UK’s withdrawal from Justice and Home Affairs: a historical institutionalist analysis of policy trajectories

Contrary to the idea that ‘Brexit means Brexit’, the article demonstrates that, in spite of leaving the Area of Freedom, Security and Justice, the UK is not automatically seeking to distance itself from the EU’s activities and approaches to these policy fields. Using the concepts of disengagement, continued engagement and re-engagement and drawing from historical institutionalism, the article further clarifies that present and future trajectories of UK positions in respect of the EU action are conditioned by a path dependence created by the evolution of UK opt-ins and opt-outs in this field, by the politicisation of the Brexit negotiations in the context of the UK–EU relations and by domestic UK politics. We explore this argument across three policy areas: (1) police and judicial cooperation, (2) immigration, borders and asylum, and (3) cybersecurity.


Introduction
At first sight, any sensible expert of Justice and Home Affairs (JHA) would argue that the UK's withdrawal is a big loss and a lose-lose situation for the EU and the UK. But is Brexit marking the end of cooperation in the field of JHA? Our assumption is that contrary to the widely spread assumption that the ideological ambition of, especially in sovereign fields such as the monopoly of force and control of borders, the UK wanting to 'take back control', the reality might be more nuanced. In order to understand the future trajectories of EU-UK cooperation in the field of JHA and identify whether we witness cases of de-Europeanisation, continued engagement or re-engagement as explained in the editorial (Wolff and Piquet 2022), we argue that it is necessary to go back to the field of JHA and to rely on a historical institutionalist analysis of the UK's participation to understand its choices pre-and post-Brexit.
The study of the JHA field has focused mostly on EU wide developments, with less single case studies on how individual countries have managed the integration of JHA acquis and Europeanisation dynamics. Indeed, so far most of the rapidly growing academic literature (e.g. Léonard and Kaunert 2020;Ripoll Servent and Trauner 2018;Bossong and Rhinard 2016;Kaunert 2011;Monar 2006;Walker 2004) has focused on what the expansion of this policy field meant for European construction theories and the nature of integration in this field. Scholars have discussed, for instance, why policy change remains quite rare in spite of an intensification of its supranationalisation (Maricut 2016;Trauner and Ripoll Servent 2016) and whether we can classify JHA as a form of new intergovernmentalism (Wolff 2015).
With a few exceptions (e.g. Geddes 2005; Adler-Nissen 2009; Carrapico et al. 2019), the scholarship in Political Science has not explored in depth the history of the UK's role in JHA, not going beyond the portrayal of the UK as a reluctant partner. This is all the more surprising considering how ambivalent the UK's approach to JHA has been. Indeed, while the UK contested the supranationalisation of these policies and negotiated opt-outs and opt-ins to accommodate its own domestic priorities over time, it has also actively cooperated with EU member states, and played a pivotal role in the development of JHA key agencies and instruments, as well as in agenda-setting over the past 20 years.
Since 2016, however, this dual approach has been deeply shaped by the discursive construction of a 'hostile environment' and the intense politicisation of immigration and ideological polarisation around the need to 'take back control'. This left a priori little room for technical and pragmatic arguments of JHA practitioners about the need to stay engaged in JHA cooperation. Brexit has thus naturally led to a sharp increase in the academic attention paid to this topic with a common wondering on whether the UK's departure is indicative of a major change of its cooperation with its European counterparts (e.g. Mitsilegas 2017;Carrapico et al. 2019;Kaunert et al. 2020).
This article contributes and expands this emerging literature by arguing that, contrary to the idea that 'Brexit means Brexit', in spite of leaving the AFSJ, the UK is not systematically seeking to distance itself from EU activities and approaches to these policy fields. It is not necessarily de-Europeanising, nor dismantling policies. Instead, we support the idea that the concepts of disengagement, continued engagement and re-engagement guiding this special issue can help to understand present and future trajectories of UK cooperation with the EU. Assuming that past legacies are relevant to understand the current state of cooperation and priorities of the UK in JHA cooperation, we draw from historical institutionalism to analyse British participation in JHA in three sub-policy areas: (i) police and judicial cooperation in criminal matters, (ii) immigration, borders and asylum, and (iii) cybersecurity. Each of these policy areas is analysed through the same sequencing of events that process-tracing has allowed us to identify (Skocpol and Pierson 2002): the intergovernmental and cherry-picking phase (1992)(1993)(1994)(1995)(1996)(1997)(1998)(1999)(2000)(2001)(2002)(2003)(2004)(2005)(2006)(2007)(2008)(2009)(2010); the contesting participation (2010)(2011)(2012)(2013)(2014)(2015)(2016); and the post-referendum sequence (2016)(2017)(2018)(2019)(2020). The analysis of policy fields through the sequencing of events helps us to identify path-dependent processes, with either 'self-reinforcing' or 'reactive' patterns.
Our main findings show that across the three policy areas legacies matter and have influenced the third phase of withdrawal/negotiations and are likely to bear some explanation in future trajectories of JHA cooperation. Indeed, in areas where the UK has traditionally been a norm exporter, like in the field of counterterrorism, police cooperation or cybersecurity, it has displayed more willingness to stay close to the EU and to cooperate in different venues, confirming patterns of either continued engagement or re-engagement. Instead, in the field of immigration, borders and asylum there is a clear disengagement, not only due to the fact that the UK has historically always been reluctant to join these structures, but also because there are fundamental differences between the EU and the UK's immigration policy philosophies. On the basis of these findings, the article aims to contribute to the academic literature on Europeanisation/De-Europeanisation (Burns et al. 2019;Copeland 2016;Radaelli 2001), and to the study of Justice and Home Affairs in a post-Brexit context (Kaunert et al. 2020;Carrapico et al. 2019) from an empirical standpoint. The article is structured as follows: the first section establishes the case for a historical institutionalist analysis of processes of disengagement, re-engagement and continued engagement and time sequencing. The second section explores the three case studies mentioned above. The final section summarises the findings and offers a reflection on the contribution the article makes towards understanding the direction of UK/EU relations post-Brexit.

Sequencing British participation to JHA: a historical institutionalist analysis of policy trajectories
In the aftermath of Brexit, JHA is at first sight an excellent case study to support a de-Europeanisation argument, i.e. the dismantling of pre-Europeanised policies and the ending of any further Europeanisation (Burns et al. 2019;Copeland 2016). Yet, following the editorial of this special issue, our analysis of JHA sub-policy areas show that contrary to the expectation that the UK would de-Europeanise in the name of sovereignty, cooperation will not necessarily stop there. Cooperation could rather take different shapes: disengagement-no dismantling but no active Europeanisation, re-engagement-continued cooperation with the EU, albeit under different forms-and continued engagement, identical cooperation. Post-Brexit paths would vary according to two post-Brexit independent variables: (i) the future of the pre-Europeanised policies and structures (dismantled or not) and (ii) the future of Europeanisation (active or not).
Historical institutionalism (HI) is well equipped to explain which scenario along this continuum is most likely to prevail in a policy field. It can help to capture how the design and trajectory of UK-EU relations in the field of JHA has happened over time and how it is likely to shape future directions. Indeed, according to HI, institutions, being the result of past developments and dynamics of resistance, are difficult to reform, and institutional resistance over time creates lock-ins that HI labels 'path dependency'.
Two types of sequencing leading to path dependency have been identified in the literature. Firstly, self-reinforcing patterns are based on the idea that after the first initial steps self-reinforcing effects appear and it becomes very difficult to reverse the gear (Mahoney 2000quoted in Daugbjerg 2009. Secondly, reactive sequencing, more dynamic, helps to explain change over time through an analysis of how new events react to past events. In this version of sequencing, some events can 'transform and perhaps reverse early events' (Mahoney 2000quoted in Daugbjerg 2009. It also acknowledges the agency of actors who 'react to events'. This means that in the field of JHA, past patterns of British participation have provided actors with opportunities to react to prior events which allow for different policy trajectories within the path. Nonetheless, HI does not fully exclude change, but it restricts it to gradual processes and 'critical junctures' (Mahoney and Villegas 2007). Essential elements of 'critical junctures' include: (i) 'periods of political uncertainty, in which the outcome from political decisions is open-ended'; (ii) 'the decisions of agents must trigger a path-dependent process that lasts much longer than the time frame within which these decisions are being undertaken'; and (iii) 'a heightened probability that change can occur does not mean that change is bound to occur' (Erdmann et al. 2011, p. 13). In other words, they are events that open the possibility for policy-makers, in a short period of time, to bring about innovative change. It is, however, often followed by long periods of stability (Daugbjerg 2009, p. 397). In addition, these critical junctures tend to produce more continuity than change through patterns of 'increasing returns' where the costs of changing paths are too high compared with continuing along the same route (Wolff 2012, p. 31).
We identify three main critical junctures in the UK's contribution to JHA. 1992 marks the beginning of a first sequence: the intergovernmental and cherry-picking phase (1992)(1993)(1994)(1995)(1996)(1997)(1998)(1999)(2000)(2001)(2002)(2003)(2004)(2005)(2006)(2007)(2008)(2009)(2010). 1992 is the introduction of the Justice and Home Affairs (JHA) pillar with the Treaty of Maastricht where the UK could still exercise its right to veto given the unanimity requirement. During this period the UK decided to 'cherry-pick' its preferred policy options as supranationalisation of JHA was introduced with the Treaty of Amsterdam (Kaunert et al. 2020). From the UK side, this form of selective participation in the AFSJ has long been perceived as delivering key benefits, namely by playing a crucial role in supporting the UK's efforts in tackling cross-border crime and enhancing its general security (Carrapico et al. 2019), without bearing the costs of community decision-making. This cherry-picking path is then further accelerated by the 2007 Lisbon Treaty with the introduction of protocol 36, offering the UK the possibility of opting out en masse from all pre-Lisbon police and judicial cooperation instruments.
2010 is another critical juncture that signals the start of the contesting participation phase (2010)(2011)(2012)(2013)(2014)(2015)(2016) with the arrival of the Conservative-Liberal Democrat Coalition government. It led to an evolution of the previous position towards selfreinforcing patterns following the Lisbon Treaty's removal of the EU pillar structure. This move meant a broad supranationalisation of these topics, a dynamic the UK has proven very reluctant to. Especially the prospect of expanding the CJEU jurisdiction in the AFSJ raised strong criticism from the UK side. Indeed, the UK announced in 2012 that it intended to activate Protocole 36 to mass opt out from the 130 pre-Lisbon police and judicial cooperation measures it was part of (Peers 2012). However, the UK opted back into 35 measures, considered essential for the continued security of the UK (Home Office 2015). This strategy shows how the UK tended to prioritise its operational needs over its more long-term ambivalent vision on European integration.
The 2016 referendum initiates the last sequence in the history of the UK in JHA, the post-referendum one (2016-2020), leading up to the withdrawal of the UK from the EU and beyond. The Brexit referendum acts a priori as a critical juncture, driven (mostly) by internal factors, as it initiates a period of significant change that will 'produce distinct legacies' (Collier and Collier 2002, p. 29). Indeed, Brexit looks at first sight like a 'revolution', as unprecedented and as creating a new window of opportunity to de-Europeanise and 'take back control' (Radaelli and Salter 2019).
Nonetheless, we argue that Brexit might not act as a critical juncture leading the UK to dismantle JHA policies in the name of sovereignty. Some examples already show that the fear of missing out on EU JHA instruments and mechanisms has quickly mobilised negotiators to guarantee access to research programs such as Horizon 2020. The de-Europeanisation scenario seems even less straightforward considering how the UK has taken an active interest in JHA since 1992, and even acted as a leader and a norm exporter rather than a laggard in the fields of counterterrorism, cybersecurity and police cooperation (Monar 2015;Christou 2016;Carrapico and Trauner 2013). The UK has also played an important role through its expertise that is highly respected in the EU and via the presence of British EU civil servants (e.g. Johnathan Faull, Director-General of DG Home Affairs from 2003 to 2010, or Rob Wainwright, director of Europol from 2009 to 2018) (Table 1).
Consequently, following the editorial of this special issue, we share the assumption that de-Europeanisation is probably not best suited to fully understand what is happening in JHA. Using the typologies of disengagement, re-engagement and continued engagement, we compare the paths of the three JHA sub-policy areas along our sequencing from 1992 to 2020 to shed light on the variety of post-Brexit (2), migration policies and border control; and (3) cybersecurity. Findings are detailed at the end of the article in Table 2.
Methodologically, the case studies were identified on the basis of purposive sampling to exemplify the existing diversity in disengagement, re-engagement and continued engagement within Justice and Home Affairs. The case studies also intend to demonstrate the diversity in outcomes relating to areas that have been politicised in the UK versus areas that have remained out of the public eye. Finally, the sampling also wanted to oppose policy areas, which the UK is known to have actively shaped, to policy areas the UK was less engaged in. For each case study, we relied on process-tracing and analysed policy documents, speeches and strategies prepared by UK decision-makers (UK Government, House of Lords, House of Commons) to identify the priorities and propositions made to the EU, as well as the arguments mobilised to justify them. This text corpus was gathered through EU institutional repositories and EUR-Lex. These primary documents were complemented by 4 semi-structured elite interviews conducted in May 2021 to gain more insights into the various phases of our sequencing, but more prominently that of the withdrawal negotiations.

Case study 1-police and judicial cooperation in criminal matters: a case of re-engagement
Since 1992, the UK has appeared both as a leader and norm exporter in the field of police and judicial cooperation in criminal matters and one of the most reluctant States to grant prerogatives to supranational institutions. The balance between these two imperatives-operational cooperation/respect of national sovereignty-has varied from the 1990s to 2020, but the UK has nevertheless established itself as a key actor in this field. The 2016 referendum, among other political attempts to regain national sovereignty, provoked strong reactions among UK practitioners to ensure the continuity of their past commitment. Although some changes are visible, path dependency is clear in this policy field due to self-reinforcing patterns.

1992-2010 intergovernmental and cherry-picking phase
British policemen and magistrates started cooperating with their European colleagues long before the Maastricht treaty institutionalised police and judicial cooperation in criminal matters. This legal change intensified the pre-existing exchanges of information, leading to the setting up of many EU agencies, mechanisms and databases. At the same time, however, the formal integration of these topics within the EU legal framework has made cooperation more complex and sensitive, raising issues of governance and sovereignty in respect of the extent of the powers granted to EU institutions. This is especially true for the UK which has traditionally opposed any move towards the transfer of sovereignty from the national to the EU level. Therefore, the UK has during the drafting of the Maastricht and Amsterdam treaties demonstrated its blocking capability of any EU mechanism conceived as detrimental to its national sovereignty.
Yet, the UK was a laggard from an operational perspective. Rather it proved in this period its willingness to initiate instruments and to cooperate through tools consistent with its vision of sovereignty. This strategy appears in the field of police cooperation with for example the UK project to set up a European Drug Intelligence Unit as an operational central coordination mechanism (König and Trauner 2021). Similarly in the field of judicial cooperation in criminal matters, the UK was one of the main architects of the mutual recognition principle (Mitsilegas 2009, p. I2). Although this principle implies some extraterritoriality (CEPS and QMUL 2018, p. 49), it aims at removing obstacles to cooperation without creating supranational additional tools. From the UK perspective, EU instruments were considered complementary and facilitating tools without dominating or replacing national authorities. When it was the case, British practitioners have intensively used EU tools, often ranking among the top users of instruments and providers of data (I1, I2, I3, I4).

2010-2016 the hostile environment and contesting participation
When the Lisbon treaty put an end to most of the exceptionalism of these two policies, the UK position became more complex as it was very reluctant to the extension of powers of the European Commission (EC), the European Parliament but mostly of the Court of Justice of the EU (CJEU). Yet the UK succeeded in obtaining a right to opt out from the EU acquis and to decide on a case-by-case basis which pre-and post-Lisbon it wanted to participate. This decision marked the beginning of a two-year process of reflection and technical discussion about the identification of measures to be opted back into. Practitioners made use of this process to voice their serious concerns over the mass opt out and urged the government to avoid a negative impact on UK security by retaining some of the most crucial measures (House of Lords 2013). Hence, in 2014 the UK opted into the police tools it had a stronger operational interest in, such as the legislation related to Europol and databases or to joint investigation teams (JIT). This was the case also in matters of judicial cooperation when the UK decided to take part in the Eurojust legal basis and the European Arrest Warrant (EAW). Yet, it did not join the instrument strengthening rights of people on trial such as Directive 2013/48/EU on having the right of access to a lawyer on criminal proceedings, nor the proposed Directive on procedural safeguards for children suspected or accused in criminal proceedings that aimed to establish common rules. The British government explained its decisions not to opt in because it did not believe that the cases had been made to demonstrate the need for EU action in this area (Home Office 2015).
The growing politicisation of JHA, starting in 2014, had some direct impacts on the UK contribution to police and criminal judicial cooperation. One relevant case study in this respect is the proposal of the new Europol regulation by the European Commission in 2013. This change of legal basis was an opportunity for the UK to opt out of the EU agency while its participation in Europol had never been questioned before. This choice was motivated by concerns in respect of the 'operational independence of UK policing' and the increase in 'the UK's obligation to provide data to Europol' (Home Office 2016a, b). Such decision was criticised by the House of Lords which, after having heard national law enforcement practitioners and Europol's director, stated that 'none of the concerns expressed by the Government in their explanatory memorandum outweigh the benefits to the UK of Europol's assistance to national police and law enforcement agencies' (House of Lords 2013).
Nonetheless, in the tools it selected the UK once again displayed a high level of commitment. This was even clearer between 2010 and 2016 when the UK managed to shape EU instruments according to its own vision. For example, during this time sequence, it succeeded in uploading its own intelligence-led policing model and strategic policy cycles at the EU level (Carrapico and Trauner 2013).

2016-2020 and beyond: re-engagement and continued europeanisation
The UK contribution to JHA after the 2016 referendum makes sense considering what happened in the previous sequences and the self-reinforcing mechanisms. Quickly after the vote, the UK government clarified that it was no longer willing to accept the jurisdiction of the CJEU and was ambiguous about continuing to respect the European Convention of Human Rights. Yet, these two instruments are part of the existing legal framework constraining EU police and judicial cooperation in criminal matters.
Nevertheless, in spite of these signs of de-Europeanisation in the name of sovereignty, the first decisions made by the Government were to opt into various EU tools and agencies. After having questioned UK participation in Europol, the May government finally decided to opt into the new regulation in November 2016, at the least expected moment. Far from being isolated (Graf von Luckner 2021), this U-turn was complemented by additional decisions of participating in the Prüm Convention in 2016, the 2014/41/EU European Investigation Order Directive (EIO) and the draft regulation on mutual recognition of confiscation and freezing orders in 2017, as well as the interoperability directive in 2018. In this time period, the UK has also been one of the drivers of the development of an EU Internet Referral Unity in order to stop the propaganda of radicalisation on the Internet (König and Trauner 2021, p. 187), pushed for the adoption of the Passenger Name Record (PNR) (Kaunert et al. 2020) and has kept using the police and judicial cooperation tools (CEPS and QMUL 2018, I4).
These opt-ins were strongly impulsed by practitioners, familiar with EU cooperation and at the heart of intense professional networks built over decades, who have seen their representations, practices and interests transformed during the years of UK participation in EU mechanisms. These changes called Europeanisation have gradually been reinforced explaining why disengagement or de-Europeanisation are the least favourite path of practitioners. Therefore, these actors have 'put a lot of pressure in Whitehall' (I1), have been 'vociferous' (I3) about the urgent need for the UK to opt in. Indeed, practitioners feared any disruption of the cooperation with the EU and conceived opt-ins as providing the best negotiating position for the years to come (I1, I2, I3, I4).
Specific attention was paid to these arguments by the UK government, the House of Lords and House of Commons (I1, I3) in a time of 'less political sensitivity' (I1), low salience of these policies that were not really predominant in the referendum debates in comparison with migration or economy (IPSOS Mori 2016). For instance, regarding the EIO, on 20 July 2017, the minister of state for security explained that '[o]pting in at this point shows our continued positive engagement with this measure, and demonstrates our commitment to work together with our European partners to fight crime and prevent terrorism now and after we leave the European Union' (quoted in CEPS and QMUL 2018).
Path dependence has also been visible since 2020 as UK contribution to police and judicial cooperation in criminal matters appears as a form of re-engagement in the continuity of previous events. Despite the growing politicisation of the negotiations under the Johnson government (I1, I2), self-reinforcement is still active. Indeed, the same Europeanised law enforcement and justice practitioners have been once again the ones asking for maintaining the status quo, afraid of losing access to EU tools and of weakening their cooperation with their EU counterparts (I1, I2, I3, I4). Their aim was 'to avoid disruption, rupture' (I1), during the transition period and after Brexit became effective, in relation with EU databases (especially SIS II, European Criminal Records Information System, PNR and Prüm), EU agencies (Eurojust and even more Europol) and operational cooperation instruments (EAW, EIO, JIT). Following this reactive entrepreneuriat and on the basis of what had been negotiated before, the UK will still have access to Prüm or PNR, and it will be able to participate in JIT. Nonetheless, practitioners' success was only partial as the EU advocated the UK could not have the same benefits as EU or Schengen member States, especially as it was still refusing the power of the CJEU in the name of national sovereignty. While it will have limited access to Europol and Eurojust, as well as some tailor-made tools replacing the EAW, the UK lost its right of access to SIS II and to ECRIS (for further details see House of Lords 2021). These results were assessed by British representatives as 'less a rupture than was generally feared, but still a rupture' (I2). Cooperation with the EU will therefore not stop, nor will Europeanisation, but they will partly take new forms and channels (e.g. bilateral networks, Interpol (I2, I3, I4)).

Case study 2-freedom of movement, immigration, asylum and borders: disengagement as a result of reactive sequencing
Analysing British participation to the EU's freedom of movement through sequencing reveals how counter-reactions unfolded and led to reverse the early events of joining a Union based on the freedom of its citizens. The final outcome of not including a chapter on mobility in the TCA is causally connected to a series of individual events which were not individually sufficient for this outcome to happen (Daugbjerg 2009, p. 399). But taken together as a series of events, these sequencing has led to the final outcome. In particular the design and terms under which the UK had reluctantly participated in mobility with the EU explain why Brexit was an inevitable opportunity to disengage and to revert to a national policy model, whatever the results of the EU-UK negotiations would be.

1992-2010 intergovernmental and cherry-picking phase
The 1992-2010 phase has been formative in understanding the path-dependent pattern regarding the selective Europeanisation of UK policies in this field. In this period, as migration, asylum and borders were shifted from the JHA Pillar to the Community Pillar, the decision-making rules applicable to this area were aligned with First Pillar rules, together with the EC's enforcement powers and the CJEU's jurisdiction. The UK managed to circumvent these rules by requesting an opt-out from all JHA issues relating to migration, asylum and borders, as well as from Schengen measures, following its incorporation into the Acquis. This cherry-picking allowed the UK to select on a case-by-case basis the migration policies in which it would take part. Also although the UK never joined the Schengen Convention as stipulated by Protocol 19 of the TEU and TFEU, it was able to decide whether it wanted to opt in to new JHA legislation according to Protocol 21.
In the field of asylum, the UK's position has quite closely, and pragmatically, matched the curb of its asylum applicant numbers. Thus, in the 1990s, because the UK was not worried about the number of asylum applications it received, it did not support the 'integrated' German proposal, inspired by Lander models, to adopt 'a system based on a 'capacity principle' to redistribute people seeking temporary protection' (Ripoll Servent 2018: 88). However, in the early 2000s, the UK experienced higher numbers of asylum applications 'partly attributed to its more liberal policies regarding cases of non-state persecution compared to Germany and France' (Zaun and Ripoll Servent 2021, p. 166). The New Labour government at the time felt that due to this increase in asylum applications, easier collective decision-making with the EU could 'alleviate the pressures on its own asylum system, and ensure a more equal distribution of asylum seekers (Fella 2006, p. 634 quoted in Zaun andRipoll Servent 2021, p. 167). It therefore decided to support a move towards qualified majority voting in the Council and thus more integration. This convergence of interest between 'Germany and the UK eventually agreed on a package deal: asylum policies were to be fully communitarised in the Constitutional Treaty' (Zaun and Ripoll Servent 2021, p. 167). Regarding the Dublin convention, the UK managed to negotiate a formal differentiated integration which had as side effect to weaken the whole integrity of the EU's asylum system (El-Enanny 2017, p. 2). Indeed although the UK decided to take part in the Dublin Convention; it managed, like Ireland and Denmark, to refuse to implement the second generation of recast asylum directives on qualification, reception conditions and procedures. (Ibid, p. 3).
In the field of immigration, the UK's position can be explained, not only by fundamental historical differences over the freedom of movement and immigration policies, but also by the politicisation of the issue since 2004. First, a series of difficulties appeared around the issue of freedom of movement. Traditionally UK immigration for non-EU citizens has adopted a 'permission-based' approach where the 'British government retains and uses the right to discriminate between would-be immigrants on grounds of nationality, skill-level and family reunification status, amongst other criteria' (Dennison andGeddes 2018, p. 1140). Second, the participation of the UK to the EU immigration regime has been explained as 'a significant racial realignment in its immigration regime' which departed from the racial and class conceptualisation of immigrants in the UK. Historically indeed, several scholars have highlighted how the UK immigration regime has been racialised along white and non-white lines (Spencer 2002 in Burrell and Schweyher 2019), but also along 'class and wealth lines' with historically a control of the mobility of the poor and who gets entitled to welfare, a link established in the 1905 Aliens Act (Burrell and Schwyyer 2019, p. 195). The alignment with EU right-based approach might have been therefore exceptional but has also been limited due to the historical legacies of the UK immigration system.

2010-2016 the hostile environment and contesting participation
This period is clearly defined by an acceleration of selective participation, but more importantly by shift with the implementation of the 'hostile environment'. Introduced under the Coalition government by Theresa May in 2010 when she was then Home Office Secretary, this policy approach guided discourse and practices in immigration in the 2010-2016 period, leading up to the conflation of anti-immigrants pro-Brexit positions. Deterring immigrants from coming to the UK, but also pushing existing immigrants to return back home is at the core of this policy, including making the life of immigrants in the UK more difficult. New Immigration acts were adopted in 2014 and 2016, public campaigns against migrants with vans carrying 'Go Home' messages. The Windrush scandal exposed to public opinion the reality of this hostile environment (Burrell and Schweyher 2019, p. 193) which asked whole sectors of society such as universities, schools, landlords to police migration leading to a 'general rise in immigration offence arrests and deportations' (Ibid).
In 2014, a strong element of contestation of British EU membership was clearly linked to the politicisation of immigration in the public debate (Wainwright in House of Commons evidence 2018). The move from 'the arena of interest group politics to the arena of mass politics' (Schimmelfennig 2018(Schimmelfennig , p. 1159 started in the UK with the 2004 enlargement (Dennison and Geddes 2018), but this process was strongly intensified by domestic political struggles. By 2014, the Conservative Party was under increased pressure from Eurosceptic Conservative backbenchers, and fringe parties were gaining political ground. The Prime Minister took the initiative of promising a referendum on EU membership as part of the 2015 elections manifesto (Cameron 2013). The expansion of the Conservative discourse on the need to control migration, its association to the EU free movement of persons and the rapid increase in the saliency of immigration in the eyes of the public changed. Self-reinforcing tendencies were noticeable with British politicians resorting to old discourses conflating these immigrants with the threats to UK social welfare they represented, including David Cameron. Accordingly 'these discourses were heavily utilised in Leave and media campaigns, stoking popular fears about the perils of EU membership, welfare integrity and freely moving foreigners' (Ibid,p. 194). Most importantly this period demonstrates a clear distancing on the side of the UK from the concept of EU citizenship.
In the field of asylum, after years of relatively good cooperation with other EU member states, the UK and Ireland chose to participate more selectively on illegal migration and asylum policies (El-Enanny 2017, p. 4). Most interestingly, the UK government did not choose to opt in into the recast of the four asylum directives in 2011 as recommended by the House of Lords European Union Committee (Ibid, p. 4) precisely because these were 'provisions which are designed to at least marginally improve protection of asylum seekers' (Ibid,p. 4). This decision according to El-Enanny was made very early in the legislative process of the recasting of the directives and was motivated by the fact that it would limit its discretion on its territory to manage asylum and that it was 'putting in jeopardy its ability to reduce the number of asylum seekers on its territory, deter 'false applicants' and control its borders' (Ibid, p. 5).

2016-2020 and beyond: disengagement and de-Europeanisation
As seen in previous sequences, the UK has evolved from being a selective partner to a reluctant one in the field of immigration, borders and asylum. It never really embraced participation in agencies such as the European Asylum Support Office (EASO) and as a non-Schengen state could not actively be involved in decisions by Frontex-the European Border and Coast Guard Agency. Given the above-mentioned political climate, there was no permissive consensus in the UK on immigration and the disengagement was 'inevitable' (I4), as well as was clearly accelerated under the Johnson government. 'The UK was mostly interested in maintaining some forms of return and access to databases, but they had little success' (I4). This disengagement in particular prompted a practitioner's support for negotiating access to the Schengen Information System (SIS). While some think that some similar databases within Interpol could be useful to the UK, SIS II is considered as 'the biggest single operational loss' for both the EU and the UK (I4).
Settled status is the last event that ends this series of counter-reaction. By ending the freedom of movement between the EU and the UK, the UK has also managed to revert to its own national philosophy of providing EU nationals a permission to stay. This also by the same token ends the mobility of British citizens in the EU. The chapter of mobility offered by the EU in the negotiations was refused by the UK, creating a series of challenges for British living and working in the EU. A point system has been introduced mostly favouring high-skilled migrants to come to the UK. British employers wishing to employ EU citizens, with no (pre)settle status, will now need to have a sponsor licence and pay the Immigration Skills Charge (Saunders 2021). Their red line was that they would not sign up to any legal framework and that the UK would leave the European Court of Justice jurisdiction. In that sense, we can here speak about a 'reversal' and a return to the British philosophy about immigration which 'permits' migrants to stay but does not provide them with rights, which was compulsory under EU law and for EU citizens.

Case study 3-cybersecurity: between re-engagement and continued engagement
The European Union cybersecurity policy is not only one of the EU's most recent policy fields, but it is also one of the most transversal and eclectic ones, whose effect is now felt across all EU policies (EC and High Representative of the Union for Foreign Affairs and Security Policy 2020). This case study proposes to explore the evolution of the UK-EU relationship in the field of cybersecurity, and to understand how this relationship, combined with the technical and depoliticised framing of this field, contributed to some sub-areas of cybersecurity being earmarked for continued engagement in the UK-EU post-Brexit relationship, at the same time as others were identified for re-engagement.

1992-2010 intergovernmental and cherry-picking phase
By the time the EU began to reflect on cybersecurity issues, the UK had already developed an advanced security interest in the topic, as well as an emerging body of legislation. Throughout the 1980s, the UK had become increasingly concerned with computer use for unauthorised purposes (The Law Commission 1989 and1988), which resulted in the adoption of the Computer Misuse Act of 1990 and positioned the UK very much at the forefront of the Member States' initiatives in this field (I3; I4). When the EC organised its first study into high-tech crime in 1998, the UK was clearly presented as a point of reference in a number of areas, including having the oldest Member State platform for industry and law enforcement to exchange best practices on cybercrime, as well as having a mechanism for the general public to report online illegal content (Bangemann 1994). It is therefore not surprising that when the EC first encouraged the approximation of Member States' national legislations on high-tech crime, namely in terms of hacking and denial of service attacks (European Commission 2001), both the UK Government and the House of Commons welcomed this initiative as being aligned with the UK's priorities and expressed interest in being involved in its future direction (Select Committee on European Scrutiny 2001). The UK was also particularly aware that the majority of cyberattacks targeting its citizens, industry and public institutions was originating from outside the country and that international cooperation was key to protecting individuals and critical databases from hackers, organised crime and terrorists using cybertools (Clark, cited in Fafinski 2008).
The UK's early interest in cybercrime, combined with its emerging leadership role in this field and its awareness of the importance of international cooperation, created a path dependence, characterised by a self-reinforcing pattern, that fostered receptiveness to EU cybersecurity initiatives, including the willingness to advance cooperation through third pillar instruments, and a general support of first pillar instruments (for example, the 2000 Council Decision to Combat Child Pornography on the Internet, and the 2005 Council Framework Decision on Attacks against Information Systems). Detailed analysis of the UK Parliament debates and of the UK Government's explanatory memoranda and letters clearly highlights the support that these instruments had during the 1992-2010 phase, as well as the UK's capacity to shape them on the basis of its domestic instruments (Select Committee on European Scrutiny 2003).

2010-2016 the (non-)contested participation phase
This second phase is characterised by a continuation of the path dependence that emerged between 1992 and 2009. In the case of cybersecurity, the change in Government and the opt-in possibility do not affect the trajectory of the field. Unlike the other case studies, this phase is not a contestation period. 2010 marks the confirmation of the UK's prioritisation of cyberattacks by organised crime, terrorist groups and other states as a Tier one security threat, the highest level awarded in the National Security Strategy, alongside terrorism, natural disasters and international military crises (HM Government 2010) (I1, I3). This understanding of cyberinsecurity as one of the most important threats facing the UK further accelerated the country's investment in cybersecurity infrastructure and governance, with the creation of the National Cyber Crime Centre in 2013 and the National Cyber Security Centre in 2016. Furthermore, there is also continuity in the UK's leadership of the field, with the 2011-2016 Cybersecurity Strategy annual report highlighting how much the UK Government has helped to shape the EU cybersecurity strategy and its implementation, providing a stronger basis for cooperation with other EU member states (Cabinet Office 2016a, b).
In line with the previous phase, the 2010-2016 period continued to be characterised by an ongoing willingness to cooperate more closely with EU partners. This trend is clearly visible given two important developments: (1) when given the opportunity to opt in to EU cybercrime legislation (following the Treaty of Lisbon expansion of the JHA opt-in to Police and Judicial Cooperation), the UK consistently chose to take part, namely in the 2013 Directive on Attacks against Information Systems (Official Journal of the European Union 2013). The position of the UK Government was that EU initiatives and legislation in this field were well aligned with UK interests and that, in fact, there was a case for more EU action in this area (House of Commons 2011);and (2) when the UK Government chose to opt out en masse from pre-Lisbon Police and Judicial Cooperation instruments in 2014, it opted for retaining the cybercrime elements (I2). Prior to the decision, Members of Parliament and practitioners were allowed to voice their concerns over the potential loss of these measures in a technical and depoliticised context (House of Lords 2013). As we approach the end of this second phase and the Brexit referendum is announced, the path dependence of cooperative and interdependent UK-EU cybersecurity relations had been reinforced throughout the 2010-2016 phase.

2016-2020 the continued engagement and re-engagement phase: cybersecurity as a priority area in the trade and cooperation agreement (TCA)
Unlike the topics of migration and the economy, cybersecurity remained very much under the radar during the referendum campaign and the negotiations of the TCA (I1). Any discussions about the possible consequences of different Brexit scenarios remained limited to practitioner and think tank fora, with private companies worried about their future capacity to attract and recruit new experts, as well as a possible regulatory gap (Curry 2019), and with law enforcement concerned about their effectiveness in addressing cybercrime in a context of reduced cooperation with the EU (Kahn 2019).
This lack of politicisation allowed UK and EU negotiators to prioritise an area that was seen as mainly escaping UK and EU red flags, as having low political salience by the general public, and as being of strategic importance for the maintenance of UK and EU security. Cybersecurity occupies a place of particular relevance in the TCA, being one of only two areas that were specifically selected for prioritisation. The Agreement foresees important cooperation elements including exchange of information and best practices, as well as the possibility to continue to cooperate with the EU Computer Emergency Response Team (CERT-EU) and the EU Agency for Cybersecurity (ENISA) (EU-UK TCA 2021: Part 4). Although there is a qualitative difference in relation to pre-Brexit cooperation (namely in terms of participation in decision-making), the panoply of instruments available for future UK-EU cybersecurity cooperation is in general indicative of continued engagement and is very much in line with the path dependence that has been developed since the 90 s. The UK's decision to keep EU cybersecurity-related measures as part of its domestic legislation (namely the Network and Information Security Directive and GDPR) is also indicative of this path dependence and of the UK's willingness to stay aligned with the EU (Walden and Michels 2021). The key importance of this relationship is also reiterated in the Draft Council Conclusions of the 9th of March, which mention UK-EU cybersecurity cooperation ahead of EU-NATO and EU-UN cooperation (Council of the European Union 2021).
Despite this optimistic outlook, there is in practice a loss of operational cooperation in cybercrime that is related to law enforcement instruments (Ni Loideáin, cited in House of Lords 2021). Even if there is overall agreement on the fact that the UK obtained more from JHA negotiations than initially expected, the loss of access to Schengen Information System real time data constitutes a real challenge that cannot be easily compensated by access to Interpol systems or bilateral UK-Member States agreements (I2). Furthermore, even those instruments for which the UK managed to negotiate some level of access, cooperation is very much dependent on the EU's data adequacy decision regarding the UK's handling of data (House of Lords 2021).
In this final phase, we can observe that, following a lack of politicisation of this area during the Referendum campaign and ensuing negotiations, cybersecurity has overall been identified as an important area for continued engagement, marked by a discourse on the crucial relevance of cooperation to ensure the security of citizens and the economy (I4). Despite this apparent continuation of the path dependence and a clear self-reinforcing pattern throughout the first and second historical phases, Brexit negotiations constitute an important critical juncture, as they mark a bifurcation in the field. Despite the prioritisation of cybersecurity as a special area of cooperation in the TCA, the sub-field of cybercrime is in practice excluded due to its reliance on law enforcement instruments. Therefore, the Brexit negotiations increase the differentiation within the UK-EU cybersecurity relation by creating a continued engagement route for cybersecurity in general (critical information infrastructure and cyberdefence) and a reengagement route for cybercrime (I2, I3).

Conclusion-future trajectories of UK JHA policies after Brexit
This article highlights the complex and differentiated evolutions of UK policies since the 2016 referendum, without any major de-Europeanisation so far. The varieties of UK engagement are mostly explained by the paths taken by the different sub-policies from 1992 to 2016 and self-reinforcing/reactive sequencing patterns. Table 2 summarises our findings across the three areas. To start with, path dependency is clear in the case of police and judicial cooperation in criminal matters. UK practitioners who have been Europeanised strongly reacted every time they were afraid of losing access to EU tools due to the sovereignty concerns expressed by the UK government. This reaction intensified following the 2016 referendum, but the low salience of these issues allowed practitioners to be listened to, leading the government to advocate continued engagement. Nonetheless, while the UK becoming a third country without being a Schengen member State nor accepting supranational authority means it has lost some of its privileges, it would maintain a strong operational link with the EU through different channels.
The case study on immigration, border controls and asylum shows that the abrupt end given to freedom of movement by the UK, refusing to insert a chapter in the TCA, is not a sudden choice but comes out of a series of prior choices, a series of temporally ordered events, which can be seen as an instance of reactive sequence. While there are some nuances and we can already identify that the UK might be eager to find in the future new ways to access EU databases such as SIS II, the discussion over immigration demonstrates that the discussion especially over EU citizens and the (pre)settled status has been marked by some backlash that has 'reversed' the initial acceptation of the UK to allow freedom of movement for EU citizens in the UK. By withdrawing from the EU, the UK could withdraw itself from this obligation and revert to its original preferred choice of giving a 'permission to stay' to EU citizens via its own national devised solution. This explains why it is a clear case of disengagement which resulted from a chain of events.
The cybersecurity case study outlined that the UK's approach to the EU-UK relationship in this policy field was always guided by its early interest in cybercrime, the opportunity to shape the direction of EU cybersecurity policy and the awareness that cooperation at EU level was key to effectively addressing cyberinsecurity. This case study highlights the existence of a self-reinforcing pattern, where external events have not resulted in critical junctures, but rather in a gradual consolidation of the path dependence of this area. Brexit negotiations, however, introduced a bifurcation into this path. Most of the areas within cybersecurity continued to be an important No dismantling of cybersecurity policy and legislation. Cybercrime follows the same path as law enforcement. Evolution through self-reinforcing pattern No dismantling and singling out of cybersecurity as an area for special cooperation in the TCA priority (continued engagement), but cooperation in the sub-field of cybercrime has now become more limited given its reliance on JHA instruments. This reduction in cooperation should not be interpreted, however, as a form of disengagement, but as re-engagement, given the UK's interest in finding alternative forms of cooperation with the EU in this field (I2, I3). Besides filling a gap in the literature on UK and JHA, three complementary contributions of this article need to be emphasised. Firstly, from a more theoretical perspective, together, these three case studies show how diverse the mechanisms underlying path dependency can be and, quite surprisingly, the dynamic nature of policy status quo as continuity is much more than a simple inertia. Secondly, this article enriches the existing scholarship on Brexit. It argues that past events can shape future trajectories, especially in complex phenomena such as the UK withdrawal, and draws the attention to the extent and nature of previous Europeanisation, the role of Europeanised actors in the domestic decision-making process and the politicisation of the cooperation with the EU to understand the different paths policies can take after the UK withdrawal. Finally, this research addresses the academic literature on Europeanisation. It sheds light on the long-lasting effects of Europeanisation, going even beyond the formal membership to the EU. It also emphasises how Europeanisation is a process and not a result, with possible forward gears as well as reverse gears. These observations call for this recent research agenda on Europeanisation to be further explored.