Northern Ireland’s constitutional arrangements are changing again. This is hardly a surprise, because change was foreshadowed in the New Decade New Approach agreement (NDNA) under which Stormont returned from its nearly three-year coma. Moreover, the change itself is backed by nearly all of Stormont’s political parties. However, the change is as unusual as the provision which is to be changed. This post examines both the change and its underlying provision: the petition of concern, a legislative mechanism unique among all the legislatures in the United Kingdom.
The Petition of Concern
The petition of concern (PoC) is a relatively simple legislative device with a powerful impact. As few as thirty MLAs can sign a PoC, on any “matter which is to be voted on by the Assembly”, and thereby require cross-community support for the matter to be voted on. Cross-community support is defined under the Northern Ireland Act 1998 (NIA) as either parallel consent (a Unionist majority and a Nationalist majority within an overall majority of MLAs present and voting) or a weighted majority (at least 40% of Unionists and 40% of Nationalists within 60% of MLAs present and voting). Failing to obtain cross-community support means that the matter in respect of which a PoC is brought fails completely, even if it has the support of a numerical majority in the Assembly.
It is important to note that there are marked differences between the original vision of the PoC and its present-day rules and practice. Two main differences concern the matters on which a PoC may be brought and the proper procedure to be followed. In the Good Friday Agreement (GFA) where the PoC originated (Strand One, paragraph 5(d)), it was provided for use when the Assembly makes “key decisions”, without detailing what these might be. Paragraph 13 of Strand One also mentions “measures” when talking about PoCs, requiring an Assembly vote on whether “the measure” may proceed without a reference to an ad-hoc Assembly Committee to examine conformity with equality requirements. The NIA expanded “key decisions” to include anything on which the Assembly votes, while Assembly Standing Order 60 appears to interpret “measure” to mean “Bill”. Meanwhile, the default position provided for in the GFA, for “measures” to proceed to the ad-hoc Committee unless the Assembly votes otherwise, has transformed into the practice whereby no Assembly vote to establish the ad-hoc Committee is required in advance of a vote on a PoC. So much for the original vision.
While the PoC acts as a powerful check against majoritarianism and dominance by a single community, the virtually limitless matters in which it can be brought, coupled with the fairly straightforward procedural requirements to bring it, has resulted in PoCs being used to block the censure of ministers and defeat the proposal to extend marital rights to same-sex couples. This use of the PoC has been criticised, with plans for its reform having been set out in no less than three political agreements under which Stormont functioned. None but the most recent of these reform proposals has been translated into draft legislation.
The changes to the PoC are contained in clause 5 of the Northern Ireland (Ministers, Elections and Petitions of Concern) Bill 2021 (Bill). While the Bill goes beyond PoC reform (see David Torrance’s research briefing), my focus is on clause 5. There are two main elements to the reforms in clause 5, which for convenience I will call the “form” and “substance” elements. The form element relates to how the PoC may be brought – maintaining the thirty-MLA threshold, but requiring MLAs of more than one party among the thirty MLAs. Moreover, a new 14-day period (the “consideration period”) is introduced between the bringing of a PoC and the consequent vote on the issue to which it relates. While the consideration period does not have a prescribed purpose, the NDNA details matters such as advice and reports from the Northern Ireland Human Rights Commission which may be considered. Undoubtedly the consideration period will also function as a cooling-off period, enabling the MLAs who initially bring a PoC to consider whether they wish to confirm their support at the end of this period. Importantly, the Speaker and Deputy Speakers may not bring or confirm a PoC (which, under the current provisions, they can).
The substance element relates to the matters in respect of which a PoC may be brought. While the current provisions do not contain any limits on such matters, the proposed changes will bar the use of PoCs during the second stage of a Bill, sanctions “in relation to” the conduct of Ministers and non-ministerial MLAs, “standards motions” which relate to MLA conduct, and “motions of questions which have no express legal or procedural effect”. This negative list of matters is found in clause 5(4) of the 2021 Bill, with the first and second matters explicitly set out in 5(4)(a) and (b) respectively, while the third and fourth matters are left to Assembly Standing Orders to be made under clause 5(6)(c).
The NDNA did not specify whether any of the form or substance elements of the proposed changes to the PoC would be done by legislative reform or reform to the Assembly’s Standing Orders. In fact, the Stormont House Agreement, which was another political agreement touching on constitutional reform and which predated the NDNA by 6 years, provided for PoC reform to be made “through a protocol agreed between the [political] parties”. This protocol was set out in greater detail in the Fresh Start proposals the following year.
The difference between reform through legislation and reform through non-legislative means is of great significance both in principle and in practice. Currently, the NIA only provides for a PoC (and prescribes its formative threshold) while leaving its substance to Assembly procedures over which the Speaker has final authority (Standing Order 1(2)). The proposed changes would place a large part of the PoC substance on a statutory footing, over which the courts have the final interpretive and adjudicatory authority. This is important when looking at the language of the proposals.
Consider, for example, that PoCs will be barred in respect of a matter which “concerns a sanction in relation to the conduct of a Minister or other member of the Assembly” (cl. 5(4)(a)). This is narrower than the language in the NDNA, which instead expressed a prohibition on the use of PoCs “where the question before the Assembly relates to a member’s conduct as a Minister or MLA”. However, even as the Bill narrows the language, it nevertheless leaves important questions open: how broadly should the words “concerns” and “in relation to” be interpreted? The NIA provides for ministers to be voted out of office for specified periods by the Assembly, while Standing Orders provide for a range of other sanctions against all MLAs, up to exclusion from proceedings and withdrawal of rights and privileges for specified periods. Would PoCs only be barred when debating actual sanctions, or in any debates around conduct which may include the possibility of debating sanctions? The answers to such questions would necessitate examining the purpose of the reformed PoC (and the political and historical reasons for such reform) and may well necessitate examining the purpose of the corresponding Assembly debate. Obviously, scrutinising executive conduct is an essential function of the Assembly (as indeed the other UK legislatures). However, in the divisive and at times febrile atmosphere of Northern Ireland politics, the possibility that such scrutiny may come from a place of hegemonic assertion rather than good faith accountability is one of the reasons why the PoC was created in the first place.
This brings us to the two main issues with the manner of PoC reform as reflected in the Bill. The first issue is with the level of judicial ingress into the Assembly process. While the devolved legislatures (including the Assembly) do not enjoy the wide immunities which the Westminster Parliament derives from Article IX of the Bill of Rights 1688, the courts may (if the proposals are enacted as they stand) enquire into the purpose of an Assembly proceeding in a way which does not currently exist. One cannot avoid such a possibility simply by pointing to the separation of powers, for the fundamental reason that authoritative statutory interpretation is a judicial function par excellence. In this regard, clause 5 is effectively a parallel to clause 1(4) of the draft bill of the Westminster Committee on Privileges dealing with sanctions for non-attendance of witnesses summoned to a Commons Select Committee. That provision expressly declares that a court may consider the “nature and purpose” of a Committee summons. While the PoC reforms are silent on what a court may or may not consider, the proposed reforms clearly make many future questions on PoC validity justiciable, with judicial scrutiny of relevant Assembly proceedings unavoidable as a result.
This leads to the second issue: the ingress is not unidirectional. If the Bill allows the courts to enter the legislative domain, then it also allows politics to seep into the judicial function. If the courts are asked to authoritatively determine whether a PoC is used in accordance with the Bill provisions, then the answer to that question may turn on the purpose of an Assembly proceeding. It is difficult (to say the least) to determine the purpose of a proceeding in a quintessentially political body, independent of the politics which underlie that body. In Cherry/Miller 2, the prorogation advice was ruled to have been unlawful on the basis of a lack of reasons for a prorogation of that length, with the Supreme Court thus able to avoid the hairy question of whether the advice had been given with improper motives. In a future challenge to the validity of a PoC, the Northern Ireland High Court might not be so lucky.
It is widely accepted that the petition of concern requires reform; that its use should be restricted to certain important matters and that it should be possible to seriously reflect upon its use as part of the process involving its use. As an important safeguard to protect the political participation of Northern Ireland’s communities, it should itself be protected from becoming an element of party-political theatre.
However, the manner of reform matters a great deal when the possibility of unintended (and consequently uncomfortable) consequences looms on the horizon. This post is a relatively modest attempt at generating debate on a constitutional reform of considerable significance to Northern Ireland. As the Bill proceeds to its second stage on 22 June, I hope its proposals will be scrutinised in greater detail.
Anurag Deb is a PhD researcher at Queen’s University Belfast.
(Suggested citation: A Deb, ‘Judicialising the Legislative Process: The Petition of Concern’, U.K. Const. L. Blog (14 June 2021) (available at https://ukconstitutionallaw.org/))