The Belfast/Good Friday Agreement and Transformative Change: Promise, Power and Solidarity

Abstract

In 2023 the 1998 Belfast/Good Friday Agreement marks its twenty-fifth anniversary. For many the Agreement projects a global image of a successfully concluded end to conflict. However, key aspects of the agreement remain under-enforced or simply undelivered: in particular, provisions related to significant and wide-ranging guarantees addressing human rights and equality of opportunity. As a result, socio-economic and cultural deficits persist, undermining the capacity to achieve a ‘positive peace’. In this article we address the question of how transformative the Agreement and associated reforms have been in addressing the root causes of the conflict and the structures that underpinned it. In doing so, we deploy Clara Sandoval's typology of different forms of societal change – ‘ordinary’, ‘structural’ and ‘fundamental’ – to guide our thinking and analysis, and tackle the most fundamental of questions in peace agreement literature and practice: whether, in fact, peace agreements can undo the fundamental causes that trigger and sustain violence. The article outlines the transformative promise of the Agreement, the multiple interlocking factors that have undermined that promise and the role of civil society in sustaining that transformative potential. Our conclusions point to a more nuanced understanding of what constitutes the ‘ordinary’ in transitional settings and a caution against the hyperbole of the transformative. We view transformative change as slothlike in its emergence, specifically grounded in progressive and cumulative re-orderings that can accompany peace processes. Rather than a moment of radical change, transformation follows from the cumulative impact of symbolic gesture, specific legal provision, procedural practice, mechanisms of accountability, and an engaged and vibrant civil society.

1. Introduction

In 2023 the 1998 Belfast/Good Friday peace agreement (the Agreement) marks its twenty-fifth anniversary.¹ For many external observers – including academics,² practitioners,³ governments⁴ and foreign relations experts⁵ – the Agreement and subsequent peace in Northern Ireland project a global image of a successfully concluded end to long-standing conflict, and the world's attention has moved on from the intricacies of the ‘troubles’.⁶ It is often cast as the global ‘poster child’ of successful peace agreement practice. Anniversaries of the Agreement prompt congregations of the great and the good in self-congratulatory mode, and Northern Irish politicians and others proselytise the success of the Agreement abroad,⁷ though domestic evangelism is generally avoided by a more sanguine local audience.⁸ In this article we address the question of how transformative the Agreement and associated reforms have been in terms of unpicking the root causes of the conflict and the structures that underpinned it. In doing so, we deploy Clara Sandoval's typology of different forms of societal change – ‘ordinary’, ‘structural’ and ‘fundamental’⁹ – to guide our thinking and analysis, and then tackle the most fundamental of questions in peace agreement literature and practice: whether, in fact, peace agreements can undo the fundamental causes that trigger and sustain violence.

In many respects the Belfast/Good Friday peace agreement is a remarkable triumph. It is a complex document, containing two unique overarching agreements. One is a comprehensive peace agreement reached in multiparty talks involving all conflict protagonists; the other is an international treaty between two sovereign states (United Kingdom and Ireland).¹⁰ The political parties involved in the multiparty talks included the largest unionist party at the time (Ulster Unionists), the then-largest nationalist party (Social Democratic and Labour Party), Sinn Féin (a nationalist party associated with the proscribed paramilitary Irish Republican Army), the cross-community Alliance Party, Northern Ireland Labour, the Northern Ireland Women's Coalition,¹¹ and two small loyalist parties associated with loyalist paramilitaries: the Progressive Unionist Party (PUP)¹² and the Ulster Democratic Party. The second largest unionist party, the Democratic Unionist Party, did not participate.

The multiparty agreement is structured into three distinct strands which address the internal democratic arrangements in Northern Ireland (Strand One), North-South relations (Strand Two), and East-West relations (relations between the British and Irish islands, Strand Three).¹³ These strands function as scaffolding in resolving the entrenched legal and political issues that were viewed as pivotal to driving the conflict. The Agreement includes significant and wide-ranging guarantees addressing human rights, equality of opportunity and rule of law safeguards, previous deficits of which had driven political alienation and exclusion and were seen as instrumental in sustaining continued violence. The Agreement, or at least much of it, has been formally implemented in the United Kingdom by the Northern Ireland Act 1998.

The Agreement constitutes part of a wider and ongoing peace process and, as such, illuminates the broader point that peace agreements are both constitutive of process and only a part of the process of peace making.¹⁴ In 1998 the peace deal was accompanied by other major reforms, most notably the reform of policing and the justice system,¹⁵ and changes to the law on religious and political discrimination in employment.¹⁶ Like many peace agreements, the implementation of this agreement is a stop-and-start affair, affirming that peace agreement success is neither linear nor easily scripted, but is worked out over time and with constant readjustment.¹⁷ The process illustrates our view that sustained peace agreement progression is generally best understood as evolution rather than transformation. This underscores Maiese's observation that ‘[t]he true nature of social and human change is obscured by the metaphor of agreement and the linear image of conflict’.¹⁸ This lesson about longevity and the interruption of peace is critical in our view to understanding the long-term ‘stickiness’ of certain peace agreements and the failures of others.¹⁹ In Northern Ireland regular crises in the wider peace process have produced periods – sometimes lengthy periods – in which the political institutions established by the Agreement have not functioned meaningfully, thus threatening the progress made and the capacity of the peace to hold. This has resulted in further political negotiations that supplement the Agreement's core arrangements (a form of mini peace agreements in their own right); these include the St Andrews Agreement (2006),²⁰ the Hillsborough Castle Agreement (2010),²¹ the Stormont House Agreement (2014)²² and the ‘New Decade, New Approach’ Agreement (2020).²³

There is much to celebrate about this peace process, most notably the ending of large-scale organised political violence and repression in the name of counter-terrorism.²⁴ However, there are important reasons for caution amidst the celebrations. Human rights groups warn of an undulating ‘rollback’ on the Agreement;²⁵ political violence has reconstituted in both different and similar patterns and remains a lived reality in Northern Ireland;²⁶ key elements of the peace agreement, particularly relating to human rights and social and economic opportunity, remain under-enforced or simply undelivered.²⁷ Disputes over the United Kingdom's exit from the European Union (Brexit) and the Ireland/Northern Ireland Protocol²⁸ have enhanced internal political disagreements, producing a situation whereby Northern Ireland has had no functioning political Executive since February 2022,²⁹ which is indicative of an enormous governance gap in a still fragile post-conflict society. Our analysis underscores the extent to which external optimism about peace process robustness often fails to see internal fault lines undermining the overall delivery of a peace agreement's promise, as well as painfully undercutting those elements which made the capacity for compromise emerge in the first place. A central concern generally and of this analysis specifically remains the serious levels of deprivation and economic disadvantage³⁰ in Northern Ireland, as well as deep disillusionment within the Protestant/Unionist community about the virtues of peace as experienced within their communities,³¹ given that such factors were conducive to the trigger for collective violence from 1969 onwards.

The realities of ‘stop-start’ transition on the ground, as well as the barriers to implementing a peace agreement in practice, dovetail with scholarly analysis calling for deeper transformative change in transitional processes and peace agreements.³² A focus on transformative transition has led to sustained calls on the ground and in the literature to redress broader economic injustice³³ and to tackle ‘structural inequalities, poverty, and social exclusion’ in the course of peace making, transition and peace enforcement.³⁴ Such demands have led to the emergence of transformative justice frameworks and discourse as an alternative to transitional justice theory and practice in post-conflict societies.³⁵ The shift in language from ‘transition’ to ‘transformation’ raises multiple questions about the deeper meaning and substance of this shift, and what it might actually mean in practice, an issue explored in this article. Language shifts activate deep-seated questions about the enterprise of transitional justice, the importance of representing change and what it means to deliver (and be seen to dispense) profound social and political changes in violent societies, as well as the ways to achieve these goals.

In parallel, ongoing failures of implementation in peace-making contexts (including but not limited to Northern Ireland) suggest that we need to consider the challenges that recent writings on transformative justice and analogous discourses have identified: specifically, despite important reforms contained in the Agreement, how transformative precisely have the Belfast/Good Friday Agreement and associated reforms been in addressing the root causes of the conflict and the structures that enabled and sustained it? To address this question, the article analyses the key legal and political texts associated with the Belfast/Good Friday peace agreement and uniquely relies on semi-structured interviews with 20 prominent leaders from a diverse representation of Northern Irish civil society.³⁶ This empirical groundwork allows us to better understand the Agreement's promise of transformation, and how and why it has not been seen to deliver in practice. The interviews provide a unique database of knowledge gleaned from key interlocutors in the conflict and its aftermath to assess the ‘health’ of the Agreement, its ongoing challenges, and the possibility for revitalisation. Our aim here is not to act as cheerleaders³⁷ for transformative change but to understand what factors, in the specific context of Northern Ireland, have impeded transformative change and identify where there nevertheless exists potential to leverage the Agreement for such change.

In reflecting on the Agreement and the changes it promises, we use Sandoval's typology of different forms of societal change – ‘ordinary’, ‘structural’ and ‘fundamental’ – to guide our thinking and analysis, though our analysis revisits and adapts some of these characteristics. For Sandoval, and adapted to this analysis, ‘[o]rdinary social change refers to everyday changes that align with dominant ideologies and structures in society’.³⁸ Structural change involves significant changes to dominant structures but without touching underlying ideologies.³⁹ Fundamental social change seeks to address both structures and ideologies and ‘occurs when various structural changes provide foundations for new dominant ideologies inspired by radically different values to those evident during the repression or conflict to flourish’.⁴⁰ Sandoval's typology is useful in bringing some order to the multiplicity of aims advocated by transformative justice scholars, which include profound social, political, legal, economic and gender reorderings at the end of conflict. Rather than focusing on any of the manifold suggestions made by transformative scholars,⁴¹ this typology helps us to reflect on what transformation might in fact entail in specific places emerging from or consolidating their post-conflict transition. A key point for us is that transformation is not a singular phenomenon in a post-conflict society, but is adduced from the sum of the many parts of change as conflict form shifts from overt and sustained violence to other forms of dispute and contestation, and ecosystems of peaceful coexistence slowly begin to take shape.⁴² We view much of the academic literature as narrowly seeking singular capture on transformation, generally identifying the binary framing of ‘working or not working’.⁴³ We see a more nuanced interaction captured by the multiple forms of change identified by Sandoval, and we take these insights further as we apply the learning gained by interviews with community leaders in Northern Ireland. We find these useful frames to apply individually and in tandem to an ongoing peace process, the promise of which is still unfolding. Moreover, our analysis identifies a more nuanced assessment of what transformation means, worked out over the decades that follow from a peace treaty being signed. In our assessment the distinctions between ‘ordinary’, ‘structural’ and ‘transformative’ often elide, and it is precisely the duration of the peace process that collapses some of these distinctions. Moreover, we are cautious in proclaiming the transformative as the essential element of profound change in a post-conflict society. Rather, this article points also towards the extraordinary potential of the ordinary in day-to-day post-conflict life.

2. The promise

The Northern Ireland peace agreement contains much that is potentially genuinely transformative both in micro and macro terms. For example, the introduction of power sharing (consociational) political institutions in Strand One is a structural change from the majoritarian winner-takes-it-all model of political democracy, which marked Northern Ireland from 1921 to 1972 (and which is still the model for central democracy in the UK).⁴⁴ The majoritarian system adopted in Northern Ireland resulted in a Unionist-dominated and controlled state,⁴⁵ the discriminatory and exclusionary practices of which were part of the conditions that were conducive to the outbreak of violence in the late 1960s.⁴⁶

The Agreement introduces substantial structural changes that move away from a sectarian identified majoritarian political system. As one of us has noted in previous work,⁴⁷ consociationalism provides significant benefits in bringing entrenched opponents into government, but consistently functions in post-conflict societies to entrench ethnic and religious divides and reward ethnic entrepreneurs. In Northern Ireland, Strand One of the Agreement provides for an Assembly elected by proportional representation and a power-sharing Executive. The recognition of the right of the people of Northern Ireland to determine whether to remain in the UK or to unite with Ireland,⁴⁸ and the creation of transnational cross-border institutions (both North-South⁴⁹ and East-West⁵⁰) demonstrates constitutional imagination and innovative international peace practice.⁵¹ The reform of policing and justice institutions – especially policing associated with systematic human rights violations for multiple decades – has resulted in a police service subject to an independent Policing Board and an independent Ombudsman. Policing reform addressed the significant problem of Catholic under-representation in the police. In 1998 less than 9% of the police were from a Catholic background; to improve this, a temporary quota to enhance Catholic police recruitment was adopted. This quota was ended in 2011 by which time Catholics represented close to 30% of the police.⁵² Some of these changes are genuinely structural, using Sandoval's terms, and hint at more fundamental change in so far as they suggest a move away from a majoritarian approach to democracy and away from assumptions about the dominant Unionist political identity in the jurisdiction. The policing change illustrates our broader point that change can appear narrow or pedantic but contains the seeds for further political and social shifts that allow for redefinition of political and legal institutions, as well as community identification within and of those structures.

Other aspects of the Agreement suggest a commitment to the realisation of substantive structural and fundamental changes in relation to economic, social and cultural issues. These portend re-orderings of societal goods and appear to firmly acknowledge that the conditions conducive to the production of violence in the first place were in the sights of the peace process. The Agreement seemed to promise much wider changes across human rights and equality, matters of economic and social justice, the need for participation and the promotion of the Irish language; it thus signalled a fundamental transformative agenda in Sandoval's terms, as will be addressed further below.⁵³ All of these elements were significant for common (if not universal) understandings of the causalities of conflict,⁵⁴ and in a long-term view of the resolution of conflict, positive and not merely negative peace appeared to be in the sights of the Agreement.⁵⁵

2.1. The power and promise of rights

Human rights and equality norms have a particular relevance and attraction in a post-conflict society. They provide a set of standards by which to assess public policy and provide mechanisms to address long-term grievances that remain unresolved from the conflict's negative human rights history.⁵⁶ This is enormously important in a conflict defined by a legacy of human rights abuses, which in turn defines the basis of division in a fractured post-conflict society:⁵⁷

… in a sense, you have the advantage if you're a human rights group that you are using relatively sort of external objective criteria and trying to convince people that these are the criteria by which we should be assessing whether we're, you know, doing the right thing or not.

During the 1998 negotiations, human rights and equality were stressed by several political parties, including Sinn Féin, the Progressive Unionist Party, the Women's Coalition⁵⁸ and (from outside the negotiation room) civil society organisations.⁵⁹ External to the negotiations civil society stakeholders were strongly engaged with those same political parties (and others) to stress the necessity of including these issues to deliver a comprehensive and lasting peace.⁶⁰

Positively, the Agreement includes substantial formal commitments to the incorporation of fundamental human rights and thus was seen to foreshadow both structural and transformative change. The inclusion of human rights is a ‘centrepiece of the deal’ and distinguishes it from earlier proposed peace texts, such as the Sunningdale Agreement in 1973, which was an early and unsuccessful attempt to resolve the issues driving the turn to political violence.⁶¹ The 1998 Agreement provides for the incorporation of the European Convention on Human Rights (ECHR)⁶² – a catalogue of largely civil and political rights,⁶³ creates a new Northern Ireland Human Rights Commission and Equality Commission of Northern Ireland,⁶⁴ provides that the new Northern Ireland Human Rights Commission would be tasked to advise on a Bill of Rights that would supplement the rights contained in the ECHR.⁶⁵ The UK government also indicated in the Agreement that it would consider signing the European Charter for Regional or Minority Languages and, in that context, would take steps to promote the Irish language while also recognising the importance of Ulster Scots.⁶⁶ In a backward view, these kinds of commitment may appear unremarkable, perhaps ordinary, and singularly any one of them might not appear to be transformative in its own right. However, both at the time and since, it is the totality of these legal and political commitments that gave a transformative direction of travel to the post-conflict space in Northern Ireland and defined the Belfast/Good Friday Agreement as unique among peer instruments.

Multiple commitments to equality and non-discrimination run through the Agreement. There are references to equality and non-discrimination in relation to human rights. There is recognition of the equality of civil, political, social and cultural rights⁶⁷ and of the right to be free from discrimination,⁶⁸ and the novel principles of parity of esteem and ‘just and equal treatment’, to underscore the necessity for non-discrimination in a society previously defined by distinctions based on religious and ethnic markers.⁶⁹ The Agreement anticipated the strengthening of the legislation addressing religious and political discrimination.⁷⁰ The deputy leader of the nationalist SDLP political party expressed the view that ‘[p]romoting equality – both individual and communal – is an essential part of the new political agenda and it will be a cornerstone of the structure we will create’.⁷¹

The Agreement included a progressive and innovative legal step by introducing an equality mainstreaming measure in the form of a statutory duty on public authorities to promote equality of opportunity.⁷² This measure would become known as the ‘section 75 duty’, as it was given legislative form in section 75 of the Northern Ireland Act 1998. The section 75 duty was intended to put formal equality at the heart of public administration in Northern Ireland and to create a more participatory model of governance.⁷³ Compared with other peace processes concluded in the same time frame, the thread of human rights institutionalisation runs firmly through the Northern Ireland peace agreement, and demonstrates a keen understanding that lofty principles of human rights in a peace agreement would be insufficient alone to address the legacy of sustained human rights violations and the transformative demands for human rights protection that defined the negotiation and its conclusion:⁷⁴

So social and economic transformation, if you want to put it that way, would have to be part of a peace settlement. … the actual text of the agreement over and over again gives emphasis to equality, to the basic division, if you like, in society between Protestants and Catholics and also to the question of economic development. So, in that sense, you can't distinguish the idea of a rights-based society from one that is fairer, at least in terms of some of the main divisions in society, which explicitly include gender in the text of the agreement.

The Agreement also addresses matters of economic and social justice which were defining of the causalities and perpetuation of conflict.⁷⁵ Here, the connection forged between the substance of the Agreement and the practice of fundamental societal change is, at face value, substantial. Many peace agreements have tended to co-opt the language of civil and political rights and it is these first-generation rights that define and shape the ‘rights content’ of the majority of peace processes.⁷⁶ The Northern Ireland peace agreement is unusual in that social and economic rights (or issues), and procedural protection for those rights, are proclaimed in the Agreement, thus anchoring human rights in concrete mechanisms to deliver them in practice. This link is illustrated by the statutory equality duty, which is a key part of the Agreement for those concerned about poverty, but it is not the only one.

The Agreement further refers to the concepts of targeting social need and objective need. Embedded in the Economic, Social and Cultural Issues section, the Agreement committed to a ‘new more Focused Targeting Social Need initiative’, measures to combat unemployment and to eliminate the unemployment differential between the two communities ‘by targeting objective need’.⁷⁷ The reference to the unemployment differential related to the fact that the 1991 census indicated that Catholics were significantly more likely to be unemployed and, in particular, to be unemployed long term.⁷⁸ This specificity of language in the Belfast peace agreement moves it, in our view, beyond the purview of ‘ordinary’ change and towards advancing structural and foundational moves in Northern Ireland. This move is not delivered in a single transformative moment, but rather follows from the cumulative impact of symbolic gesture, specific legal provision, procedural practice, mechanisms of accountability, and an engaged and vibrant civil society. It is the sum of these constituent parts that brings us to deep-seated social and economic change in Northern Ireland. Self-evidently, the process is ‘in motion’ and not yet complete, and the challenges that face the peace process are most clearly felt in this realm of social and economic change. As explored further below, the depth and delivery of those changes in practice are the subject of contention, particularly among civil society actors, but their inclusion is indicative of a stance leaning towards structural economic and social change that is distinctive in peace agreement practice.

Structural transformation is also found in the ways by which the Agreement looks to a more participatory model of governance, and the potential for transformation of government portends ways in which other substantive social and economic changes can be pressed into action. For example, the Agreement refers specifically to the ‘right of women to full and equal political participation’,⁷⁹ and the UK government undertakes to promote ‘the advancement of women in public life’.⁸⁰ As a result, many feminist scholars and practitioners have heralded the Northern Ireland peace agreement as a striking example of what the inclusion of women in the negotiation process for the ending of conflict can achieve in practice for their inclusion in post-conflict political life.⁸¹ While a highly rosy assessment of gender inclusion followed the signing of the Agreement, more sanguine and tempered analysis has demonstrated just how hard it is to undo the patriarchal and gender exclusionary politics of a conflicted society.⁸² Nonetheless, the gendered landscape of peace has been reshaped by the Agreement and recognition of the central role played by women in ending the hostilities, mostly among male combatants, and ‘holding the peace’.⁸³

Post-conflict inclusion practice is also found in other parts of the Agreement, including the section 75 duty, which could provide for wide consultation on the equality implications of public authorities’ policies, and to enable civil society organisations – including ‘community groups, pressure groups and unions’ – to contribute to policy formation.⁸⁴ Here, again, structural barriers in existing ordinary politics were challenged through the peace process as mechanisms to entry were created for new actors and institutions via innovative public policy mechanisms embedded in the formalities of the Agreement itself. In this respect we claim that widening capacity for procedural participation may be the most influential and change-embedding outcome of a peace agreement. We believe that the full influence of these procedural devices will unfold over time. This is an ‘ordinary’ move, generally not seen as transformative, but it creates the kind of ‘trickle-up’ practices that reshape discourse and regulation over the long haul of a post-conflict landscape. Finally, the Agreement also provides for institutional expression of the need for wider participation by proposing the establishment of a ‘Civic Forum’ comprising representatives of business, trade unions and voluntary groups.⁸⁵ This was intended to provide for a consultative mechanism specifically on ‘social, economic and cultural issues’. The Agreement envisaged that there might, in addition, be an all-island consultative assembly, appointed by the Dublin and Belfast administrations, to consult on social, cultural and economic issues.⁸⁶ All of these layers of inclusion and institutional reordering promised a new kind of politics and a different playing field to bring in new actors, reorder old institutional hierarchies, and enable a set of interlocking changes to advance profound and ambitious structural change.

3. Power reasserts itself

Other scholars and observers have already catalogued the non-implementation of the promise and potential of the Agreement: the absence of a Bill of Rights, the dissatisfaction with implementation of the section 75 equality mainstreaming duty, the disappearance of the Civic Forum, the lack of an anti-poverty strategy.⁸⁷ How and why have these perceived failures come to pass, and where has structural and transformative change in relation to economic, social and cultural issues gone?

Our research identifies several different problems with the implementation of the transformative promise of the Agreement: hierarchy and imprecision in the text itself; the lack of an enforcement mechanism for key provisions central to structural change; proceduralism over substance; the intricacies of a consociational power-sharing arrangement; and the failure to reform key formal and informal power structures. We also guard against overstating what is transformational and what is not, given our organic and long-term view of how transformative change consolidates and moves in unexpected ways, moving ‘in whispers and not bangs’.⁸⁸

3.1. Hierarchy and imprecision in the wording of the Agreement itself

While the Agreement contains substantial language that suggests commitment to profound and structural reforms, the precise language used, on close inspection, hints at later political problems. The Agreement's formidable and symbolic subsection on ‘Human Rights’ is oriented specifically towards the protection of civil and political rights. In a classic sense this might be seen as a triumph for human rights ascendency in a peace-agreement formula. Such rights inclusions affirm some of the historical grievances that sustained the conflict and offer important pathways to structural change. Significantly, however, the human rights subsection speaks of the importance of ‘civil rights’ and ‘religious liberties’, thus signalling prioritisation of the sectarian interpretation of the conflict with an emphasis on those rights that are coded ‘orange and green’. The subsections go on to establish that the signatories affirm eight rights and, though the affirmation is symbolic, it is understood that the rights selected were those that were the easiest for which to obtain agreement from both nationalists and unionists.⁸⁹ Most of these rights are in the classic civil and political rights tradition: free political thought, freedom of religious belief, the right to pursue political aspirations, the right to seek constitutional change, choice of residence, freedom from sectarian harassment. Distinctively, given the general lack of attention to economic and social rights in peace treaties at this time, the list mentions ‘the right to equal opportunity in all social and economic activity, regardless of class, creed, disability, gender or ethnicity’;⁹⁰ thus, the advancement of economic and social rights is included but primarily through the legal prism of non-discrimination. Furthermore, while much work has been done on advancing social and economic rights through the frame of non-discrimination, such strategies have observable limits.

Given the long history of rights contestation in the jurisdiction, allied with embedded hierarchies of victimhood, the inclusion of rights as both baseline and compromise is a marriage of pragmatism and principle, embedding the hope that such inclusion might provide the platform for deeper normative change and post-conflict transformation. However, as we discuss further below, the limited enforceability of rights generally added to the marginalisation of the social and economic, and has come to have a sizeable influence on the perception of the success of the peace process, as well as its long-term and structural implications for the rights of individuals and marginalised communities.

A critical part of the debate about human rights in post-conflict Northern Ireland has been the possibility of adopting a bespoke Northern Ireland Bill of Rights.⁹¹ While the Bill of Rights would clearly support classic civil and political rights protection, the Bill envisaged in the 1998 Agreement was to provide an opportunity for civil society to advocate economic and social rights:⁹²

A Bill of Rights would be a sort of, hopefully, a good exercise in kind of civic society building. And within that, obviously, from the outset, we argued and continue to argue that socio-economic rights would be included in that.

There was broad agreement in our interview group that the protection of economic and social rights was essential for addressing deep-rooted inequalities in society, not least gender hierarchies.⁹³ However, in the black and white text of the Agreement, little detail was provided on the possible scope of the new Bill of Rights, manifesting a constructive ambiguity on the totality of rights being affirmed and advanced by the Agreement. When deconstructed, what the Agreement primarily envisaged was a process to advance discussions within Northern Ireland society on its rights-securing future:⁹⁴

[W]hen I look back on it, the Bill of Rights, we just asked for a consultation around a Bill of Rights. We didn't put anything, we didn't argue for anything to be in the text or subsequent texts about what it should contain. And we were just grateful that it was in the agreement, reference was made to it and reference made to the Human Rights Commission, gave it [giving advice on a Bill of Rights] as a responsibility to the newly created Human Rights Commission. So, there you've got a vehicle for further debate, you've got your mechanism for bringing it forward. But no principles, no detail, nothing about what it should or shouldn't contain.

The obvious lesson we and our interlocutors take from this particular constructive ambiguity is that peace treaties work best for the issues they champion when they are specific in defining their scope of action and routes to enforceability.⁹⁵ As McAuliffe warns, much turns on the specificity of peace agreements: ‘specificity of the agreement – those that are broad, ambiguous or insufficiently tailored to the context may impose limits on transformation by making every issue a recurrent source of contention’.⁹⁶ The public can be swayed by symbolic gestures in a peace agreement but, in reality, structural changes in rights enforcement demand detail and specific obligation. The lack of detail in the Agreement fundamentally reflects the (lack of) priority given to a holistic vision of human rights in the negotiations, and the ‘haphazard’ way in which they were treated.⁹⁷ However, as many scholars have advised that (in Lyons’ words) ‘all negotiated settlements are bad, to varying degrees’,⁹⁸ the conditions of the parties during a negotiation process are focused on stopping the physical violence and the particular interests underpinning it. Thus, distinct obligations are not a priority for the negotiation table. To future-proof the capacity of peace treaties to survive, the lesson we draw here is that focused engagement on rights with negotiators and political representatives is necessary in order to make concrete commitments to enforcement translate into agreement text.

3.2. The lack of enforceability of the Agreement

A further difficulty in the rights domain concerns the legal enforceability of the Agreement. It is a text of different parts but includes an international treaty concluded between the sovereign states of Ireland and the United Kingdom. While it is a ‘binding’ international treaty, there is no conflict resolution mechanism or international forum for considering disputes that arise in respect of the treaty language or differences in interpretation as to the obligations that mutually and singly bind both sovereign states. Both states are dualist in international law, so treaties are enforceable in domestic courts only to the extent provided for in domestic legislation or other legal norms.⁹⁹ Ireland maintains a long-standing reservation about the jurisdiction of the International Court of Justice in relation to ‘any legal dispute with the United Kingdom of Great Britain and Northern Ireland in regard to Northern Ireland’.¹⁰⁰ The United Kingdom famously has a ‘political’ constitution in which the highest source of law is an Act of Parliament. This lack of redress for a failure to implement, or a gap in implementation as between sovereigns, has been seen clearly by members of civil society:¹⁰¹

I think one of the most significant weaknesses in the Good Friday agreement was the absence of a dispute resolution mechanism. So, one of the problems has been that there are still on paper a number of significant provisions within the Good Friday Agreement on economic and social and cultural rights [and it] has been very difficult to get them implemented. And there's nowhere to go to for arbitration; now that would be obviously dependent on the Irish government, which would have pushed some of the things – Bill of Rights, Irish Language Act, some of the legacy recommendations – but have nowhere to go to because there is no international court or other sort of jurisdiction. … it's no wonder the EU has insisted on a legally binding protocol and provisions within the withdrawal agreement because you just cannot trust the UK to implement the agreements that it has made … But there are so many provisions within the agreement that have either not been implemented or have been implemented, then rolled back or have been implemented in a very half-baked manner.

We believe that the failure of enforcement had three observable causes. First, the inclusion of rights protection was seen as so novel in many respects that there was a mistaken and formalist view that mere articulation would function as a guarantee of subsequent enforcement. Second, there were few models of rights enforcement readily available in the peace agreement universe for negotiators to ‘dip into’. We have identified in other work how the available heuristics of peace-agreement provisions and pathways have an outsize effect on the perceptions of what is possible in ongoing peace negotiations.¹⁰² Finally, there was deep and profound disagreement over including rights in the text of the Agreement. This meant that inclusion was often seen as a ‘win’ for one side, and explicit enforcement mechanisms were simply a task too arduous in a complex and multifaceted negotiation involving decommissioning of weapons, de facto amnesty, and other highly fraught issues for both sides.

In late 2020 the weakness of the UK's dualist approach to international law and its unwritten political constitution were highlighted by the remarkable provisions in the UK Internal Market Bill, as introduced before Parliament. As originally proposed, this Bill authorised ministers to break binding international treaties.¹⁰³ Subsequently, this approach was withdrawn, but that it was even contemplated demonstrates a casual attitude towards respecting existing international legal obligations. More recently the UK government has revisited this issue; the Northern Ireland Protocol Bill 2022, if passed, will give ministers powers to adopt regulations inconsistent with the Ireland/Northern Ireland Protocol, part of the Withdrawal Agreement between the UK and the EU.¹⁰⁴

Enforceability in a universe where the perceived value of a political agreement to end political violence diminishes over time is deeply challenging. Comparative analysis of the durability of peace agreements underscores the point that the average life cycle of such an agreement is approximately five years.¹⁰⁵ We also know that peace agreements require multiple iterations before they finally ‘stick’,¹⁰⁶ meaning that the formula for ending a conflict rarely involves a one-off negotiation and the implementation phase is likely to be a long-term process.¹⁰⁷ The challenge of external enforceability is also multiplied by the lack of domestic enforceability for key symbolic aspects of the Agreement – for example, the inclusion of women in public life is a laudable idea which has no entry point for actualisation in public policy or political practice.

3.3. Proceduralism over substance

A further problem in charting the depth of change engaged by a peace agreement has been an emphasis on procedure over substance in the enforcement of its substantive provisions. This has been a particular problem in Northern Ireland with the equality mainstreaming duty in section 75 of the Northern Ireland Act 1998. Recall that this innovative public policy device was seen as a means to bring affected communities into meaningful dialogue with the government about the proposals that would affect them as a result of government policy:¹⁰⁸

So, you had a potent tool [section 75] there that was actually able to connect the agreement to communities of marginalisation and communities that were typically seen to be alienated from the people.

And¹⁰⁹

And therefore, you needed robust protections for economic and social rights in order to remove some of the causes of conflict. Now, what the agreement provided, as we know, for the equality duty and enhanced fair employment legislation, and we did end up for a brief transitional moment with some of the most robust equality legislation anywhere on the planet.

The development of those equality measures and, in particular, the statutory duty to promote equality have been criticised as profoundly disappointing by community activists and scholars alike.¹¹⁰ The quote above continues: ‘… some of the most robust equality legislation anywhere on the planet. Now, well, that's history now. And we're well behind’.¹¹¹

Section 75 of the Northern Ireland Act imposes a duty on designated public authorities to have due regard to the need to promote equality of opportunity. There have been multiple problems with implementation (and thus with producing the profound structural changes it was seen to foreshadow) of section 75. One of the key challenges in practice has been the sense (and the illustrated reality) that ‘[s]ection 75 is a process that is technically very cumbersome’.¹¹² The nature of a highly specific statutory duty brings technical and procedural weight, but the difficulty of advancing a meaningful process without burdening those who are intended to be its recipients has proved to be a particular barrier for it to deliver on its transformative potential. Its implementation and development have been highly focused on process – leading to a form of stasis that, far from being transformational, has rather entrenched status quo decision making and outcomes for vulnerable and particularly affected communities:¹¹³

[T]he extent to which that's been implemented is problematic in terms of some of the specific measures like Section 75. There's been a tendency to reduce it to bureaucratic formalism rather than make it a dynamic tool for social change.

Section 75, while bearing the weight of symbolic human rights change, has also been hampered by a lack of effective domestic enforcement. Unlike the equivalent statutory equality duties in Great Britain, the courts in Northern Ireland are reluctant to enforce section 75 obligations in judicial review proceedings.¹¹⁴ This lack of judicial teeth means that public authorities (the putative targets of action under section 75) do not consider the equality measures to have any meaningful consequences for their day-to-day implementation of policy. This institutional prevarication undermines and reworks the peace agreement formula from a symbolic and practical vehicle of change to a tamed creature that has created resentment and perceptions of non-delivery for the Agreement as a whole:¹¹⁵

It's not even that the public authorities, the people who work in these authorities are being malicious and ignoring the Section 75 duty. It's just that this is really genuinely the accepted culture, is that there's going to be absolutely no repercussions for ignoring the section 75 duty.

This reputation for being bureaucratic and lacking in enforceability has led to profound disenchantment at the grassroots level with the peace process itself, and the sense of delivery for those communities who most expected real-time change to flow from the Agreement:¹¹⁶

But, I mean, a lot of activists that I talked to are very understandably just very frustrated with the process. And they think that even when you go through this huge, cumbersome process, you don't really get a satisfactory result at the end. And it was hard to argue with that logic.

This frustration expresses a more profound point: namely, the isolation and exclusion felt by particularly marginalised communities from the benefits of a peace process. While it would be an overstatement to connect this with a return to political violence, given all that is known about festering discontent, marginalisation and lack of delivery from the state for such communities in Northern Ireland, to ignore the articulation of detachment and frustration is to miss something very important about how and why peace processes fail. In general, we view the existing literature on transition as failing to pay sufficient attention to the faltering moment, and thus to listen keenly to dissatisfaction at the grassroots. In such murmurings lies an understanding of what goes wrong with peace, and precisely why and for whom it goes wrong.

3.4. Power sharing and its discontents

A further obstacle to transformative change lies in the tension between the power-sharing (consociational) dimensions of the Agreement and its more transformative dimensions.¹¹⁷ There is a risk that the power-sharing element of the political settlement can ‘constrain deeper aspects of political transformation’,¹¹⁸ contribute to ‘ineffective governance and paralysis of policy’,¹¹⁹ stymie progressive change or, worse, result in a tendency to return to sectarian carve-ups and pork-barrel politics.¹²⁰

Indeed, in some ways the political evolution of Northern Ireland since the signing of the Agreement has seen a greater emphasis on the protection of the interests of the two main communities, at the expense of the development of a pluralistic and multi-dimensional democracy defined around intersectional rather than sectarian axes.¹²¹ The Northern Ireland Act 1998 (as amended) establishes a system of cross-community voting, which incorporates a petition of concern mechanism (a mutual veto mechanism) going significantly beyond, for instance, the petition of concern mechanism described in the Agreement. In particular, the amended legislation includes an executive veto arrangement not anticipated in the original 1998 Agreement.¹²²

The temptation in such a system may be to keep ‘both sides’ happy but that comes at a tremendous cost to the integrity of the peace agreement and, in particular, to the change agenda (connected to rights) that many observers presumed it would deliver. One of our key stakeholders outlines the stark tension between rights and power-sharing implications:¹²³

I mean, I think that that housing situation for me in North Belfast was the starkest example of that, where it was saying the people negotiated people out of the right to housing to keep each other side, each of the other sides sort of happy. And that was, I guess, what my problem again with the peace agreement is that rights, equality and rights, what should come, take us natural, were just being side-lined for the sake of keeping each other happy.

The same interviewee refers not just to the power-sharing institutions in this regard but also the way in which the ‘good relations’ agenda has been used to undermine human rights and equality initiatives, by insisting on measures that could obtain agreement and pacify different sides rather than implement transformative change.¹²⁴

The power-sharing arrangements have been used to stymie progress on issues entirely unrelated to the fundamental constitutional interests of the two main communities but which have been important for realising rights implementation across the community. One particularly salient and unexpected feature we wish to highlight has been the gendered implications of consociational political arrangements in Northern Ireland. Theoretically, consociationalism is presumed to offer better prospects for the inclusion of women and sexual minorities in political life, because it was believed that the women's only political party (the Women's Coalition) would survive and thrive, which would further pressure other political parties to include more women in their political lists.¹²⁵ In practice, our post-conflict analysis suggests that this has not been the case in Northern Ireland. The nature of power-sharing in carved out ethnic enclaves produced by consociational peace agreements appears particularly to undermine women's participation and rights in post-conflict settings. The most obvious example was the failure of the only women's political party to thrive electorally after the signing of the Agreement. We observe women being sidelined in the rush to protect the interests of the main communities or the power blocks that represent them.¹²⁶ The agreement to exclude the right of access to abortion for women in Northern Ireland for almost two decades with strong cross-party support is a case in point. Add this to the ongoing economic and social disenfranchisement experienced by women,¹²⁷ and the total costs of under-enforcement of the Agreement can lead to deep cynicism and isolation among the sectors that one would expect to be its strongest proponents:¹²⁸

We have instead been blocked by, I believe, a misuse of the Good Friday Agreement. So, at the moment, the Minister for Health says that because abortion is a controversial issue, that it needs to go to the Executive before it can be enacted. However, this was primary legislation that came from Westminster. The Good Friday Agreement whenever it was talking about controversial issues, I don't think it was talking about basic human rights. I think it was talking about sectarian issues. And they know that, they might be using the letter of the law … or the letter of the agreement in terms of the Belfast Agreement or the Good Friday Agreement, but they're certainly not using the spirit of it. In terms of us trying to use mechanisms. We've participated in consultations. We have critiqued consultations and decided not to answer them in the way that we've been directed because of how faulty we believe that they were in the first place.

It is apparent that a mutual veto arrangement in a power-sharing system can enable a block on progressive social change in a post-conflict setting. It can enable political parties to avoid hard issues, and becomes a convenient means to do ‘ordinary politics’ by clear avoidance of the ‘transformative’ via the procedural:¹²⁹

I think one of the big things that we've always come across and the peculiarities of our own sort of peace process, but it was particularly in the previous administration … Well, essentially, what you had was a veto over how that operated when it got pushed to its most, it ended up in a petition of concern getting vetoed around things. But as a daily kind of operational thing, the way we found it working was, OK, housing inequality exists, particularly impacts that what you would call the Republican community, nationalist community. The people who are going to be naturally advocates of that would be Sinn Fein, who are the major partner in government. They know that bringing it to the DUP …, that's not going to make it to the programme for government or anything like that. So, all of the sudden, they're making cold political calculations as to whether it's worth their time bringing the issue of housing equality to the Executive table when they're not going to get anywhere with it.

We recognise that power sharing was essential for the Agreement in 1998 and has been a critical part of the formula of multiple post-cold war peace agreements.¹³⁰ Despite the limitations of consociationalism in Northern Ireland, the Agreement also included elements that could in theory counter-balance the negative effects of power sharing. Unfortunately, as we address below, it is precisely these elements that have not been implemented. For instance, the Agreement had included a Civic Forum ‘which reflected ideas about the benefits of participatory forms of politics more likely to address the concerns of minority or stigmatised identities’,¹³¹ but this Forum was quickly abandoned by the (then) new Executive and was never realised.

3.5. Failure to reform key formal and informal power structures

Paradoxically, power sharing has been an obstacle to progressive change while at the same time constituting a significant revision to existing power structures in Northern Ireland. Simultaneously, power sharing reveals how other power structures, formal and informal, have not been fundamentally reordered by the Agreement. For instance, the Agreement has nothing to say about the role of the permanent civil service in the administration of government. There has been no major reform of the civil service post-Agreement, and scholars have noted the extent to which this permanent (and unreformed) core infrastructure of government has played an essential if negative role in blocking social and legal change since the Agreement was signed.¹³² This permanent infrastructure of governance, the repository of knowledge and old practice, remains consistently in place, even as the outward appearance of business as usual has changed. According to one interviewee:¹³³

We were delighted when we saw the agreement being concluded, when we saw the language which we wanted in terms of many aspects of this included. I think we sat back, we had a big sigh of relief and we congratulated ourselves. We were, I think, very naive because what we didn't appreciate, and we learned very quickly, was that people who had opposed the inclusion of some of the things which we wanted secured in this agreement, those were the very people who were charged with the actual implementation of the agreement. So, the same civil servants who had blocked things that we wanted to see included were then, after the agreement was signed, charged with its implementation … It was naive not to pay attention to, in particular, how you bring about transformation within a civil service who are charged with implementation.

The intact and undisturbed nature of the permanent institutions and personnel of governance left some of our interviewees with the view that the traditional civil service is ill-equipped to address pressing economic and social challenges:¹³⁴

I'm going to say one word is incompetence. And I mean, I work with the department every day, and some people in the department are really lovely. And one of the things that for us it faces when we train to become an advice worker, we need to know the spectrum. So, I need to know how universal credit works, I need to know how our legacy benefits work, I need to know how ESA works. I need to know everything about all benefits. When you go into government, number one, when you go into civil service – and I brought this up with them, funnily enough, two years ago when they were recruiting for universal credit workers – they put out a call, first of all, for I think it's like customer service people. They don't even recruit into a specific role. Then, people are just applying for a generic job in the civil service, not realising that they could be on the front line of dealing with people who are going to throw the roof off if they don't get their 100 pounds at the end of the week.

Our data suggests that it is not only formal power structures that have been left untouched by the process of the Agreement.¹³⁵ In parallel, unofficial patriarchal structures and representation in governance have mostly remained static and unmoved. The work of feminist scholars, in addressing the ‘nested institutionalism’ of patriarchy in the unseen institutions of government, speak powerfully to the ways in which the institutions of governance (as opposed to government) are often entirely untouched by apparently profound institutional change.¹³⁶ Hence, peace agreements often hide a deep schizophrenia in the outward appearance of change to the gender order, and the internal consolidation and maintenance of the status quo. Thus, in the context of Northern Ireland, women's participation in public life was formally included as a provision in the Agreement and, while there has been some notable political success for women, full and equal participation in public life remains tenuous at best:¹³⁷

However, in endorsing the Agreement it also endorsed ‘the right of women to full and equal political participation’, which was included in the Rights, Safeguards and Equality of Opportunity Section … and ‘the right to equal opportunity in all social and economic participation’. [Trade union organisation] fully supported and campaigned for the right for women to full and equal participation and also for the Civic Forum. However, when the institutions were re-established after the first collapse of the Assembly, the Civic Forum disappeared off the agenda and, while the participation of women in civic society has increased as has the number of women elected to the Assembly, we still have some way to go to achieve gender equality, despite the current First and Deputy First Ministers being female.

This comment highlights that there have been some obvious and high-profile political successes for individual women.¹³⁸ There has indeed been some structural change – at the start of 2020 three of the five main political parties were led by women – but this does not reflect a fundamental reframing of gender orders and gender expectations in public life in Northern Ireland.¹³⁹ The maintenance of gender hierarchies, and even their consolidation in post-conflict settings, have been well identified in multiple conflict contexts, and the remaking of gender relations in society as a whole proves surprisingly resilient, even in the most apparently progressive of post-conflict settings.¹⁴⁰ In 2016, for instance, the Executive established a high-profile and politically sensitive 13-member Commission on Flags, Identity, Culture and Tradition, which included precisely one woman.¹⁴¹ The signal that, on the most partisan and sectarian of issues dividing the communities in Northern Ireland, a proportionate voice for women was neither necessary nor desired underscores a plethora of small and large exclusions in public life since 1998. There is no strategy or policy to ensure participation in public life at different levels and this leads to the uninterrupted reproduction of gender hierarchies, particularly in a context where the traditional community divide receives heightened attention:¹⁴²

[T]he key provision of the Good Friday Agreement was the right of women to full and equal political participation, and it's just a case in point that that particular provision has sat on the page of the Good Friday Agreement, and there's been absolutely no meaningful implementation of it, and the UK government has steadfastly resisted the implementation of the UN Security Council resolution 1325 … to Northern Ireland with a very sort of colonial and paternalistic attitude that this type of thing applies to the global south but not to the UK, whereas in fact that's exactly what we need at the over-dominance of men in the conflict resolution mechanisms and even things like go back to the discussion on objective need. One of the concerning things under the last executive mandate was the Social Investments Fund, not because it didn't fund some good projects. A number of the projects were quite good, but it was the process by which monies were allocated. We went right back to direct political involvement in decision making rather than decision making independently on the basis of objective need. If you look at some of the political social investment local area committees like, well, let's take [one in] North Belfast, the percentage of women on that committee making decisions I remember was zero.

Civil society activists note that even when women have traditionally been involved in community and voluntary work, their participation is potentially reduced when these activities become more professionalised or attract more funding or prestige:¹⁴³

I think once there is some credit given for something and some perception of importance and power and money and all of those things, and certainly the stuff that wasn't remotely glamorous or interested or beneficial seeming to other people suddenly becomes much more attractive.

A consistent pattern we observe in a post-conflict society that has adopted a power-sharing political system is to focus on the traditionally perceived main communities, entrenching male privilege and representation,¹⁴⁴ and overlooking men and women who do not easily fit into the accepted main communities. The consolidation of masculine power structures is seen not only in the exclusion of women but in the marginalisation of LGBTQI persons in public political practice.¹⁴⁵ Moreover, within those communities a range of factors, including the ongoing influence of paramilitary legacies and structures, mean that women will simply not be seen as ‘representative’ of community interests and priorities. This distracts attention from other equality issues; more attention to other equality issues may help to disrupt the temptation for these systems to fall into a ‘sectarian carve-up’:¹⁴⁶

And some people were saying vote for everybody except for Sinn Fein and Women's Coalition. I could never understand that because, you know, I thought, you vote for SDLP but not the Women's Coalition? … But I think, maybe it's about kind of a threat to their culture, something that does reach across, and it is not a thought maybe thing as much. I never really sort of questioned it at the time. I just always thought it was strange.

To conclude here, the transformative potential of the Agreement has been impeded by a combination of factors, including the imprecision and hierarchies in the language of the Agreement; lack of enforcement; a tendency towards proceduralisation; elite-driven power sharing, which has enhanced sectarian division rather than overcome it; and a fundamental failure to reform key formal and informal power structures. Thus, sustained social change in a fundamental sense remains elusive in Northern Ireland, and deep structural transformative change is well out of reach because of these embedded intersecting factors. In this view, the transformative capacity of a peace agreement shows its limits: namely, that conflict ending can reduce violence but not the structures that produce and sustain it, and that the work to undo deeply embedded power and economic structures is achingly slow, and consistently undermined. The challenges follow not from the traditional ‘spoilers’ of peace agreements,¹⁴⁷ though they can play a part, but rather in the intractable undoing of long-standing beneficiaries of the status quo, in both political and economic life.

4. Power, civil society and solidarity

The resistance to implementing transformative change speaks of the resilience of existing power structures even amidst partial transformation. Other sources of power also exist; for Arendt, power is not force: ‘power corresponds to the human ability not just to act but to act in concert’.¹⁴⁸ Northern Ireland provides lessons in the power of civil society to provide a more reliable foundation for both ordinary and transformative change even in the context of resistance; and, in so doing, civil society relies on the wording of the Agreement and subsequent agreements to continue to make claims for transformative change. Civil society advocacy, litigation and action played a significant role in ensuring that the Agreement included concepts like economic, social and cultural rights, equality mainstreaming, a Bill of Rights, and women's right to participate in the first place.¹⁴⁹ Whatever the dissatisfaction with the implementation or non-implementation of these innovations, it is important that they are rooted in the Agreement as reference points, points of continued mobilisation for civil society action.

Civil society action has taken various forms, and Northern Irish civil society organisations have been adept at selecting different forms of action and navigating the ongoing uncertainty of the post-conflict space, whether it be strategic litigation, supporting equality mainstreaming, developing public platforms and coalition building. One example of strategic litigation relates to the legal commitment to adopt an anti-poverty strategy. The St Andrews Agreement of 2006 builds on commitments in the Belfast/Good Friday Agreement relating to economic and social justice by committing the UK government to publishing ‘an Anti-Poverty and Social Exclusion strategy to tackle deprivation in both rural and urban communities based on objective need and to remedy patterns of deprivation’, with the Northern Ireland Executive intended to follow this up.¹⁵⁰ Following this, the Northern Ireland Act 1998 was amended to require the Executive to adopt such a strategy and to keep it under review.¹⁵¹ By 2015 the Executive had failed to adopt such a strategy despite the explicit legal obligation to do so. In the face of Executive inaction on this front, a local non-governmental organisation (NGO), the Committee on the Administration of Justice (CAJ),¹⁵² secured a legal ruling that the Executive had failed in its obligations under the Northern Ireland Act.¹⁵³ This confrontation has been a precedent for similar litigation, seeking to force the government to live up to the promises of the Agreement and the domestic legislation that was intended to implement it.¹⁵⁴ This kind of strategic litigation – which pressed for the enactment of essential legislation to address the poverty that was causal to the conflict, exacerbated by it, and the alleviation of which is necessary to bring about a transformed social and economic dividend for marginalised communities – is a tactic to force transformation even as it appears unwanted by the government.

Civil society has also developed more explicit political programmes as well as legalistic tactics, with plans for a Feminist Recovery¹⁵⁵ and a Manifesto for a Rights-based Return to Power Sharing.¹⁵⁶ Much of this work has been made possible through sustained solidarity and coalition building within Northern Ireland civil society.¹⁵⁷ These practices of intersectional coalition building predate the Agreement but have, in important ways, been strengthened by the process of formal political negotiations. The Human Rights Consortium, for instance, is an umbrella group comprising 160 different organisations (NGOs, community groups, trade unions and charities) working to support the advancement of human rights in Northern Ireland and, in particular, the need for a Bill of Rights.¹⁵⁸ The Equality Coalition is a network of more than 100 organisations coordinated by CAJ and a trade union, Unison. As well as Unison, the Coalition includes several other trade unions.¹⁵⁹ The involvement of trade unions in this work is significant as they are among the most powerful and well-organised elements of civil society.¹⁶⁰ Our point here is to underscore that the long-standing ‘ordinary’ work of organising, advocating and arguing for human rights has remained at the centre of propelling social and economic change in Northern Ireland as a post-conflict society. We argue that this ordinary and sustaining work, which is often invisible to the assessments of how and why change occurs in a post-conflict society, should not be ignored and needs to be better understood as the bulwark that sustains change.

The record of civil society in Northern Ireland is an impressive one,¹⁶¹ but we also acknowledge the distinct challenges here. The fact that there is so much reliance on civil society is itself an indictment of the failure of formal political institutions, and an obstacle to achieve the required social changes. We also acknowledge that a peace agreement, a time of transition, creates challenges as well as opportunities for civil society.¹⁶² Civil society organisations often face financial challenges, the loss of personnel, emotional and physical exhaustion and, in extreme cases, physical violence.¹⁶³ Without support, civil society cannot be relied on to remedy the defects of an inadequately implemented peace agreement. The experience of Northern Ireland emphasises one crucial area of support that can sustain such organisations, which consists of solidarity, organised and consistent partnership, and common cause among the various strands of civil society. The civil society's collective action might contribute to achieving the potential of the ordinary, structural and fundamental changes embedded in the peace agreements. However, this requires an effort from the state institutions to create effective spaces for civil society's participation in all phases of the process: design, adoption and implementation of the measures and policies, and ensuring political space for maintaining evaluation and accountability processes aimed at generating both ordinary and transformative changes.

5. Conclusion

The Belfast/Good Friday Agreement remains an important step in Northern Ireland's peace process and the transition from conflict. As we have seen, it is a multi-layered text with commitments across many different areas ranging from internal democratic structures, cross-border institutions, human rights and equality, to cultural issues (including the Irish language), and economic and social issues. In some respects it seems to offer a promise of post-conflict transformation in a unique and multi-dimensional way.

The promise of this kind of deep transformation has not been delivered. Ambiguities and hierarchies in the text, the lack of enforceability, the attractiveness of proceduralisation as a barrier to substance in change politics, the power-sharing systems themselves, the ambivalent attitude of one of the state parties to international obligations, and the failure to address different power structures has meant that many obligations have not been implemented or have been disappointing in their implementation. While some aspects of the Agreement have enabled structural change, especially in relation to core political institutions (and policing), and their cumulative effect has been significant in the post-conflict era, the deficit in implementing substantial provisions of the Agreement has spawned deep disappointment and a sense of opportunity lost for fundamental and meaningful transformational change. At the same time the Brexit debate and subsequent Ireland/Northern Ireland Protocol have unsettled the peace process; while the Agreement had found nuanced solutions to difficult challenges around the border, sovereignty and identity, Brexit has revived debate about those very issues.¹⁶⁴ Controversy over the Protocol has led directly to the non-functioning of the internal democratic institutions since early 2022.

Some of the problems we have identified lie in the legal nature of the Agreement, including the terminology used and the lack of formal enforceability in the text of the political agreement. One could argue that clearer and more enforceable legal rules would remedy these deficits, but we have to acknowledge two limitations. The first is that the Agreement, like all peace agreements, is itself the product of a political process and not one primarily of a legal nature; as Bell puts it, institutions ‘find their way into agreements as a direct result of a political bargaining process rather than principled design’.¹⁶⁵ Expectations as to what is possible have to be tempered by that realisation. Secondly, we acknowledge the limitations of law as a tool of transformation. Even if there are important legal reforms, as feminist scholars have long-reminded us, the law is likely to offer only piecemeal and incremental reform, often flawed by commitments to precedent and procedure over substance, and if power structures are not transformed, the law itself is an unlikely engine of transformation.¹⁶⁶ Similar criticism can be made more specifically of rights. Acknowledging these critiques, at the same time we do not reject the importance of law and rights; we endorse a critical approach to rights that does not reject their potential for supporting demands for more transformative change.¹⁶⁷

Nevertheless, the case study of Northern Ireland should not be grounds for unremitting pessimism about the potential for substantive, deep and meaningful, social and legal transformation in a scarred society where the hurt of decades sits in the realities of people's lives. Sandoval stresses that fundamental change is an intergenerational project that is long term and requires patience. In this agreement, as in others, we still see potential to deliver on the full promise made in the spirit of ending violence and committing to a new beginning for the whole of society. The most hopeful transformative message lies in the experience of Northern Ireland's civil society, which has played a significant role in securing elements of the Agreement and which, since 1998, has impressively advocated the transformative potential of the Agreement with no end-date on its work.