In an article in the Walrus on the 35th anniversary of the Canadian Charter of Rights and Freedoms, Lauren Heuser raises the issue of what is probably the least loved provision in our constitution ― the Charter‘s section 33, a.k.a. the “notwithstanding clause”. Section 33 allows Parliament and provincial legislatures to immunize legislation from judicial review and invalidation under sections 2 and 7 to 15 of the Charter ― provisions protecting, among other things, the freedoms of religion and expression, the due process rights of the accused in criminal cases, and most equality rights. Ms. Heuser wants us to reconsider the existing norm, some would even say convention (although I do not think it is one yet) against using section 33. She is wrong.
Ms. Heuser quotes Howard Anglin, who points out (correctly) that section 33 was an essential component of the political compromise that made the enactment of the Charter possible, and Emmett Macfarlane, who insists that this compromise reflects a “recognition that the courts wouldn’t always get it right”, and not only a hidebound commitment to “parliamentary sovereignty with no reason”. Ms. Heuser concludes ― it’s not clear to me whether Mr. Anglin and prof. Macfarlane share the conclusion ― that
[c]ontrary to what much of the public has been led to think, then, it is not necessarily inappropriate for government officials to push back when they believe a court gets a ruling wrong; this is as legitimate as a citizen asserting her Charter right to justify some action.
Provided that a legislature can justify itself by “explaining how a court’s policy analysis failed to take account of relevant considerations or contravened the will of the democratic majority”, resorting to section 33 should not be regarded as a political impossibility.
Ms. Heuser is not alone in trying to rehabilitate the “notwithstanding clause”. Attempts to do so are made with some regularity in nationalist circles in Québec, where the Charter is still seen as an illegitimate imposition on the province’s legislative competence. The most recent such attempts have been in response to the purported iniquity of courts giving effect to the constitutional right to be tried within a reasonable time. (Over at À qui de droit, Maxime St-Hilaire has argued convincingly that Québec could not validly override the Supreme Court’s decision in R v Jordan, 2016 SCC 27, [2016] 1 SCR 631 even if it tried, and Finn Makela has ventured what strikes me as at least a plausible political explanation for why politicians ignore these legal arguments. I would only add that this is not the first time there has been talk of invoking section 33 when it was manifestly impossible to do so.) But whoever is making this argument, and whatever their motivations, they are wrong.
Here’s what I wrote when I considered invoking section 33 in response to what I considered to be some of the worst decisions ever made by the Supreme Court under the Charter, those that elevated to constitutional status the rights to bargain collectively and to strike:
[T]he norm … against using the notwithstanding clause is, on balance[,] a very good thing. Courts sometimes make mistakes, even very bad mistakes, as [the labour rights decisions] demonstrate. But, pace Jeremy Waldron, the Supreme Court of Canada’s record on rights issues is still so much better than that of Parliament and the provincial legislatures that we’re better off with not merely a rebuttable presumption, but a bright-line [rule] against legislative corrections of perceived judicial mistakes. The likelihood of the perception being itself mistaken is simply too high.
This should, I think, take care of prof. Macfarlane’s and Mr. Anglin’s arguments. Sure the courts don’t always get it right, and section 33 was put into the Charter as a remedy against courts systematically getting it wrong (as the Charter’s framers thought ― wrongly ― the American courts had during the so-called “Lochner era”). But in light of our experience with the Charter we know that the courts get it right more often than the legislatures that would be relying on section 33 would.
As for Ms. Heuser’s suggestion that legislatures would be justified in setting aside judicial decisions whenever these “contravene[] the will of the democratic majority”, taking it seriously would make those provisions of the Charter that are subject to section 33 so many dead letters. By hypothesis, all democratically enacted legislation reflects the will of the majority, and any judicial finding that such law is unconstitutional contravenes this will. There are occasions when we may be able to show that what I have previously called a “democratic process failure” has occurred, and the law did not in fact reflect the majority’s will. But demonstrating that this has happened is not straightforward, and for obvious reasons legislators will be the last people in the world to accept such claims. Ms. Heuser would, in effect, give them carte blanche to override any judicial decision they disagreed with. This is not a crazy position, to be sure, but those who support it should recognize that they are advocating for a substantial revision of our whole approach to judicial review ― a clear change to constitutional practice rather than a return to the roots.
For the reasons outlined above, I would not support such a change. Although I disagree with the Supreme Court more than most Canadian constitutional lawyers, I still trust its judges more than I trust legislators. That section 33 was the price to pay to have the Charter at all is not a reason to use it now ― or ever. The status quo ain’t broke, and there is no need to fix it.
Double Aspect 
I’m curious as to the origin of the term “systematically” in your phrase “section 33 was put into the Charter as a remedy against courts systematically getting it wrong”. The term seems to import a requirement that judicial error persist over time (presumably in a series of (related?) decisions?) – but that isn’t supported by the text or (I think) the legislative history.
Also, I’d be interested in hearing more about your bright-line rule – what would be the consequences of breaking the rule? Entirely political?
Re “systematically”: what I meant is that I don’t think Blakeney & Co. would have insisted on it just because they thought courts would sometimes err. My impression is that the worry was not mere occasional mistakes, but (perceived) court-goes-off-the-rails situations like the Lochner era, which was on their minds a lot. But I might be wrong about that. And of course there’s textual requirement for that. (You couldn’t write one, could you?) The clause as written can be used whenever. (See also Ford v Québec.)
Re consequences: sure, it’s all political. I think there is a norm against using s 33, but not even a convention.
Your conclusion depends a lot on the premise that you trust judges more than legislators. Of course, you seem to be in the majority here, which suggests the counter-majoritarian difficulty has a counter-majoritarian difficulty.
I have a slightly different contrarian perspective on s. 33 and its effective desuetude. Section 33 created a tool which allowed for the exercise of a vestigial form of Parliamentary sovereignty – in a sense it allowed Parliament and the legislatures (within their proper fields of legislative competence) to be as “bad as they want to be.” By contrast Section 1 allowed Parliament and the legislatures to trench on Charter rights but subject to certain explicit requirements. Unfortunately, given the absence of a well developed body of political precedent allowing for the use of s. 33, courts have increasingly found themselves faced with potentially setting aside major democratically based policy initiatives that infringe Charter rights but which seem to have a fragile or non-existent evidentiary basis (which strikes me as one of the most basic forms of unreasonableness). Because of the courts’ increased concern about institutional deference this has led them (in my view) to take refuge in quite generous views of deference (see for example Newfoundland (Treasury Board) v. N.A.P.E., [2004] 3 SCR 381, 2004 SCC 66) when faced with infringing legislation. If courts were more willing to say “sorry but this does not meet the reasonableness test” and refuse to give relief under s. 1 directing legislatures to s. 33 if they just want to impose decisions, Charter rights may in fact have more robust protection than they do now. That being said, the on the ground reality is that s. 33 is really heading the way of disallowance under the Constitution Act, 1867 and barring the election of a Trump-like Prime Minister it is unlikely we will see its use again.