Northern Ireland and Brexit: where sovereignty and stability collide?

ABSTRACT Having been issues which stimulated little discussion beyond Northern Ireland before the Brexit referendum, the ‘Irish issue’ dominated and complicated the subsequent EU–UK withdrawal negotiations. The ‘taking back control’ narrative bears the hallmarks of Prime Minister Boris Johnson’s notion of a ‘recaptured sovereignty’. This conception of sovereignty, however, clashes with the post-sovereign character and content of the 1998 Belfast/Good Friday Agreement. The institutions created by the Agreement move beyond the old frameworks for political order by creating a series of interlocking territorial, cross-territorial, and cross-national institutions based on shared sovereignty. This prompts critical questions about the resilience and sustainability of the Agreement against the backdrop of a shift in sovereignty which appears incompatible with the key tenets of Northern Ireland’s peace agreement. This article examines how Northern Ireland’s approach to and experience of the EU referendum differed from the rest of the UK. It queries how the referendum result, subsequent political and electoral developments, and difficulties in agreeing a withdrawal formula and future UK–EU relationship, expose the gap between the UK’s commitments to a post-sovereign arrangement for Northern Ireland and the preference for a return to an older conception of sovereignty as implied by the Brexit vote.

Moreover, the status of Northern Ireland as a part of the UK is underpinned by the principle of consent (linked to popular sovereignty) (see Hadfield 1998). The EU framework of open borders and shared sovereignty within which these innovations were situated was an important facilitating context. In contrast, Brexit clashes with an understanding of sovereignty which appears incompatible with the key tenets of Northern Ireland's peace agreement. This prompts critical questions about the resilience and sustainability of the Agreement's institutions and principles, and the long-term prospects for stability in Northern Ireland.
This article examines how Northern Ireland's approach to and experience of the EU referendum differed from the rest of the UK. It explores how the referendum result, subsequent political and electoral developments, and the pronounced difficulties in agreeing a withdrawal formula and future UK-EU trade relationship, expose a gulf between the UK's commitments to a post-sovereign arrangement for Northern Ireland and the return to an older conception of national sovereignty implied by the pro-Brexit rhetoric and narrative. This reconfiguration of the British order which Brexit entails intersects clumsily with Northern Ireland's post-sovereign political settlement and has propelled previously latent debates and discussions in Northern Ireland about the long-term viability of the 1998 Agreement and the prospects for future constitutional change.

The Northern Ireland context
The UK's 2016 EU referendum campaign coincided with a period when Northern Ireland's historically contested politics had settled into a cycle of somewhat mundane normality. The violent campaign of paramilitary violence which began in the late 1960s and had its roots in 'the social, economic, cultural and geographical structure of Northern Ireland ' (Darby 1976, 196) had given way to a delicate political equilibrium underpinned by the provisions of the 1998 Belfast/Good Friday Agreement. The fundamental division in Northern Ireland between unionists and nationalists was not eliminated by the peace process. Political and electoral contestation in Northern Ireland continues to be based on a deep and profound constitutional cleavage where unionists support the constitutional link between Northern Ireland and the rest of the UK, while nationalists aspire to the creation of a united Ireland (see Matthews and Pow 2020). The potency of Northern Ireland's communal and political division, however, has been challenged and the UK's devolution settlement has been instrumental in that respect.
Northern Ireland's form of devolution allowed former political foes to share power, and to exercise executive and legislative functions in devolved areas. Tensions between the parties continued to exist after 1998, and were particularly evident in relation to equality issues, language rights, and conflict legacy matters (see Armstrong, Herbert, and Mustad 2019). Differences were also apparent in relation to a lack of progress on key issues, such as integrated education, welfare entitlements and shared housing (see Gray et al. 2019). Local politicians in Northern Ireland have encountered particular difficulties in confronting and reforming these highly sensitive policy domains. The challenges in discussing and agreeing progress, hinges on deeper problems related to the weaknesses inherent in the consociational settlement (see, for example , Hall 2018;Taylor 2008) and a lingering lack of trust between the unionist and nationalist blocs (Teague 2019). At the time of the calling of the EU referendum, Murphy (2018) identified the existence of a 'negative peace' in Northern Ireland epitomised by the ongoing polarisation of politics. Although a 'negative peace' environment is not unexpected for a region in transition from conflict to peace, it is nevertheless a precarious and tenuous moment for any conflict transformation process. The calling of the 2016 EU referendum coincided with such a period in Northern Ireland.

The EU referendum in Northern Ireland
The EU referendum campaign across the UK differed depending on territory. For much of the rest of the UK (and most notably in England), the key referendum themes were immigration and the rather vague notion of 'taking back control'. It included a narrative which reflected not just on UK sovereignty vis-à-vis the EU, but revealed underlying ambiguities and tensions in relation to the exercise of sovereignty vis-à-vis the UK's constituent units (see Sandford and Gormley-Heenan 2020).
In Northern Ireland, the referendum campaign was markedly different and the emphasis was less fixated on calls to recapture British sovereignty and more focused on localised political and economic concerns. These concerns included the impact of a Leave vote on the status of the border between Northern Ireland and the Republic of Ireland; the implications for Northern Ireland trade; the future of EU structural/Peace funding to Northern Ireland; the impact on the agriculture sector; and the consequences for the Northern Ireland peace process (see Murphy 2016). Nationalists, in particular, were concerned about the impact of Brexit on political stability in Northern Ireland. A key achievement of the peace process was that it took the border (between north and south) out of Irish politics (see Gormley-Heenan and Aughey 2017). This facilitated the opening up of the border and the creation of cross-border institutions; and allowed for the removal of border infrastructure and check-points. It was an especially important component of the peace process for nationalists because it legitimised their identification with the rest of Ireland and enabled institutional mechanisms which gave expression to that identity. In this context, a key success of the Belfast/Good Friday Agreement was its ability to neutralise the toxicity which had previously attached to border politics in Ireland. The Agreement, therefore, largely settled the contestedness of the border and both unionists and nationalists were at peace with its existence, its operation and its meaning. Crucially, the open border between north and south was facilitated by UK and Irish membership of the EU, and specifically by membership of the EU's single market. The single market is based on open borders and free movement of people and trade, so there was an easy and almost fortuitous way in which it enabled key aspects of the Belfast/Good Friday Agreement (see Hayward 2018).
The Northern Ireland EU referendum campaign re-opened discussions about how a Leave vote might impact on the settled status of the Irish border. Nationalists, and the Irish government, were vocal during the referendum campaign about the potential for a vote in favour of Leave to lead to the erection of a reinforced physical and metaphorical border between north and south (Aughey and Gormley-Heenan 2017). The economic ramifications of such a development were viewed as being potentially detrimental to the Northern Ireland economy (see, for example, ESRI 2015; Open University 2015; Oxford Economics 2016). Politically, the re-erection of the border was viewed by nationalists as a threat to the viability and stability of arrangements and institutions created by the Belfast/Good Friday Agreement. In contrast, unionists did not share this interpretation of how a Leave vote might impact on Northern Ireland. Given their lesser affinity with the Republic of Ireland, Unionists have traditionally been less wedded to the cross-border element of the Agreement than their nationalist counterparts (see Smithey 2011). Of Northern Ireland's political parties, the largest and pro-Brexit Democratic Unionist Party (DUP) was particularly dismissive of claims that Brexit might challenge Northern Ireland's hard-won peace.
Like all political polls in Northern Ireland, the EU referendum divided nationalists and unionists, although the division was not a clean one. All of the nationalist parties (the SDLP and Sinn Féin) and centrist parties (Alliance Party and Green Party) supported the Remain position. Among unionists, there was less consistency. The DUP and the small Traditional Unionist Voice (TUV) party both advocated for the Leave position. In contrast, the more moderate Ulster Unionist Party was 'on balance' supportive of the UK remaining in the EU. Despite the magnitude and seriousness of the issues confronting Northern Ireland however, the referendum campaign was muted. Political parties tended to shy away from discussing issues which touched on contested issues of sovereignty and which exposed tensions between nationalism and unionism (see Murphy 2016).
The UK-wide referendum result revealed different territorial preferences. Where the UK, as a whole, supported Leave, Northern Ireland (like Scotland) voted in favour of remaining in the EU (see McCann and Hainsworth 2017). The breakdown of the referendum result in Northern Ireland demonstrates how the vote divided unionists and nationalists. Research by Garry (2017, 2) notes that voter choice was largely driven by 'the deep underlying ethno-national divide in Northern Ireland politics'. In other words, 66% of self-described 'unionists' voted to exit the EU while almost 90% of self-described 'nationalists' voted Remain. Although this does not demonstrate a clear-cut unionistnationalist divide, it hints at a degree of latent division in Northern Ireland on the issue of Brexit. The depth of this division, however, crystallised and intensified during the ensuing UK-EU withdrawal negotiations. That divide between unionists and nationalists was influenced by the immense difficulties in agreeing a withdrawal formula which facilitated the maintenance of an open border between Northern Ireland and the Republic of Ireland.

The UK-EU referendum and the absence of an Irish narrative
Issues related to the Irish border, trade, and the peace process which animated the referendum campaign in Northern Ireland did not enjoy any sort of airing outside Northern Ireland. The broader UK-EU referendum campaign was essentially an Ireland free zone. As Hayward and Murphy (2018, 282) note: 'the 2016 UK referendum on EU membership highlighted . . . the low priority of Northern Ireland for the UK government's campaign and the British electorate'. There was no serious or significant attention paid to how Brexit might impact on Northern Ireland's tenuous and delicate political situation (see Burke Wood and Gilmartin 2020). There was no sense that the timing, even the very fact of the referendum, might prove challenging for a region in transition from conflict to peace. Glencross (2016, 9-10) observes that a lack of attention by Prime Minister David Cameron to the wider dynamics of the referendum beyond Britain 'illustrates how far this vote was essentially a domestic party political matter'. This obliviousness to Northern Ireland's particular situation was evident across both Remain and Leave campaign groups while few within either the Conservative Party or Labour Party broached the subject during the referendum period. And this was despite the acute concerns expressed by the Irish government 1 and nationalists in Northern Ireland about the referendum outcome and its potential implications for Ireland, north and south.
This lack of awareness of Northern Ireland's particular situation resulted in a marked lack of consideration, preparation, or contingency planning for the fallout from the Leave vote for Northern Ireland. More tellingly, it also overtly revealed an evident disconnect between Britain and Northern Ireland: a lack of basic awareness at elite (and public) level as to the potential for the Irish border issue to be problematic. This hinged not just on day-to-day detachment and disconnection of peoples, it also underlined the limited appreciation, and in some cases, the dissatisfaction with the kind of constitutional and political construct the UK had become after 1998. The EU referendum campaign, which was dominated by largely English discontents, also revealed an element of English dissatisfaction with 'the perceived unfair advantages Scotland gets from devolution, [and] the absence of an institutionalised political voice for England' (Henderson et al. 2017, 638). These tensions and misunderstandings, though evident during the EU referendum campaign, played out much more emphatically during the UK-EU Brexit negotiations which followed.

Northern Ireland and the Brexit Negotiations
Having been little more than a footnote during the wider UK referendum campaign, issues pertaining to Northern Ireland moved front and centre during the ensuing EU-UK negotiation process (see O'Rourke 2019). The withdrawal negotiations were structured around three key priorities: citizens' rights for EU-27 citizens in the UK and UK citizens in the EU-27; the settlement of the UK's financial obligations; and ensuring the Northern Ireland peace process is not compromised. Specifically, the Negotiating Directives for Article 50 Negotiations (22 May 2017) resolved that: 'Negotiations should in particular aim to avoid the creation of a hard border on the island of Ireland, while respecting the integrity of the Union legal order ' (Paragraph 14).
Given that the Republic of Ireland shared a land border with Northern Ireland and was acutely attuned to the potential pitfalls of Brexit for economic and political stability on the island of Ireland, the Irish government fought hard for issues concerning the island of Ireland to be high on the EU's negotiating agenda. The threats and challenges which Brexit posed for Ireland -north and south -were immense (see for example, Economic and Social Research Institute (ESRI) 2015). Indeed, the Republic of Ireland stood to be more impacted by Brexit than any other EU member state. With a view to protecting Irish interests: 'Ireland's objective from the outset was to transform Irish interests into those of the EU ' (Laffan 2018, 571). The Irish government deployed an extensive and sustained programme of meetings and briefings with key figures across the EU institutions and all EU capitals. This exercise aimed to communicate the implications of Brexit for Ireland, and in particular to explain the complexities of the Irish border issue for Ireland and its relationship with the UK.
The EU was receptive to Irish concerns. This was not solely due to the diplomatic persuasiveness of the Irish government, but linked to the EU's prolonged interest in Northern Ireland. The EU had long been supportive of the Northern Ireland peace process. Following the calling of paramilitary cease-fires in 1994, the European Commission (1995) moved quickly to create the Special Support Programme for Peace and Reconciliation in Northern Ireland and the Republic of Ireland worth €400 million for the period 1995-1999. The programme aimed to bolster moves towards peace by supporting economic development and peace-building projects (see Racioppi and O'Sullivan See 2007), and has been continuously renewed since 1995. The programme has also been supplemented by other supporting and networking initiatives in the intervening period (see Hayward and Murphy 2012).
Like the EU, the UK was also sensitive to specific Irish concerns. A number of these concerns were raised in a letter to Prime Minister Theresa May in August 2016 from the First Minister and Deputy First Minister of Northern Ireland (Northern Ireland Executive 2016). The DUP First Minister and Sinn Féin Deputy First Minister had taken different positions on the EU referendum, but they demonstrated some shared awareness in highlighting Brexit issues of specific concern to Northern Ireland. These included the potentially negative impact of Brexit on the Irish border, the energy sector, agriculture, EU peace funding and trade. In her letter of withdrawal, forwarded to the EU in March 2017, the Prime Minister acknowledged that Brexit posed particular challenges for Ireland. Her letter noted 'the UK's unique relationship with the Republic of Ireland and the importance of the peace process in Northern Ireland' and expressed a desire that there not be a return to a hard border on the island of Ireland (10 Downing Street 2017).
In the early period of withdrawal negotiations, British, Irish and EU protagonists evidently shared similar objectives in relation to preventing the imposition of a hard border on the island of Ireland. However, although all sides agreed on what the negotiations should achieve, there was little consensus about how to realise the outcome of a frictionless and seamless border between north and south. The situation was complicated in the period following the referendum as the Brexit debate in the UK became increasingly characterised by calls to patriotism, marked disagreement over what form Brexit should take, and pernicious discord between pro-and anti-Brexit camps. Westminster party politics became 'caught in a catastrophic equilibrium' with 'the two major party leaders . . . unable to offer clear, consistent and compelling plans for Brexit' (Wincott 2018, 15). As the negotiations progressed, the gulf between EU and UK negotiators, particularly on the Irish border issue, became mired in political dispute and disagreement. This hostile atmosphere did little to quell the anxiety and divisiveness which Brexit was stoking in Northern Ireland, and as the issue became increasingly politicised, so Brexit produced a polarising effect on Northern Ireland politics (see Murphy 2018).
The progress of the withdrawal negotiation process was also impacted by a series of unanticipated developments. In January 2017, a financial scandal in Northern Ireland (the so-called Cash for Ash scandal) resulted in a serious breakdown in relations between the DUP and Sinn Féin and effectively collapsed the devolved institutions and produced a political vacuum there (see Birrell and Heenan 2017). The fragility of Northern Ireland's political situation was complicated further by Prime Minister Theresa May's position on what form Brexit should take. In an attempt to satisfy pro-Brexit Conservative Party MPs, and as a means to secure support for a UK-EU withdrawal deal, the Prime Minister's Lancaster Speech (17 January 2017) signalled her support for a harder form of Brexit which would see the UK leaving the single market and customs union. In the absence of special arrangements for Northern Ireland, the UK being outside the single market and customs union would necessitate a border on the island of Ireland. The Prime Minister's ability to negotiate and agree special arrangements for Northern Ireland was derailed by her decision to call a general election in June 2017. The election produced an extraordinary political arithmetic and unexpectedly propelled the DUP into a Confidence and Supply Agreement supporting a minority Conservative Party government. The terms of the agreement involved approximately £2 billion of financial support for Northern Ireland and a commitment from the DUP to support the British government on all Brexit-related decisions (Tonge 2017). The DUP's political elevation and their king-making position at the heart of British politics angered Irish nationalists who accused the British government of abandoning all impartiality in relation to Northern Ireland (Birrell and Heenan 2020).
Relations between the DUP and Prime Minister May were not harmonious. May's early attempts at fashioning a bespoke arrangement for Northern Ireland by keeping it within the EU customs union and single market, while the rest of the UK remained outside, was roundly rejected by the DUP and their pro-Brexit allies in the Conservative Party (see Connelly 2018). A revised 'backstop', involving the entire UK remaining in the EU customs union (unless and until the UK and EU deem the arrangement no longer necessary), and Northern Ireland continuing to align with some single market rules, would have necessitated some checks on goods entering Northern Ireland from the rest of the UK. Unionists' perception of this arrangement creating an Irish Sea border between Northern Ireland and the rest of the UK, was viewed as undermining the constitutional integrity of the UK. May's formula which acknowledged the specificity of the Northern Ireland situation, and experimented further with post-sovereignist formulations for the region, although acceptable to nationalists, was not considerate of unionist fears and vulnerabilities. Unionists in general, and the DUP in particular, viewed the 'backstop' as anathema to unionist principles and as an afront to all British unionists for the way in which it treated Northern Ireland differently from the rest of the UK (see Murphy and Evershed 2020). Such a prospect would mean not just the application of different rules and regulations in Northern Ireland, it also meant that Northern Ireland would not be part of the return to a traditional form of British sovereignty as implied by Brexit.
This series of developments and decisions which materialised after the EU referendum demonstrates the immensely challenging political sensitivities which Brexit awoke in Northern Ireland. It also lays bare the sheer complexity of fashioning a Brexit withdrawal process which attends to the wider agenda of retrieving some semblance of Britain's traditional sovereignty character, but simultaneously respects and accommodates the special circumstances which pertain in Northern Ireland.

The clash of sovereignties: Brexit and the Belfast/Good Friday Agreement
The problems and challenges which have beset the UK's Brexit process link to what Gordon (2016, 333) describes as 'a captivation with the concept of sovereignty' which has shaped UK constitutionalism. Discomfort with the consequences of EU membership for British sovereignty has been an enduring and contested feature of the UK's relationship with the EU (see for example, Geddes 2013). The 'Take Back Control' rhetoric espoused by Leave supporters during the referendum alluded to Britain recovering its confiscated sovereignty. Before and after the EU referendum, discussion of Brexit by both Leavers and Remainers touched explicitly and implicitly on both the legal and political dimensions of sovereignty (see Gordon 2016). As Agnew (2020) observes however, the entire Brexit debate (and the subsequent UK-EU negotiation period) was largely oblivious to the meaning, implications and complexities of the sovereignty concept.
A central principle of British constitutionalism is the doctrine of parliamentary sovereignty (see Dicey 1996). This denotes that there is no higher source of authority than parliament and all other branches of the state are subordinate to parliament. In many ways however, this conception of parliamentary sovereignty is both inaccurate and outdated. Elliott (2004) notes that: 'The changing [British] constitutional landscape . . . is causing the net around parliamentary sovereignty to tighten'. This refers to developments in British politics including UK accession to the EU in 1973, the introduction of devolved powers for Scotland, Wales and Northern Ireland in 1999, and an increased emphasis on human rights which have (and are) undermining Britain's constitutional landscape in ways that do not sit comfortably with the key tenets of the parliamentary sovereignty doctrine. By way of example, the supremacy of EU law conflicts with the canons of parliamentary sovereignty because when the UK was an EU member state, Westminster was constrained by EU law. The 1998 Human Rights Act also challenges the doctrine of parliamentary supremacy by requiring that Westminster laws must, whenever possible, be compatible with the European Convention on Human Rights (ECHR). The asymmetrical devolution of powers to Northern Ireland, Scotland and Wales occasioned a new constitutional convention which dictates that Westminster, though legally permitted, will not normally legislate with regard to devolved matters as this risks undermining the legitimacy of the devolution scheme. In these instances, the supremacy of the British parliament is severely constrained by EU interdependence, human rights commitments and a programme of domestic political decentralisation.
In many ways, the EU referendum was a response and reaction to the ways in which British parliamentary sovereignty was being diluted by the obligations attached to EU membership. For many voters, leaving the EU was conceived as the only way for the British people and their parliament to regain control of national and parliamentary sovereignty. Such a view, that the means to regain sovereignty was to leave the EU and to recoil from the ECHR; however, failed to acknowledge the other ways in which the doctrine of British parliamentary sovereignty was already challenged by internal domestic developments. Bogdanor (2009) has characterised devolution as 'a very radical constitutional reform', one which transformed Britain 'from a unitary state to a quasifederal state'. In effect, the UK's devolution of powers to constituent units challenges the extent to which Westminster maintains supremacy. As noted by Bogdanor (1999): 'It is in constitutional theory alone that the supremacy of Parliament is preserved. Power devolved, far from being power retained, as implied by constitutional theory, will be power transferred, as dictated by political reality; and it will not be possible to recover that power except under pathological circumstances '. Devolution, therefore, fundamentally restructured the location and exercise of power in UK politics and undermines the extent to which the Westminster parliament is truly sovereign and supreme. This is particularly apparent in relation to Northern Ireland's devolution settlement which is partly a conventional exercise in territorial decentralisation, but more substantially designed to address a highly contested political situation and so incorporates extra-territorial, popular sovereignty and shared sovereignty concepts.
Before Brexit became a reality, Meehan (2014) noted 'despite continuing attachment to the old language of sovereignty, the UK has adapted how it "does" sovereignty in connection with [the EU and] Ireland/Northern Ireland'. Similarly, Todd (2017) determines that the Belfast/Good Friday Agreement, which contains the terms of Northern Ireland's devolution settlement, 'transcended the traditional statecentred meaning of sovereignty'. In line with a historic attachment to the supremacy of Westminster, the UK traditionally rejected any role for Ireland in addressing the Northern Ireland conflict, construing such involvement as interference in the internal affairs of the UK. However, British attitudes to sovereignty in Ireland and Northern Ireland evolved gradually during the conflict, and from the 1970s onwards, successive UK governments began to accommodate an 'Irish dimension' to attempts at resolving the conflict. This took different forms at different times, but at a basic level, it spoke to the UK government's acknowledgement that cooperation with the Republic of Ireland on the vexed issue of achieving peace in Northern Ireland was merited and justified. This morphed into a more determined role for Ireland which was elucidated in various agreements and joint documents. The 1985 Anglo-Irish Agreement heralded a new approach to sovereignty by placing the problem of Northern Ireland 'within a cooperative Anglo-Irish intergovernmental framework' (Tonge 2002, 127). The 1993 Downing Street Declaration and the 1995 Framework Documents signed by the British and Irish governments signalled some relaxation in British government attitudes to Northern Ireland's territorial autonomy by endorsing roles for the Irish government in cooperation with Northern Ireland, the British government and other parts of the UK (see McCall 1999; Meehan 2014). These developments spoke to a less absolutist notion of sovereignty being considered by the UK government vis-à-vis Northern Ireland (Tannam 2001).
The Belfast/Good Friday Agreement built on the foundations contained in previous British-Irish agreements. The 1998 Agreement engages with creative and innovative post-sovereign principles and institutions. In this context, the Agreement is grounded in the principle of consent and incorporates an element of popular sovereignty. The document recognises 'the legitimacy of whatever choice is freely exercised by a majority of the people of Northern Ireland with regard to its status, whether they prefer to continue to support the Union with Great Britain or a sovereign united Ireland ' (p. 3). In other words, it is for the people of Northern Ireland to decide the territory's constitutional future. The principle of consent performs two key functions (see Mac Ginty et al. 2001). Firstly, it serves to reassure those who felt threatened by the peace process (namely unionists) that Northern Ireland's constitutional status as part of the UK would not be undermined unless a majority in Northern Ireland favoured change. Given the historic dominance of unionism and unionist representation in Northern Ireland, the principle of consent relieves concerns about the possibility or likelihood of constitutional change. Secondly, by offering the prospect of constitutional change in the future, should circumstances and preferences shift, the principle of consent entices those on the extremes (namely Republicans) into engagement with the peace process. The 1998 Northern Ireland Act gives legal effect to the Belfast/Good Friday Agreement. There, the principle of consent is practically facilitated by a provision in that Act which loosely outlines the criteria for calling a referendum on Irish unity. The Secretary of State for Northern Ireland is enabled to call such a referendum 'if at any time it appears likely to him that a majority of those voting would express a wish that Northern Ireland should cease to be part of the United Kingdom and form part of a united Ireland' (Schedule 1).
Notwithstanding the principle of consent, Hadfield (1998) argues that, in theory, Westminster retains ultimate authority. It would, however, be egregious (in the extreme), following a majority decision in favour of Irish unity, endorsed by referendums in Northern Ireland and Ireland, if the Westminster parliament chose to override and deny the popular wishes of the electorate on the island of Ireland. The exercise of the doctrine of parliamentary sovereignty in such a purist form would pose immense political difficulties and would entail considerable risks in terms of peace and stability. Therefore, although theoretically supreme, in reality, Westminster is, in fact, constrained by the constitutional wishes of the electorate on the island of Ireland. Westminster's sovereignty is therefore qualified and must be exercised in accordance with the 1998 Agreement and its institutions (see O'Leary 2018, 233).
In addition to creating a Northern Ireland Assembly and Executive which operates according to powersharing principles, the Belfast/Good Friday Agreement also contains a novel set of cross-border and crossnational institutions which bear the hallmarks of a post-sovereign settlement. A North-South Ministerial Council and set of cross-border bodies ostensibly recognise and acknowledge nationalist identification with Ireland by providing executive opportunities for cross-border cooperation on issues of mutual interest. The East-West institutional dimension of the Agreement formalises contact between Ireland, the UK and its constituent units on policy areas of shared concern through the British-Irish Council (although this body does not have executive powers).
Crucially, the Agreement's blurring of sovereign boundaries happened against the backdrop of joint UK and Irish membership of the EU. Although the Agreement makes only minimal explicit reference to the EU, there is an assumption of Irish and UK membership of the Union forming the framework within which the Agreement is situated. By facilitating an open border, cross-border movement and North-South institutions, the EU was supporting the redefinition of relationships across the island of Ireland and between Ireland and the UK which were set out in the Agreement. As Hayward (2020) notes: 'the integrative model of sovereignty exemplified in the EU . . . has been fundamental to the Good Friday (Belfast) Agreement of 1998 and, therefore, to the political environment in Northern Ireland. This is why, in the most fundamental terms, a "clean break" Brexit was always impossible when it comes to this small region of the UK'. By severing the link between the UK and the EU, Brexit removes some of the practical and conceptual framework which supports the invisible border between Northern Ireland and the Republic of Ireland, and so challenges the fundamentals of the UK's existing post-sovereign political architecture in ways which are consequential for Northern Ireland.

The implications for Northern Ireland's constitutional settlement
Attempts to prevent a hard border on the island of Ireland were the key stumbling block throughout the entirety of the UK-EU withdrawal process. Prime Minister Theresa May's failure to win Westminster support for a withdrawal agreement (on three separate occasions) effectively ended her premiership. Her successor Boris Johnson faced similar challenges, but with a larger and more solid parliamentary majority which was not dependent on DUP support following the December 2019 general election, the Prime Minister was able to agree a formula which prevented a hard border on the island of Ireland, but created a de facto border between Northern Ireland and the rest of the UK. The Northern Ireland Protocol, agreed between the UK and the EU in October 2019, is a complex arrangement which allows Northern Ireland to remain formally part of the UK's customs territory, but required to apply EU customs rules and align with EU single market provisions (see European Commission, 17 October 2019). This effectively places a customs and regulatory border in the Irish Sea. Moreover, it is a permanent arrangement and not one which is superseded by the UK-EU trading relationship. It means that Northern Ireland will have a different trading relationship with the EU than that which applies for the rest of the UK. Inbuilt to the Protocol arrangement is a 'consent mechanism' which allows the Northern Ireland Assembly the opportunity to vote to either continue or suspend the trade and regulation provisions applicable to Northern Ireland. If the Assembly refuses consent, the possibility of a physical border between Northern Ireland and the Republic of Ireland is reopened.
The terms of the Protocol are deemed by unionism to be constitutionally and economically detrimental to Northern Ireland. Nationalists are also unenthusiastic about the arrangements, although their reservations are linked to their opposition to Brexit outright. Vociferous cross-party objections, and a Northern Ireland Assembly vote (on 20 January 2020), which enjoyed the support of Northern Ireland's five main political parties, withheld consent for the Withdrawal Agreement Bill, but did not prevent its passage through Westminster. The terms of the withdrawal bill, the manner of its achievement and the support it won at the centre was achieved on the back of a complicated relationship between Northern Ireland, the UK and Ireland, and typified by differing interpretations of Brexit and its meaning for UK sovereignty. This is evident in three key ways outlined below. Firstly, Brexit exposes a limited understanding of Northern Ireland in general, and the post-sovereign character of the Belfast/Good Friday Agreement in particular, among the British political elite and more widely. Secondly, the British government's approach to Brexit challenges key features of the UK's devolution settlement by undermining some devolved powers and by exposing the consequences of weak arrangements for intergovernmental cooperation. And thirdly, Brexit has disrupted a key plank of Northern Ireland's peace settlement, the bilateral British-Irish relationship.

British ambivalence
The long and tortuous road to agreement on a Brexit formula which maintains an open border in Ireland demonstrated the extent to which Northern Ireland political interests and concerns were either unknown (in the early phase of withdrawal negotiations) or readily overlooked and side-lined by the UK government (during the final phase). Prime Minister Theresa May's goals for Brexit and the Irish border were based on the UK leaving the single market and customs union, but without the need for infrastructure or checks at the border between Northern Ireland and the Republic of Ireland, or barriers to trade between Northern Ireland and rest of the UK. These goals were essentially incompatible with each other and ultimately unachievable in the absence of special provisions for Northern Ireland. British prescriptions for using technology to maintain an open border on the island of Ireland revealed the limited understanding in British government circles in relation to the political contestedness of the Irish border. As McCall (2018) notes, the use of technology at the border would not mean an invisible border, while border technology itself would likely be vulnerable to destruction for being a regressive marker of division and for being perceived (predominantly by nationalists) as undermining a key achievement of the Belfast/Good Friday Agreement. These British prescriptions for dealing with Northern Ireland's Brexit challenge illustrated a lack of appreciation for the immense sensitivities which attach to the border issue in Ireland. It reflects what Hayward (2020) describes as 'the simplistic [British] understanding of territorial sovereignty behind the Brexitian quest' which limits the space wherein complex understandings and applications of sovereignty, identity and multi-level governance as they pertain to Northern Ireland can be accommodated.
A further example of British ambivalence was demonstrated by Prime Minister Boris Johnson following his signing up to the Protocol in October 2019. Less than one year previously, in a speech to the DUP Annual Conference in Northern Ireland (in November 2018), Johnson promised unionists that he would never contemplate a border in the Irish Sea when he stated: 'no British Conservative government could or should sign up to any such arrangement [which places checks at the Irish Sea border] ' (Channel 4 News, 17 October 2019). Within a year however, the depth of his commitment proved malleable when set against mounting pressure from the pro-Brexit wing within the Conservative Party. His political u-turn was widely viewed as 'a betrayal of unionism' (Lowry 2019). This ambivalence intensified further following the publication of the Internal Market Bill on 9 September 2020. The bill included provisions detailing how trade between Great Britain and Northern Ireland would remain barrier-free after the Brexit transition period ended on 31 December 2020. Part 5 of the bill empowered UK ministers 'to prevent the application of, and unilaterally re-interpret and disapply parts of the Protocol, as well as ignore their legal obligations under both domestic and international law to enact the Protocol'. In other words, the bill included express provisions to override the UK's international treaty obligations by not complying with provisions of the Northern Ireland Protocol. According to a spokesperson for the UK Attorney General's Office: 'the Bill is designed to create a legal safety net and is in accordance with principles of parliamentary sovereignty' (Venis 2020). The British government's explicit prioritisation of parliamentary sovereignty clashed not just with the UK's international legal obligations, but also with the assumptions which underpin devolved power in the UK by restricting the legislative powers of the devolved regions in regulating economic activity (see Dougan et al. 2020). Although offending clauses related to the Northern Ireland Protocol were eventually dropped from the bill, the UK Internal Market Act -which came into force on 1 January 2021maintains provisions which undermine devolved powers.
The difficulties facing Northern Ireland during the Brexit period are complicated further by apparent British public apathy in relation to Northern Ireland. The complexity of the Irish border issue and the (perceived) intransigence of Northern Ireland political parties frustrated and exasperated many (pro-Brexit) British voters. This translated into weakening support for Northern Ireland's place in the UK. A YouGov (2019) poll illustrates the extent of the disconnect between Northern Ireland and the rest of the UK: 'Given the choice between having their preferred outcome on Brexit and Northern Ireland staying in the Union, a majority of 58% chose the former and only 18% the latter'. A 2020 YouGov poll notes that many British people feel distant from Northern Ireland and 54% would not care if Northern Ireland left the UK.
A desire to reclaim sovereignty subsequent to Brexit has shed light on the existence and extent of British ambivalence towards Northern Ireland at the level of both elite and the general public. The meaning of UK sovereignty in the era of Brexit is based on an apparently narrowing and simplistic conception of the territorial limits of the British state and increasingly ambivalent attitudes in relation to the very parameters of that same state.

Intergovernmental inadequacies
Northern Ireland's ability to influence the withdrawal process was severely constrained by the collapse of Northern Ireland's devolved institutions in January 2017. In the absence of a functioning Northern Ireland Assembly, Northern Ireland lost a forum wherein discussion of Brexit might have taken place between the various political parties. Northern Ireland, however, was not alone in being unable to influence the UK's withdrawal negotiations. The Scottish and Welsh devolved units expressed serious frustration with the lack of opportunities and gateways for inputting to the UK negotiation process and for limiting the hollowing out of devolved powers which materialised in the Internal Market Act (2020). In these ways, Brexit brought the weaknesses of the UK's devolution settlement into sharp focus and 'created new tensions between the UK government and the devolved administrations' (Greer 2018, 136). Despite the use of the Joint Ministerial Committee (JMC) mechanism, which brought together representatives from central government and the devolved units, the UK's asymmetrical devolution arrangement never fully accommodated voices and input from the UK's constituent units. Although this was particularly problematic for Northern Ireland which -owing to the collapse of devolved institutions there -was unable to participate (or even agree a detailed shared position on how to prevent a border on the island of Ireland), it was also challenging for the Scottish and Welsh devolved units who perceived that their voices and perspectives were not being adequately heard. In effect, the more coherent national positions developed by the Scottish and Welsh were stymied by the underdevelopment of the UK's arrangements for intergovernmental relations (see McEwen et al. 2020).
The weaknesses of the intergovernmental system pre-dated Brexit, but its inadequacies and flimsiness were illuminated by the EU referendum result. Territorially different referendum outcomes and emerging differences as to what form Brexit should take produced deepening tensions between central government and the devolved units. Devolved units are effectively caught in a clash between Brexit and the more traditional conception of national sovereignty which characterises it, and the reality of the devolved system which inherently challenges (and diminishes) those older sovereignty doctrines.

Bilateral tensions
The Northern Ireland peace process was built on the back of close British-Irish cooperation (see Tannam 2001). The provisions set out in the 1998 Belfast/Good Friday Agreement accommodate formal cooperation between the two states on issues related to Northern Ireland. This provision departs from the UK's historic adherence to the traditional tenets of national sovereignty by institutionalising an approach to Northern Ireland which encompassed elements of shared sovereignty including, for example, North-South institutions. The transformation in British-Irish relations during this period facilitated a progressive deepening of a previously strained political relationship and helped to sustain the peace process during troubled periods after 1998. The sense that historic enmity between the two states had been overcome and replaced with respect was deeply felt (although arguably moreso in Ireland). The strength and depth of that relationship, however, was challenged during the EU referendum campaign, and following the vote, relations between the two neighbouring states were unsettled as both embarked on different future trajectories -one inside the EU, the other outside (see O'Brennan 2019). Marked differences about how to maintain an open border on the island of Ireland created tensions between the two governments to the point where: 'the assumption that there is long-term robust mutual trust between both governments and predictability in the relationship has greatly weakened' (Tannam 2020). The notion of a shared future based on a shared approach to Northern Ireland no longer holds fast following the UK exit from the EU. The disparities between the two governments centred not just on the detail of the Brexit negotiations, but also reflected differences in terms of respective British and Irish interpretations of the Irish border issue, prescriptions for maintaining an open border on the island, and distinct British and Irish perspectives on the exercise of the shared sovereignty provisions contained in the Belfast/Good Friday Agreement.
Brexit begs questions about the nature and depth of altered British-Irish relations and how the relationship can be maintained and nurtured post-Brexit. Keeping the border open between north and south demonstrates some continuing acceptance of a shared sovereignty framework. However, the relationship has been fractured (at least temporarily) not just by the decision and act of the UK leaving the EU, but also by an altered perspective on UK national sovereignty which does not fully accommodate the post-sovereign framework within which Northern Ireland's stability is situated. In the aftermath of Brexit, the Irish government has signalled its support for re-energising and reinvigorating British-Irish relations by engaging more deliberately and effectively with the institutions created by the Belfast/Good Friday Agreement (see Irish Government 2020). The British government, however, appears less enthusiastic than their Irish counterparts in terms of engagement with the British-Irish Council (see Murphy 2018, 107) and is more circumspect in its commitment to British-Irish cooperation. Vocal support for enhanced cooperation post-Brexit, as expressed by Prime Minister Boris Johnson (Belfast Telegraph, 13 August 2020), has not been matched by more weighty UK proposals for developing the British-Irish relationship. In effect, Brexit's attempt to recapture British sovereignty jars with a bilateral relationship which challenges the fundamentals of that same sovereignty model.

Conclusion
The rupture which Brexit has occasioned in UK politics has had far-reaching consequences for Northern Ireland including economic upheaval, political instability, and constitutional uncertainty. These developments are taking place within the context of a wider process of dislocation and reconfiguration of the British political order related in substantial part to altered conceptions of British national and parliamentary sovereignty. The Irish dimension and the Irish border dilemma are both the driver and potentially the casualty of this Brexit-related rupture.
Brexit lends new significance to old arguments around borders, identity and sovereignty, and is feeding the forces of contestation and change in Northern Ireland. The EU referendum vote and the tortuous attempts to extract the UK from the EU revealed a pronounced disconnect between Northern Ireland and the rest of the UK where difficulties in acknowledging and accommodating the distinctiveness of Northern Ireland's situation and its peace settlement were encountered. Brexit highlights the limited knowledge and appreciation at the UK centre of highly politicised and contested issues with direct ramifications for different territories within the UK, and indeed for the British state itself. Relatedly, Brexit heralds in a period of political and constitutional instability for the UK polity. This is evident in relation to increased support for Scottish independence and an emerging conversation around Irish unity (see McEwen and Murphy forthcoming 2021). For Northern Ireland, the prospect of Irish unity would be constitutionally transformative and profound and may usher in a period of political instability and volatility.
A harking back to traditional conceptions of sovereignty which are not entirely compatible with a system of asymmetrical devolution, and more particularly with the post sovereign and shared sovereignty dimensions of Northern Ireland's political architecture, has complicated the UK's post-Brexit political situation. The changes and challenges unleashed by Brexit in Northern Ireland touch on some very fundamental considerations about how Northern Ireland is understood as a component and constituent part of the UK, and about how it is simultaneously accommodated by a system of devolution which is vulnerable to altered constructions of sovereignty. The outworking of these dynamics confronts not just Northern Ireland, but the UK as a whole, with fundamental questions about what model of sovereignty accommodates future constitutional stability. For Northern Ireland, Brexit is no longer solely about the UK's future outside the EU, it is also about Northern Ireland's place within a UK whose sovereign and constitutional moorings are being transformed.

Disclosure statement
No potential conflict of interest was reported by the author.

Funding
This work was supported by the Economic and Social Research Council [ES/P009441/1].