—Emmett Macfarlane, University of Waterloo
I am grateful that my article, Judicial Amendment of the Constitution, has received two thoughtful and productive responses by such esteemed scholars. Both responses seek to interrogate and expand upon the factors I present for distinguishing between judicial interpretation and judicial amendment, and while my focus in the article was on the empirical dimensions, both responses also (unsurprisingly) extend the discussion into more fundamental normative terrain. Given the limited space I have, I will assume readers have read the relevant pieces and avoid rehashing the arguments.
Erin Delaney and Christopher Schmidt’s primary objective is to extend my approach by introducing a temporal factor. They write, correctly, that my focus is on a decision as it is handed down, and what it purports to do, rather than an analysis of the subsequent events to identify whether the public has accepted a judicial opinion as an amendment, something that, in their view, “helps to differentiate between an illegitimate or contested opinion and a judicial amendment of the constitution.” Delaney and Schmidt explore the events following Brown v. Board of Education, a case they assert meets my criteria for judicial amendment and whose eventual acceptance by political actors and the public affirmed its place as a constitutional text.
Delaney and Schmidt thus offer an insightful path forward for identifying and – more importantly – helping to explain the legal and political reactions to decisions that cement themselves as successful instances of judicial amendment. Two of the three cases to which I apply my multi-factor test are properly understood in this light, in that Canadian actors immediately accepted the Court’s Quebec Secession reference and the Indian Court’s resolve ultimately resulted in the entrenchment of basic structure doctrine as a meta-textual rule imposed on the amending process. Yet given that the third case I examine is a dissenting opinion, my article is clear that the multi-factor test can also help us to identify an attempted judicial amendment, something which remains important in contexts where the courts may yet produce profound constitutional change.
My approach thus remains agnostic about the success of a particular attempt at judicial amendment: the goal is to identify decisions that attempt to produce a constitutional change that should only be possible through the constitution’s formal amending process. This framing is directly connected to one fundamental normative implication at stake in the distinction between interpretation and amendment. While my article leaves unattended the normative desirability of the substantive outcomes of judicial amendments, it clearly, if implicitly, contemplates judicial amendment as procedurally illegitimate.
The difference between substantive versus procedural normative understandings is helpful in answering key questions Delaney and Schmidt pose, including those about the political community’s understanding of what the constitution does or does not contain, and why that political community might come to accept judicial amendments. Simply put, judicial amendments may be desirable in substance even if they were brought into effect in a constitutionally dubious manner. For example, countries whose constitutional amending procedures are too difficult, or whose political culture surrounding constitutional reform is effectively toxic (both these arguably being the case in Canada), may find solace in judicial amendment as a safety valve to resolve increasing tensions between the established constitution and evolving societal contexts. Yet this substantive acceptance does not bestow procedural legitimacy to the change.
A key question in my approach to determining whether recourse to the formal amending procedure was required is whether there is evidence at the time of the decision about the political community’s expectations regarding what the constitution contains or whether a formal amendment is necessary to make a given change. Even an unsuccessful judicial amendment, where the political community was effective in negating a controversial court decision, violates the procedural concern. For example, the political community should not be forced to attempt a formal amendment to undo a judicial decision that exceeds appropriate limits on constitutional interpretation in the first place. For these reasons, I do not believe Delaney and Schmidt’s temporal dimension is necessary, however eminently useful it may be in other respects, for identifying a decision as a judicial amendment, while acknowledging that any decision’s status or acceptance may initially remain uncertain.
Kate Glover Berger engages in two important critiques, one of my application of the multi-factor approach to the Quebec Secession reference, and another to the fate of structural reasoning within the approach generally. Glover Berger objects to characterizing the invocation of a duty to negotiate as a change to the amending formula in Part V of the Constitution Act, 1982. She asserts that “the closed Part V system deals only with the final, formal step of Canada’s amendment process.” This strikes me as a rather narrow and textualist conception of Part V and its purpose. There are parallels to the understanding of the “method of selecting” senators that was rejected by the Supreme Court of Canada in Reference re Senate Reform. A narrow constitutional view of the method of selecting senators in Canada envisions the formal appointment by the governor general, which in practice occurs on the advice of the prime minister. Indeed, in terms of the constitution’s text, the governor general’s role is the only element mentioned in the Constitution Act, 1867 as it relates to senatorial appointment. The federal government, attempting to implement consultative elections for the Senate, argued that the change would not offend the “method of selection” provision referenced in Part V because the prime minister’s discretion over the final decision would be preserved. The Court stated that such an argument “privileges form over substance. It reduces the notion of constitutional amendment to a matter of whether or not the letter of the constitutional text is modified.” The Court went on to declare such an approach inconsistent with a broad and purposive approach to understanding the constitution. Under this view, the entire process leading up to the legislative votes on resolutions in Parliament and the provincial legislatures might be understood as part of the constitution’s amending process. Indeed, the creation of a new constitutional rule for amendments relating to secession seems to imply (incorrectly, in my view) that the constitution itself is capable of regulating that process, at least in specific contexts.
Moreover, Glover Berger’s discussion seems to imply that a process of intergovernmental negotiations is integral to the amending process. But the political sphere, upon which the Court intrudes in invoking the duty, surely leaves open the prospects of a refusal to participate, or the ability to introduce amending resolutions in one’s legislature without first consulting the other partners to Confederation. It is for this reason that I would maintain the duty to negotiate is inextricably tied to Part V, for one might imagine scenarios where adherence to the duty would spur a legislative vote on resolutions that would not have otherwise occurred.
However, even if we accept Glover Berger’s argument that the duty to negotiate does not stand as an amendment to Part V specifically (at least in relation to secession), this seems to have little impact on the application of the multi-factor test in relation to the creation of the duty to negotiate, because it is still an amendment to the constitution; it is a new constitutional rule, invented by the Court, absent any plausible foundation in the constitutional text, with no contemplation by the constitutional framers or, indeed, the purpose of any identifiable element of the written constitution, and to the complete surprise of the political community.
On a more fundamental issue, Glover Berger wonders if, given my analysis of the use of unwritten principles especially, there is a risk under the multi-factor approach that structural reasoning “will always be at least suspect in the eyes of judicial amendment.” This, she cautions, might be contrary to the spirit of my approach, which attempts to consider any broadly accepted approach to constitutional interpretation. I would admit this might be true in cases of ‘structural interpretation’ that abandons the text entirely; where unwritten principles are, without other aid, relied on to add, remove, or modify the constitution in a way that belies the text and the evident purpose of the relevant provisions, and that defies the expectations of the political community, such an approach will likely constitute judicial amendment. However, structural interpretation that informs valid interpretations of the existing text, or that works in concert with well established understandings of purpose or framer’s intent, by drawing on unwritten principles or identifying implications in the text, is perfectly valid. Indeed, I suspect Glover Berger agrees at least with this this last point, given her own superb scholarship on the topic. In that respect, structural reasoning, like any mode of interpretation, can either remain an act of interpretation or tread dangerously into judicial amendment territory.
 Emmett Macfarlane, Judicial Amendment of the Constitution, 19(5) International Journal of Constitutional Law 1894 (2021); Erin F. Delaney and Christopher W. Schmidt, There’s Something About Brown, 19(5) Int’l J. Const. L. 1934 (2021); Kate Glover Berger, Judicial Amendment and our Constitutional Lives, 19(5) Int’l J. Const. L. 1925 (2021).
 Delaney and Schmidt, supra note 1, at 1937.
 Reference re Secession of Quebec,  2 S.C.R. 217
 Glover Berger, supra note 1 at 1928.
 Reference re Senate Reform, 2014 SCC 32
 Id, at para. 52.
 Glover Berger, supra note 1, at 1931.