Blog of the International Journal of Constitutional Law

Thresholds for Interpretation and Secession

Adam Perry, Lecturer in Law, Queen Mary University of London

Three of my favourite topics are statutory interpretation, Prince Charles, and Canadian electoral politics. I never thought these topics were all that closely related. Happily, I was wrong.

For years Prince Charles has been writing letters to government ministers on everything from natural medicine to agricultural policies, and for years the Guardian has been trying to get a hold of them. In 2005, the Guardian made a request to see the letters under the Freedom of Information Act 2000, and in 2012 the Upper Tribunal ordered their disclosure (see here for my discussion of the tribunal’s decision). Everyone thought the Guardian had won. However, the Freedom of Information Act gives the Attorney General a power to veto the disclosure of information on “reasonable grounds”, which the AG immediately used to block the letters’ release. The Guardian didn’t give up, though, and applied for judicial review of the AG’s veto. The application went all the way to the Supreme Court – and the Guardian won, for good this time (the case is Evans v Attorney General). In May of this year the letters were finally disclosed (you can read them here).

The Supreme Court’s decision was a split decision. One disagreement was between Lord Neuberger (in the majority) and Lord Hughes (in dissent) about how to interpret the phrase “reasonable grounds” in the veto power provision. Simplifying a bit, there were possible two readings: (1) a wide reading, which would allow the AG to block the disclosure of information if he reasonably thought it shouldn’t be disclosed; and (2) a narrow reading, which would require the AG to know of, for example, new reasons against disclosure. Both Lord Neuberger and Lord Hughes seemed to accept that (1) would tend to undermine the rule of law, by placing the executive partly beyond the reach of the courts. Both judges accepted that there is a strong interpretive presumption that Parliament intends to respect the rule of law. Lord Neuberger thought the presumption wasn’t overcome in this case, so he opted for (2). Lord Hughes, on the other hand, thought the presumption clearly was overcome, so he opted for (1).

I imagine this disagreement seems familiar. It is like so many disagreements about statutory interpretation, especially in administrative law. One side points to an important value (like the rule of law) to avoid imputing some intention to the legislature. The other side acknowledges the value, but objects to what they see as a strained interpretation.

My suspicion is that in these disagreements the two sides are often talking past each other.

Faced with competing interpretations of a statute, one conversation we can have is about what the legislature probably meant. Is it more probable that the legislature meant A or B? Suppose we decide that it’s more probable than not that the legislature meant A. We attach a 51% probability to A, say. Should we go ahead and impute A to the legislature? Well, no, or not yet anyway. First we need to have a conversation about whether more-probable-than-not is probable enough, or whether we should demand a higher degree of certainty instead. Now we’re no longer talking about probabilities per se, but about thresholds for imputing intentions to the legislature. We’re talking not about proof, but about standards of proof.

Depending on which conversation we think we’re having – about probabilities or thresholds – we’ll be thinking of interpretive presumptions differently. Some interpretive presumptions are just factual generalizations. They set baselines that help us figure out how likely it is that the legislature meant A or B. The presumption against extraterritoriality might be an example (see here for discussion). Other interpretive presumptions seem more akin to the presumption of innocence. They establish probability thresholds, which must be exceeded before we go ahead and impute some meaning to the legislature. If the threshold is set high (i.e., higher than more-probable-than-not) we may end up refusing to impute a meaning to the legislature we know it probably intended.

My impression is that, in the Prince Charles case, Lord Hughes treats the presumption that Parliament intends to respect the rule of law as a generalization. He starts from the position that Parliament probably meant to respect the rule of law, and goes on to assess the evidence for Parliament’s intention against that background. Because he finds the evidence for a rule-of-law-imperiling interpretation persuasive, he imputes that intention to Parliament. My impression is that Lord Neuberger, on the other hand, thinks of the presumption as establishing a threshold. He would say – or so I imagine – that the rule-of-law-imperiling interpretation is probably the right one, but that “probably” isn’t good enough, not when the rule of law is at stake.

Who’s right, Lord Hughes or Lord Neuberger? I wish I knew! There’s certainly a debate to be had about whether presumptions-as-thresholds are consistent with parliamentary sovereignty. There are also interesting parallels to be drawn with other areas of law. One occurred to me the other day, when I was watching a different sort of debate – the first leaders’ debate in the run up to the Canadian general election. During the debate, there was a heated back and forth between Tom Mulcair (leader of the New Democratic Party) and Justin Trudeau (leader of the Liberals) about the conditions under which Canada ought to negotiate Quebec’s secession. The Clarity Act says that negotiations may take place only if a secession referendum is carried by a “clear majority”, meaning something more than 50% plus one. Mulcair said that he would repeal the Clarity Act, potentially making secession easier; Trudeau demurred; controversy ensued.

The “clear majority” requirement looks a bit puzzling on the face of it. If a referendum is fair, and 50% plus one voted for secession, then the best evidence we have is that Quebecers do want to leave Canada. Setting the threshold higher than a mere majority seems undemocratic. But one rationale for the high threshold – obvious to many, no doubt – is that we should demand greater certainty before taking big risks. Before breaking up the country, we should be certain, or nearly so, that it’s what Quebecers want. If that’s right, then the rationale for the Clarity Act parallels Lord Neuberger’s rationale (as I represented it) for narrowly interpreting the AG’s veto power. In both cases, a heavy ‘burden of proof’ is placed on the party seeking to justify undermining something of value – national unity or the rule of law.

Suggested Citation: Adam Perry, Thresholds for Interpretation and Secession, Int’l J. Const. L. Blog, Sept. 2, 2015, at:


One response to “Thresholds for Interpretation and Secession”

  1. […] post originally appeared on the blog of the International Journal of Constitutional Law here on 3 September […]

Leave a Reply

Your email address will not be published. Required fields are marked *