implementation has featured prominently in EU documents in this field. The most recent example is the New Pact on Migration and Asylum, which is expected to "foster trust in EU policies by closing the existing implementation gap" (European Commission, 2020, p. 2), based on the acknowledgement that "common rules are essential, but they are not enough" (European Commission, 2020, p. 2).
The faith on implementation that is implicit in the metaphor of the 'implementation gap' results in the overestimation of the potential of harmonisation. Instead, as Barry (1994, p. 18) suggests, harmonisation is an "ambition," whose goal is not the achievement of complete uniformity. Harmonisation does not aim for "the complete eradication of difference" (Barry, 2001, p. 73); it rather seeks to align standards and reduce differences, thus leaving scope for diverse interpretations and practices. The harmonisation of provisions concerning the asylum procedure, the reception conditions, and the contents of international protection does not concern the operationalisation of these provisions in practice. This inevitably leads to divergences between states as national and sometimes even local legal frameworks, traditions, and practices differ, thereby producing a situation in which the same rules are applied differently depending on the context. The limits of harmonisation are particularly evident if one considers the reception of asylum seekers. Indeed, the latter presents what Vianelli (2017) defines as an "excessive character" that impedes the creation of a level playing field where asylum seekers can be treated alike regardless of the context of reception. The reasons for such an excessive character of reception are mainly three and concern people, places, and policymaking.
First, reception is lived and embodied. It takes shape through relations involving human beings with different values, characters, training and resources. Yet, the characteristics of social workers, reception officers, and other positions who work daily with asylum seekers are not specified by the reception conditions directive. Important aspects like skills, profile, and training of those assisting asylum seekers are not covered by the harmonisation of reception, although these aspects make a difference in asylum seekers' experiences of reception. The training provided by the European Union Agency for Asylum (EUAA) is certainly an important starting point in this regard, but it is far from targeting most reception officers and social workers on the ground as it primarily involves state officials in specific positions. The type of organisations that should oversee reception measures is also unspecified. In fact, the reception of asylum seekers is managed by extremely diverse actors across the EU, such as state agencies, non-governmental organisations, or even profit-making companies, thus leading to remarkable differences in the ways facilities are run and services provided (Vianelli, 2017).
Second, reception is not only embodied in human relations but it also depends on, and is shaped by, the context in which it takes place (see Glorius & Doomernik, 2020). The place of reception makes a difference in terms of proximity to services, opportunities and infrastructures. Asylum seekers who are accommodated in urban areas are more likely to have better access to language and training courses, transport, healthcare facilities, and social networks as compared to those who live in isolated reception centres (Vianelli et al., 2019). The harmonisation of reception conditions does not reach such micro practicalities of reception, which are nonetheless extremely relevant if one aims to provide equivalent conditions to all asylum seekers. Reception conditions are also influenced by factors such as levels of wealth, development and overall living standards characterising the area where facilities are situated. These factors are beyond the scope of harmonisation, but they might have implications for the quality of services provided to asylum seekers. For instance, housing standards, specialised healthcare services, and training opportunities differ between local contexts, even within the same member state. Moreover, in some countries, subnational levels of government (i.e., regional or municipal) have some autonomy in the ways reception policies are imple-mented (Caponio et al., 2019), thus leading to different local arrangements. Hence, contextual differences informing reception practices cut across states and therefore call into question the feasibility of the creation of an EU-wide level playing field.
Third, the reception of asylum seekers is strictly linked to other policy domains that are not harmonised at the EU level, such as healthcare, housing, social welfare, and education. This further limits the potential of harmonisation in the field of reception because asylum seekers' experiences are inevitably contingent upon member states' domestic policies in the abovementioned domains. For example, article 15 of the reception conditions directive entitles asylum seekers to work no later than nine months after they lodge their asylum application. However, besides the formal recognition of the right to work, how member states try to make asylum seekers' access to the labour market effective differ significantly because employment policies and job placement measures are designed at the national or even sub-national level. Similar examples can be given about access to healthcare services and education, as these policy domains are primarily under member states' responsibility and no attempt at harmonising national differences is made by EU institutions. By separating reception from other related policies that are left under member states' responsibility, it is the current architecture of the EU that impedes that asylum seekers be provided with equivalent conditions across national jurisdictions.
These brief examples concerning reception expose some structural, constitutive limitations of harmonisation in the field of asylum policies. These limitations reveal that heterogeneous forms of reception also exist at the national level because the people assisting asylum seekers, the places where reception is provided, and the policy frameworks in which it is embedded vary within countries, and not just between them. Similar limitations have also been explored in relation to other aspects of the CEAS, such as status determination (Gill & Good, 2018a). Drawing on a multi-sited qualitative study of asylum appeal hearings in several European countries, the ERC-funded project ASYFAIR has highlighted the remarkable diversity characterising adjudication procedures across Europe. In fact, although EU law provides for the right of appeal for asylum seekers who receive a negative first instance decision, what an appeal is and how it is practically implemented depends on member states' justice systems, which are not affected by the harmonisation of asylum procedures. Accordingly, differences exist across-and often within-countries concerning the use of in-person hearings as opposed to paper procedures, the publicness of asylum hearings, and the degree of centralisation of adjudication processes (Gill et al., 2020).
All these examples caution against overestimating the scope of harmonisation. On the one hand, harmonisation should not be considered as a linear process, moving from policy formulation to implementation on the ground, but rather as an open-ended and contested endeavour that leads to heterogeneous practices. On the other hand, it needs to be reminded that harmonisation does not aim to erase differences at the national, regional, and local levels, but rather to provide a framework within which differences are tolerated. These features of harmonisation make it not possible to treat similar cases alike given the substantial extent of differences across as well as within states. Consequently, the level playing field that should underpin the CEAS is not achievable. It is utopian. And yet, asylum seekers are prevented from choosing where to present their asylum claims based on the assumption that they will be offered the same conditions across the EU.

The Performative Character of the Common European Asylum System
The image of the CEAS that emerges from the quote that opened this article is that of a promise. Only one day in the future, former Commissioner Malmström states, the CEAS will provide better access to the asylum procedure for those who seek protection, fairer, quicker, and solid asylum decisions, as well as dignified and decent conditions for asylum seekers and beneficiaries of international protection. Drawing on the work of the anthropologists Abram and Weszkalnys (2011), it is important to stress the performative character of promises and reflect on the effects of the promise of the CEAS. Abram and Weszkalnys (2011, p. 9) observed that "promises are not merely statements" as "they do more than describe: they express intention." "Promising is a performance," the authors continued, and as such, it has very concrete effects. Two possible effects of the promissory and performative character of the CEAS are particularly relevant and call for greater attention to its temporal dimension and its overly optimistic expectations of harmonisation.
The first effect is that of depoliticisation. Through the teleological discourse and the subsequent focus on implementation, the CEAS is presented in very prescriptive terms, as a process still in the making, whose outcomes are situated in an indefinite future. In this way, the CEAS is depoliticised through an endless deferral of the promise of its success. The attention is kept on the deficiencies of the system and the ways for improving it, whereas the overall rationale and policy framework are maintained. In this respect, the promise of the CEAS resembles the "promise of development" that has been discussed by Li (2007, p. 276) in relation to the "will to improve" informing the development apparatus. Drawing on the seminal work of Ferguson (1994), Li described the development apparatus as an "antipolitics machine," which presents a "prodigious capacity…to absorb critiques" and to keep "the attention of many critics focused on the deficiencies of such schemes and how to correct them" (Li, 2007, p. 276). "Although improvement seldom lives up to the billing," Li continued, "the will to improve persists" through the "endless deferral of the promise of development to the time when the ultimate strategy is devised and implementation perfected." A similar mechanism is at play in the CEAS thanks to a teleological narrative that constantly defers the outcomes of the harmonisation of asylum policies in the future. The promissory character of the CEAS serves a twofold purpose. On the one hand, it frames problems as technical issues requiring technical solutions, such as better transposition of EU directives, more effective implementation, and increased practical cooperation. On the other hand, it diverts attention from the structural limitations of the CEAS and specifically from the fact that procedural and contextual differences between and within countries will not be wiped off by harmonisation. As a result, the overall framework of EU asylum policies is never interrogated, although it has failed to provide equivalent conditions to all asylum seekers regardless of the place of residence. Notably, the very possibility of treating similar cases alike and the legitimacy of imposing the country of destination on asylum seekers are never called into question. The wider effect of depoliticisation is precisely that of dismissing alternative approaches from the debate by maintaining the attention on the process of improvement. The failure of the system is thus turned into the engine for the constant renovation of its governing practices.
The second effect of the promise of the CEAS is that it prepares the ground for a greater intervention of the EU in the field of asylum, which is based precisely on the abovementioned process of depoliticisation and on the resulting framing of the EU intervention as apolitical. This occurs in two ways. First, it takes place through an increased role of EU agencies (Scipioni, 2018). For example, the call for more practical cooperation in asylum matters, which is often repeated in EU documents, has led to a significant expansion of the role, funding and mandate of the European Asylum Support Office (EASO). Since its establishment in 2012, when it started with 18 employees and EUR 10 million budget (EASO, 2013), EASO grew so much that in 2021 it had around 500 staff, EUR 142 million budget, as well as operations in Cyprus, Greece, Italy, Malta, and Spain (European Commission, 2021). In January 2022, EASO was transformed into a fully-fledged agency-the EUAA-and given a reinforced mandate, including monitoring, case preparation, and development of operational standards (European Commission, 2021).
However, EUAA is not the only EU agency that has experienced a growing involvement in member states' asylum matters in recent years. In fact, other EU agencies, such as the European Border and Coast Guard (former Frontex), Eurojust, and Europol, could step up their activities on the ground by deploying their staff in Greece and Italy following the introduction of the hotspot approach. The latter was introduced by the European Agenda on Migration to "swiftly identify, register and fingerprint" migrants arriving at the EU's external borders (European Commission, 2015a, p. 6). As clarified by the European Commission (2015b, p. 2) itself, the hotspot approach provides "a platform for the agencies to intervene, rapidly and in an integrated manner in frontline [sic] Member States when there is a crisis due to specific and disproportionate migratory pressure at their external borders." In this respect, it is worth emphasising that the approach is still implemented and the abovementioned agencies are therefore currently operating in designated border areas in Greece and Italy, even though the current situation can hardly be described as characterised by "disproportionate migratory pressure." In fact, in 2021, the number of migrant arrivals by sea was 4,331 for Greece (UNHCR, 2022a) and 67,477 for Italy (UNHCR, 2022b), whereas in 2015, when the hotspot approach was introduced, sea arrivals were 856,723 in Greece (UNHCR, 2018) and 153,842 in Italy (UNHCR, 2017).
Second, the new legislative instruments that have been proposed by the European Commission after the completion of the second phase of the CEAS reveal an attempt to restrict member states' autonomy to enhance harmonisation. Both in the unsuccessful 2016 reform of the CEAS and the legislative proposals accompanying the more recent New Pact on Migration and Asylum, it is possible to identify the European Commission's tendency to replace directives with regulations. Being directly applicable in national legal systems, regulations leave less discretion to states compared to directives, as it is demonstrated by the 2016 proposals for the asylum procedure regulation and the qualification regulation.
For instance, in the former, the European Commission proposed mandatory rules on the maximum duration of the procedure, admissibility, use of border and accelerated procedures, and treatment of subsequent applications. Instead, the 2016 proposal for the qualification regulation sought to introduce a compulsory status review for those granted international protection prior to the renewal of their residence permits. These are all aspects in which member states currently have some degree of autonomy in the framework of the asylum procedure directive and the qualification directive. Furthermore, the 2016 proposal for the asylum procedure regulation established an EU common list of "safe countries of origin" in order to facilitate an accelerated processing of applications presented by people from these countries. The proposal also tackled specific procedural implications of the adoption of the concept of "safe country of origin" in order to "remove the current discretion regarding whether or not to use it" (European Commission, 2016, p. 10). The amended proposal for the asylum procedure regulation, presented alongside the New Pact on Migration and Asylum in 2020, maintained the approach of the 2016 proposal and even specified the cases in which member states should implement accelerated procedures (European Commission, 2020).
Although the process of reform of the CEAS is still open as the new legislative instruments have not been adopted yet, the proposals tabled by the European Commission show how more intrusive EU law-making is seen as a possible remedy against the weaknesses of the CEAS and for the invigoration of the process of harmonisation. While this approach by the European Commission might seem justified by the need to overcome disputes and disagreements between states, which are particularly marked in this field, it is important to stress how this is legitimised by the teleological discourse underpinning the CEAS.

Conclusions
This article has shown the importance of exploring the temporal aspects of the CEAS, while at the same time emphasising the teleological character informing EU asylum policies and the resulting effects in terms of depoliticisation and legitimation of further EU intervention. From a theoretical perspective, the analysis of the temporal governance of asylum in the EU opens the way for two key contributions: one concerning asylum and the other relating more broadly to the policy of harmonisation as a technique of government. With respect to the former, the focus on the teleological character of the CEAS invites one to call into question the possibility of the level playing field that underlies its whole architecture. By scaling down expectations on the possible outcomes of the harmonisation of asylum policies, the idea of the CEAS as a teleology makes clear that the homogenous space where asylum seekers can receive an equivalent treatment regardless of their place of residence is a myth. Significant differences are indeed destined to remain across and within states, notwithstanding the efforts in terms of legal harmonisation, policy implementation, and practical cooperation. Not only academic scholarship but policy reforms too should accept this irreducibility of differences within EU space to be more effective. This calls for rethinking the current architecture of EU asylum policies and most notably the principle according to which it is fair to impose the country of destination on asylum seekers because they are offered equivalent conditions across the EU.
Concerning the second contribution, the analysis of the temporal dimension of the CEAS offers an interesting opportunity to develop a theoretical reflection on harmonisation by taking it beyond the traditional attention to measures, standards, and technology, which has dominated this field of research (Barry, 1993(Barry, , 1994(Barry, , 2001. Notably, the study of harmonisation in the field of asylum policies shows how harmonisation is not only "a spatial and a political project" (Barry, 2001, p. 78), which constitutes "Europe as a governable entity" (Barry, 1993, p. 324), that can be "acted upon in a European way" (Barry, 1993, p. 322). The focus on asylum policies demonstrates that harmonisation is also a temporal project that consolidates the role of EU institutions and agencies through the endless deferral of the promise of its accomplishment. In this way, harmonisation allows the government of Europe through space as well as through time.

Acknowledgments
This project has received funding from the European Union's Horizon 2020 research and innovation programme under the Marie Skłodowska-Curie grant agreement No. 838722. Many thanks to three anonymous reviewers for their generous feedback and to the journal editor for the support throughout the publication process. I am also grateful to the editors of this thematic issue, Birgit Glorius and Jeroen Doomernik, for their invitation to reflect on the utopian/dystopian character of the CEAS.