In an important sense, the report can be seen as one of many illegitimate attempts to increase unionist influence over a northern border poll. Some commentators would deepen the unionist veto over constitutional change by altering the Good Friday Agreement’s 50% + 1 threshold for winning a border poll.[1] The Working Group wisely rejects such proposals. But the Working Group would deepen the veto in a different and equally unwarranted way, by altering the GFA’s very conception of a border poll. The final report sweeps away the settled meaning of consent and radically extends the reach of the northern veto well beyond the question of a change in constitutional status. The Working Group offers no defensible argument for its marked departure from the provisions of the Agreement.
The Working Group’s final report comes six months after its interim report (Working Group, 2020). Since the two reports are in all essentials the same, my critique of the interim report still stands (Burke, 2021). In this review, I aim both to add to and expand on some of the concerns I raised in the earlier piece.
The Working Group has 12 members based in universities in the north, south, Britain and the United States. Its members have expertise in an impressive array of relevant areas including constitutional law, democracy, elections and referendums, Irish-British relations, divided societies, devolution, power-sharing, and human rights.[2] It is an exceptionally credible academic voice in the discussion of Ireland’s future. It will be influential. The media warmly received the recent report, especially for its sensible recommendations that there needs to be detailed planning in advance of any border polls and clear rules of conduct for the referendum campaigns (Creighton, 2021; Downing, 2021; McBride, 2021; Irish Times, 2021). Sinn Féin approvingly cited the final report as a useful path forward (Finucane, 2021; McArdle, 2021).
My fear is that the Working Group will be too influential. One of its stated objectives is to stimulate informed discussion about possible referendums and everything they entail. Engaging people for knowledgeable conversation is a worthy aim. At the same time, the Working Group has established a formidable presence in a debate that is just getting off the ground. Its two reports number some 550 pages, not counting the prolific output of individual Working Group members. There is every reason to believe that these individuals will remain weighty players in the ongoing deliberations. Other participants might consider whether some major terms of the debate have been already and prematurely set.
The Working Group recognizes that its final report is not the last word in the debate. I trust it will not be the dominant word on referendum processes and sequences. People might wish to approach the final report with a critical eye: take what is good in the report—and there is much good in it—and discard what is bad. My purpose here is to highlight some of the bad.
Two related falsehoods permeate the Working Group’s final report and taint much of its analysis. Individual members of the Working Group repeat these falsehoods in their own writings. The first is the Working Group’s claim that its analysis is procedural, not political. It purports to offer solely technical advice on how best to run unification referendums (Working Group, 2021, paras 1.1, 1.6, 1.24, 10.6, 10.30, & 15.2). I disagree. The Working Group constructs its elaborate referendum procedures on a highly political foundation.
The second falsehood is more an assumption than an outright claim. For the Working Group, the GFA defines the basic framework by which a northern border poll must be held (paras 1.32 & 4.1). But because the GFA is silent on some crucial issues concerning the conduct of a poll, the Working Group presumes that it has wide latitude to insert whatever meaning it deems appropriate (paras 9.31 & 15.5; Renwick, 2021; Renwick & Hayward, 2021; Whsyall, 2021). Again, I don’t agree. The GFA is not such a blank slate. Ascribing meaning to its sometimes vague and ambiguous provisions, or filling in its elisions, needs to be consistent with a fair reading of the plain words of the text and with the political context in which the Agreement is embedded.[3] The Working Group’s understanding of the very essence of a northern border poll is seriously deficient on both these scores.
Consent to What? The Nature of a Border Poll
The Working Group examines in detail five different designs for holding unification votes. It finds that three of them—referendum configurations 2, 4 and 5—merit further consideration.
The Working Group is especially interested in procuring unionist engagement in and acceptance of the detailed design of a united Ireland. In summarizing the potential benefits of one of its favoured configurations, for example, it states:
This may be seen as the approach that, symbolically as much as materially, could maximise unionist involvement in developing the structures of the new state. It would offer a chance for unionist names to be on its title deeds (para. 9.49).
The Working Group finds a related configuration attractive because it could include “safeguards to protect northern or specifically unionist concerns” (para. 9.58).
Here, we begin to see the Working Group equating northern consent with unionist consent. Outside the final report’s chapters on historical context and the state of opinion on the constitutional question (chaps 2 & 3), northern nationalists scarcely appear. They certainly do not come into view as a distinct group that might have its own concerns about the design of a border poll. Marginalizing northern nationalists and nationalists generally is an unfortunate trend in many recent academic analyses of referendum planning.[4]
It’s reasonable for academics (and others) to consider the interests of those groups that may feel most threatened by the possibility of major change. But it’s also reasonable to recognize and be sensitive to the concerns of those who worry that their long-standing aim of constitutional change may be unfairly impeded, unjustifiably sidetracked or willfully frustrated.
The constitutional debate would benefit from a truly inclusive and more balanced consideration of the referendum concerns of various groups and constituencies. Still, it’s difficult to argue with the Working Group’s objective of enhancing unionist participation and gaining unionist acceptance. But the Working Group tries to secure this objective in an entirely unacceptable way, by undermining the Agreement’s notion of consent.
A simple question captures the core of my disagreement with the Working Group. What are people in the north being asked to consent to when they vote in a border poll? I maintain that they are being asked to consent to the principle of unification, to the basic question of sovereignty. The Working Group maintains that they are being asked to consent to the detailed arrangements of a united Ireland.
There is a huge and politically-charged difference between the two approaches. Perhaps the best way to illustrate that difference is to examine the different questions that would appear on the referendum ballot. The Working Group itself gives useful examples of such questions.[5]
In the approach I support:
This ballot question closely resembles the wording that appears in the GFA and that was used in the 1973 border poll. The Working Group explicitly rejects this option (paras 9.26-28).[6]
In the Working Group’s approach, the referendum would offer voters either a detailed model for a united Ireland or a detailed process for determining the structures of a united Ireland. All three of the Working Group’s favoured configurations involve such detailed designs. In this instance:
The principal difference between this approach and the previous one is that here the detailed arrangements of unification appear on the referendum ballot in the form of the named document.
Let’s assume that the document cited on the referendum ballot sets out detailed proposals for the model of a united Ireland.[7] What might these proposals entail? What might be the terms of unification that are on offer to referendum voters in the north?
The Working Group argues that:
Here, we begin to see the Working Group equating northern consent with unionist consent. Outside the final report’s chapters on historical context and the state of opinion on the constitutional question (chaps 2 & 3), northern nationalists scarcely appear. They certainly do not come into view as a distinct group that might have its own concerns about the design of a border poll. Marginalizing northern nationalists and nationalists generally is an unfortunate trend in many recent academic analyses of referendum planning.[4]
It’s reasonable for academics (and others) to consider the interests of those groups that may feel most threatened by the possibility of major change. But it’s also reasonable to recognize and be sensitive to the concerns of those who worry that their long-standing aim of constitutional change may be unfairly impeded, unjustifiably sidetracked or willfully frustrated.
The constitutional debate would benefit from a truly inclusive and more balanced consideration of the referendum concerns of various groups and constituencies. Still, it’s difficult to argue with the Working Group’s objective of enhancing unionist participation and gaining unionist acceptance. But the Working Group tries to secure this objective in an entirely unacceptable way, by undermining the Agreement’s notion of consent.
A simple question captures the core of my disagreement with the Working Group. What are people in the north being asked to consent to when they vote in a border poll? I maintain that they are being asked to consent to the principle of unification, to the basic question of sovereignty. The Working Group maintains that they are being asked to consent to the detailed arrangements of a united Ireland.
There is a huge and politically-charged difference between the two approaches. Perhaps the best way to illustrate that difference is to examine the different questions that would appear on the referendum ballot. The Working Group itself gives useful examples of such questions.[5]
In the approach I support:
Voters would be asked for their view on the basic principle of sovereignty. In Northern Ireland, the question, at its simplest, could be ‘Should Northern Ireland leave the United Kingdom and become part of a united Ireland?’, with response options ‘Yes’ and ‘No’ ... (para. 9.16).
This ballot question closely resembles the wording that appears in the GFA and that was used in the 1973 border poll. The Working Group explicitly rejects this option (paras 9.26-28).[6]
In the Working Group’s approach, the referendum would offer voters either a detailed model for a united Ireland or a detailed process for determining the structures of a united Ireland. All three of the Working Group’s favoured configurations involve such detailed designs. In this instance:
Voters in Northern Ireland might be asked ‘Should Northern Ireland leave the United Kingdom and become part of a united Ireland on the terms set out in [specified document]?’ ... (para. 9.29).
The principal difference between this approach and the previous one is that here the detailed arrangements of unification appear on the referendum ballot in the form of the named document.
Let’s assume that the document cited on the referendum ballot sets out detailed proposals for the model of a united Ireland.[7] What might these proposals entail? What might be the terms of unification that are on offer to referendum voters in the north?
The Working Group argues that:
The proposals put to voters would ... encompass the proposed constitutional structures and policy arrangements with which a united Ireland would begin its existence (para. 10.19).
- That is, voters in the north could be voting on many terms and conditions, including:
- the basic form of government in a united Ireland, be it unitary, fully federal or devolved;
- the status of the Irish language and national symbols like the flag and anthem;
- the standing of the capital city;
- taxation and finance arrangements, health and welfare provision, education policy, rights and equality issues, and the nature of the policing and justice system; and
- Ireland’s place in international affairs and membership of international organizations (paras 7.56-78).
Most crucially, if all these details appear on the ballot in a border poll, they are subject to northern approval. None of the governance structures and policy dispositions of a united Ireland can be set without northern consent. Northern acceptance becomes an absolute condition of maintaining current arrangements or changing them. The Working Group in effect gives to the north (read unionists) a comprehensive veto over many foundational aspects of a united Ireland. In a sense, the Working Group resurrects the general policy veto that John Hume thought the 1985 Anglo-Irish Agreement had eliminated, with the difference that now the unionist veto is exercised over Irish rather than British policy (Hume, 1989).
To be curt, the Working Group’s position violates central tenets of the Agreement.
The Missing Argument
Consent and a border poll are about the principle of unification, as the GFA says. The Working Group is simply mistaken to claim that they are, or can be, about the detailed design of a united Ireland. The Working Group’s position is considerably weakened by its failure to offer any convincing argument supporting its point of view.
In this section I examine the extent to which the favoured referendum configurations accord with the GFA, a criterion the Working Group calls legal legitimacy (paras 5.6-8). My position is that any configuration deemed acceptable must meet the minimum requirement of legal legitimacy. Later I’ll explore the important and related requirement of democratic legitimacy.
The Working Group introduces two different standards to judge if a referendum configuration is legitimate according to the GFA. One is whether the configuration “clearly falls within the terms of the 1998 Agreement” (para. 9.17). But the Working Group does not use this standard to judge the three configurations it recommends as worthy of further consideration. It never says that these configurations “clearly fall within the terms of the Agreement.” Instead, it uses a second, weaker and looser, standard. It says its three favoured configurations are “compatible” with the provisions of the GFA (paras 9.31, 9.47 & 9.59). We need to probe this elusive notion of compatibility that allows the Working Group to claim allegiance to the GFA while plainly deviating from it.
To support its position, the Working Group says that “nothing in the Agreement precludes” having a unification referendum on detailed designs (para 9.31).[8] The argument is, I suppose, that the three recommended configurations are “compatible” with the Agreement because the Agreement does not specifically rule them out. This argument is feeble in the extreme. The GFA does not specifically preclude lots of things. Can all of these things be ruled in? Is it possible even to compile a list of such non-precluded things that may someday be ruled in?
More to the point, the Working Group’s argument is meaningless because it supports two fundamentally different conceptions of a border poll: one sees the poll as a vote on the principle of unification, the other as a vote on the detailed design of a united Ireland. And these two kinds of border polls are based on two incompatible understandings of consent. The first type of poll gives the north a simple veto over a change in its constitutional status, the second bestows upon the north an all-embracing veto over the structures, institutions and policies of a united Ireland.
We must look elsewhere—past the word games of “compatibility” and “not precluding”—in search of a credible argument that can sustain the Working Group’s interpretation of consent and a border poll. In its final report, the Working Group asks other participants in the constitutional debate to produce evidence in support of their alternative interpretations of the GFA; and it rejects those interpretations that are unsupported by evidence.[9] Let’s ask the same of the Working Group. And let’s use its threshold of acceptance: a valid interpretation must be backed by sufficient evidence.
In hundreds of pages across two voluminous reports, the Working Group does not produce a single piece of evidence to substantiate its position that consent and a border poll are about the detailed arrangements of a united Ireland. Since there is no evidence at all, the question of what might constitute “sufficient” evidence does not arise. But certainly, the total absence of evidence is insufficient. The Working Group’s interpretation must be rejected.
There is nothing actually in the GFA stating, suggesting or implying that consent and a border poll pertain to the detailed terms and conditions of a united Ireland. There is not a hint of this view in any of the documents codifying consent since consent was defined in popular terms in 1972-73. And the north’s only experience of a border poll in 1973 was a vote on the basic question of sovereignty, not detailed designs.[10]
The Working Group concocts, out of nothing, the notion that border polls set the constitutional structures of a united Ireland, define its principal institutions, specify its national symbols and establish its major policies. And it manufactures the related notion that the north has a comprehensive veto over all these matters. The Working Group’s attempt to effect elemental change in the Agreement is without foundation and bereft of merit.
Text and Context
The Working Group itself presents a compelling case against its own interpretation of consent and unification referendums. It recognizes that a referendum on the basic question of sovereignty “clearly falls within the terms of the 1998 Agreement, which appears to envisage a simple process whereby the Secretary of State would call a poll, which, in the event of a vote for unification, would be followed by negotiations and implementation” (para. 9.17). In other words, the very text of the Agreement says that a referendum concerns the simple principle of unification. The vote is not about the detailed form of a united Ireland. That comes later, in the post-referendum stage of negotiations.
The passage quoted above also says something essential about the political context of the Agreement. The Working Group is absolutely correct in saying that the GFA envisages the straightforward sequence of a vote on unification followed by detailed negotiations and implementation. This depiction of the process is over-simplified and abbreviated. But it does represent a vital component of the immediate political context in which the peace negotiations were conducted. It describes the parties’ understanding of the constitutional provisions that were on the negotiating table.[11] More centrally, it describes the voters’ understanding of those provisions as they moved to ratify the Agreement in dual referendums in May 1998. Everyone at that time thought that a border poll was about the basic question of sovereignty. There was no reason for them to think otherwise.
The Working Group’s position on unification referendums has no legal legitimacy. It’s not supported by the plain language of the Agreement. But it also lacks democratic legitimacy, in two senses. First, for the Working Group to say, now, that a border poll and consent are about something completely different from what voters understood in 1998 is to upend the democratic ratification of the Agreement.[12] It is materially to change the provisions of the Agreement after the Agreement was endorsed by voters in the north and south.
The Working Group’s interpretation also lacks democratic legitimacy in a second, related sense. It is not in the gift of the Working Group to grant the north a comprehensive veto over the detailed design of a united Ireland. It is for the Irish people, not a collection of academics, to determine if such a veto is warranted. Any proposal to deepen the veto should be put explicitly to the people so that they might decide democratically. It was not so put in the GFA ratification votes in 1998, however much the Working Group wishes to smuggle in a sweeping northern veto based on a seriously flawed reading of the Agreement’s referendum clauses.
Rory Montgomery, a member of the Irish team that negotiated the GFA, notes:
The bedrock of the Agreement was an intricately interwoven and balanced set of principles, understandings and commitments regarding the constitutional status of Northern Ireland and the manner in which that status could change, and a united Ireland be established” (Montgomery, 2021, p. 84).
In a decidedly political intervention, the Working Group tips this intricate balance in favour of unionism. It fundamentally and unjustifiably rewrites what it admits are “keystone” provisions of the wider peace settlement (para. 8.5).
The Scottish Model
There is a critical difference between having the details of a united Ireland available before a referendum and having a referendum on the details. All sides currently involved in the debate about Ireland’s future want the details available. They agree that rigorous and inclusive planning about what a united Ireland might look like needs to take place in advance of any referendum. They also agree that the work to delineate clear processes to facilitate constitutional change must start now. This position is reasonable and sensible.
It’s neither reasonable nor sensible for the Working Group to propose a referendum on details that inserts a thoroughgoing northern veto over the shape of a united Ireland. The Working Group seems to take the position that the only details that matter are the ones that appear on the referendum ballot. It is propagating what might be called the myth of the ballot paper.
Scotland’s experience explodes this myth. The Scottish independence referendum in 2014 shows that it’s possible to have a vote on the basic question of sovereignty and at the same time to give voters a detailed account of what the new state might look like. The question used in Scotland was about sovereignty, pure and simple: ‘Should Scotland be an independent country?’, with voters required to vote yes or no (Electoral Commission, 2014, p. 4). Ten months before the referendum, the Scottish government published Scotland’s Future, a 670-page guide to an independent Scotland that laid out the government’s vision and priorities for action (Scottish Government, 2013). This document covered at least as many issues as the Working Group includes in its detailed designs. The government’s guide examined the transition to Scotland’s new constitutional status as an independent state. It also set out the direction in which the new state might go regarding finances and the economy, social services, education, human rights, international relations, justice and security, culture, and democratic development.
The Working Group rejects this Scottish model, unfairly I think.[13] It seriously underestimates and undervalues the detailed information that was available to Scottish voters, both through the government’s comprehensive guide and the various types of information that are naturally generated by parties, organizations and the media both before and during a referendum campaign. The Working Group seems to discount all that information because it was not entrenched in a confirmed and definitive plan that appeared on the ballot paper. But that information proved very useful to Scottish voters. The Scottish Electoral Commission found that voters had exceptionally high levels of knowledge of what the referendum was about: “This is the highest level of knowledge recorded for all major elections since we started asking this question in our post-election/referendum surveys.” The vast majority of voters felt that they had enough information “to make an informed decision” in the referendum (Electoral Commission, 2014, paras 3.71 & 3.76, respectively).
The Scottish model, or some variation of it, is appropriate for borders polls in Ireland.[14] Above all, it has both legal and democratic legitimacy. It is legally legitimate in that it fully complies with the constitutional provisions of the Agreement. It is democratically legitimate because, first, it respects the choice that voters made in ratifying the GFA in 1998 and, second, it doesn’t usurp the right of the Irish people directly to decide what role, if any, a northern veto might play in a united Ireland.
In this model, the Irish government might follow the lead of its Scottish counterpart and publish a detailed guide to the new constitutional order. Other groups and parties might endorse the government’s proposals. Or they might not, in which case there may be several proposals on the field. In any case, voters in Ireland are likely to have more than sufficient detail to make an informed decision, as had voters in Scotland. None of the proposals before voters is confirmed or definitive because final negotiations on the form of a united Ireland will occur only after concurrent referendum decisions to unify. The citizens of the newly-reunified state have the ultimate authority to set the terms and conditions of their own governance.
Just as participants in the debate about Ireland’s future urge advanced planning, so do they favour securing unionist participation in the process and unionist acceptance of a pro-unity outcome. They have extended to unionists open invitations to engage and have made generous offers in response to unionist concerns. In the development of guides and proposals about Ireland’s future, there should be ample opportunity for diverse and extensive forms of participation. Despite these attempts at inclusivity, it’s likely that a certain segment of unionism will never engage or accept. But the Scottish model might encourage some unionist participation because the decision in favour of unity is made before final negotiations about the shape of the Irish state take place. If Ireland is going to be unified in any event on the basis of democratic votes north and south, unionists might wish to become involved in negotiations in order to have some say on the nature of the state’s political structures, institutions and policies.[15]
The Scottish model might achieve this aim without making the mistake, as the Working Group does, of holding a referendum on detailed designs that confers a far-reaching and GFA-busting veto on the north.
Notes
[1] See Burke (2017 & 2020) for critiques of proposals to alter the majority consent rule. Brendan O’Leary (2021) recently made a strong case against such proposals.
[2] The members of the Working Group are: Prof. Oran Doyle, Prof. John Garry, Dr Paul Gillespie, Prof. Cathy Gormley-Heenan, Prof. Katy Hayward, Prof. Robert Hazell, Dr David Kenny, Prof. Christopher McCrudden CBE, Prof. Brendan O’Leary, Dr Alan Renwick, Dr Etain Tannam, and Alan Whysall (pp. viii-ix).
[3] See McCrudden’s (2021) useful outline of the different interpretive principles that lawyers, historians and political scientists bring to bear on the text of the Agreement. There are, of course, many variations within each of these three categories: lawyers, historians and political scientists will disagree, sometimes vehemently, among themselves. McCrudden recognizes these internal differences but doesn’t examine them.
[4] In her recent examination of unionist concerns about Irish unity and how nationalists might accommodate them, Jennifer Todd (2021) cites conversations with loyalists and everyday unionists. She does not reference conversations with nationalists or republicans, on whom the burden of accommodation will disproportionately fall. Brendan O’Leary (2021) uses the concept of “losers’ consent” to emphasize the centrality of the concerns of unionists, cultural Protestants and British citizens in the process of constitutional change. In his response to Rory Montgomery’s (2021) scheme for a unification referendum, Oran Doyle (2021) focuses almost exclusively on unionists’ possible reactions. Alan Whysall (2021) wants to spare no effort to gain unionist involvement and consent. See Burke (2021) for other examples of how academics are privileging unionist fears in their analyses of Ireland’s future.
[5] The Working Group sets out basic questions in chapter 9 and explores variations of those questions in chapter 13.
[6] The Working Group labels this design configuration 1. It gives a rather stark presentation of this option. I think configuration 1 could provide much more detail to voters than the Working Groups seems to allow. I address this question in more detail in the text.
[7] I’m examining what the Working Group calls configuration 2, its “Maximum Plan” option (para. 9.29). As I argue in the text, a comprehensive northern veto figures prominently in this configuration. The northern veto also plays an important part in configurations 4 and 5, the other two designs the Working Group recommends. For that reason, the general argument I make in the text against the veto in configuration 2 also applies to configurations 4 and 5.
In configurations 4 and 5, there are two sets of referendums. The first is a referendum on the principle of unification that also spells out the detailed process for designing a united Ireland and sets the default or interim terms on which unity would take place. The second set of referendums is on the detailed form of a united Ireland. In the initial unification referendum, the north has a veto over both the detailed process and the default/interim arrangements. The Working Group notes that these arrangements need to be carefully considered as they may become the permanent structures of a united Ireland. If they do become permanent, the northern veto in these configurations is as powerful as it is in configuration 2. In configurations 4 and 5, the northern veto also plays a part in the second set of referendums. This second-level veto appears to be structural in configuration 4 and optional in configuration 5 (paras 9.46, 9.58, 10.41-42 & 10.47). Configuration 5 demonstrates especially well how deeply and complexly the northern veto is embedded in the Working Group’s framework. In the first referendum in configuration 5, the north can veto any proposed process that does not give it a veto in the second referendum.
[8] The Working Group uses the exact same argument in its interim report (2020, para. 9.29).
[9] The Working Group uses the lack-of-evidence argument to contest Justice Humphreys’s interpretation that devolved government in the north must endure after unity (para 7.48). It also uses this argument to reject the view that the Agreement requires that the north’s position within a future united Ireland would be conditional on continuing expressions of majority consent (para. 7.52).
[10] The legal and political contexts of the 1973 border poll were, of course, very different from those of 1998. The 1973 poll was, nevertheless, the only direct historical precedent available to the GFA negotiators and the public in 1998.
[11] Once more, the Working Group’s reasoning can be used against it. To adapt an argument it employs against Humphreys: “There is no evidence that the participants in the 1996–98 talks envisaged” a border poll as a way of “mapping out in binding detail the shape of a united Ireland” (para. 7.48). See note 9.
Peace negotiations were rushed and chaotic in the final phase, with constitutional issues receiving little if any attention. There is nothing unusual in this set of circumstances: eleventh-hour bargaining is often disorderly and selectively focused. I’ve argued elsewhere that Sinn Féin in particular should have paid closer attention to the GFA’s border-poll process and the powers it granted to the Secretary of State and the British government (Burke, 2020). The main parties in the GFA talks accepted (sometimes with hesitation) the constitutional principles that the British and Irish governments had set out in a series of important documents, including the Joint (Downing Street) Declaration of 1993 and the Joint Framework Document of 1995. These documents served as models for the simple constitutional process described in the Agreement (Montgomery, 2021).
[12] I’m making the same point against the Working Group that Brendan O’Leary, a member of the Working Group, makes against those who propose to change the 50% + 1 rule. Altering the meaning of a border poll and consent in the manner the Working Group suggests “would profoundly disrespect what voters, north and south, concurrently endorsed in the 1998 referendums” (O’Leary, 2021, p. 7).
[13] This model is configuration 1 in the Working Group’s analysis (paras 9.16-28), which I introduced when examining the different questions that might appear on a referendum ballot. See note 6.
[14] Montgomery (2021) proposes a three-stage process consistent with this model: concurrent referendums on unity in the north and south, detailed negotiations, and a Constitutional amendment incorporating provisions agreed in negotiations.
[15] The Working Group repeatedly makes the argument that waiting to design the final shape of a united Ireland until after pro-unity referendum decisions is a structure that might encourage unionists’ participation and thus facilitate their eventual agreement. It sees such unionist inclusivity as real potential strengths of two of its favoured configurations, namely configurations 4 and 5 (paras 9.45 & 9.59). For reasons that are not clear to me, the Working Group is much less enthusiastic about a similar structure encouraging unionist engagement in configuration 1 (the Scottish model), a design it rejects (para 9.23-24). The Working Group’s very different treatment of generally similar structures seems ad hoc.
References
Burke, M. (2017). “Deepening the Unionist Veto.” The Pensive Quill. 1 November. Retrieved from http://thepensivequill.am/2017/11/deepening-unionist-veto.html
Burke, M. (2020). “Stealing Irish Unity: The Repertoire of Thieves.” The Pensive Quill. [This piece is posted in five parts on September 5, 12, 19, 26 and October 3. All five parts can be retrieved from https://www.thepensivequill.com].
Burke, M. (2021). “Academics Discussing Border Polls.” The Pensive Quill. 27 March. Retrieved from https://www.thepensivequill.com/2021/03/academics-discussing-border-polls.html
Creighton, S. (2021). “Unionism needs to be smart on who it talks to in unification debate and can’t bury its head in the sand.” Belfast Telegraph. 26 May. Retrieved from https://www.belfasttelegraph.co.uk/opinion/news-analysis/unionism-needs-to-be-smart-on-who-it-talks-to-in-unification-debate-and-cant-bury-its-head-in-the-sand-40470793.html
The Scottish Model
There is a critical difference between having the details of a united Ireland available before a referendum and having a referendum on the details. All sides currently involved in the debate about Ireland’s future want the details available. They agree that rigorous and inclusive planning about what a united Ireland might look like needs to take place in advance of any referendum. They also agree that the work to delineate clear processes to facilitate constitutional change must start now. This position is reasonable and sensible.
It’s neither reasonable nor sensible for the Working Group to propose a referendum on details that inserts a thoroughgoing northern veto over the shape of a united Ireland. The Working Group seems to take the position that the only details that matter are the ones that appear on the referendum ballot. It is propagating what might be called the myth of the ballot paper.
Scotland’s experience explodes this myth. The Scottish independence referendum in 2014 shows that it’s possible to have a vote on the basic question of sovereignty and at the same time to give voters a detailed account of what the new state might look like. The question used in Scotland was about sovereignty, pure and simple: ‘Should Scotland be an independent country?’, with voters required to vote yes or no (Electoral Commission, 2014, p. 4). Ten months before the referendum, the Scottish government published Scotland’s Future, a 670-page guide to an independent Scotland that laid out the government’s vision and priorities for action (Scottish Government, 2013). This document covered at least as many issues as the Working Group includes in its detailed designs. The government’s guide examined the transition to Scotland’s new constitutional status as an independent state. It also set out the direction in which the new state might go regarding finances and the economy, social services, education, human rights, international relations, justice and security, culture, and democratic development.
The Working Group rejects this Scottish model, unfairly I think.[13] It seriously underestimates and undervalues the detailed information that was available to Scottish voters, both through the government’s comprehensive guide and the various types of information that are naturally generated by parties, organizations and the media both before and during a referendum campaign. The Working Group seems to discount all that information because it was not entrenched in a confirmed and definitive plan that appeared on the ballot paper. But that information proved very useful to Scottish voters. The Scottish Electoral Commission found that voters had exceptionally high levels of knowledge of what the referendum was about: “This is the highest level of knowledge recorded for all major elections since we started asking this question in our post-election/referendum surveys.” The vast majority of voters felt that they had enough information “to make an informed decision” in the referendum (Electoral Commission, 2014, paras 3.71 & 3.76, respectively).
The Scottish model, or some variation of it, is appropriate for borders polls in Ireland.[14] Above all, it has both legal and democratic legitimacy. It is legally legitimate in that it fully complies with the constitutional provisions of the Agreement. It is democratically legitimate because, first, it respects the choice that voters made in ratifying the GFA in 1998 and, second, it doesn’t usurp the right of the Irish people directly to decide what role, if any, a northern veto might play in a united Ireland.
In this model, the Irish government might follow the lead of its Scottish counterpart and publish a detailed guide to the new constitutional order. Other groups and parties might endorse the government’s proposals. Or they might not, in which case there may be several proposals on the field. In any case, voters in Ireland are likely to have more than sufficient detail to make an informed decision, as had voters in Scotland. None of the proposals before voters is confirmed or definitive because final negotiations on the form of a united Ireland will occur only after concurrent referendum decisions to unify. The citizens of the newly-reunified state have the ultimate authority to set the terms and conditions of their own governance.
Just as participants in the debate about Ireland’s future urge advanced planning, so do they favour securing unionist participation in the process and unionist acceptance of a pro-unity outcome. They have extended to unionists open invitations to engage and have made generous offers in response to unionist concerns. In the development of guides and proposals about Ireland’s future, there should be ample opportunity for diverse and extensive forms of participation. Despite these attempts at inclusivity, it’s likely that a certain segment of unionism will never engage or accept. But the Scottish model might encourage some unionist participation because the decision in favour of unity is made before final negotiations about the shape of the Irish state take place. If Ireland is going to be unified in any event on the basis of democratic votes north and south, unionists might wish to become involved in negotiations in order to have some say on the nature of the state’s political structures, institutions and policies.[15]
The Scottish model might achieve this aim without making the mistake, as the Working Group does, of holding a referendum on detailed designs that confers a far-reaching and GFA-busting veto on the north.
Notes
[1] See Burke (2017 & 2020) for critiques of proposals to alter the majority consent rule. Brendan O’Leary (2021) recently made a strong case against such proposals.
[2] The members of the Working Group are: Prof. Oran Doyle, Prof. John Garry, Dr Paul Gillespie, Prof. Cathy Gormley-Heenan, Prof. Katy Hayward, Prof. Robert Hazell, Dr David Kenny, Prof. Christopher McCrudden CBE, Prof. Brendan O’Leary, Dr Alan Renwick, Dr Etain Tannam, and Alan Whysall (pp. viii-ix).
[3] See McCrudden’s (2021) useful outline of the different interpretive principles that lawyers, historians and political scientists bring to bear on the text of the Agreement. There are, of course, many variations within each of these three categories: lawyers, historians and political scientists will disagree, sometimes vehemently, among themselves. McCrudden recognizes these internal differences but doesn’t examine them.
[4] In her recent examination of unionist concerns about Irish unity and how nationalists might accommodate them, Jennifer Todd (2021) cites conversations with loyalists and everyday unionists. She does not reference conversations with nationalists or republicans, on whom the burden of accommodation will disproportionately fall. Brendan O’Leary (2021) uses the concept of “losers’ consent” to emphasize the centrality of the concerns of unionists, cultural Protestants and British citizens in the process of constitutional change. In his response to Rory Montgomery’s (2021) scheme for a unification referendum, Oran Doyle (2021) focuses almost exclusively on unionists’ possible reactions. Alan Whysall (2021) wants to spare no effort to gain unionist involvement and consent. See Burke (2021) for other examples of how academics are privileging unionist fears in their analyses of Ireland’s future.
[5] The Working Group sets out basic questions in chapter 9 and explores variations of those questions in chapter 13.
[6] The Working Group labels this design configuration 1. It gives a rather stark presentation of this option. I think configuration 1 could provide much more detail to voters than the Working Groups seems to allow. I address this question in more detail in the text.
[7] I’m examining what the Working Group calls configuration 2, its “Maximum Plan” option (para. 9.29). As I argue in the text, a comprehensive northern veto figures prominently in this configuration. The northern veto also plays an important part in configurations 4 and 5, the other two designs the Working Group recommends. For that reason, the general argument I make in the text against the veto in configuration 2 also applies to configurations 4 and 5.
In configurations 4 and 5, there are two sets of referendums. The first is a referendum on the principle of unification that also spells out the detailed process for designing a united Ireland and sets the default or interim terms on which unity would take place. The second set of referendums is on the detailed form of a united Ireland. In the initial unification referendum, the north has a veto over both the detailed process and the default/interim arrangements. The Working Group notes that these arrangements need to be carefully considered as they may become the permanent structures of a united Ireland. If they do become permanent, the northern veto in these configurations is as powerful as it is in configuration 2. In configurations 4 and 5, the northern veto also plays a part in the second set of referendums. This second-level veto appears to be structural in configuration 4 and optional in configuration 5 (paras 9.46, 9.58, 10.41-42 & 10.47). Configuration 5 demonstrates especially well how deeply and complexly the northern veto is embedded in the Working Group’s framework. In the first referendum in configuration 5, the north can veto any proposed process that does not give it a veto in the second referendum.
[8] The Working Group uses the exact same argument in its interim report (2020, para. 9.29).
[9] The Working Group uses the lack-of-evidence argument to contest Justice Humphreys’s interpretation that devolved government in the north must endure after unity (para 7.48). It also uses this argument to reject the view that the Agreement requires that the north’s position within a future united Ireland would be conditional on continuing expressions of majority consent (para. 7.52).
[10] The legal and political contexts of the 1973 border poll were, of course, very different from those of 1998. The 1973 poll was, nevertheless, the only direct historical precedent available to the GFA negotiators and the public in 1998.
[11] Once more, the Working Group’s reasoning can be used against it. To adapt an argument it employs against Humphreys: “There is no evidence that the participants in the 1996–98 talks envisaged” a border poll as a way of “mapping out in binding detail the shape of a united Ireland” (para. 7.48). See note 9.
Peace negotiations were rushed and chaotic in the final phase, with constitutional issues receiving little if any attention. There is nothing unusual in this set of circumstances: eleventh-hour bargaining is often disorderly and selectively focused. I’ve argued elsewhere that Sinn Féin in particular should have paid closer attention to the GFA’s border-poll process and the powers it granted to the Secretary of State and the British government (Burke, 2020). The main parties in the GFA talks accepted (sometimes with hesitation) the constitutional principles that the British and Irish governments had set out in a series of important documents, including the Joint (Downing Street) Declaration of 1993 and the Joint Framework Document of 1995. These documents served as models for the simple constitutional process described in the Agreement (Montgomery, 2021).
[12] I’m making the same point against the Working Group that Brendan O’Leary, a member of the Working Group, makes against those who propose to change the 50% + 1 rule. Altering the meaning of a border poll and consent in the manner the Working Group suggests “would profoundly disrespect what voters, north and south, concurrently endorsed in the 1998 referendums” (O’Leary, 2021, p. 7).
[13] This model is configuration 1 in the Working Group’s analysis (paras 9.16-28), which I introduced when examining the different questions that might appear on a referendum ballot. See note 6.
[14] Montgomery (2021) proposes a three-stage process consistent with this model: concurrent referendums on unity in the north and south, detailed negotiations, and a Constitutional amendment incorporating provisions agreed in negotiations.
[15] The Working Group repeatedly makes the argument that waiting to design the final shape of a united Ireland until after pro-unity referendum decisions is a structure that might encourage unionists’ participation and thus facilitate their eventual agreement. It sees such unionist inclusivity as real potential strengths of two of its favoured configurations, namely configurations 4 and 5 (paras 9.45 & 9.59). For reasons that are not clear to me, the Working Group is much less enthusiastic about a similar structure encouraging unionist engagement in configuration 1 (the Scottish model), a design it rejects (para 9.23-24). The Working Group’s very different treatment of generally similar structures seems ad hoc.
References
Burke, M. (2017). “Deepening the Unionist Veto.” The Pensive Quill. 1 November. Retrieved from http://thepensivequill.am/2017/11/deepening-unionist-veto.html
Burke, M. (2020). “Stealing Irish Unity: The Repertoire of Thieves.” The Pensive Quill. [This piece is posted in five parts on September 5, 12, 19, 26 and October 3. All five parts can be retrieved from https://www.thepensivequill.com].
Burke, M. (2021). “Academics Discussing Border Polls.” The Pensive Quill. 27 March. Retrieved from https://www.thepensivequill.com/2021/03/academics-discussing-border-polls.html
Creighton, S. (2021). “Unionism needs to be smart on who it talks to in unification debate and can’t bury its head in the sand.” Belfast Telegraph. 26 May. Retrieved from https://www.belfasttelegraph.co.uk/opinion/news-analysis/unionism-needs-to-be-smart-on-who-it-talks-to-in-unification-debate-and-cant-bury-its-head-in-the-sand-40470793.html
Downing, J. (2021). “Rules on referendum campaigns ‘must be updated’ ahead of any border poll.” Belfast Telegraph. 26 May. Retrieved from https://www.belfasttelegraph.co.uk/news/northern-ireland/rules-on-referendum-campaigns-must-be-updated-ahead-of-any-border-poll-40470047.html
Doyle, O. (2021). “Configuring Irish Unification Processes: A Response to ‘The Good Friday Agreement and a United Ireland’ by Rory Montgomery.” Irish Studies in International Affairs 32:2, 111-114. Retrieved from https://www.jstor.org/stable/10.3318/irisstudinteaffa.32.issue-2
Electoral Commission. (2014). Scottish Independence Referendum: Report on the referendum held on 18 September 2014. ELC/2014/02. December. Retrieved from https://www.electoralcommission.org.uk/sites/default/files/pdf_file/Scottish-independence-referendum-report.pdf
Finucane, J. (2021). “Preparations for unity referendum must be stepped up – Finucane.” Sinn Féin. 26 May. Retrieved from https://www.sinnfein.ie/contents/60924
Hume, J. (1989). “John Hume on the end of the Unionist veto in Ulster.” London Review of Books 11:13 (2 February). Retrieved from https://www.lrb.co.uk/the-paper/v11/n03/john-hume/john-hume-on-the-end-of-the-unionist-veto-in-ulster
Irish Times. (2021). “The Irish Times view on unification referendums: democratic dilemmas.” 27 May. Retrieved from https://www.irishtimes.com/opinion/editorial/the-irish-times-view-on-unification-referendums-democratic-dilemmas-1.4577151
McArdle, E. (2021). “Planning for the Unity Referendum.” New Ireland/éire Nua. 1 (Summer): 16-18. Retrieved from https://www.sinnfein.ie/irish-unity
McBride, S. (2021). “Few have yet realised the staggering complexity behind a border poll.” News Letter. 5 June. Retrieved https://www.newsletter.co.uk/news/politics/sam-mcbride-few-have-yet-realised-the-staggering-complexity-behind-a-border-poll-3261373
McCrudden, C. (2021). “The Hermeneutics of the Good Friday Agreement: A Response to ‘Getting Ready’ by Brendan O’Leary. Irish Studies in International Affairs 32:2, 39-43. Retrieved from https://www.jstor.org/stable/10.3318/irisstudinteaffa.32.issue-2
Montgomery, R. (2021). “The Good Friday Agreement and a United Ireland.” Irish Studies in International Affairs 32:2, 83-110. Retrieved from https://www.jstor.org/stable/10.3318/irisstudinteaffa.32.issue-2
O’Leary, B. (2021). “Getting Ready: The Need to Prepare for a Referendum on Reunification.” Irish Studies in International Affairs 32:2, 1-38. Retrieved from https://www.jstor.org/stable/10.3318/irisstudinteaffa.32.issue-2
Renwick, A. (2021). “Working Group on Unification Referendums on the Island of Ireland: Final Report.” Constitution Unit, University College London. 26 May. Retrieved from https://constitution-unit.com/2021/05/26/the-working-group-on-unification-referendums-on-the-island-of-ireland-final-report/
Renwick, A. and K. Hayward. (2021). “A Referendum on Irish Unification: Why it Needs Attention.” Political Insight 12:2 (June): 16-19.
Scottish Government. (2013). Scotland’s Future: Your Guide to an Independent Scotland. Edinburgh. November. Retrieved from https://www.gov.scot/publications/scotlands-future/
Todd, J. (2021). “Unionism, Identity and Irish Unity: Paradigms, Problems and Paradoxes.” Irish Studies in International Affairs 32:2, 53-77. Retrieved from https://www.jstor.org/stable/10.3318/irisstudinteaffa.32.issue-2
Whysall. A. (2021). “Irish unity: why Britain should reflect.” Constitution Unit, University College London. 8 June. Retrieved from https://constitution-unit.com/2021/06/08/irish-unity-why-britain-should-reflect/
Working Group. (2020). Interim Report. Working Group on Unification Referendums on the Island of Ireland. The Constitution Unit, School of Public Policy, University College London. November. Retrieved from https://www.ucl.ac.uk/constitution-unit/research/elections-and-referendums/working-group-unification-referendums-island-ireland
Working Group. (2021). Final Report. Working Group on Unification Referendums on the Island of Ireland. The Constitution Unit, School of Public Policy, University College London. May. Retrieved from https://www.ucl.ac.uk/constitution-unit/working-group-unification-referendums-island-ireland
⏮ Mike Burke has lectured in Politics and Public Administration in Canada for over 30 years.
An excellent piece that cuts to the chase.
ReplyDeleteIt is like so many other pieces by Mike on this topic.
I cannot recommend highly enough the collected writings on the subject by Mike Burke.
My one regret is that they merit a much wider airing that TPQ is able to offer.
Nevertheless, we are delighted that he has chosen to post here.
Richard O'Rawe comments
ReplyDeleteMike Burke's article should be read by all republicans and Nationalists, and dare I suggest Unionists and Loyalists, because:
1/ it is hugely important
and,
2/ It provides an invaluable service in dissecting the reports of The Working Group on Unification Referendums on the Island of Ireland.
Burke correctly identifies fundamental flaws in the Group's latest report, and highlights where it hopes to subvert one of the fundamental tenets of the GFA i.e. that any vote on the future constitutional direction of Ireland should be conducted on the basis of a simple 50+1% vote. To say that the ballot paper should include a comprehensive outline of what a new Ireland would look like is as absurd as it is dangerous.
As Burke says: The Working Group concocts out of nothing, the notion that border polls set the constitutional structure of a united Ireland, define its principle institutions, specify its national symbols, and establish its major policies. And it manufactures the related notion that, the North has a comprehensive veto over all these matters.
I agree entirely with the Group position that the electorate should know what type of a new Ireland is being proposed, but who determines what those proposals may be? And what of counter-proposals, do they go on the ballot paper too? Do the varying political party policies go on the ballot paper? Where does it end?
A referendum on the future constitutional position of Ireland is coming; it is only a matter of timing. I have already heard voices advocate that 50+1% is too narrow a majority to bring about the unification of Ireland (but 50-1% is enough to keep the North in the United Kingdom). I suspect that this Group's report will be studied in-depth by parliamentarians, north and south. I also suspect that some of those same politicians may use it to undermine the concept of a 50+1 referendum, and, I wouldn't be at all surprised if the ended up telling us that times have changed and the GFA needs a re-write. Perhaps 60+1% might do the trick?
Richard - you nail what many people have come to expect and that nationalists in the North are going to be awarded the status of unequal citizenship, with their vote counting less than votes for the union.
DeleteMike has been publishing on TPQ for quite a while. He identifies the elephant in the room and is determined not to call it a mouse. His work deserves a much wider reading.
Anthony and Richard,
ReplyDeleteThanks for the kind comments.
In the July issue of Fortnight, Tom Hadden uses the Working Group’s final report to argue that the Good Friday Agreement should be ignored or changed and that the British government should delay calling a border poll. This is an obvious implication of the Working Group’s analysis. I think London and Dublin will draw the same conclusion as Hadden. The Working Group undermines the GFA’s constitutional provisions and cheapens the democratic ratification of the Agreement. You can find Hadden’s article at https://fortnightmagazine.org/wp-content/uploads/2021/07/fortnight_482-1.pdf