COURT OF APPEAL FOR ONTARIO

CITATION: Law Society of Ontario v. AA, 2026 ONCA 47¹

DATE: 20260126

DOCKET: M56164 & COA-25-CV-0462

Sossin, Favreau and Wilson JJ.A.

BETWEEN

Law Society of Ontario

Respondent (Appellant/Responding Party)

and

AA

Applicant (Respondent/Moving Party)

and

Justice for Children and Youth

Intervener (Intervener)

Amy Block, Benjamin Kates and Andrea Luey, for the appellant/responding party

James Melnick, for the respondent/moving party

Mary Birdsell and Samira Ahmed, for the intervener Justice for Children and Youth (COA-25-CV-0462)

Justin Manoryk, for the intervener Toronto Star Newspapers Limited (M56164)

Matthew Grace, for the intervener Law360 Canada (M56164)²

Heard: August 29, 2025

On appeal from the order of the Divisional Court (Justices David L. Corbett, Shaun O’Brien, and Janet Leiper), dated October 29, 2024, with reasons reported at 2024 ONSC 5971, dismissing an application for judicial review of a decision of the Law Society Tribunal Appeal Division, dated March 15, 2024.

TABLE OF CONTENTS

OVERVIEW.. 3

PART I: DID THE DIVISIONAL COURT ERR IN UPHOLDING THE TRIBUNAL’S GOOD CHARACTER DECISION?. 8

A. Decisions Below. 8

1. Law Society Tribunal Hearing Division (2023 ONLSTH 99) 8

2. Law Society Tribunal Appeal Division (2024 ONLSTA 6) 9

3. Divisional Court (2024 ONSC 5971) 10

4. Stay decisions. 11

B. Issues. 12

C. Standard of Review. 12

D. Analysis. 13

1. The legislative Scheme. 13

2. The Tribunal’s reasons. 15

3. The relevant constraints. 19

4. The reasonableness of the Tribunal’s decision in light of the text, context, and purpose of s. 27. 21

5. Remedy. 54

PART II: SHOULD THE ANONYMIZATION MOTION BE GRANTED?. 55

A. Overview. 55

B. Decisions Below. 57

1. Law Society Tribunal Hearing Division (2022 ONLSTH 9) 57

2. Law Society Tribunal Appeal Division (2024 ONLSTA 6) 58

3. Divisional Court: Temporary Anonymization Decision (2024 ONSC 3102) 58

4. Divisional Court: Permanent Anonymization Decision (2024 ONSC 5971) 59

C. Issues. 60

D. Standard of Review. 61

E. Analysis. 62

1. Legal framework. 62

2. The Divisional Court’s anonymization order does not bind this court, however its publication ban is still in effect 62

3. The motion should be granted on a fresh Sherman Estate analysis. 68

4. The appeal: the Divisional Court should have conducted a fresh Sherman Estate analysis, but its decision should stand. 84

DISPOSITION. 86

Sossin J.A.:

OVERVIEW

1This appeal requires the court to apply the framework for judicial review articulated in Canada (Minister of Citizenship and Immigration) v. Vavilov, 2019 SCC 65, [2019] 4 S.C.R. 653to a licensing decision of the Law Society Tribunal involving the good character requirement.

2In 2009, on three occasions over the course of a two-month period, AA sexually abused three children, including his eldest daughter, while he and his family lived abroad in a close-knit religious community. These instances of abuse involved clothed sexual touching. After being confronted by the father of one of his victims, AA disclosed his sexual abuse to various people, including his then-spouse, a religious leader, and the local child protection agency. He was not criminally charged, and the family moved back to Canada later that year. AA was subsequently diagnosed with paedophilic disorder, assessed to be in remission. At present, none of AA’s children, including his eldest daughter, know of his past sexual abuse.

3Between 2009 and 2017, AA was dishonest with medical practitioners, child protection authorities, and the Law Society of Ontario (the “Law Society”) about his past actions. He failed to disclose the nature and extent of his sexual abuse to various treating professionals, and instead, disclosed lesser issues. Furthermore, when contacted by Canadian child protection authorities about his misconduct, he misrepresented it, disclosing only one instance of sexual abuse rather than three.

4In 2012, AA first sought to be licensed as a lawyer. During this initial application process, AA did not disclose the facts of his sexual abuse. Instead, an anonymous letter alerting the Law Society to the incidents prompted it to launch an investigation. During that investigation, AA was uncooperative, withholding medical records and other information. In 2017, AA withdrew this licensing application.

5In 2019, AA again applied to be licensed. Pursuant to s. 27(4) of the Law Society Act, R.S.O. 1990, c. L.8 (the “Act”), the Law Society referred the matter to the Law Society Tribunal Hearing Division for a hearing into whether AA was “of good character” within the meaning of s. 27(2) and was thus entitled to a licence to practice law. Prior to the hearing, a panelist of the Hearing Division granted AA an order imposing a publication ban, and anonymizing and sealing the Tribunal’s record. At the hearing, AA was determined to be of good character. The Hearing Division also imposed a condition on AA’s licence, requiring that he be supervised in any meeting with “minor children”.

6The Law Society appealed this decision to the Tribunal’s Appeal Division, which upheld both the Hearing Division’s decision on AA’s good character and the licensing condition it imposed.

7The Law Society then sought judicial review of this decision at the Divisional Court, which upheld the decision of the Appeal Division, again including the licensing condition. The Divisional Court also issued an interim order imposing a publication ban with respect to AA’s identity (and those of his family members) and anonymizing its own record, which it subsequently made permanent in its decision on the merits of the judicial review application.

8On April 14, 2025, this court granted the Law Society leave to appeal. As part of the leave process, in my case management capacity, I issued an order anonymizing this court’s record, to the date of the appeal hearing. This order was not opposed.

9AA brought a motion, heard at the same time as the appeal of the Divisional Court’s decision on the merits, seeking to impose a publication ban with respect to proceedings in this court, and an anonymization order with respect to this court’s record, on the same terms as the Divisional Court’s decision on anonymization and non-publication. AA characterizes this order as an extension of the Hearing Division’s original order. While we reserved our decision on the motion, at the hearing, the panel extended this court’s interim anonymization order to the date of the release of this decision. Further, as part of its appeal of the Divisional Court’s decision on the judicial review application, the Law Society argues that the Divisional Court’s anonymization and non-publication decision should be set aside.

10For the reasons set out below, I would allow the appeal and remit the question of AA’s good character back to the Law Society Tribunal Hearing Division. The Appeal Division’s good character decision, upholding that of the Hearing Division, was unreasonable because it failed to grapple with key aspects of the good character threshold under s. 27 of the Act. As is discussed below, for ease of reference, I refer to this as the Tribunal’s decision. In particular, the Tribunal’s failure to consider whether finding AA to be of good character was consistent with public trust in the legal professions³ ignored key elements of s. 27’s text, context, and purpose. Additionally, the licensing condition imposed by the Tribunal introduced internal inconsistency into the underlying good character decision and is inconsistent with the text, context, and purpose of s. 27. It too is unreasonable.

11With respect to AA’s motion for an anonymization and non-publication order relating to his identity and that of his family, I would grant the motion, though for different reasons than those set out by the Tribunal and the Divisional Court. Applying the framework for discretionary anonymization orders from Sherman Estate v. Donovan, 2021 SCC 25, [2021] 2 S.C.R. 75, I conclude that the harm to AA’s daughter that would result from disclosing AA’s identity, and thus potentially revealing to the public that she was sexually assaulted by him, is a sufficiently serious harm to an important public interest that can only be guarded against by imposing an anonymization and non-publication order. Additionally, at this stage, the benefits of anonymization outweigh the harm to the open court principle. In order to give effect to my decision on the motion, I would also decline to interfere with the Divisional Court’s anonymization and non-publication decision, which remains in effect.

12My reasons proceed in five parts. With respect to the main appeal, first, I outline the various decisions below and their reasoning; second, I consider the relevant constraints on the Tribunal’s decision: the text, context and purpose of the good character assessment required by ss. 27(2) and 27(4) of the Act, including how the Tribunal has historically interpreted these provisions; and third, I examine the reasons of the Tribunal in light of these constraints, to assess the reasonableness of its decision. With respect to the motion, I first consider the decisions below on this issue, before applying the Supreme Court’s analysis in Sherman Estateto determine if anonymization is warranted. In that analysis, I also deal with the Law Society’s appeal of the Divisional Court’s decision with respect to anonymization and non-publication.

PART I: DID THE DIVISIONAL COURT ERR IN UPHOLDING THE TRIBUNAL’S GOOD CHARACTER DECISION?

A. DECISIONS BELOW

1. Law Society Tribunal Hearing Division (2023 ONLSTH 99)

13The Hearing Division structured its analysis around the oft-cited test for good character from Armstrong v. Law Society of Upper Canada, 2009 ONLSHP 29, at para. 29, which considers:

(1) The nature and duration of the misconduct;

(2) Whether the applicant is remorseful;

(3) What rehabilitative efforts have been taken and their success;

(4) The applicant’s conduct since the misconduct; and

(5) The passage of time since the misconduct.

14The Hearing Division found that AA’s misconduct was serious. However, it also found that AA was deeply remorseful and that evidence from medical and psychological practitioners and other support persons established significant and successful rehabilitative efforts. The Hearing Division held that AA’s post-2017 conduct was free from dishonesty and was characterized by AA pursuing rehabilitation and making amends for his actions, such that his risk of reoffending against minors was low. Furthermore, the Hearing Division held that a significant amount of time had passed since AA’s sexual abuse without him reoffending. The Hearing Division also held that AA’s actions since 2017 showed a sincere and concerted attempt to address not only his historical sexual misconduct but also to address his failure to be candid, forthright, and open about that misconduct.

15The Hearing Division then went on to impose a licensing condition that AA be supervised in any meeting with “minor children”. This condition evolved from an undertaking offered by AA himself at the hearing that was accepted by the panel. Citing the Tribunal’s leading authority on the propriety of imposing licensing conditions, Howard Steven Levenson v. Law Society of Upper Canada, 2009 ONLSHP 98, the Hearing Division reasoned that this condition was not meant to improperly raise AA to the status of having good character. Instead, it was meant to enhance public confidence in the regulation of lawyers and paralegals.

2. Law Society Tribunal Appeal Division (2024 ONLSTA 6)

16The Appeal Division rejected all three of the Law Society’s grounds of appeal. First, the Appeal Division rejected the Law Society’s argument that the Hearing Division committed any palpable and overriding errors in weighing the evidence before it. Second, the Appeal Division rejected the argument that the Hearing Division imposed the licensing condition in a manner that violated the Law Society’s right to procedural fairness, given that the Law Society had ample opportunity to provide its position on the condition’s propriety.

17Finally, the Appeal Division concluded that the Law Society could not demonstrate on a standard of palpable and overriding error that the licensing condition and the good character finding were inconsistent. In the Appeal Division’s view, the licensing condition in this case was not meant to “bootstrap” AA up to the level of good character. All it did was provide extra protection to the public, thus promoting public confidence. The Appeal Division was also not convinced that the condition was unenforceable.

3. Divisional Court (2024 ONSC 5971)

18The Divisional Court unanimously dismissed the Law Society’s application for judicial review. It first rejected the Law Society’s argument that the Tribunal’s decision was unreasonable because it failed to consider the public interest in granting AA’s licence. The court reasoned that the Hearing Division conducted a careful analysis of the Armstrong factors in a manner that adequately accounted for the seriousness of AA’s misconduct and paid careful attention to the extent of his rehabilitation, which in turn spoke to his risk of reoffending against minors. Thus, it adequately considered the public interest in granting AA’s licence, including the best interests of children. According to Law Society of Ontario v. Colangelo, 2024 ONSC 2446, 497 D.L.R. (4th) 676 (Div. Ct.), aff’d 2025 ONCA 341, 38 Admin. L.R. (7th) 4, this public interest is subsumed into the Armstrong factors and need not be considered separately.

19The court then rejected the Law Society’s contention that the Hearing Division’s licensing condition was inconsistent with its good character finding, rendering the decision as a whole unintelligible. First, because the Tribunal correctly recognized that licensing conditions cannot be used to “bootstrap” an applicant up to the level of good character. Second, the court concluded that upholding the licensing condition was reasonable in this case because it arose in circumstances that promoted rather than undermined AA’s good character. AA offered the Hearing Division an undertaking not to meet with children unsupervised, and the court reasoned that it may have caused a loss of public confidence for the Hearing Division not to have accepted it. Third, the court concluded that the licensing condition was not unenforceable, as contended by the Law Society.

4. Stay decisions

20The Law Society has been successful in the Appeal Division, Divisional Court, and this court at staying the orders of the lower court/Tribunal. Thus, the Hearing Division’s order is not in effect and AA is not currently licensed to practice law.

B. ISSUES

21The Law Society argues that the Appeal Division’s decision is unreasonable because it did not have due regard for the overarching public interest function of the licensing process, manifesting in three ways:

(1) The Appeal Division erred by failing to appreciate the nature of AA’s misconduct as a matter of law and in view of the evidence before it;

(2) The Appeal Division erred by assessing good character myopically, without assessing the result against the statutory purpose that animates the good character requirement; and

(3) The Appeal Division erred by upholding a practice condition that is ineffective, unenforceable, and cannot serve its purpose.

C. STANDARD OF REVIEW

22This court must first determine whether the Divisional Court correctly identified and applied the standard of review relevant to the Appeal Division’s decision to uphold the Hearing Division’s finding of good character: Agraira v. Canada (Public Safety and Emergency Preparedness), 2013 SCC 36, [2013] 2 S.C.R. 559, at paras. 45-47.

23There is no dispute that the Divisional Court correctly identified reasonableness as the appropriate standard of review. Accordingly, this court must next step into the shoes of the Divisional Court and determine if it correctly applied this standard in holding that the Appeal Division’s decision, affirming the Hearing Division’s decision, was reasonable. For ease of reference, I will refer to this inquiry as whether the Tribunal’s decision in this case was reasonable. Where the Hearing Division and the Appeal Division of the Tribunal framed the issues or its analysis differently, I refer to each level individually.

24As the Supreme Court explained in Vavilov, a decision will be unreasonable where it displays a failure of internal rationality in its reasoning, or a failure of justification in light of the constraints bearing on the decision: at para. 101. Accordingly, the content of reasonableness review “will always depend on the constraints imposed by the legal and factual context of the particular decision under review”: Vavilov, at para. 90. This demands a restrained, but robust “reasons first” approach to judicial review that focuses on the decision the administrative decision maker actually made, with special attention paid to the justification offered in light of these constraints: Mason v. Canada (Citizenship and Immigration), 2023 SCC 21, 485 D.L.R. (4th) 583, at para. 8, citing Vavilov, at paras. 12,15, 24 and 84-85.

D. ANALYSIS

1. The legislative Scheme

25The Act and the by-laws passed thereunder contain several provisions that are relevant to a good character assessment.

26Sections 4.1 and 4.2 of the Act set out the functions of the Law Society, and how it is required to discharge those functions. Section 4.1(a) defines one of the Law Society’s core functions as ensuring that:

(a) all persons who practise law in Ontario or provide legal services in Ontario meet standards of learning, professional competence and professional conduct that are appropriate for the legal services they provide.

27Section 4.2 provides that, in “carrying out its functions, duties and powers” under the Act, the Law Society, “shall have regard to” certain principles, including that:

1. The Society has a duty to maintain and advance the cause of justice and the rule of law.

2. The Society has a duty to act so as to facilitate access to justice for the people of Ontario.

3. The Society has a duty to protect the public interest. [Emphasis added.]

28Thus, in ensuring the competence and professionalism of lawyers, which necessarily includes their licensure, the Law Society must have due regard for the public interest.

29Section 27(2) sets out the requirement that all licensees be “of good character”, and s. 27(4) tasks the Law Society Tribunal with determining whether an applicant is of good character after a hearing upon referral by the Law Society. These sections provide:

27(2) It is a requirement for the issuance of every licence under this Act that the applicant be of good character.

(4) An application for a licence may be refused only after a hearing by the Hearing Division, on referral of the matter by the Society to the Tribunal.

30The Law Society’s By-Law 4, passed under s. 62(0.1)4.1 of the Act, governs the “licensing of persons to practise law in Ontario as barristers and solicitors”. Section 8(1) of the By-Law incorporates the good character requirement in s. 27(2) of the Act as a condition of licensing:

8.(1) The following are the requirements for the issuance of any licence under the Act:

3. The applicant must be of good character.

2. The Tribunal’s reasons

31As Vavilov directs, I begin my analysis with the reasons offered by the Tribunal: at para. 84; Mason, at para. 60. In deciding whether AA was of good character under s. 27(2) of the Act, the Hearing Division adopted a statutory interpretation of that provision based on earlier Tribunal good character decisions.

32I highlight the following aspects of the Hearing Division’s reasons.

33First, I note that the Hearing Division referred to the oft-invoked description of good character from Mary Southin, Q.C.’s (as she then was) article on the good character requirement in British Columbia:

“[G]ood character” means those qualities which might reasonably be considered in the eyes of reasonable men and women to be relevant to the practice of law.

Character…comprises in my opinion at least these qualities:

1. An appreciation of the difference between right and wrong;

2. The moral fibre to do that which is right, no matter how uncomfortable the doing may be and not to do that which is wrong no matter what the consequences may be to oneself;

3. A belief that the law at least in so far as it forbids things which are malum in se must be upheld and the courage to see it is upheld:

Mary F. Southin, Q.C., "What is ‘Good Character’" (1977), 35:2 Advocate 129, at p. 129.

34I also note that the Hearing Division applied the test for good character from Armstrong, noting that the Tribunal itself views the factors articulated therein as the general analytical framework for assessing good character. It then set out those factors, which I reproduce again below for convenience:

(a) the nature and duration of the past misconduct;

(b) the passage of time since the past misconduct;

(c) the applicant’s conduct since the past misconduct;

(d) what rehabilitative efforts, if any, have been taken, and the success of such efforts; and

(e) whether the applicant is remorseful.

35The Hearing Division explained, at para. 54 of its reasons, that:

[T]he Tribunal’s case law shows that the application of the Armstrong factors is not a mechanical or formulaic exercise and that the issue of whether the applicant is currently of good character is the central question that must be answered.

36The Hearing Division then made findings on each of the Armstrong factors.

37With respect to the nature and extent of the misconduct, the Hearing Division recognized the serious consequences of AA’s sexual assaults on three children in 2009. It noted that child sexual assaults have “devastating” and “lasting” impacts on the children involved. The Hearing Division concluded its analysis of this factor, stating, at para. 59, “[w]e find, however, that since [his misconduct] the applicant has been open and diligent in acknowledging his past conduct and its impact and has actively sought treatment for his sexual dysfunction.”

38With respect to remorse, the Hearing Division held, at paras. 60-61,

The applicant acknowledged that his sexual offences have a human cost and he spoke with regret about his actions. He described being unbalanced and filled with self-deception until 2017, when he acknowledged his actions and his dishonesty to himself and those affected by his actions.

We accept that since then he has confronted his past, been candid and that he lives with remorse.

39With respect to rehabilitation, the Hearing Division concluded, at para. 69:

We are persuaded that the applicant has a genuine commitment to continuing therapy and vigilance to ensure that he does not reoffend against minors. We accept that his actions show that he has recognized the harm that deception, both of himself and those around him, has caused and would cause.

40With respect to AA’s conduct since the incidents, the Hearing Division found, at para. 70:

The last of the incidents involving sexual misconduct took place in 2009. There is no evidence of any allegations of him committing similar acts since that time. He has not been the subject of any investigation flowing from any allegations of a similar nature, nor has he been charged or convicted by any authorities of any offences or improper conduct involving minors.

41Finally, with respect to the passage of time since the misconduct, the Hearing Division concluded, at para. 77:

We find that significant time has passed since the serious misconduct that took place in 2009 and that his actions since 2017 show a sincere and concerted attempt to address not only the historical sexual misconduct but also to address his failure to be candid, forthright, and open about that misconduct with his family, medical practitioners, and the Law Society during the initial licensing process.

42After this analysis of each Armstrong factor, the Hearing Division stated its conclusion, at para. 80, that “[i]n the present case, we are persuaded that the applicant has established he is currently a person of good character and that his application for an L1 licence should be granted.”

43The Hearing Division then stated, at para. 80, its conclusion with respect to whether AA’s license should be subject to conditions:

We have also considered his offer of an undertaking not to meet in unsupervised settings with minor children and conclude that public confidence in the regulation of lawyers and paralegals would be enhanced by a term that requires that he not do so.

44The Hearing Division did not define what is meant by “minor children”, but I take this phrase simply to include all individuals under the age of 18.

45In the Appeal Division’s view, the Hearing Division imposed this condition, not to impermissibly raise AA to the level of good character, but rather to further enhance public confidence.

3. The relevant constraints

46I now turn to the relevant constraints on the Tribunal’s decision, which inform the determination of whether this decision was reasonable.

47Vavilov identifies a list of non-exhaustive constraints that bear on whether an administrative decision is intelligible and justified: at para. 106. Here, the Tribunal’s reasons concern its interpretation and application of the phrase “good character” in s. 27(2) of the Act. In my view, this means that the two most important constraints at play in this case are the statutory scheme under the Act and the principles of statutory interpretation. According to the Supreme Court in Vavilov, “because administrative decision makers receive their powers by statute, the governing statutory scheme is likely to be the most salient aspect of the legal context relevant to a particular decision”: at para. 108. It follows that where a decision is made pursuant to a particular provision, as here, the administrative decision maker’s interpretation of that provision is particularly relevant to the reasonableness analysis: see e.g., Pepa v. Canada (Citizenship and Immigration), 2025 SCC 21, 504 D.L.R. (4th) 1, at paras. 62-65.

48In Vavilov, the Supreme Court elaborated that the decision maker’s task is to interpret the contested provision in a manner that is consistent with its text, context and purpose: at para. 121. Further, it must demonstrate that it was alive to each of these essential elements: Mason, at para. 69.

49Within this framework, a decision will be unreasonable where the decision maker, “in interpreting a statutory provision, fails entirely to consider a pertinent aspect of its text, context or purpose” in a manner that “causes the reviewing court to lose confidence in the outcome reached by the decision maker”: Vavilov, at para. 122. This can happen where “it is clear that the administrative decision maker may well, had it considered a key element of a statutory provision’s text, context or purpose, have arrived at a different result”: Vavilov, at para. 122.

4. The reasonableness of the Tribunal’s decision in light of the text, context, and purpose of s. 27

a. Text

50The text of the Act in general, and s. 27 in particular, must be the “anchor” of the Tribunal’s analysis: Québec (Commission des droits de la personne et des droits de la jeunesse) v. Directrice de la protection de la jeunesse du CISSS A, 2024 SCC 43, at para. 24, citing Mark Mancini, “The Purpose Error in the Modern Approach to Statutory Interpretation” (2022) 59 Alta. L. Rev. 919, at p. 927. Close attention must be paid to the text because it “specifies, among other things, the means chosen by the legislature to achieve its purposes”: CISSS A, at para. 24.

51As is clear from the outline of the legislative framework set out above, the text of s. 27 is broad and open-ended. While s. 4.2 signals a “public interest” approach to licensure, there are no statutorily prescribed factors in the Act, nor defined terms in By-Law 4, that serve to constrain or even guide the Tribunal’s good character assessment pursuant to s. 27, nor is the concept of the public interest in s. 4.2 a defined term under the Act.

i. The Hearing Division’s approach to the text

52As the text is the most important constraint on the Hearing Division’s interpretation of s. 27, it is necessary to consider the Hearing Division’s approach to the text to determine whether it was alive and responsive to the words the legislature actually used in setting out the requirements for licensure. The Hearing Division did not interpret and apply the phrase “good character” in isolation, but rather within the framework it had earlier elaborated in Armstrong. That is, it incorporated Armstrong’s interpretation of the meaning of the words “good character” by reference. It is thus necessary to consider whether this approach is sufficiently anchored in the text to be reasonable.

53As set out above, in this case, the Hearing Division referred to the description of good character from Mary Southin, Q.C. (as she then was). This description was also adopted by the Hearing Division in Armstrong: at para. 24. It includes the notion that good character means an appreciation of the difference between what is right and what is wrong, the “moral fibre” to do that which is right, and a belief that the law must be upheld, insofar as the law prohibits acts which are inherently wrong.

54In Armstrong, the Hearing Division rooted the various factors set out above in the broader role played by the public trust in the good character assessment. There, at para. 28, the Hearing Division adopted the following passage from its earlier decision in Preyra, Re, 2000 14383 (Ont. L.S.T.H.), at p. 6:

The onus is on the applicant to prove that he is of good character at the time of the hearing of the application. The standard of proof is the balance of probabilities. The relevant test is not whether there is too great a risk of future abuse by the applicant of the public trust, but whether the applicant has established his good character at the time of the hearing on a balance of probabilities. The test does not require perfection of (sic) certainty. The applicant need not provide a warranty or assurance that he will never again breach the public trust. The issue is his character today, not the risk of his re-offending.

55In other words, while the applicant need not provide a warranty that they will never again breach the public trust, they must demonstrate, on a balance of probabilities, their good character at the time of the hearing.

56After citing this excerpt from Re Preyra, the Tribunal in Armstrong turned to the question of the factors relevant to a good character assessment, and set out the factors reproduced above.

57At para. 25 of its reasons, the Hearing Division in Armstrong also adopted practitioner Gavin MacKenzie’s statement of the objectives of the good character requirement, which he views as synonymous with those of discipline, namely to: “protect the public, to maintain high ethical standards, to maintain public confidence in the legal profession and its ability to regulate itself, and to deal fairly with persons whose livelihood and reputation are affected”: Gavin MacKenzie, Lawyers & Ethics: Professional Responsibility (Toronto: Carswell, 2004) (loose-leaf updated June 2025, release 2), at § 23:2.

58Overall, the Hearing Division in Armstrong defined good character, at para. 23 of its reasons, as:

[T]hat combination of qualities or features distinguishing one person from another. Good character connotes moral or ethical strength, distinguishable as an amalgam of virtuous attributes or traits which undoubtedly include, among others, integrity, candour, empathy and honesty.

59By incorporating the analysis in Armstrong by reference, this is evidently the interpretation of the text of s. 27 adopted by the Hearing Division in this case.

60The breadth of the plain meaning of s. 27 makes this a plausible reading of the text. However, this breadth also makes it necessary to examine the context and purpose of s. 27 in order to determine whether the Tribunal’s interpretation of the good character requirement in the circumstances of this case was reasonable.

b. Context

61The good character requirement under s. 27 of the Act does not exist in a legislative vacuum. A key context for the good character assessment is that of the self-regulation of the legal professions in the public interest, consistent with s. 4.2(3) of the Act. For the Tribunal’s decision that AA is of good character to be reasonable, its reasons must demonstrate that it was alive to and appropriately constrained by this broader context.

62A review of the case law on the regulation of the legal profession reveals that the concepts of public trust and public confidence are central to this public interest context. This is both in the sense of the requirement that lawyers be worthy of their clients’ trust, and in the sense that the Law Society itself maintain the public’s trust and confidence. Indeed, the Supreme Court of Canada has consistently offered a justification for lawyers’ self-regulation in the public interest that has foregrounded the need for law societies to safeguard relationships of trust between individual lawyers and clients, such that the public can maintain confidence in the legal profession in general.

63For the Tribunal’s decision to be reasonable, it must demonstrate that it was alive to this broader context.

i. The Supreme Court’s emphasis on public trust and confidence as fundamental to self-regulation in the public interest

64The Supreme Court underscored the centrality of trust in the regulation of the legal profession in Fortin v. Chrétien, 2001 SCC 45, [2001] 2 S.C.R. 500, at para. 17:

The special rules governing the practice of the legal profession are justified by the importance of the acts that advocates engage in, the vulnerability of litigants who entrust their rights to them, and the need to preserve the relationship of trust between advocates and their clients.

65This relationship of trust flows directly from the fact that law societies have been granted by legislatures the authority to self-regulate in the public interest. As Wagner J. (as he then was) stated in Green v. Law Society of Manitoba, 2017 SCC 20, [2017] 1 S.C.R. 360, while considering the validity of rules enacted by the Law Society of Manitoba, law societies are given “broad discretion to regulate the legal profession on the basis of a number of policy considerations related to the public interest”: at para. 22. Thus, law societies must be afforded “considerable latitude in making rules based on [their] interpretation of the ‘public interest’ in the context of [their] enabling statute[s]”: Green, at para. 24; see also, Law Society of British Columbia v. Trinity Western University, 2018 SCC 32, [2018] 2 S.C.R. 293, at para. 36. Even where the legislation governing a law society lacks a public interest clause of the kind found in s. 4.2(3) of the Act, courts have held that the principle of self-regulation in the public interest nonetheless applies: see e.g., Song v. The Law Society of Alberta, 2025 ABKB 525, at paras. 88-89.

66The Supreme Court has also affirmed that the Law Society’s mandate to protect the “public interest”, as framed in s. 4.2(3), is a broad one that infuses all of its functions and includes safeguarding public confidence in the professions’ licensing process: Trinity Western University v. Law Society of Upper Canada, 2018 SCC 33, [2018] 2 S.C.R. 453, at paras. 13-27 and 48.

67The majority in Trinity Western v. LSUC also treats the integrity of and societal trust in the professions as conceptually identical to the public confidence purpose underlying s. 4.2(3), indicating that these terms should be interpreted synonymously. For example, at para. 27, the majority says that the Law Society was entitled to conclude that Trinity Western University’s discriminatory admissions policy could “undermine public confidence in the [Law Society’s] ability to self-regulate in the public interest.” The majority makes the same point earlier, at para. 21, but uses the “societal trust enjoyed by the legal profession” and its integrity as stand-in values for public confidence.

68The Tribunal itself recognized this broader context in Armstrong, when it quoted from Gavin Mackenzie’s book, and in particular that the good character requirement is meant to “maintain public confidence in the legal profession and its ability to regulate itself”.

ii. Public trust and confidence are dependent on the trustworthiness of individual lawyers

69As is clear from the excerpt from Fortin above, the Supreme Court has also repeatedly stressed that the trustworthiness component of self-regulation stems from the fact that those who seek out legal services have important interests at stake and are in a position of vulnerability vis-à-vis their lawyers. For example, in A.G. Can. v. Law Society of B.C., 1982 29 (SCC), [1982] 2 S.C.R. 307, at p. 335, Estey J. stated:

[M]embers [of the legal profession] are officers of the provincially-organized courts; they are the object of public trust daily; the nature of the services they bring to the public makes the valuation of those services by the unskilled public difficult; the quality of service is the most sensitive area of service regulation and the quality of legal services is a matter difficult of judgment. [Emphasis added.]

70Moreover, the court’s jurisprudence supports the claim that the public’s trust and confidence in the legal profession at large is an aggregation of individual clients’ trust in their particular lawyers. For example, in Law Society of Saskatchewan v. Abrametz, 2022 SCC 29, [2022] 2 S.C.R. 220, Rowe J., in the context of law society discipline proceedings, linked the trust members of the public place in their lawyers to “public confidence in the profession”: at para. 53. Citing the Court of Appeal of Alberta’s decision in Adams v. Law Society of Alberta, 2000 ABCA 240, 266 A.R. 157, Rowe J. stated that there is a “public dimension” to law society discipline proceedings which considers the effect of individual misconduct on the professions in general: at para. 88.

71In Adams, the court derived this “public dimension” from the “overarching trust that the profession and each member of the profession accepts” and which forms its “very foundation”: at paras. 6-10. This reasoning is not specific to the discipline context. Indeed, in Hill v. Church of Scientology of Toronto, 1995 59 (SCC), [1995] 2 S.C.R. 1130, Cory J. opined, in the defamation context, that a lawyer’s reputation for “professional integrity and trustworthiness” is both the “cornerstone of a lawyer’s professional life” and of “paramount importance to … other members of the profession”: at para. 118.

72The Law Society has itself acknowledged the primacy of public trust in the context of licensing individual applicants. In its submission on the Federation of Law Societies of Canada’s National Suitability to Practise Standard Consultation Report,⁴ the Law Society stated, at p. 70:

It is important to convey to the public and the profession that licencees are required to comply with standards of professional conduct. One of the ways of doing so is to licence those who, at the time of licensing, have demonstrated … respect for the rule of law and the administration of justice, honesty, governability and financial responsibility. Underlying these behaviours is the principle that the profession must be worthy of clients’ and the public’s trust. If an applicant’s past conduct has raised some question about his or her respect for the behaviours integral to the profession, it is valuable for law societies to make further inquiries and determine whether the applicant should be licenced. In this way, the Law Society’s commitment to maintaining standards of professional conduct is demonstrated. [Emphasis added.]

iii. Conclusion on context

73The primacy of trust and confidence sheds contextual light on the constraints that governed the Tribunal’s decision that AA is of good character. It is also necessary to examine the purpose of s. 27 in order to understand the link between the trustworthiness of individual applicants and the public’s trust and confidence in the legal professions on the one hand, and the good character requirement on the other.

c. Purpose

74As noted above, one of the key purposes of the Act as a whole is defined in s. 4.2(3), which provides that, “in carrying out its functions, duties and powers under this Act, the Society shall have regard to” its “duty to protect the public interest.”

75AA argues that because s. 4.2 of the Act only references the “Society,” not the “Tribunal”, which are defined separately in s. 1 of the Act, the Tribunal is not under the same duty to advance the public interest in its adjudicative function as the Law Society is in its broader regulatory function. I would reject this reading of the Act. It would make little sense, for example, for the Law Society to be under a duty to consider the public interest in referring a matter for a hearing, while the Tribunal does not have to consider it in presiding over that hearing.

76Additionally, while the Tribunal is an independent adjudicative body, it is an adjudicative body with a regulatory purpose, established to advance a policy goal: see e.g., Ocean Port Hotel Ltd. v. British Columbia (General Manager, Liquor Control and Licensing Branch), 2001 SCC 52, [2001] 2 S.C.R. 781, at para. 24. Further, the Tribunal’s Rules of Practice and Procedure describe their purpose as providing for “fair processes that consider the interests of the public” and the “timely determination of proceedings in accordance with the public interest”: rr. 1.1(a)-(b) (emphasis added).

77The Law Society submits that the objectives of the good character requirement, as interpreted by the Tribunal, are driven by the overriding public interest in protecting the public, maintaining high ethical standards, maintaining public confidence in the legal professions and their ability to self-regulate, and dealing fairly with applicants: see Gaya v. Law Society of Ontario, 2022 ONLSTH 53, at para. 8, citing Armstrong, at para. 25. See also, Mundulai v. Law Society of Ontario, 2024 ONSC 959 (Div. Ct.), at paras. 38-39, motion for extension of time to file a motion for leave to appeal denied, 2025 ONCA 68.

78I agree. In my view, good character cannot be reasonably understood or applied in isolation from the broader objectives of the Act, which take the questions of individual licensing applicants’ trustworthiness, and the public’s trust and confidence in the legal professions generally, as points of departure for the Law Society’s self-regulation in the public interest: see also, Law Society of Upper Canada v. Abbott, 2017 ONCA 525, 139 O.R. (3d) 290, at para. 78, leave to appeal refused, [2017] S.C.C.A. No. 355. The Tribunal was required to keep the purposes underlying s. 27 of the Act in mind when it decided AA is of good character.

d. The Hearing Division’s approach to context and purpose: the Armstrong framework on good character

79The Hearing Division’s approach to these key aspects of context and purpose was centred around its application of the Armstrong framework for assessing good character. Thus, it is important to consider the requirements of that framework as set out by the Tribunal in other good character cases as well. As Vavilov makes clear, a tribunal’s own jurisprudence interpreting a particular provision represents an important constraint on its decision making: at para. 131.

80In general, the Tribunal’s approach has been to view the good character requirement as rooted in the public trust vested in the Law Society. As the Hearing Division put it in Rad v. Law Society of Ontario, 2023 ONLSTH 67, at para. 56, aff’d 2023 ONLSTA 26:

Lawyers and paralegals are the holders of a societal trust and the Law Society of Ontario is the gatekeeper of the legal professions in Ontario. In part, the Law Society fulfils this role through the good character requirement. [Citations omitted.]

81Additionally, as the Tribunal explained in Yeager v. Law Society of Upper Canada, 2016 ONLSTH 42, at para. 5, the Armstrong framework is tied to public confidence in whether a person is able to be a lawyer despite prior misconduct:

Under s. 27 of the Law Society Act, RSO 1990, c. L.8 and the Tribunal’s jurisprudence, a person who has engaged in previous misconduct must show that he or she is of good character. It is our job to take a hard look at what happened with the public’s interest in mind. We must look at what went wrong, whether the person feels remorse for it, what he or she has done since to address the issues, how they have behaved since and how long ago it was. At the end of the day a key question is whether the public can be confident that the person is able to be a paralegal or lawyer despite what happened before. Our caselaw recognizes that people make mistakes, but also overcome them to become an excellent and ethical professional …. [Emphasis added.]

82While the Tribunal has consistently applied the factors from Armstrong, it also has consistently emphasized that they are not to be applied as a mere score card: Polanski v. Law Society of Upper Canada, 2020 ONLSTH 115, at para. 172, aff’d 2021 ONLSTA 26, appeal quashed, 2022 ONSC 1428 (Div. Ct.), leave to appeal to Ont. C.A. refused, M53298 (August 16, 2022); Birkett v. Law Society of Ontario, 2023 80139, at para. 55. As Gavin Mackenzie puts it: “[t]hese factors are not a code to be applied mechanically”: Mackenzie, at § 23:3.

83Rather, the Tribunal has been clear that these factors must be weighed holistically based on the facts of the case, in order to reach a conclusion on good character, and always bearing in mind the purposes of the good character requirement: Birkett, at para. 55; Law Society of Ontario v. Colangelo, 2023 ONLSTA 16, at paras. 24-25, aff’d 2024 ONSC 2446, 497 D.L.R. (4th) 676 (Div. Ct.), aff’d 2025 ONCA 341, 38 Admin. L.R. (7th) 4; Pachai v. Law Society of Ontario, 2021 ONLSTH 18, at para. 88; Yeager, at para. 5. Additionally, as the Tribunal’s decisions in cases such as Gaya and Mundulai show, it is particularly alive to this overarching public interest in its own interpretation of the good character requirement.

84To be clear, as emphasized by the Tribunal in the cases referred to above, this analytic framework is a holistic one. The concern for the public’s trust and confidence in the legal professions is embedded in the analysis of whether a particular person meets the good character threshold under s. 27 of the Act.

85Put differently, reasonably applied, the individual Armstrong factors cannot be seen as ends unto themselves.

86The question, therefore, is not simply whether a person has shown that they are remorseful, or that they have been successful in their rehabilitation, or that their conduct since the misconduct at issue has been exemplary, or that a significant amount of time has passed since the misconduct. These inquiries are undertaken for the purpose of determining whether the applicant seeking to be licensed is of good character. This purpose, in turn, requires the Tribunal to step back and engage in a broader assessment. That assessment will generally include the seriousness of the prior conduct, and the impact on the public’s trust and confidence in the legal professions, if any, of finding that an applicant is of good character notwithstanding that conduct and licensing them.

87In Rad, at paras. 59-60, the Hearing Division explained the link between the purposes of the good character requirement and the analysis of the Armstrong factors:

The good character requirement at its core has four purposes. It shields members of the public from those who cannot be entrusted with the power to preserve and protect the rights and freedoms of others. It maintains high ethical professional standards. It enhances the respect for, and the legitimacy of the legal professions, which are vital to upholding the rule of law and the continuation of self-regulation. Finally, it provides a mechanism “to deal fairly with persons whose livelihood and reputation are affected”: Armstrong v. Law Society of Upper Canada, 2009 ONLSHP 29 at para. 25, Gaya at para. 8.

In furtherance of these purposes, the requirement of good character emphasizes the public’s expectations that lawyers and paralegals “have qualities such as honesty, integrity, empathy, candour, moral or ethical strength, knowing the difference between right and wrong, having the moral fibre to do what is right no matter the consequences, and following the law”: Nsamba v. Law Society of Ontario, 2020 ONLSTH 62 at para. 11, Armstrong at paras. 23-25.

88The Hearing Division in Rad applied a two-step analytic approach to its good character assessment, noting that, “[i]n assessing whether the applicant has satisfied us that he is currently of good character, we considered these purposes and qualities, and the five factors as set out in Armstrong”: at para. 61 (emphasis added).

89In this sense, the Armstrong framework, as developed by the Tribunal, may be said to include two distinct analytic steps: first, an analysis of the various Armstrong factors, which may differ in significance and relevance in the context of particular cases; and second, an assessment of good character in light of those findings and the broader purposes of the good character analysis. Whether these two analytic steps are separated out by the Tribunal into a two-stage framework, or considered holistically together is of no moment, as discussed below.

90As the Hearing Division observed in John Blackburn v. Law Society of Upper Canada, 2010 ONLSHP 112, at paras. 52-53, the applicant’s misconduct leading to the good character hearing may be understood through the metaphor of the applicant, by virtue of the misconduct, having dug themselves into a hole. The deeper the hole, “the more difficult it is to climb out of”: Blackburn, at para. 52. The purpose of the good character hearing is to determine whether, at the time of the hearing, the applicant has successfully climbed out of that hole. That determination, in turn, requires an overall assessment, and not simply conclusions on each of the particular factors that may be relevant to the assessment.

91With this metaphor in mind, the relevance of the Armstrong factors is that they enable the Tribunal to determine if the applicant has climbed out of the hole they have dug for themselves at the time of the hearing. The seriousness of the prior misconduct assists the Tribunal in determining how deep that hole is. The relevance of public trust and confidence in the legal professions also may be understood as part of the assessment of how steep the climb will be for the applicant to reach licensure. The other Armstrong factors relating to remorse, rehabilitation, conduct since the misconduct, and the length of time since the misconduct, all enable the Tribunal to determine whether the applicant has in fact climbed out of the hole, and to explain that conclusion.

92While the Armstrong framework developed by the Tribunal may not be the only reasonable way to assess an applicant’s good character within the constraints of s. 27 of the Act, I see no basis on which to conclude this framework is unreasonable. It is responsive to the text, context and purpose of s. 27.

e. Was the Hearing Division’s decision reasonable?

93I turn now to the reasonableness of the Appeal Division’s decision to uphold the Hearing Division’s application of the Armstrong framework.

94The Law Society argues the Appeal Division’s decision was unreasonable for three reasons:

(1) The Appeal Division failed to appreciate the nature of AA’s misconduct;

(2) The Appeal Division assessed good character myopically, without assessing the result against the statutory purpose that animates the good character requirement; and

(3) The Appeal Division erred by upholding a licensing condition that is ineffective, unenforceable, and cannot serve its purpose.

95I address each ground of appeal in turn.

i. The Tribunal did not fail to appreciate AA’s misconduct

96AA objects, on a preliminary basis, to the ground of appeal relating to the Appeal Division’s appreciation of his misconduct because it was not raised before the Appeal Division itself. He submits that this court should decline to entertain it. He also argues that the intervener, Justice for Children and Youth’s (“JFCY”) submission that the Appeal Division’s decision is inconsistent with the best interests of the child principle is an argument that was not before the Appeal Division and should similarly not be entertained.

97I am not persuaded by these submissions. This is first because the Divisional Court exercised its discretion to hear both of these novel issues on judicial review: Alberta (Information and Privacy Commissioner) v. Alberta Teachers’ Association, 2011 SCC 61, [2011] 3 S.C.R. 654, at paras. 22-29. This court should not interfere with that discretion.

98The Divisional Court noted, at para. 12, that:

[T]he Law Society’s submissions to the Appeal Division do not appear to have raised the issue of a failure to consider the public interest including an express consideration of the best interests of children.

99The court went on to say:

To the extent the Law Society and intervener’s submission is that the Appeal Division had a proactive obligation to intervene in the Hearing Division’s decision to protect the public interest including the rights of children, I disagree that intervention was required in this case.

100However, the Divisional Court then went on to consider whether the Appeal Division appreciated the nature of AA’s misconduct in the context of the record before it, and whether its application of the Armstrong factors adequately protected the public interest, including the best interests of children: at paras. 13-20.

101Further, the Law Society raised these issues generally in its amended notice of appeal to the Appeal Division, noting that the Hearing Division failed to “consider relevant factors” in its good character analysis and failed to place sufficient weight on “the public interest in maintaining high ethical standards.”

102That said, at the substantive level, the Law Society’s argument on this issue effectively asks this court to re-weigh the Hearing Division’s conclusions on each of the Armstrong factors, which the Appeal Division upheld. This would be improper and I decline to do so: see Vavilov, at para. 125.

103As Vavilov instructs, reasonableness review must be attentive to the reasons the decision maker actually gave, in order to see whether they are coherent and attentive to factual constraints and the submissions of the parties: at paras. 105-7, 125-28. I see no errors in this regard with respect to the Tribunal’s reasons.

104The Hearing Division fully acknowledged the seriousness of AA’s past misconduct and held, at para. 55 of its reasons, that the “harm to a child who is the victim of sexual abuse is undeniable and severe.”

105The Law Society further argues that the Tribunal failed to grapple with the fact that the sexual abuse of children is one of society’s “greatest scourges.” The Law Society relies on sentencing case law (which holds that the penalties for sexual offences against children must increase) to suggest that the Tribunal should have recognized the severity of AA’s actions as a matter of law: see e.g., R. v. Friesen, 2020 SCC 9, [2020] 1 S.C.R. 424. I do not view these precedents as applicable to licensing, given that punishment is not an objective of a good character hearing.

106The findings of the Hearing Division on the individual Armstrong factors are entitled to deference, and the Appeal Division acted reasonably in deciding not to interfere with them.

107Finally, in response to JFCY’s submissions on the alleged failure of the Tribunal to consider the best interests of the child, I view this question as more appropriate to address as part of the second ground of appeal, discussed below. To the extent it was raised as a basis to conclude the Hearing Division failed to appreciate AA’s misconduct as part of its analysis of the Armstrong factors per se, I am not persuaded this constituted an error.

108Therefore, I would reject this ground of appeal.

ii. The Tribunal failed to consider the overarching public interest in the good character assessment

109The Law Society argues that the Tribunal failed to consider that the Law Society’s public interest role required the Tribunal to assess AA’s character with regard to his “heinous misconduct” and its impact on the reputation of the professions.

110The parties in this case disagree as to whether the Armstrong factors “subsume” the public interest. AA argues that this issue was settled by the Divisional Court in Colangelo, where Edwards R.S.J. stated, at para. 50:

There is a strong public interest in ensuring that only persons of good character are licenced to provide legal services in this Province. In my view, the public interest, including the public interest in rehabilitation is subsumed within the Armstrong factors. As put by the Tribunal at para. 53:

Our criminal justice system is not premised just on denunciation and punishment but also Society on correction and rehabilitation […] Requiring demonstration of current good character, as we have done, protects the public. Both support public confidence in a fair and just legal system. [Emphasis added.]

111AA submits, as the Divisional Court’s decision in Colangelo was recently upheld by this court, it is settled that no separate “public interest analysis” is necessary. This was also the interpretation of Colangelo adopted by the Divisional Court in this case.

112While this court’s reasons in Colangelo did not comment specifically on the question of a separate public interest analysis, in my view, the question of whether one is required in the good character assessment is a red herring.

113The statement by Edwards R.S.J., excerpted above, in my view, is accurate as far as it goes. The public interest underlying the good character requirement is precisely why the specific factors applied in Armstrong and subsequent cases are relevant. However, this does not mean that the public interest analysis in a good character assessment is exhausted by the mere analysis of each of the Armstrong factors individually, nor does Colangelo explicitly say this. As discussed above, those factors always must be balanced and assessed in light of the Law Society’s public interest mandate, and with the public’s trust and confidence in the legal professions in mind. While it may not be necessary in every case to include a separate reference to public trust and confidence (as the Armstrong framework is not to be applied mechanically), those broader considerations cannot be ignored if the Tribunal is operating reasonably, within the parameters of s. 27, in light of the text, context and purpose of that provision.

114Returning to the metaphor set out above, the question for the Tribunal in the second step of its analysis, stepping back to consider whether an applicant has established that they meet the good character threshold, may be framed as whether AA had climbed out of the hole in which he found himself due to his admitted misconduct. Here, the misconduct at issue included both AA’s sexual assault of three children, and the years he deceived others, including the Law Society, about the number and nature of those assaults.

115The Hearing Division referred to this concern in responding to the Law Society’s submission on the seriousness of AA’s misconduct and his dishonesty with respect to that misconduct, stating at para. 58 of its reasons:

[T]he public is entitled to expect that licensees will be held to a high standard of honesty and trustworthiness. Any misconduct involving deception, lack of integrity or intent to mislead is troubling as it has the effect of eroding public trust in the profession.

116While the Hearing Division identified the importance both of AA’s trustworthiness and the public’s trust and confidence in the legal professions, it failed to grapple with these questions in light of its findings on each of the Armstrong factors.

117The Hearing Division made findings on each of the Armstrong factors, which as I noted above are entitled to deference, and thus completed the first analytic step in the Armstrong analysis. However, the Hearing Division then summarily stated its conclusion that AA was of good character. The Hearing Division failed to step back and explain how it had balanced and assessed the Armstrong factors to reach this conclusion.

118Further, the Hearing Division’s analysis does not reflect any consideration of whether granting AA a licence would be consistent with public trust and confidence in the legal professions. For example, the Hearing Division did not explain why a period of five years of honesty, in its view, could overcome a period of eight years of dishonesty, so as to demonstrate AA had “climbed out of the hole” of his prior criminal activity.

119While the Hearing Division acknowledged the seriousness of AA’s sexual assault of children, it did not address how the seriousness of this conduct affected its analysis and weighing of the other Armstrong factors in light of the Law Society’s public interest mandate.

120Additionally, the Law Society, supported by JFCY, argues that the best interests of the child principle should have been considered in the Tribunal’s good character analysis. JFCY refers in particular to the United Nations Convention on the Rights of the Child,⁵ to which Canada is a signatory, and which specifies that the best interests of the child analysis is triggered by state actions “concerning children”: art. 3(1).

121While neither the Law Society nor JFCY were prepared to say AA’s admitted assaults would preclude him from ever being found to have the good character necessary to be licensed, both argue that the Hearing Division’s reasons paid mere lip service to the best interests of the child principle.

122In my view, it was not necessary for the Hearing Division to employ a best interests of the child analysis per se, particularly in this context where there was not a specific child at risk in the record before the Hearing Division. That said, the failure to consider the seriousness of AA’s assaults on children in relation to the public trust and public confidence in the legal professions constituted another gap in the Hearing Division’s analysis.

123These gaps cannot be filled by the specific findings with respect to the specific Armstrong factors. The Hearing Division’s error was not in how it weighed the evidence as part of its conclusions on each of the Armstrong factors. As noted above, the Hearing Division’s findings on that evidence are entitled to deference. Rather, the Hearing Division erred in failing to assess AA’s good character in light of those findings, in a manner consistent with the text, context, and purpose of s. 27. While reasons are to be read “holistically and contextually”, and reviewing courts are allowed to “connect the dots on the page where the lines, and the direction they are headed, may be readily drawn”, here, the Hearing Division’s reasons were entirely unmoored from the key legal constraints that govern the good character requirement under s. 27 of the Act: Vavilov, at para. 97, citing Komolafe v. Canada (Minister of Citizenship and Immigration), 2013 FC 431, 16 Imm. L.R. (4th) 267, at para. 11.

124In my view, it is far from clear the Hearing Division would have reached the same outcome with respect to good character if it had grappled with these questions. For this reason, I conclude this court cannot be confident in the outcome reached: Vavilov, at para. 122; Mason, at para. 69. Indeed, the absence of reasons on a particular constraint, here key aspects of statutory context, can render a decision unreasonable: Paul Daly, “The Scope and Meaning of Reasonableness Review After Vavilov” (2025) 63:2014 ABQB 496, 1 Alta. L.R. 1, at pp. 19-20, citing Mason; Ontario (Attorney General) v. Ontario (Information and Privacy Commissioner), 2024 SCC 4, 488 D.L.R. (4th) 1; Commission scolaire francophone des Territoires du Nord-Ouest v. Northwest Territories (Education, Culture and Employment), 2023 SCC 31, 487 D.L.R. (4th) 631.

125For these reasons, on this ground, I would conclude that the Hearing Division failed to grapple with key aspects of the good character assessment. Therefore, in my view, the Appeal Division’s decision upholding the Hearing Division was unreasonable.

iii. The licensing condition is inconsistent with a finding of good character

126While my conclusion above that the Tribunal’s good character decision was unreasonable would be sufficient to justify allowing the appeal, a significant portion of oral submissions dealt with the condition imposed by the Hearing Division after it found AA to be of good character. In my view, it is important for completeness to address this ground of appeal as well.

127AA argues that the reasonableness of the Hearing Division’s condition was not raised below. I would reject this submission. In my view, this is not a new issue. At the Appeal Division, the Law Society argued that the condition “is inherently inconsistent with a finding of good character”: Appeal Division Reasons, at para. 21. At the Divisional Court, it made essentially the same argument: Divisional Court Reasons, at paras. 21-22.

128The Tribunal has broad authority to attach conditions to the licensing of a lawyer. Section 49.26 of the Act provides, “[a]n order of the Hearing Division may include such terms and conditions as the Division considers appropriate.”

129It is well settled that terms and conditions cannot be imposed to “bootstrap” the applicant up to the level of good character: see e.g., Levenson, at para. 81; Nsamba v. Law Society of Upper Canada, 2020 ONLSTH 62, at para. 71. As the Appeal Division put it in this case, at para. 96:

The panel must first make a finding of good character, under the usual tests, which the hearing panel applied to AA. Only then, and to achieve the specific goal of public confidence, conditions may be attached.

130The Hearing Division stated its conclusion with respect to the condition attached to its finding that AA is of good character at para. 80:

In the present case, we are persuaded that the applicant has established he is currently a person of good character and that his application for an L1 licence should be granted. We have also considered his offer of an undertaking not to meet in unsupervised settings with minor children and conclude that public confidence in the regulation of lawyers and paralegals would be enhanced by a term that requires that he not do so.

131The Appeal Division, in rejecting the argument that the condition was inconsistent with a finding of good character, referred to two earlier precedents: Sheps v. Law Society of Upper Canada, 2016 ONLSTH 124, and Colangelo. In Sheps, the applicant suffered from a serious mental health condition. The Hearing Division made a good character finding and attached numerous restrictions to his licence, including ongoing therapy, practice monitoring, and that the applicant practice for two years as an employee and not a sole practitioner. In Colangelo, the Appeal Division found no inherent contradiction between a finding of good character and a restriction that effectively delayed the applicant’s practice of law until she completed her criminal sentence.

132Before this court, the Law Society has focused on the alleged unenforceability of the condition. The Law Society objects, as it did at the Appeal Division and Divisional Court, to the absence of any ongoing reporting or monitoring requirements as part of the condition.

133The Tribunal clearly has broad discretionary authority under s. 49.26 of the Act to attach a wide variety of conditions to the licence of an applicant it has deemed is of good character. Despite this broad statutory authority, it is obvious that the Tribunal cannot impose conditions that are unreasonable. This may occur where the condition fails to respect the constraints imposed by the text, context, and purpose of s. 27, which inform the parameters of reasonable conditions under s. 49.26. It may also occur where the condition introduces internal incoherence into the underlying good character decision: see Vavilov, at paras. 102-7. Indeed, there is no such thing as untrammeled discretion, and discretionary decisions must still be made reasonably: Vavilov, at para. 108; see also, Universal Ostrich Farms Inc. v. Canada (Food Inspection Agency), 2025 FCA 147, at paras. 49-50, leave to appeal refused, [2025] S.C.C.A. No. 344.

134The question on this ground of appeal is whether the condition imposed was reasonable, and, in particular, whether it was consistent with a decision that AA was of good character. Here, the salient contextual constraint that the Tribunal failed to respect was the notion that public trust in the legal professions is dependent on the trustworthiness of individual lawyers. Put another way, the Tribunal failed to recognize that public trust is eroded by imposing a demographic condition that suggests that AA cannot be trusted with all of the responsibilities of licensure.

135Both the examples of Sheps and Colangelo are distinguishable because neither suggest that the applicant in those cases lacked trustworthiness with respect to a particular demographic group. The condition imposed on AA, however, does suggest so.

136Sheps, in my view, is particularly instructive, as conditions like those imposed in that case are entirely consistent with a finding of good character. Someone who lives with a mental health condition does not lack trustworthiness as a result. However, it is open to the Tribunal to conclude that person’s trustworthiness is linked to their ongoing commitment to appropriate treatment or that temporary restrictions on the type of professional context within which they may practice are appropriate. The diligence and vigilance reflected in such conditions is consistent with the public’s trust and confidence in the legal professions.

137While the Hearing Division asserted its view that public confidence in the regulation of lawyers and paralegals would be enhanced by the demographic condition it imposed on AA, the condition itself suggests that AA cannot be trusted to be alone with children. This is fundamentally at odds with the conclusion that AA presently meets the threshold of good character. Further, by concluding AA was of good character but could not be trusted to be alone with children, public trust and confidence in the legal professions was more likely to be eroded than enhanced. This introduces internal incoherence into the Tribunal’s decision and constitutes a fatal flaw in the “overarching logic” of its interpretation and application of s. 27: Vavilov, at paras. 101-4.

138The parties have not raised any case in which a demographic restriction was placed on an applicant’s licence after a finding of good character. This seems to be the first and only time such a demographic condition has been imposed as part of a good character disposition.

139That said, AA argues there is precedent for demographic conditions imposed by the Tribunal as part of disciplinary dispositions. He relies on a number of such cases, including: Law Society of Ontario v. Splinter, 2021 ONLSTH 58, Law Society of Ontario v. Schulz, 2021 ONLSTH 178, and Law Society of Ontario v. Lesieur, 2021 ONLSTH 144.

140These cases are distinguishable and do not assist AA. The disciplinary context is distinct from the licensing context, and in particular, the good character requirement as part of the criteria for licensure. Once a licensed lawyer engages in misconduct, the consideration of the Tribunal in determining an appropriate remedy in the face of that misconduct asks different questions than those which are relevant to a good character determination. The Tribunal is not concerned with the trustworthiness of an applicant, but rather how trust in the subject lawyer might be restored, and how the protection of the public in the face of the risk posed by the lawyer’s misconduct might be assured: Mackenzie, at § 26:1; see also, Splinter, at para. 30; Schulz, at paras. 69-70; Law Society of Ontario v. King, 2022 ONSLTH 30, at paras. 10-12; Law Society of Upper Canada v. Aron, 2011 ONLSHP 31, at para. 5.

141A demographic condition in the disciplinary context is remedial by its very nature. A panel imposing such a condition may be understood as doing so with a view to restoring a licensed lawyer to a position of trust. It is not assessing an applicant’s prospective trustworthiness for the first time.

142Finally, the Divisional Court held that the Appeal Division’s decision to uphold the condition was reasonable because it was proposed by AA himself. The Divisional Court explained its reasoning as follows, at para. 23:

Indeed, upholding the condition in this case was reasonable because the condition arose in circumstances that promoted rather than undermined A.A.’s good character. This is because A.A. offered the Law Society and Tribunal an undertaking not to meet in an unsupervised setting with minor children. The Hearing Division accepted that offer and imposed it as a term on his licence. Given that A.A. proactively offered the undertaking as a good faith effort to provide further reassurance, it may have caused a loss of public confidence to decide not to add the condition to his licence. It therefore was reasonable for the Appeal Division to decline to intervene in the decision to impose the condition.

143I approach the origin of the condition from a different perspective. In my view, the fact the condition arose as a proposal of AA, and not as a result of the Hearing Division’s own analysis, renders it less likely to enhance the public’s trust and confidence.

144While there is no doubt the undertaking was offered by AA as a good faith effort to reassure the Tribunal that he would not pose a risk, if AA himself does not believe he can be trusted to be alone with children, this raises the question of why the Hearing Division did not address this admission in its good character analysis. In my view, this is a failure of the Hearing Division to address a key submission of the parties, a hallmark of unreasonableness under Vavilov: at paras. 127-28.

145Further, if the Hearing Division took AA’s suggestion seriously enough to include it in its order, then it must reflect the Hearing Division’s recognition of an ongoing risk to children resulting from AA’s licensure. The absence of any monitoring or enforcement mechanisms attached to the condition raises troubling questions as to how the Hearing Division envisioned addressing this ongoing risk through such a condition.

146Given my conclusion that the condition was unreasonable, however, it is not necessary to address these additional questions here.

147In my view, the Hearing Division’s imposition of the condition on AA as part of its good character assessment was unreasonable, as was the Appeal Division’s decision to uphold this aspect of the Hearing Division’s decision. The condition cannot be justified in relation to the text, context, and purpose of s. 27, which inform the parameters of reasonable conditions under s. 49.26. It also introduces incoherence into the Tribunal’s underlying good character decision. The unreasonableness of the condition is a separate and distinct basis for allowing the appeal.

f. Conclusion

148For all of these reasons, the Appeal Division’s decision, upholding the Hearing Division’s good character finding and the licensing condition it imposed on AA, was unreasonable. Accordingly, the Divisional Court erred in its application of the standard of review. The appeal must be allowed.

5. Remedy

149The Law Society argues that, if successful, the proper remedy is for this court to substitute a finding that AA is not of good character, rather than remitting the matter back to the Hearing Division.

150The general rule is that the appropriate remedy when allowing an application for judicial review is remitting the matter back to the tribunal of first instance: Vavilov, at para. 142. It is only in limited circumstances, such as where there is a single reasonable decision or where the outcome is otherwise inevitable, such that remitting the matter would serve no useful purpose, where the reviewing court may substitute its own decision for that of the tribunal: see e.g., Pepa, at para. 121; Mason, at para. 120.

151If I was of the view that certain conduct, such as child sexual assault, could forever preclude a person from being found to be of good character, then this remedy would make sense. Why remit a matter whose outcome is inevitable?: see Mobil Oil Canada Ltd. v. Canada-Newfoundland Offshore Petroleum Board, 1994 114 (SCC), [1994] 1 S.C.R. 202, at pp. 227-28. However, the Law Society has not taken that position, nor is it consistent with the scheme of s. 27 of the Act. If certain misconduct precluded an applicant from demonstrating good character, there would be no need to refer every good character matter for a hearing under s. 27(4).

152This is not a case like Mason or Pepa where there is a single reasonable interpretation of s. 27 that leads inexorably to a single reasonable outcome. Rather, it is a case where the application of s. 27, interpreted reasonably, leaves room for different outcomes.

153For this reason, in my view, the proper remedy is to quash the Appeal Division’s decision and remit the matter back to the Hearing Division to consider the question of AA’s good character afresh, in accordance with the principles set out above.

PART II: SHOULD THE ANONYMIZATION MOTION BE GRANTED?

A. OVERVIEW

154AA brings this motion seeking to anonymize his name and any information that would tend to identify his children or his ex-spouse in this court’s record. He also asks this court to impose a publication ban over this same information. For ease of reference, I refer to this motion as one for an anonymization and non-publication order. In AA’s submission, this is merely a continuation of the order granted by the Hearing Division in January 2022, and subsequently upheld by both the Appeal Division and the Divisional Court. The existing orders protect the identity of AA and his family, particularly his four children, one of whom was a victim of his sexual assaults, and three of whom, according to AA, are currently in treatment for mental health concerns.

155AA submits that the anonymization and non-publication order will protect his children from serious psychological harm, exposure, and stigma, especially in light of growing online hostility and threats directed at AA and his family. Accordingly, anonymization of this court’s record is necessary to preserve this protection, especially because, in the absence of such an order, AA’s children could learn of his abuse through this court’s reasons.

156AA further argues the order would not impair public transparency, as AA’s licence condition would be published, together with his actual name, on the Law Society website directory.

157The Law Society opposes the motion and argues that an anonymization and non-publication order is not warranted in these circumstances. The Law Society also appeals from the anonymization and non-publication decision made by the Divisional Court.

158I set out the decisions on anonymization and non-publication below, together with the reasons why I conclude a fresh Sherman Estate analysis is required before this court on the motion, followed by the outcome of that analysis.

159While I do not accept all of AA’s submissions above, and for different reasons than the Divisional Court, I would nonetheless grant AA’s motion at this time.

160I also conclude that the Divisional Court erred by failing to conduct a fresh Sherman Estateanalysis when it made its temporary anonymization and non-publication order permanent. However, interfering with this decision would defeat the purpose of the decision to anonymize this court’s record and to prohibit publication of its proceedings. Consequently, I would leave the Divisional Court’s decision intact.

B. DECISIONS BELOW

1. Law Society Tribunal Hearing Division (2022 ONLSTH 9)

161Prior to his good character hearing, AA brought a motion seeking an order from a panelist of the Hearing Division, anonymizing his name, and those of his children, victims, and ex-spouse, sealing the Tribunal’s file, and imposing a publication ban on any information that would identify him and his family members. The Law Society opposed anonymizing AA’s name, but accepted that any reference to the victims of AA’s assaults or his other children should be anonymized.

162The panelist applied the test from Sherman Estate and concluded that the order AA sought was justified because any harm to the open court principle was outweighed by the harm to AA’s children of learning that AA had sexually assaulted one of them through a Tribunal proceeding.

163The panelist declined to make this order permanent. Instead, it was to remain in effect until varied or cancelled on a motion based on a material change in circumstances, or by the panel presiding at the good character hearing.

2. Law Society Tribunal Appeal Division (2024 ONLSTA 6)

164The Appeal Division did not engage in a fresh assessment of the grounds for anonymization and non-publication, but rather included a footnote indicating that the earlier anonymization and non-publication order remained in effect.

3. Divisional Court: Temporary Anonymization Decision (2024 ONSC 3102)

165Prior to the Divisional Court hearing the application for judicial review, AA brought a motion before Davies J., seeking a declaration that the Tribunal’s anonymization and non-publication order applied to proceedings in the Divisional Court, or in the alternative, that the court make an order on the same terms as that of the Tribunal. The Law Society opposed the motion.

166The motion judge concluded that the Tribunal’s order did not apply to proceedings in the Divisional Court. However, on a fresh Sherman Estate analysis, she imposed a temporary anonymization and non-publication order on the same terms as the Tribunal’s, effective to the date of the Divisional Court’s decision on the merits, unless extended by the panel. However, the motion judge noted that if AA was successful on the application for judicial review, the Law Society would then have a very strong interest, consistent with its statutory mandate, in having the anonymization and non-publication order lifted, so the public could make an informed decision about whether to retain AA with the full knowledge of his misconduct.

4. Divisional Court: Permanent Anonymization Decision (2024 ONSC 5971)

167In its merits decision, the Divisional Court concluded, in three brief paragraphs and without conducting a fresh Sherman Estate analysis, that the Appeal Division’s continuation of the Hearing Division’s order was reasonable and that Davies J.’s temporary anonymization and publication ban order should be continued. The court stated, at paras. 58-60:

The Law Society submits it was unreasonable for the Appeal Division to extend the anonymization order. In its submission, while the condition prohibiting unsupervised access would be attached to A.A.’s name on the Law Society website, the decision explaining the basis for the condition and A.A.’s history would not. This absence of transparency would constitute a failure to protect the public interest.

I would not interfere with the Appeal Division’s continuation of the anonymization order. The applicant’s name will also continue to be anonymized in the publication of this court’s decision. It was reasonable for the Appeal Division to continue the order, considering the need to balance public protection with the interests of A.A.’s children.

The Law Society also submits circumstances have changed because of the passage of time and the results in this court. However, the appropriate forum to raise those issues at first instance is before the Tribunal: Dr. Q v. College of Physicians and Surgeons of British Columbia, 2003 SCC 19, [2003] 1 S.C.R. 226, at para. 46. The record before this court does not satisfy me that the interests of A.A.’s children have changed materially. If the Law Society raises the issue, it will remain open to the Tribunal to determine whether a variation of the anonymization order is warranted. [Emphasis added.]

168While the Divisional Court addressed the issue of anonymization and non-publication in its decision, the formal order of the Divisional Court on appeal makes no mention of anonymization or non-publication.

169Before this court, the Law Society’s notice of appeal seeks to set aside the order of the Divisional Court. There is an argument, therefore, that the decision of the Divisional Court with respect to anonymization and non-publication is not properly before this court on appeal.

170That said, the Supreme Court has confirmed that confidentiality decisions do not necessarily need to be formalized in orders: see e.g., Canadian Broadcasting Corp. v. Manitoba, 2021 SCC 33, [2021] 2 S.C.R. 785, at paras. 41-42.

171In this case, the Law Society’s notice of appeal clearly asserts that the Divisional Court erred by “continuing its anonymization order and/or by upholding that of the Tribunal without regard to the principles set out in Sherman Estate”.

172Therefore, the Divisional Court’s decision on anonymization and non-publication is before this court on appeal.

C. ISSUES

173AA raises three issues to be determined on this motion:

(1) Is there a presumptive continuity of the anonymization order granted by the Tribunal and affirmed by the Divisional Court, such that the Court of Appeal should maintain those protections in the absence of a material change in circumstances?

(2) Has the Law Society waived or forfeited its right to challenge the anonymization order at the Court of Appeal stage, given its express position before the Tribunal Appeal Division that it would not seek to vary the order unless the matter were remitted for a new hearing?

(3) If the court deems it necessary to revisit the merits, should the anonymization order be continued under the Sherman Estate framework?

D. STANDARD OF REVIEW

174As is discussed below, AA’s motion to anonymize this court’s record and impose a publication ban is a first-instance decision. No standard of review applies.

175The standard of review with respect to the Law Society’s appeal of the Divisional Court’s anonymization and non-publication decision is less clear. Some courts have held that anonymization decisions are exercises of judicial discretion, subject to a deferential standard of review unless the court below erred in principle, ignored or misapplied a relevant factor, or was so clearly wrong so as to amount to an injustice: see e.g., Eghtesad v. British Columbia (Director of Civil Forfeiture), 2024 BCCA 32, at para. 16. However, there is authority from this court holding that confidentiality orders which interfere with court openness are subject to appellate review on a standard of correctness: P1 v. XYZ School, 2022 ONCA 571, at paras. 37-38.

176It is unnecessary to delve further into the standard of review because, as discussed further below, the error alleged in this case would meet either standard.

E. ANALYSIS

1. Legal framework

177A party asking the court to exercise its discretion to limit court openness through an anonymization order must satisfy the three criteria set out in Sherman Estate, at para. 38:

(1) [C]ourt 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.

2. The Divisional Court’s anonymization order does not bind this court, however its publication ban is still in effect

178There is substantial authority for the proposition that the anonymization and non-publication order of a lower court does not bind a higher court. This stems from the fact that appellate courts have the power to control their own process and court record. This power is necessarily implied by the legislative grant of adjudicative authority given to appellate courts: CBC v. Manitoba, at paras. 62-63.

179The power of a court to control its own process is just that: the power to control its own process. Applying a lower court order restricting the openness of its record to an appellate record would offend the fundamental principle that courts must control their own process and be empowered to exercise other powers that are practically necessary to accomplish the role the law assigns them: CBC v. Manitoba, at para. 62; see also, R. v. Cunningham, 2010 SCC 10, [2010] 1 S.C.R. 331, at para. 19; Lochner v. Ontario Civilian Police Commission, 2020 ONCA 720, at para. 27.

180The Divisional Court has routinely held that the anonymization orders made by subordinate tribunals do not bind it. For example, in her temporary anonymization decision, Davies J., relying on G-L. v. OHIP (General Manager), 2014 ONSC 5392, 327 O.A.C. 53 (Div. Ct.), at paras. 6-8, leave to appeal to Ont. C.A. refused, M43614 (August 22, 2014), concluded that parties are required to bring a separate motion if they wish to be anonymized in proceedings before the Divisional Court, even if they were anonymized in proceedings before a subordinate body: see also, Nahas v. Health Professions Appeal and Review Board, 2021 ONSC 6940 (Div. Ct.). It follows that this court is similarly not bound by the anonymization and non-publication orders of lower courts.

181The authorities AA cites for the contrary proposition are distinguishable. This is particularly so because several of those authorities, like R.A.R. v. College of Physicians and Surgeons of Ontario (2006), 2006 37118 (ON CA), 275 D.L.R. (4th) 275 (Ont. C.A.), deal with mandatory anonymization/non-publication orders (or orders where proceedings were presumed not to be public) and the Sherman Estate framework only applies to discretionary orders: Sherman Estate, at paras. 37-38.

182It is helpful to make a comment in this context about the distinction between the different types of orders limiting court openness and the finding that the applicability of such orders is limited to the court that made them.

183Accepted limits on court openness include publication bans, in camera or closed hearings, sealing orders, and orders allowing anonymization or the use of pseudonyms: James Rossiter, Law of Publication Bans, Private Hearings and Sealing Orders (Toronto: Thomson Reuters, 2006) (loose-leaf updated May 2025, release 1), at § 1:6. Both conceptually and as a practical matter, these types of orders operate differently.

184Publication bans prohibit the disclosure of information, usually in a publication or broadcast, which is the subject of the ban: Rossiter, at § 1:7. This prohibition imposes a duty on the world to refrain from publication. Sealing orders, redaction orders, and anonymization orders, by contrast, treat information in a court file as confidential and outside of the public realm, prohibiting the public from physically accessing, reading, or reviewing the information: Rossiter, at § 1:12, 1:13. More fundamentally, sealing, redaction and anonymization orders pertain to a physical court file, whereas publication bans pertain to intangible information: Rossiter, at § 1:12.

185Since each court has the power to decide whether to make an order limiting court openness in its own proceeding, information may be treated differently in one court than it is in another. For example, one court may decide that a document in its court record should be sealed. Another court may take the opposite approach. The practical result is that the document will be accessible if a copy can only be obtained from one court, but not the other.

186Publication bans, on the other hand, operate differently. For example, if a lower court imposes a publication ban over certain information and an appellate court declines to impose a ban over the same information, and the order below is not set aside, varied, stayed, or appealed, the practical result is that disclosure of the information is still prohibited by operation of the first ban: see e.g., CBC v. Chief of Police, 2021 ONSC 6935, 158 O.R. (3d) 401 (Div. Ct.), at paras. 41-55; Laity v. The College of Physicians and Surgeons of Ontario, 2018 ONSC 4557 (Div. Ct.), at para. 11.

187There is a potential for inconsistent results between different levels of court or administrative tribunals. This may seem like an unusual outcome, but it is, in fact, well-founded in the open court principle. Whether to impose a discretionary order limiting court openness is always a contextual balancing exercise, based on the circumstances before the court at the time it is considering the requested order, including whether the information at issue is peripheral or central to the judicial process: Sherman Estate, at paras. 105-6; XYZ School, at para. 36. Often, the circumstances advanced in support of a confidentiality order are different before an appellate court than they were before a lower court. The balancing exercise in the Sherman Estate analysis will be different and may therefore appropriately lead to a different result.

188This inconsistency can also be remedied, if the circumstances warrant it. Courts retain supervisory jurisdiction over their own records to make or vary court openness orders after a formal order is entered on the merits: CBC v. Manitoba, at para. 38. The landscape that led a lower court to impose a confidentiality order may be entirely different when it comes time for an appellate court to consider a motion for the same order. As discussed below, that may well be the case in these proceedings. Additionally, the change in circumstances that led an appellate court to make a different order than the lower court may provide a basis for an interested party to seek to set aside or vary the original confidentiality order made by the lower court.

189The facts of this case illustrate this point. In this case, AA moves for an order anonymizing his name and the names of his former spouse and children from the publicly accessible court file. He also moves in this court for a publication ban over any information that would tend to identify any of them.

190However, the confidentiality orders imposed by the Tribunal and the Divisional Court remain intact (subject to the Law Society’s appeal of the Divisional Court’s confidentiality order, which I address below). Accordingly, the publication ban first imposed by the Hearing Division continues to prohibit disclosure of information that would identify AA or his family, but that is not because the order applies in this court. Instead, it is a feature of the Hearing Division’s order not having been set aside, varied, appealed, or stayed. The sealing and anonymization orders imposed by those bodies also remain intact, but because they cover different physical court files, this has no practical impact on this court, which must decide whether to exercise its powers to control its own process and impose an order sealing material in its own file.⁶

a. Conclusion

191Thus, there are three “orders” relevant to AA’s motion and the Law Society’s appeal: first, the anonymization, sealing, and publication ban order of the Tribunal, which is not before this court and remains in effect; second, the Divisional Court’s decision granting anonymization and a publication ban for the purposes of the judicial review of the Tribunal’s decision before the Divisional Court, which I deal with below; and third, the order AA seeks on his motion before this court, anonymizing his name and any information that might identify his ex-spouse and children in this court’s record, and imposing a publication ban over that information for the purposes of his appeal. It is to this motion that I now turn.

3. The motion should be granted on a fresh Sherman Estate analysis

a. The Law Society has not waived its objection to anonymization and non-publication

192According to the Law Society, AA misstates the record with respect to its position on the anonymization and non-publication order AA seeks. Reviewing the record, the Law Society’s argument has some force. It has not waived any objection to AA’s motion for an anonymization and non-publication order in this court.

193The Law Society expressly put appealing the anonymization and non-publication order on the table in its notice of appeal to the Appeal Division. One of its grounds of appeal was that “the Hearing Panel failed to consider the public interest concern of the anonymization of an order”. This position was confirmed in its submissions to the Appeal Division. For example, in oral submissions before the Appeal Division, counsel for the Law Society submitted, “[s]o another error committed … is that [the Hearing Division] didn't consider when viewing the mandate of the Law Society, the public confidence … [regarding] anonymization of the identity of the candidate”.

194It is clear from the record that the Law Society never waived its challenge of anonymization and non-publication. I would add that while the Law Society did not oppose the interim orders relating to anonymization and non-publication in this court, counsel for the Law Society stipulated at each stage that the Law Society reserved its right to challenge the anonymization and non-publication order at the hearing of the motion.

195Further, as the Law Society notes, AA did not raise the issue of waiver before the Divisional Court and should not be able to bring this new argument on appeal: see e.g., R. v. Reid, 2016 ONCA 524, 132 O.R. (3d) 26, at para. 39, leave to appeal refused, [2017] S.C.C.A. No. 432, citing R. v. Warsing, 1998 775 (SCC), [1998] 3 S.C.R. 579, at para. 16, per L’Heureux-Dubé J. (dissenting in part).

196I also agree with the Law Society that it is doubtful that a party can waive the application of the open court principle. It cannot be said to be the privilege of a particular litigant: see e.g., L.C.F. v. G.F., 2016 ONSC 6732, 406 D.L.R. (4th) 750, at para. 21. Rather, it is a right of the public at large, emanating from the “constitutionally-entrenched right of freedom of expression” and designed to protect a central feature of liberal democracies, being the “public scrutiny” of open courts: Sherman Estate, at paras. 1-2.

197For these reasons, I would reject AA’s contention that his motion should be granted on the basis of any purported waiver of the Law Society.

b. The fresh Sherman Estate analysis

198With respect to conducting a fresh Sherman Estate analysis, the Law Society argues that there is a weak evidentiary record on which to conclude that revealing AA’s name would cause specific and demonstrable harm to his children.

199Because AA submits that he merely wishes to continue the existing order of the Tribunal, his record in support of the motion is mostly comprised of information that was before the Tribunal. Thus, he argues that since there has been no material change to the Tribunal record, no modification of the anonymization and non-publication order is warranted, and no new Sherman Estate analysis is needed or appropriate.

200I would reject this argument. The Tribunal, the Divisional Court, and this court each have a separate duty as custodians of their own records to consider the Sherman Estate analysis in response to AA’s wish to keep his identity from being revealed in the record and in any published decision.

201Accordingly, below, I apply the Sherman Estate analysis in the context of this motion.

i. Is an important public interest put at risk if anonymization and non-publication are not granted?

202As noted above, the first stage of the test in Sherman Estate requires AA to identify an important public interest that would be put at serious risk if the open court principle is not restricted in this appeal. Here, AA argues that disclosure of his name will cause his children, one of whom is the victim of his abuse, serious psychological harm because they will learn of his misconduct through a court process. He also argues, seemingly for the first time, that the online threats he has received heighten the risk to his family and create a risk of harm to him personally that can only be remedied with an anonymization and non-publication order.

203Each of these arguments relates to the public interest in privacy. As Kasirer J. observed in Sherman Estate, at para. 34:

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.

204The Supreme Court of Newfoundland and Labrador applied Sherman Estate on an application for a publication ban in a criminal proceeding brought by a practicing lawyer charged with sexual assault: R.R. v. Newfoundland and Labrador, 2022 NLSC 46, 417 C.C.C. (3d) 154. The court rejected the argument that disclosing the identity of the lawyer charged with the sexual assault met the public interest threshold under the Sherman Estate test, explaining, at para. 72:

I reject the Applicant’s assertion that the release of his identity on the facts of this case attack his biographical core. The information which would be revealed is his name and the fact that he is being charged with a number of sexual offences. While it would undoubtedly be embarrassing, it does not impact his biographical core. It does not raise a serious risk to an important public interest. It does not reach the “high bar” established in Sherman Estate and the authorities cited therein. The interests of the Applicant engaged in this case are purely private and personal.

205A similar analysis is applicable here.

206I would adopt the description of the open court principle in the good character context given by the panelist of the Hearing Division at paras. 4-5 of his reasons:

As the Appeal Division has expressed, it is a basic principle of Canadian law that proceedings of courts and administrative Tribunals be open to the public, with the ability to be publicized and reported upon. The open court principle protects democracy by ensuring that the exercise of decision-making power can be scrutinized.

These principles are particularly apt to a good character hearing. The public is entitled to know why an applicant for a licence who has engaged in serious misconduct in the past is entitled to be licensed in the future. Hearing panels engaged in examination of good character frequently have to examine the details of personal misconduct which may involve criminality, moral turpitude, ethical lapses, and errors of judgment. Good character hearings by definition involve a public examination of events which may cause embarrassment, distress or humiliation for the applicant and others involved. In the typical good character hearing the pressing need for Tribunal processes to be open, transparent and accountable far outweigh any concern regarding its impingement on personal privacy for the applicant, witnesses or third parties.

207Accordingly, I would reject AA’s argument that the risk of harm to him personally constitutes a matter of public interest.

208I would also reject AA’s argument that concern for his children’s well-being, generally, could constitute a serious risk to an important public interest. In my view, while children are inherently vulnerable and this court is entitled to identify “objectively discernible harm on the basis of logical inferences”, this “inferential reasoning is not a license to engage in impermissible speculation”: A.B. v. Bragg Communications Inc., 2012 SCC 46, [2012] 2 S.C.R. 567, at para. 17; Sherman Estate, at para. 97. The lack of any contemporaneous evidence in this case renders a finding that revealing AA’s name would cause harm to his children, in general, mere speculation.

209AA couched his request for anonymization and non-publication before the Tribunal in terms of the harm his children would suffer if they learned of his misconduct “prematurely”. Much of the contemporaneous evidence of this potential harm came from statements in AA and his former spouse, BB’s, affidavits. The panelist of the Hearing Division accepted that “protecting vulnerable children from suffering trauma caused by their parents' legal proceedings” met the first prong of the Sherman Estate test: at para. 68.

210These affidavits do not specify the harm that the children would suffer if they learned of AA’s conduct, apart from vague references to the children’s mental health. In H. (M.E.) v. Williams, 2012 ONCA 35, 108 O.R. (3d) 321, this court held that convincing evidence was needed for a non-publication order protecting the identity of a serial murderer’s spouse, and that the affidavit of a treating psychiatrist alone was insufficient for that purpose. The same absence of specific evidentiary foundation exists here. The updated affidavits filed before this court shed no contemporaneous light on the record with respect to harm.

211The Law Society points out that courts have been hesitant to impose publication bans in cases where mere stigma may attach to the relatives of serious offenders: see e.g., R. v. Hosannah, 2015 ONSC 380, at para. 25; R. v. Jha, 2015 ONSC 1064, at paras. 14-16. Indeed, if violent misconduct does not form part of the perpetrator’s biological core, as R.R. makes clear, it is hard to see how it would form part of their relatives’.

212Moreover, the decision of AA and BB to tell or not tell their children about AA’s assaults, including the assault on their daughter, has little bearing on the Sherman Estate analysis. To put it bluntly, it is not up to individuals coming before the courts to dictate whether the open court principle applies based on their parenting preferences.

213In my view, the only public interest put at risk if anonymization and non-publication are not granted relates to the identity of AA’s daughter, as well as the other victims of his assaults. Recent jurisprudence shows that the existence and details of an assault that a sexual abuse survivor has suffered is information so sensitive that its publicity could be an affront to dignity that the public would not tolerate, even in service of open proceedings.

214For example, in Fedeli v. Brown, 2020 ONSC 994, Faieta J. held that disclosing the identity of a sexual assault complainant posed a serious risk to the public interest in maintaining the privacy and dignity of survivors of sexual violence:

The privacy interests of a person who makes an allegation of sexual assault or sexual harassment in a civil proceeding is high, particularly when she has not initiated the civil proceeding. A complainant may be subject to unnecessary trauma and embarrassment, both for herself and her family, if she is identified. Without protection of her privacy interests, a person who has been sexually assaulted or sexually harassed may be unwilling to come forward. Further, the failure to afford such protection to a person alleging sexual assault or sexual harassment may deter other persons from coming forward to report sexual misconduct. Such interests are recognized and protected in a criminal proceeding as s. 486.4 of the Criminal Code, R.S.C. 1985, c. C-46, provides that an order banning publication of any information that could identify a victim of sexual assault is mandatory if sought by the Crown or victim. In my view, the policy reflected by s. 486.4 of the Criminal Code is equally applicable in these civil proceedings.

215The Supreme Court, in Sherman Estate, at para. 77, specifically refers to Fedeli to conclude that the “range of sensitive personal information that gives rise to a serious risk” to an important public interest includes whether a person has been subject to “sexual assault or harassment”.

216The Supreme Court returned to this question in R. v. T.W.W., 2024 SCC 19, 437 C.C.C. (3d) 1, at para. 74, concluding:

The sexual nature of the evidence in this case touches on the complainant’s dignity and right of privacy, and that publication of this type of information gives rise to a serious risk of affront to the public interests of personal privacy and dignity (Sherman Estate, at para. 77; Fedeli v. Brown, 2020 ONSC 994, 60 C.P.C. (8th) 417, at para. 9). Privacy and personal dignity are public interests that have been recognized in our jurisprudence (see MacIntyre, at pp. 185-87; C.B.C. v. New Brunswick) including in cases involving sexual offences (see R. v. Jarvis, 2019 SCC 10, [2019] 1 S.C.R. 488, at para. 82). Protecting the complainant’s privacy and personal dignity, as far as practicable, promotes the objectives of the Criminal Code’s statutory protections for complainants in encouraging reporting of offences, participation in the process, and overall confidence in the administration of justice (R. v. Kirkpatrick, 2022 SCC 33, at para. 30). [Emphasis added.]

217Favreau J. (as she then was) adopted a similar line of reasoning on a motion for a publication ban and sealing order in Ricard v. The University of Windsor, 2021 ONSC 5877 (Div. Ct.), at para 7:

With respect to the first part of the test, in Sherman Estate, the Supreme Court recognized that preservation of an individual’s dignity is a matter of public interest. At para. 75, the Court held that a person’s dignity can be at risk if sensitive personal information relevant to core aspects of that person’s life are made public through court proceedings. At para 77, the Court specifically identified “subjection to sexual assault or harassment” as the type of personal sensitive information that, if exposed, could pose a serious risk to a person’s dignity. Keeping the identity of complainants confidential in the context of cases involving allegations of sexual assault is also consistent with sealing orders and publication bans made in civil cases that predate the Sherman Estate decision.

218While this reasoning strikes me as applicable to this motion, it is worth reiterating that AA’s daughter is not a complainant, and there has been no criminal or civil proceeding in relation to AA’s admitted sexual assaults. Nevertheless, AA’s daughter is, by his own admission, a victim of his sexual abuse. In these unique circumstances, the absence of criminal or civil proceedings in which AA’s daughter is a complainant or a party does not diminish the public interest in maintaining confidentiality over that information. In my view, the Supreme Court in Sherman Estate made it clear that “subjection to sexual abuse or harassment” is a piece of sensitive personal information that, if exposed, could give rise to a serious risk to an important public interest: at para. 77. I do not take this statement to be restricted to complainants in criminal proceedings or parties to civil litigation.

219Indeed, the issue of the public interest risk in relation to the sexual assault of children has also been considered in the family context. In S.E.L. v. O.V.P., 2022 ONSC 1390, at paras. 29-30 and 61-69, the court recognized a public interest risk in relation to children who had been victims of sexual abuse by a third party (i.e. where the children were not complainants). The record in the family proceedings disclosed details about the abuse, CAS records, and interviews with the children: at paras. 18-20. The trial judge anonymized the identities of the parents and the children, observing that Sherman Estate recognized sexual abuse as an example of personal information that would give rise to a serious risk to an important public interest if exposed: at paras. 61-80.

220On the basis of the principles recognized in Sherman Estate and these lines of cases, I would accept there is a public interest risk in disclosing the identity of AA or his daughter because doing so could have the effect of disclosing that AA sexually assaulted her.

ii. Is the order necessary to prevent the serious risk?

221At this stage, the court’s focus is on minimally impairing the open court principle by considering reasonably alternative measures that restrict openness less significantly without sacrificing the prevention of risk: XYZ School, at para. 46, citing Sherman Estate, at para. 105.

222AA takes the position that his name is uncommon, and that his family is part of a close-knit religious community. In that context, he says that if he were identified, his children’s identities would be readily discoverable.

223I am not able to determine on the record before the court if it is accurate to describe AA’s name as so distinct that identifying AA will invariably identify his children.

224For the purposes of the second stage of the Sherman Estate analysis, however, I would accept, based on their familial relationship, that identifying AA could lead to identifying his daughter, and therefore the order is necessary to prevent the public interest risk in identifying a victim of a child sexual assault. In other words, it is not possible to guard against this risk without anonymizing AA’s name and imposing the publication ban sought, and so the order sought is minimally impairing.

iii. Do the benefits of the order outweigh its negative effects?

225At this stage, I am satisfied that the benefits of the anonymization and publication ban order AA seeks outweigh the negative effects on the open court principle. As part of this exercise, I must consider whether the information the order seeks to protect from public access is central or peripheral to the judicial process, acknowledging that the more central the information is, the more serious the deleterious effects are in protecting it: Sherman Estate, at para. 106. AA and his daughter’s identities are peripheral issues in this litigation, and I am satisfied that the orders sought sensitively balance protecting the important privacy interests at play in this case with the need to maintain court openness.

226However, this calculus likely changes if AA is eventually licensed. In light of the conclusion set out above that a new good character assessment by the Hearing Division is needed, AA’s status as a licensed lawyer remains to be decided.

227There is an argument that irrespective of whether AA is licensed to be a lawyer, there is a negative effect in shielding his identity in these proceedings. As the intervener Toronto Star Newspapers Limited argues:

While AA has not been convicted of a criminal offence, he has admitted to sexually assaulting several children, and his efforts to withhold or conceal those actions form the basis of the good character proceedings that underlie this motion. His identity and actions are undoubtedly central to those proceedings. In this context, there is significant public interest in that information being known, and the deleterious effects of the order sought outweigh the benefits it might provide.

228That negative effect on the public interest is no doubt greater if AA is licensed to be a lawyer. In that scenario, the intervener Law360 Canada submits that:

Keeping A.A.’s identity secret would prevent the public, including future clients, colleagues, law firms, employers and volunteer organizations, from knowing about A.A.’s sexual abuse of children, including his own—a fact that people might consider relevant to their interactions with him. Tribunal-mandated secrecy could prevent members of the public from exercising their choice, and ability to make informed decisions about, interactions with A.A., including in his role as a lawyer—contrary to the public interest.

Keeping A.A.’s identity secret would make it impossible for members of the bar, bench and governments to decide whether he is worthy of trust as a lawyer, especially in matters involving children, and whether they should refer vulnerable clients to him, work with him or indeed appoint him to the judiciary.

229I agree with the interveners and accept that the harm to the open court principle resulting from an anonymization and non-publication order is present irrespective of whether AA is licensed, but is heightened if AA is licensed as a lawyer. In that scenario, the harm includes the importance that members of the public know the identity of AA as a lawyer licensed notwithstanding his prior misconduct, and that members of the public have the ability to make an informed choice regarding selecting AA as their lawyer.

230I note that during oral submissions, AA argued this aspect of the public interest in his identity might be satisfied by the fact the condition imposed by the Tribunal was a distinguishing feature of his licensure and would be published together with his actual name on the Law Society’s website registry of licensed lawyers. Therefore, his past misconduct would be clear to potential clients. However, in light of my conclusion that this condition would be inconsistent with a finding of good character, if he were licensed after a future, positive good character assessment, presumably this would occur without the same identifying condition.

231Therefore, with no other way for the public and potential clients to identify AA and connect him to and review the proceedings leading to his licensure, the public interest in AA’s identity being known, in my view, would be sufficiently significant so as to potentially outweigh the risk that his daughter’s identity may be discoverable.

232To summarize, at this time, while AA remains merely an applicant seeking to be licensed as a lawyer, the benefits of the anonymization and non-publication order in protecting the identity of AA’s daughter, as a victim of his sexual assault, outweigh its negative effects. I would therefore impose a publication ban and anonymization order over the names of AA, his ex-spouse and their children, along with any other information that might identify them.

233As set out above, this balancing exercise may well lead to a different result if AA were licensed as a lawyer. Even if AA could still satisfy the first and second steps of the Sherman Estate test, absent a further record not currently before the court, AA may well fail at the third step. In other words, at the point AA is licensed as a lawyer, the benefits of an anonymization and non-publication order may be outweighed by its negative effects, including the harm to the open court principle and to public confidence in the administration of justice, at least based on the information that is currently before this court.

234To that end, while I would impose the requested anonymization order and publication ban at this stage, it would be open to any party to return to this court to request that they be set aside or varied following the resolution of the good character proceedings, consistent with the court’s ongoing supervisory jurisdiction over its own record: CBC v. Manitoba, at paras. 37-42. A decision of this court to set aside or vary the order being issued as a result of this motion would not affect the existing orders at the Divisional Court or Tribunal, which would remain in effect.

235While I set out above how the balancing exercise might change if AA were licensed as a lawyer, I emphasize that this issue is not before the court at the moment. The balancing exercise with respect to AA as a mere licensing applicant leads to the conclusion, set out above, that the benefits of the anonymization and non-publication order in protecting the identity of AA’s daughter outweigh the negative effects of the order on the open court principle.

c. Conclusion

236In light of this analysis, AA has established that an anonymization order with respect to this court’s record of the appeal, and a publication ban with respect to information which could identify AA, his ex-spouse or his children, are warranted.

237This analysis does not bind the Tribunal, which must make its own determination with respect to its record.

238It is neither necessary nor advisable to predetermine the result of the Tribunal’s proceedings, as there may be other arguments or facts which arise at that time which could affect its Sherman Estate analysis.

4. The appeal: the Divisional Court should have conducted a fresh Sherman Estate analysis, but its decision should stand

239Turning now to the Law Society’s appeal of the Divisional Court’s decision on confidentiality, it follows from the analysis above that I agree with the Law Society that the Divisional Court should have conducted a fresh Sherman Estate analysis when deciding to make its anonymization and non-publication order permanent. Its failure to do so constituted a reversible error on either standard of review discussed above. Indeed, failing to consider and apply the legal test applicable to a discretionary decision is an error of law or an error in principle that justifies a reviewing court’s intervention: Canada (Transportation Safety Board) v. Carroll-Byrne, 2022 SCC 48, 475 D.L.R. (4th) 274, at para. 41; Airia Brands Inc. v. Air Canada, 2017 ONCA 792, 417 D.L.R. (4th) 467, at para. 39, leave to appeal refused, [2017] S.C.C.A. No. 476.

240To the extent that the Divisional Court, in making a permanent anonymization and non-publication decision, relied on the reasoning in previous orders, this reliance was misplaced. Each of the prior orders was made before any final determination on AA’s good character. The motion before the panelist of the Hearing Division resulted in an order that was to remain effective until varied or cancelled on a separate motion or at the hearing of the good character proceeding itself. Similarly, the order of Davies J. was expressly temporary in nature.

241However, as is discussed above, the Sherman Estate balancing exercise is fundamentally altered in the context of a permanent order following a finding of good character. As Madsen J.A. noted in Kirby v. Woods, 2025 ONCA 437, at para. 24, the balancing exercise is “fact-specific”. The factual matrix fundamentally shifts, and with it the nature of the balancing exercise, where the order sought will shield the public from knowing that the lawyer they are retaining is, in fact, AA.

242Given the permanent nature of the decision made by the Divisional Court and its consequences for the parties (and the public), a fresh Sherman Estate analysis was needed, just as, in my view, a fresh Sherman Estate analysis was required by this court.

243While I have concluded that the Divisional Court erred in not conducting a fresh Sherman Estate analysis before making its anonymization order permanent, I would not set aside or vary the Divisional Court’s anonymization and non-publication decision with respect to its record, as to do so would defeat the conclusion set out above that an anonymization and non-publication order is presently warranted.

244Therefore, I would decline to set aside the Divisional Court’s decision with respect to anonymization and non-publication. That decision should remain in effect with respect to its record, until and unless it is varied by the Divisional Court.

DISPOSITION

245For the reasons set out above, I would allow the Law Society’s appeal of the Divisional Court’s decision upholding the Tribunal’s finding that AA is of good character. I would remit the matter back to the Hearing Division for a fresh good character assessment to be decided in accordance with these reasons.

246With respect to the motion, I would grant AA’s motion for a publication ban with respect to information that could identify him, his ex-spouse or his children, and anonymization of this court’s record with respect to the appeal, with the proviso that, in my view, if AA is granted a licence to practice law, that may well warrant revisiting this order for the reasons set out above.

247The Law Society is entitled to costs on the merits appeal from AA, in the agreed-upon, all-inclusive amount of $7,500.

248AA is entitled to costs on the motion from the Law Society, in the agreed-upon, all-inclusive amount of $2,500.

249There are no costs orders in relation to any of the interveners on the merits or the motion.

Footnotes

1. 1This appeal is subject to an order prohibiting publication of any information that could identify AA, AA’s former spouse, or AA’s children. This appeal is also subject to an order anonymizing in the court’s record any information that could identify AA, AA’s former spouse, or AA’s children.
2. 2Mr. Grace did not appear but made written submissions on behalf of the intervener.
3. 3Since 2007, the Law Society has had regulatory authority over both lawyers and paralegals and thus its regulatory mandate extends to both legal professions. When referring to case law in Ontario prior to 2007, or to settings outside of Ontario, I refer simply to “the legal profession.”
4. 4See Law Society of Upper Canada, “Submission on the Federation of Law Societies of Canada’s National Suitability to Practise Standard Consultation Report, November 2013, Professional Regulation Committee Report to Convocation” (2013), Tab 4.1.
5. 5November 20, 1989, 1577 U.N.T.S. 3, Can. T.S. 1992 No. 3 (entered into force 2 September 1990, accession by Canada 13 December 1991).
6. 6The Law Society did raise the Appeal Division's failure to set aside the original confidentiality order in its notice of application for judicial review before the Divisional Court. However, the Divisional Court held at para. 29 that “[i]t was reasonable for the Appeal Division to continue the order”, and the original confidentiality order imposed by the panelist of the Hearing Division was never appealed directly. It therefore remains intact.