I’m writing to respond to Etain Tannam’s recent posting to the Constitution Unit, University College London, in which she discusses British-Irish institutional cooperation during a border poll (Tannam, 2020).
Tannam says that British-Irish cooperation is key to the successful navigation of any movement toward Irish reunification. She notes that this intergovernmental approach:
began in 1985 with the Anglo-Irish Agreement and contributed to a gradual shift in Ireland from traditional nationalism, whereby Irish unification was a territorial concept and zero sum, to the SDLP-influenced concept of unification of peoples, not territory that eventually would lead to unification with consent across unionist and nationalist communities. The Belfast/Good Friday Agreement was the culmination of this strategic British-Irish intergovernmental approach that began 15 years before the Agreement’s ratification. (p. 3).
This passage throws together a lot of different elements that may be usefully considered separately, even if we allow for some obvious interrelationships. At various points below, I comment on Tannam’s discussion of the intergovernmental approach. Now, I want to examine what she says, or implies, about consent.
The question of consent is crucial to the process of constitutional change, and it’s essential that we get it right. The need for precision is magnified by the long list of commentators who are trying to alter retrospectively the meaning of consent as entrenched in the Good Friday Agreement and related documents. Tannam skates perilously close to the proposals of this amending crew, who would grant unionists a veto over constitutional change in the north.
Neither the Anglo-Irish Agreement nor the GFA requires that unification emerge from “consent across unionist and nationalist communities,” to use Tannam’s phrase. The SDLP’s notion of unification of peoples is not part of the GFA’s provisions for constitutional change, much to the chagrin of the late Seamus Mallon. There are sections of the GFA that do require cross-community consent, but they are limited to certain key decisions of the Assembly. They do not apply to constitutional issues.
In fact, the Anglo-Irish Agreement and the GFA use identical language to define the kind of consent needed for constitutional change, namely, “consent of a majority of the people of Northern Ireland”. According to the architecture of the peace process, the movement toward Irish unity requires not the parallel consent of unionists and nationalists in the north, but the concurrent consent of majorities in the north and the south.
The majority-consent formula for constitutional arrangements in the north was driven by the concerns of unionism and the British state. And it did not begin with the intergovernmental approach of the 1980s. It emerged much earlier, with the gerrymandering of the borders of a “statutory Ulster” that could both escape Home Rule and house a comfortable unionist majority, which, of course, resulted in the partition of Ireland (Rankin, 2009). The notion of majority consent for change to the constitutional status of the north was implied in the Government of Ireland Act 1920, the 1921 Treaty and the Ireland Act 1949. It appeared in incipient form in many constitution-related documents over the next two decades. It was made explicit in a series of constitutional initiatives in 1973, in much the same language as appears in the GFA.
Majority consent has been the linchpin of constitutional change ever since. Talk of “consent across unionist and nationalist communities” muddies the waters of what is a conceptually clear formulation for altering the constitutional status of the north. Taken literally, such talk subverts long-established and democratically-affirmed protocols for bringing about a united Ireland.
I also have concerns with Tannam’s limited depiction of the Secretary of State’s role in the calling of a border poll. She states that: “Only the Secretary of State for Northern Ireland has the statutory power to call a referendum on Irish unification, if they perceive there to be evidence of majority support in Northern Ireland for unification” (Tannam, 2020, p. 1). This description encompasses only one-half of the Secretary of State’s power. In addition to the mandatory duty to call a poll where majority support is likely, the Secretary of State has the discretionary power to call a poll at any time and for virtually any reason. The sole constraint on this discretion is that seven years must separate the holding of any one poll and the next. In the McCord case in June 2018, the Belfast High Court clarified the dual character of the Secretary of State’s authority over a border poll and underlined the open-ended nature of the discretionary power (NIQB 106, 2018).
Tannam’s analysis here contributes to the sometimes ill-informed debate about whether the conditions for calling a border poll have been met. This discussion has been given new life by the just-published surveys by the University of Liverpool and the Detail/LucidTalk, which indicate that between 29 and 45.4 percent of people in the north would vote for a united Ireland if there were a border poll. The common assumption of this debate is that the Secretary of State can call a border poll only if it appears likely that a majority in the north would support Irish unity.  That assumption is not only flawed but irrelevant to the full power exercised by the Secretary of State. There are absolutely no conditions that have to be met before the Secretary of State can exercise the discretionary power to call an initial border poll. The McCord ruling states explicitly that the Secretary of State can order a poll even if majority support for a united Ireland were unlikely, or if there were doubt about which side might win a majority. There are, to be sure, many political constraints affecting the use of the discretionary power, but they are completely independent of the Secretary of State’s essentially unfettered legal authority to call a border poll at any moment.
My third concern is Tannam’s confusing analysis of Dublin’s part in the Secretary of State’s mandatory duty to call a border poll. She is correct to say that the Irish government will be centrally involved in the broader constitutional referendum process. Both the British-Irish Agreement accompanying the GFA and the amended version of Article 3 of the Irish Constitution assume that referendums on Irish unity will take place in the north and the south. The McCord ruling confirmed this view:
It is clear that a border poll in Northern Ireland to produce the outcome of a united Ireland would have to be replicated by a poll in the Republic of Ireland producing a concurrent expression of a majority wish in the Republic to bring about a United Ireland. (NIQB 106, 2018, para. 5).
Tannam is incorrect, however, to suggest that Dublin might have a role in the Secretary of State’s border-poll calculations. She says that the Secretary of State’s determination that there is likely majority support for Irish unity depends, in part, “on whether the Irish government perceives evidence of majority support in Ireland for unification” (p. 4). Quite the opposite, the level of support for Irish unity in the south has little if anything to do with the Secretary of State’s legal duty, which is to assess the likelihood of majority support in the north. 
The McCord ruling states unequivocally that if the Secretary of State were to believe a majority in the north would vote in favour of a united Ireland “then she has no choice but to call a border poll” (NIQB 106, 2018, para. 20). Richard Humphreys, appointed to the Irish High Court in 2015, concurs:
if it can be demonstrated to the secretary of state that it is likely that a majority would vote to change the constitutional position of Northern Ireland, then the holding of the poll becomes a mandatory obligation on the secretary of state . . . and he is required . . . to make an order [for the poll to be held]” (Humphreys, 2009, p. 122).
To my mind, it would be a clear breach of the GFA and the Northern Ireland Act 1998 if the Secretary of State were to think that a northern majority for Irish unity is likely but decides not to call a border poll because Dublin supposes there is no corresponding majority in the south. That is, contrary to what Tannam suggests, the Secretary of State’s legal obligation to call a poll does not depend on what Dublin thinks. 
It’s to be expected that the British and Irish governments would be in close contact as the majority-support condition in the north appears to be approaching. Some part of the extensive intergovernmental coordination Tannam examines would most probably be put in place. At the very least, both governments would work together, as the McCord ruling suggests, to draw up plans for the holding of concurrent referendums in each jurisdiction. Dublin might be pleased or apprehensive that a southern poll would fail; it might be fearful or relieved that it would pass. No matter Dublin’s feelings, and intergovernmental cooperation aside, the duty to call a border poll in the north rests exclusively with the Secretary of State’s reading of support levels in the north.
My final concern is with Tannam’s analysis of the British-Irish Intergovernmental Conference (B-IIGC). As she suggests, the B-IIGC is probably the institution of choice through which the two governments can cooperate on the holding of constitutional referendums. But I think Tannam seriously underestimates the potential for widespread and persistent unionist opposition to the Conference.
She is certainly aware of unionism’s fears:
Unionists have long been wary of the B-IIGC, as it gives a consultative role to the Irish government. There were unionist suspicions in the early years of the [Good Friday] Agreement that the B-IIGC could be a vehicle for joint UK and Irish sovereignty over Northern Ireland and eventually a path to unification. (p. 4)
She suggests that such doubts might be assuaged by re-positioning the B-IIGC as part of a broader network of East-West institutions that promote intergovernmental coordination outside the EU. And, crucially, she believes that emphasizing the B-IIGC as an institutional means of protecting unionist rights and identity during a border poll process could make it more acceptable to unionists.
I disagree. It’s highly unlikely that unionism will warm to the B-IIGC. On the face of it, Tannam’s argument is problematic. Unionists generally do not trust British governments; they tend to see all Irish governments as foreign interlopers. And Brexit has had a sharply negative impact on unionism’s assessments of London and Dublin. Why would unionists ever accept both governments acting together in the B-IIGC, presiding over a process that could break up the Union?
Tannam is mistaken to suggest that unionist suspicion of the B-IIGC was confined to “the early years of the Agreement”. On the contrary, that suspicion was heightened considerably during the recent three-year collapse of devolved government that ended only in January. There was sharp disagreement between unionism and nationalism over the type of governance regime that should step into the breach left by a non-functioning Stormont. The DUP and UUP wanted direct rule from London; Sinn Féin, the SDLP, and the Irish government wanted intergovernmental oversight through the B-IIGC. Sinn Féin openly referred to the B-IIGC as joint stewardship and joint authority. The DUP quickly rejected Sinn Féin’s interpretation, dismissing the B-IIGC as a “talking shop” without any executive functions and insisting that it had no role in the internal affairs of the north.
Unionist parties and the British government have always been vigilant in guarding what they see as the impermeable boundaries separating the three strands of the peace process. Whenever the political parties and two governments are about to commence discussions to resolve yet another devolution crisis, the Northern Ireland Office is careful to point out that the talks will be held “in full accordance with the well-established three-stranded approach.” All participants understand that such language is code for: “don’t worry unionists, London retains full sovereignty here, and the government won’t allow Dublin to meddle in the internal affairs of Northern Ireland.”
The extended period without devolved government has intensified unionists’ concerns with the B-IIGC, as a contemporary form of Dublin “meddling.” Those concerns will not be easily set aside. The B-IIGC will be expected to play a fulsome part in any cross-border constitutional referendum process. To be effective, it should be prepared to address what will in all probability be significant unionist resistance to its activities. I doubt that it will be prepared.
 In an article accompanying the release of the Detail/LucidTalk survey, Rohan Naik is careful to examine both aspects of the Secretary of State’s authority to call a poll (Naik, 2020). Others have too, including Alan Whysall of the Constitution Unit (Whysall, 2019a & 2019b). But this distinction is mostly lost in the public commentary on the imminence, or not, of a border poll. And it’s blurred in Tannam’s analysis.
 This interpretation is supported by the analyses of Alan Whysall and Mark Bassett and Colin Harvey. Whysall’s list of factors that could trigger a border poll is concerned entirely with evidence of majority support in the north (Whysall, 2019b, pp. 6-7). Similarly, Bassett and Harvey’s discussion of what might prompt a unity referendum focuses on the northern situation. The last prompt they list does give a role to the Irish government in advocating for concurrent referendums, but this urging occurs in the context of the British government (Secretary of State) ignoring evidence of majority support for constitutional change in the north (Bassett & Harvey, 2019, para. 53). Of course, all such discussion of triggers and prompts is necessarily speculative, because the Secretary of State has infinitely wide latitude in considering evidence of majority support (see note 3).
Given that latitude, I suppose there is some small possibility that the Secretary of State could hold that the lack of majority support in the south somehow nullifies evidence of majority support in the north. But this kind of reasoning would surely violate the McCord ruling’s stipulation that the Secretary of State “must honestly reflect on the evidence available” (NIQB 106, 2018, para. 20).
 That it’s impossible to know how or when the Secretary of State’s mandatory duty comes into effect surely complicates matters. We don’t know what kind of evidence would make the Secretary of State believe that majority support for a united Ireland is likely and thereby trigger the holding of a border poll. As the McCord ruling emphasized, it’s completely up to the Secretary of State to determine what kind of evidence to consider, how to weigh that evidence, and how the importance of evidence might change in a very fluid situation. And the Secretary of State is under no obligation to specify any of the elements of the decision-making process. The conceptually clear notion of a northern majority as a condition of constitutional change becomes irrevocably ambiguous when tied to this legal but mystical capacity of the Secretary of State.
We need to ask how a duty could be “mandatory” when no one can know the real circumstances in which the Secretary of State would be compelled to act by calling a border poll. In the body of the posting, I am, by necessity, speaking hypothetically, as if it is possible to know that the Secretary of State believes the majority-support condition has been met. I do this to point out the problems in Tannam’s logic about Dublin’s role in the calling of a border poll.
There are also important questions to be asked about what might happen should a northern border poll reveal majority support for a united Ireland. I have serious doubts that the British government would move to implement Irish unity if there were significant unionist opposition to the results of the poll. The same could be said about the Irish government after a southern poll. When legality clashes with expedience, the law often loses.
Humphreys, R. (2009). Countdown to unity: debating Irish reunification. Dublin and Portland: Irish Academic Press.
Naik, R. (2020). “What are the mechanics behind a border poll being called?” The Detail. 24 February. Retrieved from https://www.thedetail.tv/articles/border-poll-mechanics?utm_source=mailinglist&utm_medium=email&utm_campaign=473-what-are-the-mechanics-behind-a-border-p
NIQB 106 (2018). In the matter of an application by Raymond McCord for leave to apply for judicial review and in the matter of Section 1 and Schedule 1 of the Northern Ireland Act 1998. In the High Court of Justice in Northern Ireland, Queen’s Bench Division (Judicial Review). Rt Hon Sir Paul Girvan. 28 June. Retrieved from https://judiciaryni.uk/judicial-decisions/2018-niqb-106
Rankin, K.J. (2009). “The Search for ‘Statutory Ulster’.” History Ireland 17:3 (May/June): 28-32.
Tannam, E. (2020). “The Secretary of State’s power to call a border poll in Northern Ireland: why British-Irish institutional cooperation is essential.” The Constitution Unit, University College London. 8 February. Retrieved from https://constitution-unit.com/2020/02/08/the-secretary-of-states-power-to-call-a-border-poll-in-northern-ireland-why-british-irish-institutional-cooperation-is-essential/
Whysall, A. (2019a). “Holding a border poll in Northern Ireland: when does it need to happen and what questions need to be answered?” The Constitution Unit, University College London. 6 March. Retrieved from https://constitution-unit.com/2019/03/06/holding-a-border-poll-in-northern-ireland-when-does-it-need-to-happen-and-what-questions-need-to-be-answered/
Whysall, A. (2019b). A Northern Ireland border poll. The Constitution Unit, University College London. 21 March. Retrieved from https://www.ucl.ac.uk/constitution-unit/sites/constitution-unit/files/185_a_northern_ireland_border_poll.pdf.
⏮ Mike Burke has lectured in Politics and Public Administration in Canada for over 30 years.
⏮ Mike Burke has lectured in Politics and Public Administration in Canada for over 30 years.