By: Alice Woolley
PDF Version: The Dangers of Inconsistency (and Consistency) in Supreme Court Jurisprudence
Case Commented On: Alberta (Information and Privacy Commissioner) v. University of Calgary, 2016 SCC 53 (CanLII)
I can’t decide whether I am more excited that the Supreme Court issued a decision dealing with two legal issues of great interest to me – administrative law standard of review and statutory incursions into solicitor-client privilege – or irritated that the Court’s handling of both issues is so annoying. Because it is the end of term, and I’m as grumpy as any other professor at the end of term, I am mostly irritated. Irritated because on standard of review the Court seems literally incapable of a consistent and practical approach, while on solicitor-client privilege the Court has been so consistent that it risks fetishizing the significance of solicitor-client confidentiality to the point of jeopardizing other important legal interests.
On standard of review the Court needs to stop. It needs to stop trying to articulate and apply a set of rules for judicial deference to administrative decision-makers. It should instead let administrative judicial review be a matter of practice and the appropriate judicial attitude, one of respectful attention to any decision-maker’s reasons for a particular decision, while recognizing that judges provide a sober second thought through judicial review, particularly on matters of legal interpretation. Along with significantly shifting every decade or so, the rules identified end up being misleading at best and unhelpful at worst, failing to capture the basic and in the end relatively straightforward idea that standard of review reflects. The Court’s attempt to articulate rules governing standard of review is like a baseball coach trying to develop a set of rules for players to use when deciding whether to swing, when the appropriate advice is both simple and incapable of more precise articulation: swing at a strike; don’t swing at a ball (or, alternately, swing at a pitch you have the skill to hit, and leave the rest alone).
On solicitor-client privilege, the Supreme Court can certainly claim to have been consistent: solicitor-client privilege is generously defined and strenuously protected. On the whole, that seems to me a good thing. But this decision raises the possibility that that consistent and vigorous protection may go beyond what is necessary for protection of the privilege, and may occur at the expense of other values of importance to the legal system.
The Decision
The specific legal issue in this case was whether, pursuant to s 56(3) of the Freedom of Information and Protection of Privacy Act, RSA 2000, c F-25, (FOIPP), the Alberta Privacy Commissioner could compel the University of Calgary to produce documents over which the University claimed solicitor-client privilege, in order for the Commissioner to assess the legitimacy of that claim.
The Court unanimously dismissed the Privacy Commissioner’s appeal of the judgment of the Alberta Court of Appeal, agreeing with the Court of Appeal that the Commissioner ought not to review documents over which the University had claimed privilege. All of the judges agreed that the University had sufficiently justified its claim to privilege such that it was improper for the Privacy Commissioner to seek to review the documents pursuant to s 56(3) of FOIPP, whether or not that provision permitted it to review a public body’s claims to solicitor-client privilege (at para 70, Côté J for the majority; para 127, Cromwell J, concurring; para 137, Abella J, concurring). The Court divided, however, both on the question of the appropriate standard of review of the Privacy Commissioner’s decision and on the question of whether s 56(3) of FOIPP did in fact permit the Privacy Commissioner to review documents to assess the legitimacy of a public body’s claim of solicitor-client privilege.
Writing for the majority, Justice Côté held that FOIPP does not empower the Privacy Commissioner to require a public body to produce documents with respect to which it has claimed solicitor-client privilege (at para 2). She held that the Commissioner’s decision that it had the power to compel production was reviewable on a standard of correctness because the question was one of “central importance to the legal system as a whole” (at para 20). Solicitor-client privilege is fundamental to the legal system and has constitutional dimensions, and the assessment of “what statutory language is sufficient to authorize administrative tribunals to infringe solicitor-client privilege is a question that has potentially wide implications on other statutes” (at para 20). Specifically, determining whether the phrase “privilege of the law of evidence” in s 56(3) includes solicitor-client privilege “necessitates an inquiry into both the substantive and evidentiary qualities of the privilege” (at para 25). The Commissioner also had no special expertise with respect to privilege (at para 22).
Section 56(3) of FOIPP provides:
(3) Despite any other enactment or any privilege of the law of evidence, a public body must produce to the Commissioner within 10 days any record or a copy of any record required under subsection (1) or (2).
Justice Côté held that this language was not sufficient to compel production of documents claimed to be privileged. Statutes purporting to set aside privilege will be “interpreted restrictively” and the legislative intent must be “clear and unambiguous” (at para 28); an inference is not sufficient. Imposing this standard does not involve “strict construction” of FOIPP and is consistent with the “modern approach to statutory interpretation;” it simply assumes “legislative respect for fundamental values” (at para 29).
Although s 56(3) only requires production to the Commissioner, not to the party applying for information, it still “constitutes an infringement of the privilege” (at para 35), particularly because the Commissioner is “not an impartial adjudicator of the same nature as a court” (at para 36). The question is: does the statutory language allow the Privacy Commissioner to require production of documents over which a public body claims solicitor-client privilege?
Justice Côté said no: the phrase “privilege of the law of evidence” is “not sufficiently clear and precise to set aside or permit an infringement of solicitor-client privilege” (at para 37). Solicitor-client privilege is a substantive rule with quasi-constitutional status, not merely a law of evidence (at para 38). Further, the evidentiary aspect of the privilege is not engaged here; the FOIPP regime produces documents without connection to any ongoing legal proceeding (at para 42). The substantive privilege should be “as close to absolute as possible and should not be interfered with unless absolutely necessary” (at para 43).
Justice Côté also noted that the statutory context supported this position, and in particular the legislation’s reference to “legal privilege” in the provision allowing a public body to refuse to disclose privileged information – i.e., “information that is subject to any type of legal privilege, including solicitor-client privilege” (at para 52, citing FOIPP s 27). That the power of the Privacy Commissioner to require disclosure uses different language than the power of the public body to refuse production is significant, and suggests that the provisions “must be understood to have different meanings” (at para 53). Solicitor-client privilege is a legal privilege but, here, is “not clearly a ‘privilege of the law of evidence’”, with the result that a public body may refuse to disclose documents over which it claims solicitor-client privilege, and the Commissioner cannot require that they be disclosed for the Commissioner to review (at para 57). Justice Côté noted that while it was possible that under British Columbia’s parallel legislation “privilege of the law of evidence” could include solicitor-client privilege, the differing statutory contexts meant that interpretation could not be “imported into the Alberta statute with equivalent effect” (at para 65).
In his concurring reasons, Justice Cromwell rejected Justice Côté’s interpretation of s 56(3), holding that solicitor-client privilege is both a legal privilege and a privilege of the law of evidence, and that no principle of statutory interpretation requires a different approach to seeing the privilege as within both of those terms (at para 73). The legislature intended to allow the Commissioner to assess claims of solicitor-client privilege in appropriate cases; that intention can be identified from the “grammatical and ordinary meaning of the words ‘any privilege of the law of evidence’” and from contextual factors (at para 79). Solicitor-client privilege has a substantive component, but it is also an evidentiary issue, and it is the evidentiary aspect that is raised by s 56(3) (at para 81); the public body is seeking “protection from disclosure required by legal authority, a matter falling squarely within the evidentiary privilege” (at para 87). The use of the phrase “legal privilege” in the statute does not preclude the interpretation of solicitor-client privilege as a privilege of the law of evidence elsewhere in the statute; all privileges of the law of evidence are legal privileges (at para 92). The Commissioner may not share information it reviews and identifies as properly subject to privilege, but it has the power to rule over claims of privilege (at para 104). Justice Cromwell also viewed the legislative history as supporting this interpretation.
Justice Cromwell “assume[d] without deciding” that correctness review applied to the decision in this case (at para 75). Justice Abella disagreed in her concurring reasons. In her view, this case was fundamentally a matter of statutory interpretation; that it touched on an important legal question did not put make it one of the exceptional cases to which correctness review ought to apply (at para 130). Ultimately, the Privacy Commissioner was not explaining “the content of solicitor-client privilege for the whole legal system, she is being asked to apply it in the context of one provision” (at para 136); that decision is properly reviewed deferentially.
Commentary
Standard of review
The Court’s discussion of standard of review in this case, and the application of the correctness standard by the majority, should be seen in the context of the recent 5-4 split on the Court in Edmonton (City) v Edmonton East (Capilano) Shopping Centres Ltd, 2016 SCC 47 (Can LII) (discussed by Shaun Fluker here, by Paul Daly here and by Leonid Sirota here). Together the two cases suggest that a part of the Court – here the majority – is uncomfortable with the presumption of reasonableness that in recent years has been granted to administrative tribunals interpreting their own statutes. Because, with all due respect to Côté J’s argument that the issue in this case relates to a matter of general importance to the legal system as a whole (solicitor-client privilege), it is hard to accept the characterization of this decision as about solicitor-client privilege, rather than about the interpretation of the Privacy Commissioner’s home statute. In her own decision, Côté J rejects the relevance of the British Columbia statute for interpreting Alberta’s, and relies significantly on inconsistent language used within the Alberta statute to justify her interpretation of the wording of s 56(3). Her decision is, for the most part, a straight up exercise in statutory interpretation.
It is true that to interpret its statute the Privacy Commissioner had to consider whether solicitor-client privilege falls within the expression “privilege of the law of evidence” as that term is employed in FOIPP s 56(3). But that decision has only as much relevance to the legal system as a whole as would any administrative decision that considers or employs legal terms used more generally in the legal system when interpreting or applying its statutory mandate. Fundamentally, the question here was the proper interpretation of s 56(3) of FOIPP. It was not “what does solicitor-client privilege mean”. That the meaning of solicitor-client privilege was relevant to the interpretation of s 56(3) of FOIPP did not make the question something other than a matter of statutory interpretation.
The decision may end up having broader significance for interpretation of statutes other than FOIPP, but that is because it is a decision of the Supreme Court of Canada, not because of the nature of the issue. A decision by the Privacy Commissioner on this question seems likely to have had few consequences for anything other than the meaning of s 56(3) of FOIPP. The broader significance – the general importance to the legal system as a whole – arises from the Supreme Court’s judgment. And surely the fact of judicial review cannot make a matter of statutory interpretation something of general significance, or every administrative decision that went to the Supreme Court would have to be reviewed on a correctness basis.
Don’t get me wrong. My point (here at least) is not to object to the use of a correctness standard in cases such as this one. My point is that doing so seems irreconcilable with the presumption of reasonableness for interpretation of a decision-maker’s home statute. At best, the Court continues to be inconsistent and unpredictable in its identification of the standard of review in specific cases and, at worst, the Court may be about to re-articulate again how the standard of review is to be identified. This judgment and Edmonton East (Capilano) suggest a Court at the brink of a major reconsideration of the issue. Further, the Court’s inconsistency and disagreement on the presumption of reasonableness exist alongside the observation that, even when the Court does consistently identify a deferential standard, how it and other courts “do deference” in any given case varies significantly (see Shaun Fluker on this point here and Paul Daly here). The Court has not truly settled on when a deferential standard ought to be applied or, when it is, what deference should look like.
And yet what does this confusion give us? Does it elucidate the fundamental tension? Does it yield insights beyond those offered by Dickson J (as he then was) back in 1979 in CUPE v New Brunswick Liquor Corporation, [1979] 2 SCR 227, 1979 CanLII 23 (SCC)? From the perspective of an administrative law dabbler and teacher like myself, the answer feels like “no, not at all”. I am literally at a loss as to how I can give my students any sort of useful understanding of administrative law when I teach it next term. And, yet, it feels like the confusion is unnecessary. Why can’t the answer be, simply, that a court should always pay attention to what an administrative decision-maker knows and the answers that it gives, but that it should do so critically, with awareness that the court also has knowledge that it should bring to bear to matters brought before it on judicial review, particularly when they involve interpretation of the law.
A court shouldn’t need a formula, a series of “factors” to consider, a set of presumptions or rules, or even a requirement that sometimes the standard of review is correctness and sometimes it is reasonableness. As a judge, just look at what you’ve been asked to consider in light of the relative knowledge of you as a judge and the administrative body as a decision-maker, always give respectful attention to the administrative body’s reasons for decision, and then decide whether the administrative body’s decision ought to stand given its statutory authority and the matter at issue in the case. If all judges did that, they’d be fine, and a lot less judicial and academic ink could be spilled in the process. Maybe sometimes courts would interfere when they shouldn’t, and maybe sometimes courts wouldn’t interfere when they should – but that happens now despite all the drama of standard of review; it’s not like we’re preventing it. And after all, appellate courts review trial judgments without so much hassle and confusion; I just do not accept that administrative law couldn’t be the same.
Now I have to acknowledge that I may be – in fact I’m likely to be – wildly off base in this suggestion. A question that the Supreme Court has struggled with for decades seems unlikely to be resolved by the end-of-term grumpy ramblings of a legal ethics professor who dabbles in administrative law. But I do believe that there is something to my basic point, which is that more of the same – more rules, formulas, factors or tests – is unlikely to fix the conundrum of standard of review. And it is certainly unlikely to make the thought of teaching administrative law next term any more palatable.
Solicitor-client privilege
On solicitor-client privilege, the Court has no such doubts or backtracking. Its approach is generally entirely consistent, Justice Cromwell’s dissent on this issue notwithstanding: solicitor-client privilege is a central aspect of the legal system and fundamental justice; it must be jealously protected and rarely interfered with; legislative incursions on solicitor-client privilege are subject to constitutional scrutiny where s 7 or s 8 of the Charter are at play (for a more fulsome discussion, see Adam Dodek’s terrific book, Solicitor-Client Privilege (Toronto: LexisNexis Canada, 2014)).
Generally speaking, I agree with the Court’s approach to solicitor-client privilege (see chapter 5 of Understanding Lawyers’ Ethics in Canada, 2d ed (Toronto: LexisNexis Canada, 2017)). But given both this case and the Court’s 2015 decision holding that money-laundering legislation was unconstitutional in part due to its effect on solicitor-client privilege (Attorney General (Canada) v Federation of Law Societies, 2015 SCC 7 (Can LII); my blog post here), I do worry whether the Court is losing perspective on what truly constitutes an interference with solicitor-client privilege and confidentiality.
In this case, what would allowing s 56(3) to apply to solicitor-client privilege actually do? It would mean that when a public body did not provide sufficient detail to support a claim to solicitor-client privilege, the Privacy Commissioner could review the documents to determine whether or not they were legitimately classified as privileged. Assuming that the Privacy Commissioner does not exercise that power unlawfully (which I think has to be assumed in assessing the law’s interpretive scope), how much of an incursion on the solicitor-client relationship would this really be? The point is not to deprive the public body of solicitor-client privilege. The point is only to require the public body to properly justify its claims to privilege, and to provide some check on abuse by a public body that claims privilege improperly or dishonestly. I confess that I find it difficult to see that mild incursion as requiring the kind of handwringing that the Court engages in here, and see the Court’s imposition of an onerous approach to statutory interpretation for a legislature to justify that sort of mild incursion on the privilege as a bit overwrought.
If s 56(3) applied to solicitor-client privilege I do not think officials of a public body would be more reluctant to confide in their lawyers. I do not think that privileged communications would improperly fall into the public domain – I am willing to assume the Privacy Commissioner will generally act lawfully. I do think that public bodies would be more careful to provide appropriate justification for privilege claims. And that, it seems to me, is an important aspect of the general legislative scheme that received insufficient attention from the majority of the Court.
In the case of privilege, the Court’s consistency has, I think, clouded its ability to see that not every incursion into the privilege is the same, and that protecting the privilege to this extent has real consequences for other legitimate concerns of the legal system. And I worry that the Court’s approach may lead to an unhealthy counter-reaction to the power this gives to lawyers and their clients to obfuscate the truth – particularly if those clients are powerful and dishonest. The University of Calgary properly justified its claim to privilege. But if another public body does not do so, the Privacy Commissioner will, presumably, have only the recourse of an application to the court to address that deficiency. That result seems unfortunate and unwarranted by the harm (if any) prevented by the Court’s approach to s 56(3).
This post may be cited as: Alice Woolley “The Dangers of Inconsistency (and Consistency) in Supreme Court Jurisprudence” (7 December, 2016), online: ABlawg, https://ablawg.ca/wp-content/uploads/2016/12/Blog_AW_PrivacyCommissioner.pdf
To subscribe to ABlawg by email or RSS feed, please go to https://ablawg.ca
Follow us on Twitter @ABlawg