Sherman Estate v. Donovan, 2021 SCC 25, [2021] 2 SCR 75

SUPREME COURT OF CANADA
Citation: Sherman Estate v. Donovan, 2021 SCC 25, [2021] 2 S.C.R. 75		Appeal Heard: October 6, 2020 Judgment Rendered: June 11, 2021 Docket: 38695
Between: Estate of Bernard Sherman and Trustees of the Estate and Estate of Honey Sherman and Trustees of the Estate Appellants   and   Kevin Donovan and Toronto Star Newspapers Ltd. Respondents   - and -   Attorney General of Ontario, Attorney General of British Columbia, Canadian Civil Liberties Association, Income Security Advocacy Centre, Ad IDEM/Canadian Media Lawyers Association, Postmedia Network Inc., CTV, a Division of Bell Media Inc., Global News, a division of Corus Television Limited Partnership, The Globe and Mail Inc., Citytv, a division of Rogers Media Inc., British Columbia Civil Liberties Association, HIV & AIDS Legal Clinic Ontario, HIV Legal Network and Mental Health Legal Committee Interveners
Coram: Wagner C.J. and Moldaver, Karakatsanis, Brown, Rowe, Martin and Kasirer JJ.

Reasons for Judgment: (paras. 1 to 108)

Kasirer J. (Wagner C.J. and Moldaver, Karakatsanis, Brown, Rowe and Martin JJ. concurring)

 

Estate of Bernard Sherman and Trustees of the Estate and

Estate of Honey Sherman and Trustees of the Estate                               Appellants

v.

Kevin Donovan and

Toronto Star Newspapers Ltd.                                                                 Respondents

and

Attorney General of Ontario,

Attorney General of British Columbia,

Canadian Civil Liberties Association,

Income Security Advocacy Centre,

Ad IDEM/Canadian Media Lawyers Association,

Postmedia Network Inc., CTV, a Division of Bell Media Inc.,

Global News, a division of Corus Television Limited Partnership,

The Globe and Mail Inc., Citytv, a division of Rogers Media Inc.,

British Columbia Civil Liberties Association,

HIV & AIDS Legal Clinic Ontario,
HIV Legal Network and Mental Health Legal Committee                      Interveners

Indexed as: Sherman Estate v. Donovan

2021 SCC 25

File No.: 38695.

2020: October 6; 2021: June 11.

Present: Wagner C.J. and Moldaver, Karakatsanis, Brown, Rowe, Martin and Kasirer JJ.

on appeal from the court of appeal for ontario

                    Courts — Open court principle — Sealing orders — Discretionary limits on court openness — Important public interest — Privacy — Dignity — Physical safety — Unexplained deaths of prominent couple generating intense public scrutiny and prompting trustees of estates to apply for sealing of probate files — Whether privacy and physical safety concerns advanced by estate trustees amount to important public interests at such serious risk to justify issuance of sealing orders.

                    A prominent couple was found dead in their home. Their deaths had no apparent explanation and generated intense public interest. To this day, the identity and motive of those responsible remain unknown, and the deaths are being investigated as homicides. The estate trustees sought to stem the intense press scrutiny prompted by the events by seeking sealing orders of the probate files. Initially granted, the sealing orders were challenged by a journalist who had reported on the couple’s deaths, and by the newspaper for which he wrote. The application judge sealed the probate files, concluding that the harmful effects of the sealing orders were substantially outweighed by the salutary effects on privacy and physical safety interests. The Court of Appeal unanimously allowed the appeal and lifted the sealing orders. It concluded that the privacy interest advanced lacked a public interest quality, and that there was no evidence of a real risk to anyone’s physical safety.

                    Held: The appeal should be dismissed.

                    The estate trustees have failed to establish a serious risk to an important public interest under the test for discretionary limits on court openness. As such, the sealing orders should not have been issued. Open courts can be a source of inconvenience and embarrassment, but this discomfort is not, as a general matter, enough to overturn the strong presumption of openness. That said, personal information disseminated in open court can be more than a source of discomfort and may result in an affront to a person’s dignity. Insofar as privacy serves to protect individuals from this affront, it is an important public interest and a court can make an exception to the open court principle if it is at serious risk. In this case, the risks to privacy and physical safety cannot be said to be sufficiently serious.

                    Court proceedings are presumptively open to the public. Court openness is protected by the constitutional guarantee of freedom of expression and is essential to the proper functioning of Canadian democracy. Reporting on court proceedings by a free press is often said to be inseparable from the principle of open justice. The open court principle is engaged by all judicial proceedings, whatever their nature. Matters in a probate file are not quintessentially private or fundamentally administrative. Obtaining a certificate of appointment of estate trustee in Ontario is a court proceeding engaging the fundamental rationale for openness — discouraging mischief and ensuring confidence in the administration of justice through transparency — such that the strong presumption of openness applies.

                    The test for discretionary limits on court openness is directed at maintaining the presumption while offering sufficient flexibility for courts to protect other public interests where they arise. In order to succeed, the person asking a court to exercise discretion in a way that limits the open court presumption must establish that (1) court openness poses a serious risk to an important public interest; (2) the order sought is necessary to prevent this serious risk to the identified interest because reasonably alternative measures will not prevent this risk; and (3) as a matter of proportionality, the benefits of the order outweigh its negative effects.

                    The recognized scope of what interests might justify a discretionary exception to open courts has broadened over time and now extends generally to important public interests. The breadth of this category transcends the interests of the parties to the dispute and provides significant flexibility to address harm to fundamental values in our society that unqualified openness could cause. While there is no closed list of important public interests, courts must be cautious and alive to the fundamental importance of the open court rule when they are identifying them. Determining what is an important public interest can be done in the abstract at the level of general principles that extend beyond the parties to the particular dispute. By contrast, whether that interest is at serious risk is a fact‑based finding that is necessarily made in context. The identification of an important interest and the seriousness of the risk to that interest are thus theoretically separate and qualitatively distinct operations.

                    Privacy has been championed as a fundamental consideration in a free society, and its public importance has been recognized in various settings. Though an individual’s privacy will be pre‑eminently important to that individual, the protection of privacy is also in the interest of society as a whole. Privacy therefore cannot be rejected as a mere personal concern: some personal concerns relating to privacy overlap with public interests.

                    However, cast too broadly, the recognition of a public interest in privacy could threaten the strong presumption of openness. The privacy of individuals will be at risk in many court proceedings. Furthermore, privacy is a complex and contextual concept, making it difficult for courts to measure. Recognizing an important interest in privacy generally would accordingly be unworkable.

                    Instead, the public character of the privacy interest involves protecting individuals from the threat to their dignity. Dignity in this sense involves the right to present core aspects of oneself to others in a considered and controlled manner; it is an expression of an individual’s unique personality or personhood. This interest is consistent with the Court’s emphasis on the importance of privacy, but is tailored to preserve the strong presumption of openness.

                    Privacy as predicated on dignity will be at serious risk in limited circumstances. Neither the sensibilities of individuals nor the fact that openness is disadvantageous, embarrassing or distressing to certain individuals will generally on their own warrant interference with court openness. Dignity will be at serious risk only where the information that would be disseminated as a result of court openness is sufficiently sensitive or private such that openness can be shown to meaningfully strike at the individual’s biographical core in a manner that threatens their integrity. The question is whether the information reveals something intimate and personal about the individual, their lifestyle or their experiences.

                    In cases where the information is sufficiently sensitive to strike at an individual’s biographical core, a court must then ask whether a serious risk to the interest is made out in the full factual context of the case. The seriousness of the risk may be affected by the extent to which information is disseminated and already in the public domain, and the probability of the dissemination actually occurring. The burden is on the applicant to show that privacy, understood in reference to dignity, is at serious risk; this erects a fact‑specific threshold consistent with the presumption of openness.

                    There is also an important public interest in protecting individuals from physical harm, but a discretionary order limiting court openness can only be made where there is a serious risk to this important public interest. Direct evidence is not necessarily required to establish a serious risk to an important public interest, as objectively discernable harm may be identified on the basis of logical inferences. But this process of inferential reasoning is not a licence to engage in impermissible speculation. It is not just the probability of the feared harm, but also the gravity of the harm itself that is relevant to the assessment of serious risk. Where the feared harm is particularly serious, the probability that this harm materialize need not be shown to be likely, but must still be more than negligible, fanciful or speculative. Mere assertions of grave physical harm are therefore insufficient.

                    In addition to a serious risk to an important interest, it must be shown that the particular order sought is necessary to address the risk and that the benefits of the order outweigh its negative effects as a matter of proportionality. This contextual balancing, informed by the importance of the open court principle, presents a final barrier to those seeking a discretionary limit on court openness for the purposes of privacy protection.

                    In the present case, the risk to the important public interest in privacy, defined in reference to dignity, is not serious. The information contained in the probate files does not reveal anything particularly private or highly sensitive. It has not been shown that it would strike at the biographical core of the affected individuals in a way that would undermine their control over the expression of their identities. Furthermore, the record does not show a serious risk of physical harm. The estate trustees asked the application judge to infer not only the fact that harm would befall the affected individuals, but also that a person or persons exist who wish to harm them. To infer all this on the basis of the deaths and the association of the affected individuals with the deceased is not a reasonable inference but is speculation.

                    Even if the estate trustees had succeeded in showing a serious risk to privacy, a publication ban — less constraining on openness than the sealing orders — would have likely been sufficient as a reasonable alternative to prevent this risk. As a final barrier, the estate trustees would have had to show that the benefits of any order necessary to protect from a serious risk to the important public interest outweighed the harmful effects of the order.

Cases Cited

                    Applied: Sierra Club of Canada v. Canada (Minister of Finance), 2002 SCC 41, [2002] 2 S.C.R. 522; referred to: Canadian Broadcasting Corp. v. New Brunswick (Attorney General), [1996] 3 S.C.R. 480; Vancouver Sun (Re), 2004 SCC 43, [2004] 2 S.C.R. 332; Khuja v. Times Newspapers Ltd., [2017] UKSC 49, [2019] A.C. 161; Edmonton Journal v. Alberta (Attorney General), [1989] 2 S.C.R. 1326; Dagenais v. Canadian Broadcasting Corp., [1994] 3 S.C.R. 835; R. v. Mentuck, 2001 SCC 76, [2001] 3 S.C.R. 442; Lavigne v. Canada (Office of the Commissioner of Official Languages), 2002 SCC 53, [2002] 2 S.C.R. 773; Dagg v. Canada (Minister of Finance), [1997] 2 S.C.R. 403; R. v. Henry, 2009 BCCA 86, 270 B.C.A.C. 5; Attorney General of Nova Scotia v. MacIntyre, [1982] 1 S.C.R. 175; A.B. v. Bragg Communications Inc., 2012 SCC 46, [2012] 2 S.C.R. 567; Toronto Star Newspapers Ltd. v. Ontario, 2005 SCC 41, [2005] 2 S.C.R. 188; Re Southam Inc. and The Queen (No.1) (1983), 41 O.R. (2d) 11; R. v. Oakes, [1986] 1 S.C.R. 103; Otis v. Otis (2004), 7 E.T.R. (3d) 221; H. (M.E.) v. Williams, 2012 ONCA 35, 108 O.R. (3d) 321; F.N. (Re), 2000 SCC 35, [2000] 1 S.C.R. 880; R. v. Dyment, [1988] 2 S.C.R. 417; Alberta (Information and Privacy Commissioner) v. United Food and Commercial Workers, Local 401, 2013 SCC 62, [2013] 3 S.C.R. 733; Toronto Star Newspaper Ltd. v. R., 2012 ONCJ 27, 289 C.C.C. (3d) 549; Douez v. Facebook, Inc., 2017 SCC 33, [2017] 1 S.C.R. 751; R. v. Paterson (1998), 102 B.C.A.C. 200; S. v. Lamontagne, 2020 QCCA 663; Himel v. Greenberg, 2010 ONSC 2325, 93 R.F.L. (6th) 357; A.B. v. Canada (Citizenship and Immigration), 2017 FC 629; R. v. Pickton, 2010 BCSC 1198; Lac d’Amiante du Québec Ltée v. 2858‑0702 Québec Inc., 2001 SCC 51, [2001] 2 S.C.R. 743; 3834310 Canada inc. v. Chamberland, 2004 CanLII 4122; R. v. Spencer, 2014 SCC 43, [2014] 2 S.C.R. 212; Coltsfoot Publishing Ltd. v. Foster‑Jacques, 2012 NSCA 83, 320 N.S.R. (2d) 166; Goulet v. Transamerica Life Insurance Co. of Canada, 2002 SCC 21, [2002] 1 S.C.R. 719; Godbout v. Longueuil (Ville de), [1995] R.J.Q. 2561, aff’d [1997] 3 S.C.R. 844; A. v. B., 1990 CanLII 3132; R. v. Plant, [1993] 3 S.C.R. 281; R. v. Tessling, 2004 SCC 67, [2004] 3 S.C.R. 432; R. v. Cole, 2012 SCC 53, [2012] 3 S.C.R. 34; Work Safe Twerk Safe v. Her Majesty the Queen in Right of Ontario, 2021 ONSC 1100; Fedeli v. Brown, 2020 ONSC 994; R. v. Marakah, 2017 SCC 59, [2017] 2 S.C.R. 608; R. v. Quesnelle, 2014 SCC 46, [2014] 2 S.C.R. 390; R. v. Mabior, 2012 SCC 47, [2012] 2 S.C.R. 584; R. v. Chanmany, 2016 ONCA 576, 352 O.A.C. 121; X. v. Y., 2011 BCSC 943, 21 B.C.L.R. (5th) 410; R. v. Esseghaier, 2017 ONCA 970, 356 C.C.C. (3d) 455.

Statutes and Regulations Cited

Bill C‑11, An Act to enact the Consumer Privacy Protection Act and the Personal Information and Data Protection Tribunal Act and to make consequential and related amendments to other Acts, 2nd Sess., 43rd Parl., 2020.

Canadian Charter of Rights and Freedoms , ss. 2(b) , 8 .

Charter of Human Rights and Freedoms, CQLR, c. C‑12, s. 5.

Civil Code of Québec, arts. 35 to 41.

Code of Civil Procedure, CQLR, c. C‑25.01, art. 12.

Freedom of Information and Protection of Privacy Act, R.S.O. 1990, c. F.31.

Personal Information Protection and Electronic Documents Act , S.C. 2000, c. 5 .

Privacy Act , R.S.C. 1985, c. P‑21 .

Authors Cited

Ardia, David S. “Privacy and Court Records: Online Access and the Loss of Practical Obscurity” (2017), 4 U. Ill. L. Rev. 1385.

Austin, Lisa M. “Re‑reading Westin” (2019), 20 Theor. Inq. L. 53.

Bailey, Jane, and Jacquelyn Burkell. “Revisiting the Open Court Principle in an Era of Online Publication: Questioning Presumptive Public Access to Parties’ and Witnesses’ Personal Information” (2016), 48 Ottawa L. Rev. 143.

Cockfield, Arthur J. “Protecting the Social Value of Privacy in the Context of State Investigations Using New Technologies” (2007), 40 U.B.C. L. Rev. 41.

Eltis, Karen. Courts, Litigants, and the Digital Age, 2nd ed. Toronto: Irwin Law, 2016.

Eltis, Karen. “The Judicial System in the Digital Age: Revisiting the Relationship between Privacy and Accessibility in the Cyber Context” (2011), 56 McGill L.J. 289.

Ferland, Denis, et Benoît Emery. Précis de procédure civile du Québec, vol. 1, 6^e éd. Montréal: Yvon Blais, 2020.

Gewirtz, Paul. “Privacy and Speech”, [2001] Sup. Ct. Rev. 139.

Guillemard, Sylvette, et Séverine Menétrey. Comprendre la procédure civile québécoise, 2^e éd. Montréal: Yvon Blais, 2017.

Hughes, Kirsty. “A Behavioural Understanding of Privacy and its Implications for Privacy Law” (2012), 75 Mod. L. Rev. 806.

Matheson, David. “Dignity and Selective Self‑Presentation”, in Ian Kerr, Valerie Steeves and Carole Lucock, eds., Lessons from the Identity Trail: Anonymity, Privacy and Identity in a Networked Society. New York: Oxford University Press, 2009, 319.

McIsaac, Barbara, Kris Klein, and Shaun Brown. The Law of Privacy in Canada, vol. 1. Toronto: Thomson Reuters, 2000 (loose‑leaf updated 2020, release 11).

McLachlin, Beverley. “Courts, Transparency and Public Confidence – To the Better Administration of Justice” (2003), 8 Deakin L. Rev. 1.

Paton‑Simpson, Elizabeth. “Privacy and the Reasonable Paranoid: The Protection of Privacy in Public Places” (2000), 50 U.T.L.J. 305.

Perell, Paul M., and John W. Morden. The Law of Civil Procedure in Ontario, 4th ed. Toronto: LexisNexis, 2020.

Québec. Ministère de la Justice. Commentaires de la ministre de la Justice: Code de procédure civile, chapitre C‑25.01. Montréal: SOQUIJ, 2015.

Rochette, Sébastien, et Jean-François Côté. “Article 12”, dans Luc Chamberland, dir. Le grand collectif: Code de procédure civile — Commentaires et annotations, vol. 1, 5^e éd. Montréal: Yvon Blais, 2020.

Rossiter, James. Law of Publication Bans, Private Hearings and Sealing Orders. Toronto: Thomson Reuters, 2006 (loose‑leaf updated 2020, release 2).

Solove, Daniel J. “Conceptualizing Privacy” (2002), 90 Cal. L. Rev. 1087.

                    APPEAL from a judgment of the Ontario Court of Appeal (Doherty, Rouleau and Hourigan JJ.A.), 2019 ONCA 376, 47 E.T.R. (4th) 1, [2019] O.J. No. 2373 (QL), 2019 CarswellOnt 6867 (WL Can.), setting aside a decision of Dunphy J., 2018 ONSC 4706, 417 C.R.R. (2d) 321, 41 E.T.R. (4th) 126, 28 C.P.C. (8th) 102, [2018] O.J. No. 4121 (QL), 2018 CarswellOnt 13017 (WL Can.). Appeal dismissed.

                    Chantelle Cseh and Timothy Youdan, for the appellants.

                    Iris Fischer and Skye A. Sepp, for the respondents.

                    Peter Scrutton, for the intervener the Attorney General of Ontario.

                    Jaqueline Hughes, for the intervener the Attorney General of British Columbia.

                    Ryder Gilliland, for the intervener the Canadian Civil Liberties Association.

                    Ewa Krajewska, for the intervener the Income Security Advocacy Centre.

                    Robert S. Anderson, Q.C., for the interveners Ad IDEM/Canadian Media Lawyers Association, Postmedia Network Inc., CTV, a Division of Bell Media Inc., Global News, a division of Corus Television Limited Partnership, The Globe and Mail Inc. and Citytv, a division of Rogers Media Inc.

                    Adam Goldenberg, for the intervener the British Columbia Civil Liberties Association.

                    Khalid Janmohamed, for the interveners the HIV & AIDS Legal Clinic Ontario, the HIV Legal Network and the Mental Health Legal Committee.

                    The judgment of the Court was delivered by

                    Kasirer J. —

I.               Overview

[1]                             This Court has been resolute in recognizing that the open court principle is protected by the constitutionally‑entrenched right of freedom of expression and, as such, it represents a central feature of a liberal democracy. As a general rule, the public can attend hearings and consult court files and the press — the eyes and ears of the public — is left free to inquire and comment on the workings of the courts, all of which helps make the justice system fair and accountable.

[2]                             Accordingly, there is a strong presumption in favour of open courts. It is understood that this allows for public scrutiny which can be the source of inconvenience and even embarrassment to those who feel that their engagement in the justice system brings intrusion into their private lives. But this discomfort is not, as a general matter, enough to overturn the strong presumption that the public can attend hearings and that court files can be consulted and reported upon by the free press.

[3]                             Notwithstanding this presumption, exceptional circumstances do arise where competing interests justify a restriction on the open court principle. Where a discretionary court order limiting constitutionally‑protected openness is sought — for example, a sealing order, a publication ban, an order excluding the public from a hearing, or a redaction order — the applicant must demonstrate, as a threshold requirement, that openness presents a serious risk to a competing interest of public importance. That this requirement is considered a high bar serves to maintain the strong presumption of open courts. Moreover, the protection of open courts does not stop there. The applicant must still show that the order is necessary to prevent the risk and that, as a matter of proportionality, the benefits of that order restricting openness outweigh its negative effects.

[4]                             This appeal turns on whether concerns advanced by persons seeking an exception to the ordinarily open court file in probate proceedings — the concerns for privacy of the affected individuals and their physical safety — amount to important public interests that are at such serious risk that the files should be sealed. The parties to this appeal agree that physical safety is an important public interest that could justify a sealing order but disagree as to whether that interest would be at serious risk, in the circumstances of this case, should the files be unsealed. They further disagree whether privacy is in itself an important interest that could justify a sealing order. The appellants say that privacy is a public interest of sufficient import that can justify limits on openness, especially in light of the threats individuals face as technology facilitates widespread dissemination of personally sensitive information. They argue that the Court of Appeal was mistaken to say that personal concerns for privacy, without more, lack the public interest component that is properly the subject‑matter of a sealing order.

[5]                             This Court has, in different settings, consistently championed privacy as a fundamental consideration in a free society. Pointing to cases decided in other contexts, the appellants contend that privacy should be recognized here as a public interest that, on the facts of this case, substantiates their plea for orders sealing the probate files. The respondents resist, recalling that privacy has generally been seen as a poor justification for an exception to openness. After all, they say, virtually every court proceeding entails some disquiet for the lives of those concerned and these intrusions on privacy must be tolerated because open courts are essential to a healthy democracy.

[6]                             This appeal offers, then, an occasion to decide whether privacy can amount to a public interest in the open court jurisprudence and, if so, whether openness puts privacy at serious risk here so as to justify the kind of orders sought by the appellants.

[7]                             For the reasons that follow, I propose to recognize an aspect of privacy as an important public interest for the purposes of the relevant test from Sierra Club of Canada v. Canada (Minister of Finance), 2002 SCC 41, [2002] 2 S.C.R. 522. Proceedings in open court can lead to the dissemination of highly sensitive personal information that would result not just in discomfort or embarrassment, but in an affront to the affected person’s dignity. Where this narrower dimension of privacy, rooted in what I see as the public interest in protecting human dignity, is shown to be at serious risk, an exception to the open court principle may be justified.

[8]                             In this case, and with this interest in mind, it cannot be said that the risk to privacy is sufficiently serious to overcome the strong presumption of openness. The same is true of the risk to physical safety here. The Court of Appeal was right in the circumstances to set aside the sealing orders and I would therefore dismiss the appeal.

II.            Background

[9]                             Prominent in business and philanthropic circles, Bernard Sherman and Honey Sherman were found dead in their Toronto home in December of 2017. Their deaths had no apparent explanation and generated intense public interest and press scrutiny. In January of the following year, the Toronto Police Service announced that the deaths were being investigated as homicides. As the present matter came before the courts, the identity and motive of those responsible remained unknown.

[10]                         The couple’s estates and estate trustees (collectively the “Trustees”)^[1] sought to stem the intense press scrutiny prompted by the events. The Trustees hoped to see to the orderly transfer of the couple’s property, at arm’s length from what they saw as the public’s morbid interest in the unexplained deaths and the curiosity around apparently great sums of money involved.

[11]                         When the time came to obtain certificates of appointment of estate trustee from the Superior Court of Justice, the Trustees sought a sealing order so that the estate trustees and beneficiaries (“affected individuals”) might be spared any further intrusions into their privacy and be protected from what was alleged to be a risk to their safety. The Trustees argued that if the information in the court files was revealed to the public, the safety of the affected individuals would be at risk and their privacy compromised as long as the deaths were unexplained and those responsible for the tragedy remained at large. In support of their request, they argued that there was a real and substantial risk that the affected individuals would suffer serious harm from the public exposure of the materials in the circumstances.

[12]                         Initially granted, the sealing orders were challenged by Kevin Donovan, a journalist who had written a series of articles on the couple’s deaths, and Toronto Star Newspapers Ltd., for which he wrote (collectively the “Toronto Star”).^[2] The Toronto Star said the orders violated its constitutional rights of freedom of expression and freedom of the press, as well as the attending principle that the workings of the courts should be open to the public as a means of guaranteeing the fair and transparent administration of justice.

III.         Proceedings Below

A.           Ontario Superior Court of Justice, 2018 ONSC 4706, 41 E.T.R. (4th) 126 (Dunphy J.)

[13]                         In addressing whether the circumstances warranted interference with the open court principle, the application judge relied on this Court’s judgment in Sierra Club. He noted that a confidentiality order should only be granted when: “(1) such an order is necessary . . . to prevent a serious risk to an important interest because reasonable alternative measures will not prevent the risk; and (2) the salutary effects of the confidentiality order outweigh its deleterious effects, including the effects on the right to free expression and the public interest in open and accessible court proceedings” (para. 13(d)).

[14]                         The application judge considered whether the Trustees’ interests would be served by granting the sealing orders. In his view, the Trustees had correctly identified two legitimate interests in support of making an exception to the open court principle: “protecting the privacy and dignity of victims of crime and their loved ones” and “a reasonable apprehension of risk on behalf of those known to have an interest in receiving or administering the assets of the deceased” (paras. 22‑25). With respect to the first interest, the application judge found that “[t]he degree of intrusion on that privacy and dignity has already been extreme and . . . excruciating” (para. 23). For the second interest, although he noted that “it would have been preferable to include objective evidence of the gravity of that risk from, for example, the police responsible for the investigation”, he concluded that “the lack of such evidence is not fatal” (para. 24). Rather, the necessary inferences could be drawn from the circumstances notably the “willingness of the perpetrator(s) of the crimes to resort to extreme violence to pursue whatever motive existed” (ibid.). He concluded that the “current uncertainty” was the source of a reasonable apprehension of the risk of harm and, further, that the foreseeable harm was “grave” (ibid.).

[15]                         The application judge ultimately accepted the Trustees’ submission that these interests “very strongly outweigh” what he called the proportionately narrow public interest in the “essentially administrative files” at issue (paras. 31 and 33). He therefore concluded that the harmful effects of the sealing orders were substantially outweighed by the salutary effects on the rights and interests of the affected individuals.

[16]                         Finally, the application judge considered what order would protect the affected individuals while infringing upon the open court principle to the minimum extent possible. He decided no meaningful part of either file could be disclosed if one were to make the redactions necessary to protect the interests he had identified. Open‑ended sealing orders did not, however, sit well with him. The application judge therefore sealed the files for an initial period of two years, with the possibility of renewal.

B.            Court of Appeal for Ontario, 2019 ONCA 376, 47 E.T.R. (4th) 1 (Doherty, Rouleau and Hourigan JJ.A.)

[17]                         The Toronto Star’s appeal was allowed, unanimously, and the sealing orders were lifted.

[18]                         The Court of Appeal considered the two interests advanced before the application judge in support of the orders to seal the probate files. As to the need to protect the privacy and dignity of the victims of violent crime and their loved ones, it recalled that the kind of interest that is properly protected by a sealing order must have a public interest component. Citing Sierra Club, the Court of Appeal wrote that “[p]ersonal concerns cannot, without more, justify an order sealing material that would normally be available to the public under the open court principle” (para. 10). It concluded that the privacy interest for which the Trustees sought protection lacked this quality of public interest.

[19]                         While it recognized the personal safety of individuals as an important public interest generally, the Court of Appeal wrote that there was no evidence in this case that could warrant a finding that disclosure of the contents of the estate files posed a real risk to anyone’s physical safety. The application judge had erred on this point: “the suggestion that the beneficiaries and trustees are somehow at risk because the Shermans were murdered is not an inference, but is speculation. It provides no basis for a sealing order” (para. 16).

[20]                         The Court of Appeal concluded that the Trustees had failed the first stage of the test for obtaining orders sealing the probate files. It therefore allowed the appeal and set aside the orders.

C.            Subsequent Proceedings

[21]                         The Court of Appeal’s order setting aside the sealing orders has been stayed pending the disposition of this appeal. The Toronto Star brought a motion to adduce new evidence on this appeal, comprised of land titles documents, transcripts of the cross‑examination of a detective on the murder investigation, and various news articles. This evidence, it says, supports the conclusion that the sealing orders should be lifted. The motion was referred to this panel.

IV.         Submissions

[22]                         The Trustees have appealed to this Court seeking to restore the sealing orders made by the application judge. In addition to contesting the motion for new evidence, they maintain that the orders are necessary to prevent a serious risk to the privacy and physical safety of the affected individuals and that the salutary effects of sealing the court probate files outweigh the harmful effects of limiting court openness. The Trustees argue that two legal errors led the Court of Appeal to conclude otherwise.

[23]                         First, they submit the Court of Appeal erred in holding that privacy is a personal concern that cannot, without more, constitute an important interest under Sierra Club. The Trustees say the application judge was right to characterize privacy and dignity as an important public interest which, as it was subject to a serious risk, justified the orders. They ask this Court to recognize that privacy in itself is an important public interest for the purposes of the analysis.

[24]                         Second, the Trustees submit that the Court of Appeal erred in overturning the application judge’s conclusion that there was a serious risk of physical harm. They argue that the Court of Appeal failed to recognize that courts have the ability to draw reasonable inferences by applying reason and logic even in the absence of specific evidence of the alleged risk.

[25]                         The Trustees say that these errors led the Court of Appeal to mistakenly set aside the sealing orders. In answer to questions at the hearing, the Trustees acknowledged that an order redacting certain documents in the file or a publication ban could assist in addressing some of their concerns, but maintained neither is a reasonable alternative to the sealing orders in the circumstances.

[26]                         The Trustees submit further that the protection of these interests outweighs the deleterious effects of the orders. They argue that the importance of the open court principle is attenuated by the nature of these probate proceedings. Given that it is non‑contentious and not strictly speaking necessary for the transfer of property at death, probate is a court proceeding of an “administrative” character, which diminishes the imperative of applying the open court principle here (paras. 113‑14).

[27]                         The Toronto Star takes the position that the Court of Appeal made no mistake in setting aside the sealing orders and that the appeal should be dismissed. In the Toronto Star’s view, while privacy can be an important interest where it evinces a public component, the Trustees have only identified a subjective desire for the affected individuals in this case to avoid further publicity, which is not inherently harmful. According to the Toronto Star and some of the interveners, the Trustees’ position would allow that measure of inconvenience and embarrassment that arises in every court proceeding to take precedence over the interest in court openness protected by the Canadian Charter of Rights and Freedoms  in which all of society has a stake. The Toronto Star argues further that the information in the court files is not highly sensitive. On the issue of whether the sealing orders were necessary to protect the affected individuals from physical harm, the Toronto Star submits that the Court of Appeal was right to conclude that the Trustees had failed to establish a serious risk to this interest.

[28]                         In the alternative, even if there were a serious risk to one or another important interest, the Toronto Star says the sealing orders are not necessary because the risk could be addressed by an alternative, less onerous order. Furthermore, it says the orders are not proportionate. In seeking to minimize the importance of openness in probate proceedings, the Trustees invite an inflexible approach to balancing the effects of the order that is incompatible with the principle that openness applies to all court proceedings. In any event, there is a public interest in openness specifically here, given that the certificates sought can affect the rights of third parties and that openness ensures the fairness of the proceedings, whether they are contested or not.

V.           Analysis

[29]                          The outcome of the appeal turns on whether the application judge should have made the sealing orders pursuant to the test for discretionary limits on court openness from this Court’s decision in Sierra Club.

[30]                          Court openness is protected by the constitutional guarantee of freedom of expression and is essential to the proper functioning of our democracy (Canadian Broadcasting Corp. v. New Brunswick (Attorney General), [1996] 3 S.C.R. 480, at para. 23; Vancouver Sun (Re), 2004 SCC 43, [2004] 2 S.C.R. 332, at paras. 23‑26). Reporting on court proceedings by a free press is often said to be inseparable from the principle of open justice. “In reporting what has been said and done at a public trial, the media serve as the eyes and ears of a wider public which would be absolutely entitled to attend but for purely practical reasons cannot do so” (Khuja v. Times Newspapers Ltd., [2017] UKSC 49, [2019] A.C. 161, at para. 16, citing Edmonton Journal v. Alberta (Attorney General), [1989] 2 S.C.R. 1326, at pp. 1339‑40, per Cory J.). Limits on openness in service of other public interests have been recognized, but sparingly and always with an eye to preserving a strong presumption that justice should proceed in public view (Dagenais v. Canadian Broadcasting Corp., [1994] 3 S.C.R. 835, at p. 878; R. v. Mentuck, 2001 SCC 76, [2001] 3 S.C.R. 442, at paras. 32‑39; Sierra Club, at para. 56). The test for discretionary limits on court openness is directed at maintaining this presumption while offering sufficient flexibility for courts to protect these other public interests where they arise (Mentuck, at para. 33). The parties agree that this is the appropriate framework of analysis for resolving this appeal.

[31]                          The parties and the courts below disagree, however, about how this test applies to the facts of this case and this calls for clarification of certain points of the Sierra Club analysis. Most centrally, there is disagreement about how an important interest in the protection of privacy could be recognized such that it would justify limits on openness, and in particular when privacy can be a matter of public concern. The parties bring two settled principles of this Court’s jurisprudence to bear in support of their respective positions. First, this Court has often observed that privacy is a fundamental value necessary to the preservation of a free and democratic society (Lavigne v. Canada (Office of the Commissioner of Official Languages), 2002 SCC 53, [2002] 2 S.C.R. 773, at para. 25; Dagg v. Canada (Minister of Finance), [1997] 2 S.C.R. 403, at paras. 65‑66, per La Forest J. (dissenting but not on this point); New Brunswick, at para. 40). Courts have invoked privacy, in some instances, as the basis for an exception to openness under the Sierra Club test (see, e.g., R. v. Henry, 2009 BCCA 86, 270 B.C.A.C. 5, at paras. 11 and 17). At the same time, the jurisprudence acknowledges that some degree of privacy loss — resulting in inconvenience, even in upset or embarrassment — is inherent in any court proceeding open to the public (New Brunswick, at para. 40). Accordingly, upholding the presumption of openness has meant recognizing that neither individual sensibilities nor mere personal discomfort associated with participating in judicial proceedings are likely to justify the exclusion of the public from court (Attorney General of Nova Scotia v. MacIntyre, [1982] 1 S.C.R. 175, at p. 185; New Brunswick, at para. 41). Determining the role of privacy in the Sierra Club analysis requires reconciling these two ideas, which is the nub of the disagreement between the parties. The right of privacy is not absolute; the open court principle is not without exceptions.

[32]                          For the reasons that follow, I disagree with the Trustees that the ostensibly unbounded privacy interest they invoke qualifies as an important public interest within the meaning of Sierra Club. Their broad claim fails to focus on the elements of privacy that are deserving of public protection in the open court context. That is not to say, however, that privacy can never ground an exceptional measure such as the sealing orders sought in this case. While the mere embarrassment caused by the dissemination of personal information through the open court process does not rise to the level justifying a limit on court openness, circumstances do exist where an aspect of a person’s private life has a plain public interest dimension.

[33]                          Personal information disseminated in open court can be more than a source of discomfort and may result in an affront to a person’s dignity. Insofar as privacy serves to protect individuals from this affront, it is an important public interest relevant under Sierra Club. Dignity in this sense is a related but narrower concern than privacy generally; it transcends the interests of the individual and, like other important public interests, is a matter that concerns the society at large. A court can make an exception to the open court principle, notwithstanding the strong presumption in its favour, if the interest in protecting core aspects of individuals’ personal lives that bear on their dignity is at serious risk by reason of the dissemination of sufficiently sensitive information. The question is not whether the information is “personal” to the individual concerned, but whether, because of its highly sensitive character, its dissemination would occasion an affront to their dignity that society as a whole has a stake in protecting.

[34]                          This public interest in privacy appropriately focuses the analysis on the impact of the dissemination of sensitive personal information, rather than the mere fact of this dissemination, which is frequently risked in court proceedings and is necessary in a system that privileges court openness. It is a high bar — higher and more precise than the sweeping privacy interest relied upon here by the Trustees. This public interest will only be seriously at risk where the information in question strikes at what is sometimes said to be the core identity of the individual concerned: information so sensitive that its dissemination could be an affront to dignity that the public would not tolerate, even in service of open proceedings.

[35]                          I hasten to say that applicants for an order making exception to the open court principle cannot content themselves with an unsubstantiated claim that this public interest in dignity is compromised any more than they could by an unsubstantiated claim that their physical integrity is endangered. Under Sierra Club, the applicant must show on the facts of the case that, as an important interest, this dignity dimension of their privacy is at “serious risk”. For the purposes of the test for discretionary limits on court openness, this requires the applicant to show that the information in the court file is sufficiently sensitive such that it can be said to strike at the biographical core of the individual and, in the broader circumstances, that there is a serious risk that, without an exceptional order, the affected individual will suffer an affront to their dignity.

[36]                          In the present case, the information in the court files was not of this highly sensitive character that it could be said to strike at the core identity of the affected persons; the Trustees have failed to show how the lifting of the sealing orders engages the dignity of the affected individuals. I am therefore not convinced that the intrusion on their privacy raises a serious risk to an important public interest as required by Sierra Club. Moreover, as I shall endeavour to explain, there was no serious risk of physical harm to the affected individuals by lifting the sealing orders. Accordingly, this is not an appropriate case in which to make sealing orders, or any order limiting access to these court files. In the circumstances, the admissibility of the Toronto Star’s new evidence is moot. I propose to dismiss the appeal.

A.           The Test for Discretionary Limits on Court Openness

[37]                         Court proceedings are presumptively open to the public (MacIntyre, at p. 189; A.B. v. Bragg Communications Inc., 2012 SCC 46, [2012] 2 S.C.R. 567, at para. 11).

[38]                         The test for discretionary limits on presumptive court openness has been expressed as a two‑step inquiry involving the necessity and proportionality of the proposed order (Sierra Club, at para. 53). Upon examination, however, this test rests upon three core prerequisites that a person seeking such a limit must show. Recasting the test around these three prerequisites, without altering its essence, helps to clarify the burden on an applicant seeking an exception to the open court principle. In order to succeed, the person asking a court to exercise discretion in a way that limits the open court presumption must establish that:

(1)   court openness poses a serious risk to an important public interest;

(2)   the order sought is necessary to prevent this serious risk to the identified interest because reasonably alternative measures will not prevent this risk; and,

(3)   as a matter of proportionality, the benefits of the order outweigh its negative effects.

Only where all three of these prerequisites have been met can a discretionary limit on openness — for example, a sealing order, a publication ban, an order excluding the public from a hearing, or a redaction order — properly be ordered. This test applies to all discretionary limits on court openness, subject only to valid legislative enactments (Toronto Star Newspapers Ltd. v. Ontario, 2005 SCC 41, [2005] 2 S.C.R. 188, at paras. 7 and 22).

[39]                         The discretion is structured and controlled in this way to protect the open court principle, which is understood to be constitutionalized under the right to freedom of expression at s. 2 (b) of the Charter (New Brunswick, at para. 23). Sustained by freedom of expression, the open court principle is one of the foundations of a free press given that access to courts is fundamental to newsgathering. This Court has often highlighted the importance of open judicial proceedings to maintaining the independence and impartiality of the courts, public confidence and understanding of their work and ultimately the legitimacy of the process (see, e.g., Vancouver Sun, at paras. 23‑26). In New Brunswick, La Forest J. explained the presumption in favour of court openness had become “‘one of the hallmarks of a democratic society’” (citing Re Southam Inc. and The Queen (No.1) (1983), 41 O.R. (2d) 113 (C.A.), at p. 119), that “acts as a guarantee that justice is administered in a non‑arbitrary manner, according to the rule of law . . . thereby fostering public confidence in the integrity of the court system and understanding of the administration of justice” (para. 22). The centrality of this principle to the court system underlies the strong presumption — albeit one that is rebuttable — in favour of court openness (para. 40; Mentuck, at para. 39).

[40]                         The test ensures that discretionary orders are subject to no lower standard than a legislative enactment limiting court openness would be (Mentuck, at para. 27; Sierra Club, at para. 45). To that end, this Court developed a scheme of analysis by analogy to the Oakes test, which courts use to understand whether a legislative limit on a right guaranteed under the Charter is reasonable and demonstrably justified in a free and democratic society (Sierra Club, at para. 40, citing R. v. Oakes, [1986] 1 S.C.R. 103; see also Dagenais, at p. 878; Vancouver Sun, at para. 30).

[41]                         The recognized scope of what interests might justify a discretionary exception to open courts has broadened over time. In Dagenais, Lamer C.J. spoke of a requisite risk to the “fairness of the trial” (p. 878). In Mentuck, Iacobucci J. extended this to a risk affecting the “proper administration of justice” (para. 32). Finally, in Sierra Club, Iacobucci J., again writing for a unanimous Court, restated the test to capture any serious risk to an “important interest, including a commercial interest, in the context of litigation” (para. 53). He simultaneously clarified that the important interest must be expressed as a public interest. For example, on the facts of that case, a harm to a particular business interest would not have been sufficient, but the “general commercial interest of preserving confidential information” was an important interest because of its public character (para. 55). This is consistent with the fact that this test was developed in reference to the Oakes jurisprudence that focuses on the “pressing and substantial” objective of legislation of general application (Oakes, at pp. 138‑39; see also Mentuck, at para. 31). The term “important interest” therefore captures a broad array of public objectives.

[42]                         While there is no closed list of important public interests for the purposes of this test, I share Iacobucci J.’s sense, explained in Sierra Club, that courts must be “cautious” and “alive to the fundamental importance of the open court rule” even at the earliest stage when they are identifying important public interests (para. 56). Determining what is an important public interest can be done in the abstract at the level of general principles that extend beyond the parties to the particular dispute (para. 55). By contrast, whether that interest is at “serious risk” is a fact‑based finding that, for the judge considering the appropriateness of an order, is necessarily made in context. In this sense, the identification of, on the one hand, an important interest and, on the other, the seriousness of the risk to that interest are, theoretically at least, separate and qualitatively distinct operations. An order may therefore be refused simply because a valid important public interest is not at serious risk on the facts of a given case or, conversely, that the identified interests, regardless of whether they are at serious risk, do not have the requisite important public character as a matter of general principle.

[43]                         The test laid out in Sierra Club continues to be an appropriate guide for judicial discretion in cases like this one. The breadth of the category of “important interest” transcends the interests of the parties to the dispute and provides significant flexibility to address harm to fundamental values in our society that unqualified openness could cause (see, e.g., P. M. Perell and J. W. Morden, The Law of Civil Procedure in Ontario (4th ed. 2020), at para. 3.185; J. Bailey and J. Burkell, “Revisiting the Open Court Principle in an Era of Online Publication: Questioning Presumptive Public Access to Parties’ and Witnesses’ Personal Information” (2016), 48 Ottawa L. Rev. 143, at pp. 154‑55). At the same time, however, the requirement that a serious risk to an important interest be demonstrated imposes a meaningful threshold necessary to maintain the presumption of openness. Were it merely a matter of weighing the benefits of the limit on court openness against its negative effects, decision-makers confronted with concrete impacts on the individuals appearing before them may struggle to put adequate weight on the less immediate negative effects on the open court principle. Such balancing could be evasive of effective appellate review. To my mind, the structure provided by Dagenais, Mentuck, and Sierra Club remains appropriate and should be affirmed.

[44]                         Finally, I recall that the open court principle is engaged by all judicial proceedings, whatever their nature (MacIntyre at pp. 185‑86; Vancouver Sun, at para. 31). To the extent the Trustees suggested, in their arguments about the negative effects of the sealing orders, that probate in Ontario does not engage the open court principle or that the openness of these proceedings has no public value, I disagree. The certificates the Trustees sought from the court are issued under the seal of that court, thereby bearing the imprimatur of the court’s authority. The court’s decision, even if rendered in a non‑contentious setting, will have an impact on third parties, for example by establishing the testamentary paper that constitutes a valid will (see Otis v. Otis (2004), 7 E.T.R. (3d) 221 (Ont. S.C.), at paras. 23‑24). Contrary to what the Trustees argue, the matters in a probate file are not quintessentially private or fundamentally administrative. Obtaining a certificate of appointment of estate trustee in Ontario is a court proceeding and the fundamental rationale for openness — discouraging mischief and ensuring confidence in the administration of justice through transparency — applies to probate proceedings and thus to the transfer of property under court authority and other matters affected by that court action.

[45]                         It is true that other non‑probate estate planning mechanisms may allow for the transfer of wealth outside the ordinary avenues of testate or intestate succession — that is the case, for instance, for certain insurance and pension benefits, and for certain property held in co‑ownership. But this does not change the necessarily open court character of probate proceedings. That non-probate transfers keep certain information related to the administration of an estate out of public view does not mean that the Trustees here, by seeking certificates from the court, somehow do not engage this principle. The Trustees seek the benefits that flow from the public judicial probate process: transparency ensures that the probate court’s authority is administered fairly and efficiently (Vancouver Sun, at para. 25; New Brunswick, at para. 22). The strong presumption in favour of openness plainly applies to probate proceedings and the Trustees must satisfy the test for discretionary limits on court openness.

B.            The Public Importance of Privacy

[46]                         As mentioned, I disagree with the Trustees that an unbounded interest in privacy qualifies as an important public interest under the test for discretionary limits on court openness. Yet in some of its manifestations, privacy does have social importance beyond the person most immediately concerned. On that basis, it cannot be excluded as an interest that could justify, in the right circumstances, a limit to court openness. Indeed, the public importance of privacy has been recognized by this Court in various settings, and this sheds light on why the narrower aspect of privacy related to the protection of dignity is an important public interest.

[47]                         I respectfully disagree with the manner in which the Court of Appeal disposed of the claim by the Trustees that there is a serious risk to the interest in protecting personal privacy in this case. For the appellate judges, the privacy concerns raised by the Trustees amounted to “[p]ersonal concerns” which cannot, “without more”, satisfy the requirement from Sierra Club that an important interest be framed as a public interest (para. 10). The Court of Appeal in our case relied, at para. 10, on H. (M.E.) v. Williams, 2012 ONCA 35, 108 O.R. (3d) 321, in which it was held that “[p]urely personal interests cannot justify non‑publication or sealing orders” (para. 25). Citing as authority judgments of this Court in MacIntyre and Sierra Club, the court continued by observing that “personal concerns of a litigant, including concerns about the very real emotional distress and embarrassment that can be occasioned to litigants when justice is done in public, will not, standing alone, satisfy the necessity branch of the test” (para. 25). Respectfully stated, the emphasis that the Court of Appeal placed on personal concerns as a means of deciding that the sealing orders failed to meet the necessity requirement in this case and in Williams is, I think, mistaken. Personal concerns that relate to aspects of the privacy of an individual who is before the courts can coincide with a public interest in confidentiality.

[48]                         Like the Court of Appeal, I do agree with the view expressed particularly in the pre‑Charter case of MacIntyre, that where court openness results in an intrusion on privacy which disturbs the “sensibilities of the individuals involved” (p. 185), that concern is generally insufficient to justify a sealing or like order and does not amount to an important public interest under Sierra Club. But I disagree with the Court of Appeal in this case and in Williams that this is because the intrusion only occasions “personal concerns”. Certain personal concerns — even “without more” — can coincide with important public interests within the meaning of Sierra Club. To invoke the expression of Binnie J. in F.N. (Re), 2000 SCC 35, [2000] 1 S.C.R. 880, at para. 10, there is a “public interest in confidentiality” that is felt, first and foremost, by the person involved and is most certainly a personal concern. Even in Williams, the Court of Appeal was careful to note that where, without privacy protection, an individual would face “a substantial risk of serious debilitating emotional . . . harm”, an exception to openness should be available (paras. 29‑30). The means of discerning whether a privacy interest reflects a “public interest in confidentiality” is therefore not whether the interest reflects or is rooted in “personal concerns” for the privacy of the individuals involved. Some personal concerns relating to privacy overlap with public interests in confidentiality. These interests in privacy can be, in my view, important public interests within the meaning of Sierra Club. It is true that an individual’s privacy is pre‑eminently important to that individual. But this Court has also long recognized that the protection of privacy is, in a variety of settings, in the interest of society as a whole.

[49]                         The proposition that privacy is important, not only to the affected individual but to our society, has deep roots in the jurisprudence of this Court outside the context of the test for discretionary limits on court openness. This background helps explain why privacy cannot be rejected as a mere personal concern. However, the key differences in these contexts are such that the public importance of privacy cannot be transposed to open courts without adaptation. Only specific aspects of privacy interests can qualify as important public interests under Sierra Club.

[50]                         In the context of s. 8  of the Charter and public sector privacy legislation, La Forest J. cited American privacy scholar Alan F. Westin for the proposition that privacy is a fundamental value of the modern state, first in R. v. Dyment, [1988] 2 S.C.R. 417, at pp. 427‑28 (concurring), and then in Dagg, at para. 65 (dissenting but not on this point). In the latter case, La Forest J. wrote: “The protection of privacy is a fundamental value in modern, democratic states. An expression of an individual’s unique personality or personhood, privacy is grounded on physical and moral autonomy — the freedom to engage in one’s own thoughts, actions and decisions” (para. 65 (citations omitted)). That statement was endorsed unanimously by this Court in Lavigne, at para. 25.

[51]                         Further, in Alberta (Information and Privacy Commissioner) v. United Food and Commercial Workers, Local 401, 2013 SCC 62, [2013] 3 S.C.R. 733 (“UFCW”), decided in the context of a statute regulating the use of information by organizations, the objective of providing an individual with some control over their information was recognized as “intimately connected to individual autonomy, dignity and privacy, self‑evidently significant social values” (para. 24). The importance of privacy, its “quasi‑constitutional status” and its role in protecting moral autonomy continues to find expression in our recent jurisprudence (see, e.g., Lavigne, at para. 24; Bragg, at para. 18, per Abella J., citing Toronto Star Newspaper Ltd. v. R., 2012 ONCJ 27, 289 C.C.C. (3d) 549, at paras. 40‑41 and 44; Douez v. Facebook, Inc., 2017 SCC 33, [2017] 1 S.C.R. 751, at para. 59). In Douez, Karakatsanis, Wagner (as he then was) and Gascon JJ. underscored this same point, adding that “the growth of the Internet, virtually timeless with pervasive reach, has exacerbated the potential harm that may flow from incursions to a person’s privacy interests” (para. 59).

[52]                         Privacy as a public interest is underlined by specific aspects of privacy protection present in legislation at the federal and provincial levels (see, e.g., Privacy Act , R.S.C. 1985, c. P‑21 ; Personal Information Protection and Electronic Documents Act , S.C. 2000, c. 5  (“PIPEDA ”); Freedom of Information and Protection of Privacy Act, R.S.O. 1990, c. F.31; Charter of Human Rights and Freedoms, CQLR, c. C‑12, s. 5; Civil Code of Québec, arts. 35 to 41).^[3] Further, in assessing the constitutionality of a legislative exception to the open court principle, this Court has recognized that the protection of individual privacy can be a pressing and substantial objective (Edmonton Journal, at p. 1345, per Cory J.; see also the concurring reasons of Wilson J., at p. 1354, in which “the public interest in protecting the privacy of litigants generally in matrimonial cases against the public interest in an open court process” was explicitly noted). There is also continued support for the social and public importance of individual privacy in the academic literature (see, e.g., A. J. Cockfield, “Protecting the Social Value of Privacy in the Context of State Investigations Using New Technologies” (2007), 40 U.B.C. L. Rev. 41, at p. 41; K. Hughes, “A Behavioural Understanding of Privacy and its Implications for Privacy Law” (2012), 75 Mod. L. Rev. 806, at p. 823; P. Gewirtz, “Privacy and Speech” (2001), Sup. Ct. Rev. 139, at p. 139). It is therefore inappropriate, in my respectful view, to dismiss the public interest in protecting privacy as merely a personal concern. This does not mean, however, that privacy generally is an important public interest in the context of limits on court openness.

[53]                         The fact that the case before the application judge concerned individuals who were advancing their own privacy interests, which were undeniably important to them as individuals, does not mean that there is no public interest at stake. In F.N. (Re), this was the personal interest that young offenders had in remaining anonymous in court proceedings as a means of encouraging their personal rehabilitation (para. 11). All of society had a stake, according to Binnie J., in the young person’s personal prospect for rehabilitation. This same idea from F.N. (Re) was cited in support of finding the interest in Sierra Club to be a public interest. That interest, rooted first in an agreement of personal concern to the contracting parties involved, was a private matter that evinced, alongside its personal interest to the parties, a “public interest in confidentiality” (Sierra Club, at para. 55). Similarly, while the Trustees have a personal interest in preserving their privacy, this does not mean that the public has no stake in this same interest because — as this Court has made clear — it is related to moral autonomy and dignity which are pressing and substantial concerns.

[54]                         In this appeal, the Toronto Star suggests that legitimate privacy concerns would be effectively protected by a discretionary order where there is “something more” to elevate them beyond personal concerns and sensibilities (R.F., at para. 73). The Income Security Advocacy Centre, by way of example, submits that privacy serves the public interests of preventing harm and of ensuring individuals are not dissuaded from accessing the courts. I agree that these concepts are related, but in my view care must be taken not to conflate the public importance of privacy with that of other interests; aspects of privacy, such as dignity, may constitute important public interests in and of themselves. A risk to personal privacy may be tied to a risk to psychological harm, as it was in Bragg (para. 14; see also J. Rossiter, Law of Publication Bans, Private Hearings, and Sealing Orders (loose-leaf), s. 2.4.1). But concerns for privacy may not always coincide with a desire to avoid psychological harm, and may focus instead, for example, on protecting one’s professional standing (see, e.g., R. v. Paterson (1998), 102 B.C.A.C. 200, at paras. 76, 78 and 87‑88). Similarly, there may be circumstances where the prospect of surrendering the personal information necessary to pursue a legal claim may deter an individual from bringing that claim (see S. v. Lamontagne, 2020 QCCA 663, at paras. 34‑35 (CanLII)). In the same way, the prospect of surrendering sensitive commercial information would have impaired the conduct of the party’s defence in Sierra Club (at para. 71), or could pressure an individual into settling a dispute prematurely (K. Eltis, Courts, Litigants and the Digital Age (2nd ed. 2016), at p. 86). But this does not necessarily mean that a public interest in privacy is wholly subsumed by such concerns. I note, for example, that access to justice concerns do not apply where the privacy interest to be protected is that of a third party to the litigation, such as a witness, whose access to the courts is not at stake and who has no choice available to terminate the litigation and avoid any privacy impacts (see, e.g., Himel v. Greenberg, 2010 ONSC 2325, 93 R.F.L. (6th) 357, at para. 58; see also Rossiter, s. 2.4.2(2)). In any event, the recognition of these related and valid important public interests does not answer the question as to whether aspects of privacy in and of themselves are important public interests and does not diminish the distinctive public character of privacy, considered above.

[55]                         Indeed, the specific harms to privacy occasioned by open courts have not gone unnoticed nor been discounted as merely personal concerns. Courts have exercised their discretion to limit court openness in order to protect personal information from publicity, including to prevent the disclosure of sexual orientation (see, e.g., Paterson, at paras. 76, 78 and 87‑88), HIV status (see, e.g., A.B. v. Canada (Citizenship and Immigration), 2017 FC 629, at para. 9 (CanLII)), and a history of substance abuse and criminality (see, e.g., R. v. Pickton, 2010 BCSC 1198, at paras. 11 and 20 (CanLII)). This need to reconcile the public interest in privacy with the open court principle has been highlighted by this Court (see, e.g., Edmonton Journal, at p. 1353, per Wilson J.). Writing extra‑judicially, McLachlin C.J. explained that “[i]f we are serious about peoples’ private lives, we must preserve a modicum of privacy. Equally, if we are serious about our justice system, we must have open courts. The question is how to reconcile these dual imperatives in a fair and principled way” (“Courts, Transparency and Public Confidence – To the Better Administration of Justice” (2003), 8 Deakin L. Rev. 1, at p. 4). In seeking that reconciliation, the question becomes whether the relevant dimension of privacy amounts to an important public interest that, when seriously at risk, would justify rebutting the strong presumption favouring open courts.

C.            The Important Public Interest in Privacy Bears on the Protection of Individual Dignity

[56]                         While the public importance of privacy has clearly been recognized by this Court in various settings, caution is required in deploying this concept in the test for discretionary limits on court openness. It is a matter of settled law that open court proceedings by their nature can be a source of discomfort and embarrassment and these intrusions on privacy are generally seen as of insufficient importance to overcome the presumption of openness. The Toronto Star has raised the concern that recognizing privacy as an important public interest will lower the burden for applicants because the privacy of litigants will, in some respects, always be at risk in court proceedings. I agree that the requirement to show a serious risk to an important interest is a key threshold component of the analysis that must be preserved in order to protect the open court principle. The recognition of a public interest in privacy could threaten the strong presumption of openness if privacy is cast too broadly without a view to its public character.

[57]                          Privacy poses challenges in the test for discretionary limits on court openness because of the necessary dissemination of information that openness implies. It bears recalling that when Dickson J., as he then was, wrote in MacIntyre that “covertness is the exception and openness the rule”, he was explicitly treating a privacy argument, returning to and dismissing the view, urged many times before, “that the ‘privacy’ of litigants requires that the public be excluded from court proceedings” (p. 185 (emphasis added)). Dickson J. rejected the view that personal privacy concerns require closed courtroom doors, explaining that “[a]s a general rule the sensibilities of the individuals involved are no basis for exclusion of the public from judicial proceedings” (ibid.).

[58]                          Though writing before Dagenais, and therefore not commenting on the specific steps of the analysis as we now understand them, to my mind, Dickson J. was right to recognize that the open court principle brings necessary limits to the right to privacy. While individuals may have an expectation that information about them will not be revealed in judicial proceedings, the open court principle stands presumptively in opposition to that expectation. For example, in Lac d’Amiante du Québec Ltée v. 2858-0702 Québec Inc., 2001 SCC 51, [2001] 2 S.C.R. 743, LeBel J. held that “a party who institutes a legal proceeding waives his or her right to privacy, at least in part” (para. 42). MacIntyre and cases like it recognize — in stating that openness is the rule and covertness the exception — that the right to privacy, however defined, in some measure gives way to the open court ideal. I share the view that the open court principle presumes that this limit on the right to privacy is justified.

[59]                         The Toronto Star is therefore right to say that the privacy of individuals will very often be at some risk in court proceedings. Disputes between and concerning individuals that play out in open court necessarily reveal information that may have otherwise remained out of public view. Indeed, much like the Court of Appeal in this case, courts have explicitly adverted to this concern when concluding that mere inconvenience is insufficient to cross the initial threshold of the test (see, e.g., 3834310 Canada inc. v. Chamberland, 2004 CanLII 4122 (Que. C.A.), at para. 30). Saying that any impact on individual privacy is sufficient to establish a serious risk to an important public interest for the purposes of the test for discretionary limits on court openness could render this initial requirement moot. Many cases would turn on the balancing at the proportionality stage. Such a development would amount to a departure from Sierra Club, which is the appropriate framework and one which must be preserved.

[60]                         Further, recognizing an important interest in privacy generally could prove to be too open‑ended and difficult to apply. Privacy is a complex and contextual concept (Dagg, at para. 67; see also B. McIsaac, K. Klein and S. Brown, The Law of Privacy in Canada (loose‑leaf), vol. 1, at pp. 1‑4; D. J. Solove, “Conceptualizing Privacy” (2002), 90 Cal. L. Rev. 1087, at p. 1090). Indeed, this Court has described the nature of limits of privacy as being in a state of “theoretical disarray” (R. v. Spencer, 2014 SCC 43, [2014] 2 S.C.R. 212, at para. 35). Much turns on the context in which privacy is invoked. I agree with the Toronto Star that a bald recognition of privacy as an important interest in the context of the test for discretionary limits on court openness, as the Trustees advance here, would invite considerable confusion. It would be difficult for courts to measure a serious risk to such an interest because of its multi-faceted nature.

[61]                         While I acknowledge these concerns have merit, I disagree that they require that privacy never be considered in determining whether there is a serious risk to an important public interest. I reach this conclusion for two reasons. First, the problem of privacy’s complexity can be attenuated by focusing on the purpose underlying the public protection of privacy as it is relevant to the judicial process, in order to fix precisely on that aspect which transcends the interests of the parties in this context. That narrower dimension of privacy is the protection of dignity, an important public interest that can be threatened by open courts. Indeed, rather than attempting to apply a single unwieldy concept of privacy in all contexts, this Court has generally fixed on more specific privacy interests tailored to the particular situation (Spencer, at para. 35; Edmonton Journal, at p. 1362, per Wilson J.). That is what must be done here, with a view to identifying the public aspect of privacy that openness might inappropriately undermine.

[62]                         Second, I recall that in order to pass the first stage of the analysis one must not simply invoke an important interest, but must also overcome the presumption of openness by showing a serious risk to this interest. The burden of showing a risk to such an interest on the facts of a given case constitutes the true initial threshold on the person seeking to restrict openness. It is never sufficient to plead a recognized important public interest on its own. The demonstration of a serious risk to this interest is still required. What is important is that the interest be accurately defined to capture only those aspects of privacy that engage legitimate public objectives such that showing a serious risk to that interest remains a high bar. In this way, courts can effectively maintain the guarantee of presumptive openness.

[63]                          Specifically, in order to preserve the integrity of the open court principle, an important public interest concerned with the protection of dignity should be understood to be seriously at risk only in limited cases. Nothing here displaces the principle that covertness in court proceedings must be exceptional. Neither the sensibilities of individuals nor the fact that openness is disadvantageous, embarrassing or distressing to certain individuals will generally on their own warrant interference with court openness (MacIntyre, at p. 185; New Brunswick, at para. 40; Williams, at para. 30; Coltsfoot Publishing Ltd. v. Foster-Jacques, 2012 NSCA 83, 320 N.S.R. (2d) 166, at para. 97). These principles do not preclude recognizing the public character of a privacy interest as important when it is related to the protection of dignity. They merely require that a serious risk be shown to exist in respect of this interest in order to justify, exceptionally, a limit on openness, as is the case with any important public interest under Sierra Club. As Professors Sylvette Guillemard and Séverine Menétrey explain, [translation] “[t]he confidentiality of the proceedings may be justified, in particular, in order to protect the parties’ privacy . . . . However, the jurisprudence indicates that embarrassment or shame is not a sufficient reason to order that proceedings be held in camera or to impose a publication ban” (Comprendre la procédure civile québécoise (2nd ed. 2017), at p. 57).

[64]                         How should the privacy interest at issue be understood as raising an important public interest relevant to the test for discretionary limits on court openness in this context? It is helpful to recall that the orders below were sought to limit access to documents and information in the court files. The Trustees’ argument on this point focused squarely on the risk of immediate and widespread dissemination of the personally identifying and other sensitive information contained in the sealed materials by the Toronto Star. The Trustees submit that this dissemination would constitute an unwarranted intrusion into the privacy of the affected individuals beyond the upset they have already suffered as a result of the publicity associated with the death of the Shermans.

[65]                         In my view, there is value in leaving individuals free to restrict when, how and to what extent highly sensitive information about them is communicated to others in the public sphere, because choosing how we present ourselves in public preserves our moral autonomy and dignity as individuals. This Court has had occasion to underscore the connection between the privacy interest engaged by open courts and the protection of dignity specifically. For example, in Edmonton Journal, Wilson J. noted that the impugned provision which would limit publication about matrimonial proceedings addressed “a somewhat different aspect of privacy, one more closely related to the protection of one’s dignity . . . namely the personal anguish and loss of dignity that may result from having embarrassing details of one’s private life printed in the newspapers” (pp. 1363‑64). In Bragg, as a further example, the protection of a young person’s ability to control sensitive information was said to foster respect for “dignity, personal integrity and autonomy” (para. 18, citing Toronto Star Newspaper Ltd., at para. 44).

[66]                          Consistent with this jurisprudence, I note by way of example that the Quebec legislature expressly highlighted the preservation of dignity when the Sierra Club test was codified in the Code of Civil Procedure, CQLR, c. C‑25.01 (“C.C.P.”), art. 12 (see also Ministère de la Justice, Commentaires de la ministre de la Justice: Code de procédure civile, chapitre C‑25.01 (2015), art. 12). Under art. 12 C.C.P., a discretionary exception to the open court principle can be made by the court if “public order, in particular the preservation of the dignity of the persons involved or the protection of substantial and legitimate interests”, requires it.

[67]                          The concept of public order evidences flexibility analogous to the concept of an important public interest under Sierra Club yet it recalls that the interest invoked transcends, in importance and consequence, the purely subjective sensibilities of the persons affected. Like the “important public interest” that must be at serious risk to justify the sealing orders in the present appeal, public order encompasses a wide array of general principles and imperative norms identified by a legislature and the courts as fundamental to a given society (see Goulet v. Transamerica Life Insurance Co. of Canada, 2002 SCC 21, [2002] 1 S.C.R. 719, at paras. 42‑44, citing Godbout v. Longueuil (Ville de), [1995] R.J.Q. 2561 (C.A.), at p. 2570, aff’d [1997] 3 S.C.R. 844). As one Quebec judge wrote, referring to Sierra Club prior to the enactment of art. 12 C.C.P., the interest must be understood as defined [translation] “in terms of a public interest in confidentiality” (see 3834310 Canada inc., at para. 24, per Gendreau J.A. for the Court of appeal). From among the various considerations that make up the concept of public order and other legitimate interests to which art. 12 C.C.P. alludes, it is significant that dignity, and not an untailored reference to either privacy, harm or access to justice, was given pride of place. Indeed, it is that narrow aspect of privacy considered to be a fundamental right that courts had fixed upon before the enactment of art. 12 C.C.P. — [translation] “what is part of one’s personal life, in short, what constitutes a minimum personal sphere” (Godbout, at p. 2569, per Baudouin J.A.; see also A. v. B., 1990 CanLII 3132 (Que. C.A.), at para. 20, per Rothman J.A.).

[68]                          The “preservation of the dignity of the persons involved” is now consecrated as the archetypal public order interest in art. 12 C.C.P. It is the exemplar of the Sierra Club important public interest in confidentiality that stands as justification for an exception to openness (S. Rochette and J.‑F. Côté, “Article 12”, in L. Chamberland, ed., Le grand collectif: Code de procédure civile — Commentaires et annotations (5th ed. 2020), vol. 1, at p. 102; D. Ferland and B. Emery, Précis de procédure civile du Québec (6th ed. 2020), vol. 1, at para. 1‑111). Dignity gives concrete expression to this public order interest because all of society has a stake in its preservation, notwithstanding its personal connections to the individuals concerned. This codification of Sierra Club’s notion of important public interest highlights the superordinate importance of human dignity and the appropriateness of limiting court openness on this basis as against an overbroad understanding of privacy that might be otherwise unsuitable to the open court context.

[69]                         Consistent with this idea, understanding privacy as predicated on dignity has been advanced as useful in connection with challenges brought by digital communications (K. Eltis, “The Judicial System in the Digital Age: Revisiting the Relationship between Privacy and Accessibility in the Cyber Context” (2011), 56 McGill L.J. 289, at p. 314).

[70]                         It is also significant, in my view, that the application judge in this case explicitly recognized, in response to the relevant arguments from the Trustees, an interest in “protecting the privacy and dignity of victims of crime and their loved ones” (para. 23 (emphasis added)). This elucidates that the central concern for the affected individuals on this point is not merely protecting their privacy for its own sake but privacy where it coincides with the public character of the dignity interests of these individuals.

[71]                         Violations of privacy that cause a loss of control over fundamental personal information about oneself are damaging to dignity because they erode one’s ability to present aspects of oneself to others in a selective manner (D. Matheson, “Dignity and Selective Self-Presentation”, in I. Kerr, V. Steeves and C. Lucock, eds., Lessons from the Identity Trail: Anonymity, Privacy and Identity in a Networked Society (2009), 319, at pp. 327‑28; L. M. Austin, “Re-reading Westin” (2019), 20 Theor. Inq. L. 53, at pp. 66‑68; Eltis (2016), at p. 13). Dignity, used in this context, is a social concept that involves presenting core aspects of oneself to others in a considered and controlled manner (see generally Matheson, at pp. 327‑28; Austin, at pp. 66‑68). Dignity is eroded where individuals lose control over this core identity‑giving information about themselves, because a highly sensitive aspect of who they are that they did not consciously decide to share is now available to others and may shape how they are seen in public. This was even alluded to by La Forest J., dissenting but not on this point, in Dagg, where he referred to privacy as “[a]n expression of an individual’s unique personality or personhood” (para. 65). 

[72]                          Where dignity is impaired, the impact on the individual is not theoretical but could engender real human consequences, including psychological distress (see generally Bragg, at para. 23). La Forest J., concurring, observed in Dyment that privacy is essential to the well‑being of individuals (p. 427). Viewed in this way, a privacy interest, where it shields the core information associated with dignity necessary to individual well-being, begins to look much like the physical safety interest also raised in this case, the important and public nature of which is neither debated, nor, in my view, seriously debatable. The administration of justice suffers when the operation of courts threatens physical well‑being because a responsible court system is attuned to the physical harm it inflicts on individuals and works to avoid such effects. Similarly, in my view, a responsible court must be attuned and responsive to the harm it causes to other core elements of individual well‑being, including individual dignity. This parallel helps to understand dignity as a more limited dimension of privacy relevant as an important public interest in the open court context.

[73]                         I am accordingly of the view that protecting individuals from the threat to their dignity that arises when information revealing core aspects of their private lives is disseminated through open court proceedings is an important public interest for the purposes of the test.

[74]                         Focusing on the underlying value of privacy in protecting individual dignity from the exposure of private information in open court overcomes the criticisms that privacy will always be at risk in open court proceedings and is theoretically complex. Openness brings intrusions on personal privacy in virtually all cases, but dignity as a public interest in protecting an individual’s core sensibility is more rarely in play. Specifically, and consistent with the cautious approach to the recognition of important public interests, this privacy interest, while determined in reference to the broader factual setting, will be at serious risk only where the sensitivity of the information strikes at the subject’s more intimate self.

[75]                         If the interest is ultimately about safeguarding a person’s dignity, that interest will be undermined when the information reveals something sensitive about them as an individual, as opposed to generic information that reveals little if anything about who they are as a person. Therefore the information that will be revealed by court openness must consist of intimate or personal details about an individual — what this Court has described in its jurisprudence on s. 8  of the Charter as the “biographical core” — if a serious risk to an important public interest is to be recognized in this context (R. v. Plant, [1993] 3 S.C.R. 281, at p. 293; R. v. Tessling, 2004 SCC 67, [2004] 3 S.C.R. 432, at para. 60; R. v. Cole, 2012 SCC 53, [2012] 3 S.C.R. 34, at para. 46). Dignity transcends personal inconvenience by reason of the highly sensitive nature of the information that might be revealed. This Court in Cole drew a similar line between the sensitivity of personal information and the public interest in protecting that information in reference to the biographical core. It held that “reasonable and informed Canadians” would be more willing to recognize the existence of a privacy interest where the relevant information cuts to the “biographical core” or, “[p]ut another way, the more personal and confidential the information” (para. 46). The presumption of openness means that mere discomfort associated with lesser intrusions of privacy will generally be tolerated. But there is a public interest in ensuring that openness does not unduly entail the dissemination of this core information that threatens dignity — even if it is “personal” to the affected person.

[76]                         The test for discretionary limits on court openness imposes on the applicant the burden to show that the important public interest is at serious risk. Recognizing that privacy, understood in reference to dignity, is only at serious risk where the information in the court file is sufficiently sensitive erects a threshold consistent with the presumption of openness. This threshold is fact specific. It addresses the concern, noted above, that personal information can frequently be found in court files and yet finding this sufficient to pass the serious risk threshold in every case would undermine the structure of the test. By requiring the applicant to demonstrate the sensitivity of the information as a necessary condition to the finding of a serious risk to this interest, the scope of the interest is limited to only those cases where the rationale for not revealing core aspects of a person’s private life, namely protecting individual dignity, is most actively engaged.

[77]                          There is no need here to provide an exhaustive catalogue of the range of sensitive personal information that, if exposed, could give rise to a serious risk. It is enough to say that courts have demonstrated a willingness to recognize the sensitivity of information related to stigmatized medical conditions (see, e.g., A.B., at para. 9), stigmatized work (see, e.g., Work Safe Twerk Safe v. Her Majesty the Queen in Right of Ontario, 2021 ONSC 1100, at para. 28 (CanLII)), sexual orientation (see, e.g., Paterson, at paras. 76, 78 and 87‑88), and subjection to sexual assault or harassment (see, e.g., Fedeli v. Brown, 2020 ONSC 994, at para. 9 (CanLII)). I would also note the submission of the intervener the Income Security Advocacy Centre, that detailed information about family structure and work history could in some circumstances constitute sensitive information. The question in every case is whether the information reveals something intimate and personal about the individual, their lifestyle or their experiences.

[78]                         I pause here to note that I refer to cases on s. 8  of the Charter above for the limited purpose of providing insight into types of information that are more or less personal and therefore deserving of public protection. If the impact on dignity as a result of disclosure is to be accurately measured, it is critical that the analysis differentiate between information in this way. Helpfully, one factor in determining whether an applicant’s subjective expectation of privacy is objectively reasonable in the s. 8 jurisprudence focuses on the degree to which information is private (see, e.g., R. v. Marakah, 2017 SCC 59, [2017] 2 S.C.R. 608, at para. 31; Cole, at paras. 44‑46). But while these decisions may assist for this limited purpose, this is not to say that the remainder of the s. 8 analysis has any relevance to the application of the test for discretionary limits on court openness. For example, asking what the Trustees’ reasonable expectation of privacy was here could invite a circular analysis of whether they reasonably expected their court files to be open to the public or whether they reasonably expected to be successful in having them sealed. Therefore, it is only for the limited purpose described above that the s. 8 jurisprudence is useful.

[79]                         In cases where the information is sufficiently sensitive to strike at an individual’s biographical core, a court must then ask whether a serious risk to the interest is made out in the full factual context of the case. While this is obviously a fact‑specific determination, some general observations may be made here to guide this assessment.

[80]                         I note that the seriousness of the risk may be affected by the extent to which information would be disseminated without an exception to the open court principle. If the applicant raises a risk that the personal information will come to be known by a large segment of the public in the absence of an order, this is a plainly more serious risk than if the result will be that a handful of people become aware of the same information, all else being equal. In the past, the requirement that one be physically present to acquire information in open court or from a court record meant that information was, to some extent, protected because it was “practically obscure” (D. S. Ardia, “Privacy and Court Records: Online Access and the Loss of Practical Obscurity” (2017), 4 U. Ill. L. Rev. 1385, at p. 1396). However, today, courts should be sensitive to the information technology context, which has increased the ease with which information can be communicated and cross‑referenced (see Bailey and Burkell, at pp. 169‑70; Ardia, at pp. 1450‑51). In this context, it may well be difficult for courts to be sure that information will not be broadly disseminated in the absence of an order.

[81]                         It will be appropriate, of course, to consider the extent to which information is already in the public domain. If court openness will simply make available what is already broadly and easily accessible, it will be difficult to show that revealing the information in open court will actually result in a meaningful loss of that aspect of privacy relating to the dignity interest to which I refer here. However, just because information is already accessible to some segment of the public does not mean that making it available through the court process will not exacerbate the risk to privacy. Privacy is not a binary concept, that is, information is not simply either private or public, especially because, by reason of technology in particular, absolute confidentiality is best thought of as elusive (see generally R. v. Quesnelle, 2014 SCC 46, [2014] 2 S.C.R. 390, at para. 37; UFCW, at para. 27). The fact that certain information is already available somewhere in the public sphere does not preclude further harm to the privacy interest by additional dissemination, particularly if the feared dissemination of highly sensitive information is broader or more easily accessible (see generally Solove, at p. 1152; Ardia, at p. 1393‑94; E. Paton‑Simpson, “Privacy and the Reasonable Paranoid: The Protection of Privacy in Public Places” (2000), 50 U.T.L.J. 305, at p. 346).

[82]                         Further, the seriousness of the risk is also affected by the probability that the dissemination the applicant suggests will occur actually occurs. I hasten to say that implicit in the notion of risk is that the applicant need not establish that the feared dissemination will certainly occur. However, the risk to the privacy interest related to the protection of dignity will be more serious the more likely it is that the information will be disseminated. While decided in a different context, this Court has held that the magnitude of risk is a product of both the gravity of the feared harm and its probability (R. v. Mabior, 2012 SCC 47, [2012] 2 S.C.R. 584, at para. 86).

[83]                         That said, the likelihood that an individual’s highly sensitive personal information will be disseminated in the absence of privacy protection will be difficult to quantify precisely. It is best to note as well that probability in this context need not be identified in mathematical or numerical terms. Rather, courts may merely discern probability in light of the totality of the circumstances and balance this one factor alongside other relevant factors.

[84]                         Finally, and as discussed above, individual sensitivities alone, even if they can be notionally associated with “privacy”, are generally insufficient to justify a restriction on court openness where they do not rise above those inconveniences and discomforts that are inherent to court openness (MacIntyre, at p. 185). An applicant will only be able to establish that the risk is sufficient to justify a limit on openness in exceptional cases, where the threatened loss of control over information about oneself is so fundamental that it strikes meaningfully at individual dignity. These circumstances engage “social values of superordinate importance” beyond the more ordinary intrusions inherent to participating in the judicial process that Dickson J. acknowledged could justify curtailing public openness (pp. 186‑87).

[85]                         To summarize, the important public interest in privacy, as understood in the context of the limits on court openness, is aimed at allowing individuals to preserve control over their core identity in the public sphere to the extent necessary to preserve their dignity. The public has a stake in openness, to be sure, but it also has an interest in the preservation of dignity: the administration of justice requires that where dignity is threatened in this way, measures be taken to accommodate this privacy concern. Although measured by reference to the facts of each case, the risk to this interest will be serious only where the information that would be disseminated as a result of court openness is sufficiently sensitive such that openness can be shown to meaningfully strike at the individual’s biographical core in a manner that threatens their integrity. Recognizing this interest is consistent with this Court’s emphasis on the importance of privacy and the underlying value of individual dignity, but is also tailored to preserve the strong presumption of openness.

D.           The Trustees Have Failed to Establish a Serious Risk to an Important Public Interest

[86]                          As Sierra Club made plain, a discretionary order limiting court openness can only be made where there is a serious risk to an important public interest. The arguments on this appeal concerned whether privacy is an important public interest and whether the facts here disclose the existence of serious risks to privacy and safety. While the broad privacy interest invoked by the Trustees cannot be relied on to justify a limit on openness, the narrower concept of privacy understood in relation to dignity is an important public interest for the purposes of the test. I also recognize that a risk to physical safety is an important public interest, a point on which there is no dispute here. Accordingly, the relevant question at the first step is whether there is a serious risk to one or both of these interests. For reasons that follow, the Trustees have failed to establish a serious risk to either. This alone is sufficient to conclude that the sealing orders should not have been issued.

(1)          The Risk to Privacy Alleged in this Case Is Not Serious

[87]                         As I have said, the important public interest in privacy must be understood as one tailored to the protection of individual dignity and not the broadly defined interest the Trustees have asked this Court to recognize. In order to establish a serious risk to this interest, the information in the court files about which the Trustees are concerned must be sufficiently sensitive in that it strikes at the biographical core of the affected individuals. If it is not, there is no serious risk that would justify an exception to openness. If it is, the question becomes whether a serious risk is made out in light of the facts of this case.

[88]                         The application judge never explicitly identified a serious risk to the privacy interest he identified but, to the extent he implicitly reached this conclusion, I respectfully do not share his view. His finding was limited to the observation that “[t]he degree of intrusion on that privacy and dignity [i.e., that of the victims and their loved ones] has already been extreme and, I am sure, excruciating” (para. 23). But the intense scrutiny faced by the Shermans up to the time of the application is only part of the equation. As the sealing orders can only protect against the disclosure of the information in these court files relating to probate, the application judge was required to consider the sensitivity of the specific information they contained. He made no such measure. His conclusion about the seriousness of the risk then focused entirely on the risk of physical harm, with no indication that he found that the Trustees met their burden as to the serious risk to the privacy interest. Said very respectfully and with the knowledge that the application judge did not have the benefit of the above framework, the failure to assess the sensitivity of the information constituted a failure to consider a required element of the legal test. This warranted intervention on appeal.

[89]                         Applying the appropriate framework to the facts of this case, I conclude that the risk to the important public interest in the affected individuals’ privacy, as I have defined it above in reference to dignity, is not serious. The information the Trustees seek to protect is not highly sensitive and this alone is sufficient to conclude that there is no serious risk to the important public interest in privacy so defined.

[90]                         There is little controversy in this case about the likelihood and extent of dissemination of the information contained in the estate files. There is near certainty that the Toronto Star will publish at least some aspects of the estate files if it is provided access. Given the breadth of the audience of its media organization, and the high‑profile nature of the events surrounding the death of the Shermans, I have no difficulty in concluding that the affected individuals would lose control over this information to a significant extent should the files be open.

[91]                         With regard to the sensitivity of the information, however, the information contained in these files does not reveal anything particularly private about the affected individuals. What would be revealed might well cause inconvenience and perhaps embarrassment, but it has not been shown that it would strike at their biographical core in a way that would undermine their control over the expression of their identities. Their privacy would be troubled, to be sure, but the relevant privacy interest bearing on the dignity of the affected persons has not been shown to be at serious risk. At its highest, the information in these files will reveal something about the relationship between the deceased and the affected individuals, in that it may reveal to whom the deceased entrusted the administration of their estates and those who they wished or were deemed to wish to be beneficiaries of their property at death. It may also reveal some basic personal information, such as addresses. Some of the beneficiaries might well, it may fairly be presumed, bear family names other than Sherman. I am mindful that the deaths are being investigated as homicides by the Toronto Police Service. However, even in this context, none of this information provides significant insight into who they are as individuals, nor would it provoke a fundamental change in their ability to control how they are perceived by others. The fact of being linked through estate documents to victims of an unsolved murder is not in itself highly sensitive. It may be the source of discomfort but has not been shown to constitute an affront to dignity in that it does not probe deeply into the biographical core of these individuals. As a result, the Trustees have failed to establish a serious risk to an important public interest as required by Sierra Club.

[92]                         The fact that some of the affected individuals may be minors is also insufficient to cross the seriousness threshold. While the law recognizes that minors are especially vulnerable to intrusions of privacy (see Bragg, at para. 17), the mere fact that information concerns minors does not displace the generally applicable analysis (see, e.g., Bragg, at para. 11). Even taking into account the increased vulnerability of minors who may be affected individuals in the probate files, there is no evidence that they would lose control of information about themselves that reveals something close to the core of their identities. Merely associating the beneficiaries or trustees with the Shermans’ unexplained deaths is not enough to constitute a serious risk to the identified important public interest in privacy, defined in reference to dignity.

[93]                         Further, while the intense media scrutiny on the family following the deaths suggests that the information would likely be widely disseminated, it is not in itself indicative of the sensitivity of the information contained in the probate files.

[94]                         Showing that the information that would be revealed by court openness is sufficiently sensitive and private such that it goes to the biographical core of the affected individual is a necessary prerequisite to showing a serious risk to the relevant public interest aspect of privacy. The Trustees did not advance any specific reason why the contents of these files are more sensitive than they may seem at first glance. When asserting a privacy risk, it is essential to show not only that information about individuals will escape the control of the person concerned — which will be true in every case — but that this particular information concerns who the individuals are as people in a manner that undermines their dignity. This the Trustees have not done.

[95]                         Therefore, while some of the material in the court files may well be broadly disseminated, the nature of the information has not been shown to give rise to a serious risk to the important public interest in privacy, as appropriately defined in this context in reference to dignity. For that reason alone, I conclude that the Trustees have failed to show a serious risk to this interest.

(2)          The Risk to Physical Safety Alleged in this Case is Not Serious

[96]                         Unlike the privacy interest raised in this case, there was no controversy that there is an important public interest in protecting individuals from physical harm. It is worth underscoring that the application judge correctly treated the protection from physical harm as a distinct important interest from that of the protection of privacy and found that this risk of harm was “foreseeable” and “grave” (paras. 22‑24). The issue is whether the Trustees have established a serious risk to this interest for the purpose of the test for discretionary limits on court openness. The application judge observed that it would have been preferable to include objective evidence of the seriousness of the risk from the police service conducting the homicide investigation. He nevertheless concluded there was sufficient proof of risk to the physical safety of the affected individuals to meet the test. The Court of Appeal says that was a misreading of the evidence, and the Toronto Star agrees that the application judge’s conclusion as to the existence of a serious risk to safety was mere speculation.

[97]                         At the outset, I note that direct evidence is not necessarily required to establish a serious risk to an important interest. This Court has held that it is possible to identify objectively discernable harm on the basis of logical inferences (Bragg, at paras. 15‑16). But this process of inferential reasoning is not a licence to engage in impermissible speculation. An inference must still be grounded in objective circumstantial facts that reasonably allow the finding to be made inferentially. Where the inference cannot reasonably be drawn from the circumstances, it amounts to speculation (R. v. Chanmany, 2016 ONCA 576, 352 O.A.C. 121, at para. 45).

[98]                         As the Trustees correctly argue, it is not just the probability of the feared harm, but also the gravity of the harm itself that is relevant to the assessment of serious risk. Where the feared harm is particularly serious, the probability that this harm materialize need not be shown to be likely, but must still be more than negligible, fanciful or speculative. The question is ultimately whether this record allowed the application judge to objectively discern a serious risk of physical harm.

[99]                         This conclusion was not open to the application judge on this record. There is no dispute that the feared physical harm is grave. I agree with the Toronto Star, however, that the probability of this harm occurring was speculative. The application judge’s conclusion as to the seriousness of the risk of physical harm was grounded on what he called “the degree of mystery that persists regarding both the perpetrator and the motives” associated with the deaths of the Shermans and his supposition that this motive might be “transported” to the trustees and beneficiaries (para. 5; see also paras. 19 and 23). The further step in reasoning that the unsealed estate files would lead to the perpetrator’s next crime, to be visited upon someone mentioned in the files, is based on speculation, not the available affidavit evidence, and cannot be said to be a proper inference or some kind of objectively discerned harm or risk thereof. If that were the case, the estate files of every victim of an unsolved murder would pass the initial threshold of the test for a sealing order.

[100]                     Further, I recall that what is at issue here is not whether the affected individuals face a safety risk in general, but rather whether they face such a risk as a result of the openness of these court files. In light of the contents of these files, the Trustees had to point to some further reason why the risk posed by this information becoming publicly available was more than negligible.

[101]                     The speculative character of the chain of reasoning leading to the conclusion that a serious risk of physical harm exists in this case is underlined by differences between these facts and those cases relied on by the Trustees. In X. v. Y., 2011 BCSC 943, 21 B.C.L.R. (5th) 410, the risk of physical harm was inferred on the basis that the plaintiff was a police officer who had investigated “cases involving gang violence and dangerous firearms” and wrote sentencing reports for such offenders which identified him by full name (para. 6). In R. v. Esseghaier, 2017 ONCA 970, 356 C.C.C. (3d) 455, Watt J.A. considered it “self‑evident” that the disclosure of identifiers of an undercover operative working in counter‑terrorism would compromise the safety of the operative (para. 41). In both cases, the danger flowed from facts establishing that the applicants were in antagonistic relationships with alleged criminal or terrorist organizations. But in this case, the Trustees asked the application judge to infer not only the fact that harm would befall the affected individuals, but also that a person or persons exist who wish to harm them. To infer all this on the basis of the Shermans’ deaths and the association of the affected individuals with the deceased is not reasonably possible on this record. It is not a reasonable inference but, as the Court of Appeal noted, a conclusion resting on speculation.

[102]                     Were the mere assertion of grave physical harm sufficient to show a serious risk to an important interest, there would be no meaningful threshold in the analysis. Instead, the test requires the serious risk asserted to be well grounded in the record or the circumstances of the particular case (Sierra Club, at para. 54; Bragg, at para. 15). This contributes to maintaining the strong presumption of openness.

[103]                     Again, in other cases, circumstantial facts may allow a court to infer the existence of a serious risk of physical harm. Applicants do not necessarily need to retain experts who will attest to the physical or psychological risk related to the disclosure. But on this record, the bare assertion that such a risk exists fails to meet the threshold necessary to establish a serious risk of physical harm. The application judge’s conclusion to the contrary was an error warranting the intervention of the Court of Appeal.

E.            There Would Be Additional Barriers to a Sealing Order on the Basis of the Alleged Risk to Privacy

[104]                     While not necessary to dispose of the appeal, it bears mention that the Trustees would have faced additional barriers in seeking the sealing orders on the basis of the privacy interest they advanced. I recall that to meet the test for discretionary limits on court openness, a person must show, in addition to a serious risk to an important interest, that the particular order sought is necessary to address the risk and that the benefits of the order outweigh its negative effects as a matter of proportionality (Sierra Club, at para. 53).

[105]                     Even if the Trustees had succeeded in showing a serious risk to the privacy interest they assert, a publication ban — less constraining on openness than the sealing orders — would have likely been sufficient as a reasonable alternative to prevent this risk. The condition that the order be necessary requires the court to consider whether there are alternatives to the order sought and to restrict the order as much as reasonably possible to prevent the serious risk (Sierra Club, at para. 57). An order imposing a publication ban could restrict the dissemination of personal information to only those persons consulting the court record for themselves and prohibit those individuals from spreading the information any further. As I have noted, the likelihood and extent of dissemination may be relevant factors in determining the seriousness of a risk to privacy in this context. While the Toronto Star would be able to consult the files subject to a publication ban, for example, which may assist it in its investigations, it would not be able to publish and thereby broadly disseminate the contents of the files. A publication ban would seem to protect against this latter harm, which has been the focus of the Trustees’ argument, while allowing some access to the file, which is not possible under the sealing orders. Therefore, even if a serious risk to the privacy interest had been made out, it would likely not have justified a sealing order, because a less onerous order would have likely been sufficient to mitigate this risk effectively. I hasten to add, however, that a publication ban is not available here since, as noted, the seriousness of the risk to the privacy interest at play has not been made out.

[106]                     Further, the Trustees would have had to show that the benefits of any order necessary to protect from a serious risk to the important public interest outweighed the harmful effects of the order, including the negative impact on the open court principle (Sierra Club, at para. 53). In balancing the privacy interests against the open court principle, it is important to consider whether the information the order seeks to protect is peripheral or central to the judicial process (paras. 78 and 86; Bragg, at paras. 28‑29). There will doubtless be cases where the information that poses a serious risk to privacy, bearing as it does on individual dignity, will be central to the case. But the interest in important and legally relevant information being aired in open court may well overcome any concern for the privacy interests in that same information. This contextual balancing, informed by the importance of the open court principle, presents a final barrier to those seeking a discretionary limit on court openness for the purposes of privacy protection.

VI.         Conclusion

[107]                     The conclusion that the Trustees have failed to establish a serious risk to an important public interest ends the analysis. In such circumstances, the Trustees are not entitled to any discretionary order limiting the open court principle, including the sealing orders they initially obtained. The Court of Appeal rightly concluded that there was no basis for asking for redactions because the Trustees had failed at this stage of the test for discretionary limits on court openness. This is dispositive of the appeal. The decision to set aside the sealing orders rendered by the application judge should be affirmed. Given that I propose to dismiss the appeal on the existing record, I would dismiss the Toronto Star’s motion for new evidence as being moot.

[108]                     For the foregoing reasons, I would dismiss the appeal. The Toronto Star requests no costs given the important public issues in dispute. As such, there will be no order as to costs.

                    Appeal dismissed.

                    Solicitors for the appellants: Davies Ward Phillips & Vineberg, Toronto.

                    Solicitors for the respondents: Blake, Cassels & Graydon, Toronto.

                    Solicitor for the intervener the Attorney General of Ontario: Attorney General of Ontario, Toronto.

                    Solicitor for the intervener the Attorney General of British Columbia: Attorney General of British Columbia, Vancouver.

                    Solicitors for the intervener the Canadian Civil Liberties Association: DMG Advocates, Toronto.

                    Solicitors for the intervener the Income Security Advocacy Centre: Borden Ladner Gervais, Toronto.

                    Solicitors for the interveners Ad IDEM/Canadian Media Lawyers Association, Postmedia Network Inc., CTV, a Division of Bell Media Inc., Global News, a division of Corus Television Limited Partnership, The Globe and Mail Inc. and Citytv, a division of Rogers Media Inc.: Farris, Vancouver.

                    Solicitors for the intervener the British Columbia Civil Liberties Association: McCarthy Tétrault, Toronto.

                    Solicitors for the interveners the HIV & AIDS Legal Clinic Ontario, the HIV Legal Network and the Mental Health Legal Committee: HIV & AIDS Legal Clinic Ontario, Toronto.